From 6c3b105dd113490a1ccd31a6f8046ac723f5cd1d Mon Sep 17 00:00:00 2001
From: github-actions <github-actions@github.com>
Date: Thu, 3 Dec 2020 16:51:49 +0000
Subject: [PATCH] Deployed fca0528 to 0.1 with MkDocs 1.1.2 and mike 0.5.5

---
 0.1/file-structure/index.html    |  12 +++---
 0.1/img/files.png                | Bin 539484 -> 321183 bytes
 0.1/search/search_index.json     |   2 +-
 0.1/sitemap.xml                  |  72 +++++++++++++++----------------
 0.1/sitemap.xml.gz               | Bin 213 -> 214 bytes
 latest/file-structure/index.html |  12 +++---
 latest/img/files.png             | Bin 539484 -> 321183 bytes
 latest/search/search_index.json  |   2 +-
 latest/sitemap.xml               |  72 +++++++++++++++----------------
 latest/sitemap.xml.gz            | Bin 213 -> 214 bytes
 10 files changed, 88 insertions(+), 84 deletions(-)

diff --git a/0.1/file-structure/index.html b/0.1/file-structure/index.html
index 66b2f900..fb682026 100644
--- a/0.1/file-structure/index.html
+++ b/0.1/file-structure/index.html
@@ -895,13 +895,15 @@
                 <h1 id="file-structure">File Structure<a class="headerlink" href="#file-structure" title="Permanent link">&para;</a></h1>
 <div class="admonition tip">
 <p class="admonition-title">Tip</p>
-<p>Read this page if you want to learn more about how RAPIDS is structured. If you want to start using it go to <a href="../setup/installation/">Installation</a>, then to <a href="../setup/configuration/">Configuration</a>, and then to <a href="../setup/execution/">Execution</a></p>
+<ul>
+<li>Read this page if you want to learn more about how RAPIDS is structured. If you want to start using it go to <a href="../setup/installation/">Installation</a>, then to <a href="../setup/configuration/">Configuration</a>, and then to <a href="../setup/execution/">Execution</a></li>
+<li>All paths mentioned in this page are relative to RAPIDS&rsquo; root folder.</li>
+</ul>
 </div>
-<p>All paths mentioned in this page are relative to RAPIDS&rsquo; root folder.</p>
-<p>If you want to extract the behavioral features that RAPIDS offers, you will only have to create or modify the <a href="../setup/configuration/#database-credentials"><code>.env</code> file</a>, <a href="../setup/configuration/#participant-files">participants files</a>, <a href="../setup/configuration/#time-segments">time segment files</a>, and the <code>config.yaml</code> file. The <code>config.yaml</code> file is the heart of RAPIDS and includes parameters to manage participants, data sources, sensor data, visualizations and more.</p>
+<p>If you want to extract the behavioral features that RAPIDS offers, you will only have to create or modify the <a href="../setup/configuration/#database-credentials"><code>.env</code> file</a>, <a href="../setup/configuration/#participant-files">participants files</a>, <a href="../setup/configuration/#time-segments">time segment files</a>, and the <code>config.yaml</code> file as instructed in the <a href="../setup/configuration">Configuration page</a>. The <code>config.yaml</code> file is the heart of RAPIDS and includes parameters to manage participants, data sources, sensor data, visualizations and more.</p>
 <p>All data is saved in <code>data/</code>. The <code>data/external/</code> folder stores any data imported or created by the user, <code>data/raw/</code> stores sensor data as imported from your database, <code>data/interim/</code> has intermediate files necessary to compute behavioral features from raw data, and <code>data/processed/</code> has all the final files with the behavioral features in folders per participant and sensor.</p>
-<p>All the source code is saved in <code>src/</code>. The <code>src/data/</code> folder stores scripts to download, clean and pre-process sensor data, <code>src/features</code> has scripts to extract behavioral features organized in their respective subfolders , <code>src/models/</code> can host any script to create models or statistical analyses with the behavioral features you extract, and <code>src/visualization/</code> has scripts to create plots of the raw and processed data.</p>
-<p>There are other important files and folders but only relevant if you are interested in extending RAPIDS (e.g. virtual env files, docs, tests, Dockerfile, the Snakefile, etc.). In the figure below, we represent the interactions between users and files. After a user modifies <code>config.yaml</code> and <code>.env</code> the <code>Snakefile</code> file will decide what Snakemake rules have to be executed to produce the required output files (behavioral features) and what scripts are in charge of producing such files. In addition, users can add or modifiy files in the <code>data</code> folder (for example to configure the <a href="../setup/configuration/#participant-files">participants files</a> or the <a href="../setup/configuration/#time-segments">time segment files</a>).</p>
+<p>RAPIDS source code is saved in <code>src/</code>. The <code>src/data/</code> folder stores scripts to download, clean and pre-process sensor data, <code>src/features</code> has scripts to extract behavioral features organized in their respective sensor subfolders , <code>src/models/</code> can host any script to create models or statistical analyses with the behavioral features you extract, and <code>src/visualization/</code> has scripts to create plots of the raw and processed data. There are other files and folders but only relevant if you are interested in extending RAPIDS (e.g. virtual env files, docs, tests, Dockerfile, the Snakefile, etc.). </p>
+<p>In the figure below, we represent the interactions between users and files. After a user modifies the configuration files mentioned above, the <code>Snakefile</code> file will search for and execute the Snakemake rules that contain the Python or R scripts necessary to generate or update the required output files (behavioral features, plots, etc.).</p>
 <figure>
   <img src="../img/files.png" width="600" />
   <figcaption>Interaction diagram between the user, and important files in RAPIDS</figcaption>
diff --git a/0.1/img/files.png b/0.1/img/files.png
index 58ee5dec797ec1e0d3a9a7ffbcd2ea53c6b92d5d..1fef7eebd110c8f0a3c69025c16892469c9da3c8 100644
GIT binary patch
literal 321183
zcmeEvcR1Gl`?o|H8IcG@nHh;fwh}_2?2(zhSIFwBJBo}uQORr|dsA7NNk+<+Eqm|%
zoF9$5>*D+S-QVMQj_04Jj=Q7lzQ%i;?{mLj=lQv(C@)P!Kuv&!g+(ND;rtaWEW8CQ
zEP`&lJ@C$n6HJouKRsp=5{fbs67-7pHpXU_Mp#%E?u9(stE{9!eRI-n*dGf&<^p~J
zegi%|(;Sv;>NBzj;^*+WkFvIeyxUJpasGVTJ|o5R?>_AhN~b@V#L73en}?NkRPUYl
z2T?uzQ<)~tbsGZwi$a^Lle!bH7b7-ziR1UkG9F6f!FtkbX?Ff5|EUwAsC%o2gKnOn
z$9B>y9jIa)OG_iY`F?Iq&DD!ioP3ru5np$6W39-SO_#oJ7Z%g8@Q;tCdG=dNSFlH(
z4<o`FIr)rVJB;&5cRSwamltl~k;vPQ9PoP5-7h1>uF!T~a*5s&2P^X9fUMFk`gZyo
z#>89{(-OoIEm9vT@9o#Qowiru{Q$%LGTh8lr!tvcsPs>j>0o2s2{_5VtRSki*(;{e
z6IFdlWlH&K@U2CPD#k+{9h(jo8@bnv-^Up~roYYcaE}!IsokwbZo?Yy2oAE;kwh5p
z_iFHbaA&4?$n0B{J6Y_Lx_Khiq?*TcJT=FUc?Fax@UfVO;qnx{Hw!hEWHoqs!K3B#
z$B^2xkT_}fxpTXud&Cr|D1>SRo;_av%pn)JFEV}l^3cl}Jjo4gyPJy_wY?<ApJa$N
zviGt!852IorpR_MnW4PfdP>yD&bMO;Tes-YQ`0*IjO|+c<sR&hO{2Vq!#bA{exbb0
zvfs8giAZTopPHZ5G7QBvwfpG%K>TMYk+j`8#kha%4&qb~6}fz0A(T;#O_sv=RO@?x
z@{<Avk@J?Z5&nnmXP74@zx?e&!|lp-wBpsFtE!y#54#wgV_l>kX%mRc?ZToaF?N)4
z!;1d4yT0RHu$)7z<Ip{^eKV@j*n;)vJ4w8Y_o*Lk71+h!edii2&bhM(?(X%&BEdV$
z9CYiY9<Dkef%yIwf}Gm~X_Q}Xu6L83!PYFsx80+SrB+O2yH^;eKo2V)hoM`9>NY*z
z-plmVFYt%YRs=C!+nY{L@}7hMbu8M26<_vFSrDz}u4EtIXtH?hgj*vqQ89h>hla?$
z5PkJoi9R|+G<Qhq-L=QWqs7cqw2yH!d-To_=-|bb$aM>!q3hiDsI+nFY~CT!-MXmB
z;({qTmjk$pj0bkV49qx}t|%I>sEo~q%Osh6j^fz|#m0CkWts$XC&ETQYAKI%hM!KK
zCbznCU1IOE=bwb<$mZDQm>qF%h`l-FLF_^J)#u1ruTsU6ZB%mi*}ZSyD}7-QYCv$(
z=&|5O?vJh?4}Zk^CUVfvNbgd5ZHu_2Fdxxd+PC;`Y1;p?s=RKvn*RI@tvtEOJ%`FW
zy?J?-+Pkb-Cg>+lPcU0!E$&w+8SNRL+E~KLrz#@W^%lD+cBkl`R{$LYEBSr=l)n_Q
zqk`mG&c!oj?N1@!gJ*c2;@R;}5mb^F8B&-+$!YEl-V3~IBg6P$SJj;Z@??KE1$~h5
zk)f3Fc%g7jB?GtBC;UQaFn{pk-|bb4RW`>D_~MuQJrJ*owT%6uRC^^-QIze|!=#Wi
zvhFV%WkY0ZWcfpMF5@d3vORrsH%yxI`H`18am4K84=+9Rc=-B5;zRb3K=wt}z=V2b
zOO<mfJSu#zXcb?-;EDHD;=MetG?4f$u_-=MWjrp%)cZ6Ybzn}Us+^DlO|;=~tRu0A
zzs`BF^OMg4<6k8R#Fxi&DkUlly=tl!=e|fqE49avun({AL|^8c070#*(S?@_F0aR~
z?Df~6)}hAsKj+WvFGL;1`y|XWY&fhuj591SyI%94W{;*^_JV0dGj*h7)Bv9!f1+*E
z-x?%Y`c{PFc;f?YylraP(oT*}*p7^XLQb_K`$l;j3miW?$~h?v6Ayoy6P`PN<K>OG
zd7YDuk1dV&WQMy22s}$ori<rlPHmWMpgVPv_b#8hN&U5{mLSvjJQuk+c*joCayOg`
zGs!B6e>a5Ue)rgj<8-HfhuPY@vM-EH*KW)BRYkw@K3;jcF6-Ug--pUSezv@)k=<(L
zkR(to{M9{onTW8Z<kLCR%Vu_8$#NVs4rol|j79oC8RyXwetYA3AVsUMRfu7uUZY7(
z8gaUg$ybAQ^MFs`e+iU32oAbu+GSOxy~=hqO*E&u_O>>sX2Cz=&f&V0hKU@J^3E>L
zY>&isi@kDtQKd(G+yYKlx(3kD3egq`3JA$WP)B%&`-caII}0qP4JDmyVs4V8yziG=
zZT^V-M3i=fcIMT<yp0~?fzXNT#a6Z{!`kmU+Fq|u?3vJBWLwlBiIsj(8F-{1awSr`
z&b%&8$6LoeRWNnX(&0MwD*b9mu7B>A)yIo_>C~b^M8^wRyRf>>b)67XDKaf0C}P>z
zv+-(!XoF!_z^>}uZM(Tq?@;9^5!%DVfl76U@a>5;4>_KAOiO!6d;i+$vSa6|8~8{f
zDNoz9=S$D}5Bd$}{&h=HekiBDQa@H-uEdk5(%a{bnzxu=n!oz~Plsguv}s-GXFMJ{
ze+iBE`DFZQW+;Hl{G!1{kB>g;T`dn<I`ahV*!xn-EXw_Qwa9Jpx$&LwpA)VfwA4Kr
zH1t?K$i+EkxWtH%i?c!VVfs+N!HB^`?{k^QFIrygmeI|N=`f!#<zaDEo58E|$-TY!
zV6vQRfCRq-m+98v@rmP1$8W!yc-4Nz=dZ{To}RS#(#l6~L>zpa!#^%@!*r!^Zu)-9
z{n(4{##g@Cr6-)Z>V86w-$Fw*x9rLl`r)g+Z?3IA*l(7WCBLWY;0Mjj;S0|lRjipe
z<vSns{jKz5*3Rr>!}Dt8(0h-r)?~l``gm2iPHZx6U}zwJ;4u4~TGG`|xep35%qg<i
zTvaN~OB<IH_cq}qwHAfu@p+7AJMoRmHWg^my)ER_W%0V?x#_&Ci}wI4#Rb9}e0f$a
zt0F6IpBAsujPp-)EF}l&y1afL8=o{(&D<GVE!MK3xKX<F?P1MTt?-<&R^P8|oi}n@
zGF`mx$;#-+Z05MOhsiy8dFSP14p*n=tB*03lf=Em%K@D;K{HJeQbM^CYgdjY<tYtc
zZOxK%Yp%FTcYW5mtD`SF&MvjK^I1WZ*MW7HRJQ1Z*KIG}B<;}(R+>%Xi_Mv3X;PSX
z(sYt^(p|i_I6o2C;+4|*x^Uy>0*aH!_IAGhkY4XQ8XtF(Y(7n1RbK8eU*g-PGo@Uf
z@+->|e|Iw+me!D_@#5YvYUXZBFENf6$aQ@o+0xKX-PPtHJmez%RI5E*H`(i}^SRCQ
z12WmcTJrl92BN6NBAjnzU2PC`XP&?4RoNThW71dKDN5;?GU`6H>_@rA6+V8=IkIrb
zbGnSFS+0^R^F*$;-J1sY#_5rCi-6_;Cjlp3UFX8i8>ItMT8kYMO8g==k*^2bMoX+}
z$Jw2~bqyENcr5BD7${7NY!tj(J+!DR_TKg6f^7}M$LA%L9|Ef*O~qbq)U4&k>cn*h
zdlY)8MM;TzwA77yEx#RG<l4Pl$6EIxa#SRhE{jH(yI;()ulyF}ZZ0OSN^ioAn3=$<
zsULOLXMCFJ)TDc*_UOpz1{Sr>sm_iUIulO{N|j1Q<=453EyPSc|5mvkr5+_HR_W;}
znlnc@A-AkHn`+f{t4M^pj+X9`?aPiTRjfPv&M@KL<X;WiICBK6jR@;5VeHDN3@Ww|
zpBaCRC355)OF90{up?N{LU)te6WxrUxFF}Z6nCr(S6})3<Q_dYxALV&@2`xF$GZP5
z@^@4J^wN^nDJ<J^EO*W$baaV_=r%QPGp>@}T<CaJ=y~oT3AUn<C3)`wd);pAjT1^D
zn^ER#tNWOFY_YteHt{n(Xaera>t%t@>uaPgV|?in78|_A!@}7`jfD%ZcEMlKT{OR5
zOYLIC+P(EYHWrqT85Yiu@5sZy$e$qii~Q#Mzq=oJW9^0iBZI##(bzw~jkgfJ`{!$d
zZukuAtdfL`4E(EPXm4a>?O<x-C@#TY0B_*iUeIvB!aB%={M{vUg>eeTZ!uF=cT~T0
zQP9xF>V*C^8v~;gE>^b4I9S3ig7DJH$Wfo(#mds!LC{6y*w!}$;WhF$*D?C7uQ*zW
z98<rfNH1YyZ$!^`;^c{w$3zL}>FI^-uNezoIWP6&=kR|b$4ng^Z3VfwoSmIdIG;LU
zV{gL6Eg&Gkb&`jRhldls!Rg>??Wpg<Y3;!9eUP8yoHueXv^TSLG_$d$N5<7RuyJw}
zId%-0=+~d`>ojsP`)4L=hacO54RRr$aB-hF$@Ob&_^B}RuArisi;<<qc{3}Z8O$Nd
z!+++q@YWBuee}<izx`DGx1aI|p#Ai>k1&6#;$UPiVPgf8I*R^tU_a*l{lgzW6y`$q
z{u?d66ME}iptLA~FxRhB6D9adls^{^lGg0Jf-?LIY=->7MUI#C`@irS`&j>8pGFBR
zEO9KE^JkS^c8z^?&QocO;Gd#XQPwNlgKfaI|A0jq;~om$?(@>u$Rx_jblx3|xEU`q
zw&ywP2qUxlUEGM4XZvYnIT)I{>kVw8&!*icl@xdB@`zlV>Z~rHJYzdiI5@W6TDN;J
zJiA%PVqJ7}W7(t0I;VeAr{|dXO{`tmIC!M=|I>@;XE<T_3(8+S)%X4HU%rWjx^+`8
zPz>jP9Tu6If|b5S3H8)K=6_qm);e+A)`<S+rT*AXy!ftnkpvwF<o?wL>0A12{tX>&
zisKUCEhu-#s+0X2I_!FP#B2Az+83Lmcv>7bfBA#|<siN<5^rIN=6^h|pL<b6Vc!y=
z$U7c-{^vXY(<zImRd)RgW`Of0I6yk~s`yem<G(uMTOu2K{?)#eqi^ad;9C1K?f*Bs
zBOROO|DU$=Ys=BtxHxWXwn`lL{>|=yb&LMh;r?OWEeii(U4&cyAG5A<V)0%&chb~?
z1x@$0=}K<riH?#YuEM2bO_~<Trp+l=d%A9nCR1wJ6gW9fbiCHA&a&P?RA1rYExahP
zBt!owNCefIX;S$YK6&qSC0!t;j!O*bkW7g%yBpKlD8}y0cI&t0qN`)%ZsXZ%mTB+y
zu&?HM?x5s&xaeDgmpN7G(S?aMs`%rRQ+-WZHva84MIt(`10m`e*JN7v*fb?+%#F34
z+CxlB<#YJDy5n%eOG`tVyW3?}oEK5TeJL8J(1*UD(Y>)c8F=}*RQb|$pW$?Gxxc8}
z?6;cd3Q?lm*WL%6nJV5*;~`4Eg8_*@59UI)gTeyCTu{&l;kazf#fg=dQ|fA;iHLg@
z`Y^oLVWHKevS@R)NSbWPZU<fU-VgMc&UMWQbh1R_pze{gRLHPI9PnCg?~-*%uzNd^
z^;zETt5dUNs!jp5^RegT+cjT2DPqc6@v1RC3Zb8>Vu>oN^V*oD61m~Xp^@=|EL7Qo
zs3wfp&oDrM`9#57gSzuqCyTGMw|5|76Aq4BM$Yq*qv$#`k#%bBvqEJtYPVs<{X>ao
z&fN%f8gJvZE?Qq$s&y_5?tEd4K_g_7K^d^)58bg0=!1nY2ZeuhXf=9N{(3Ny#_u3!
zLe>CpUSuezRs}Alw&EFmR)MNY$zO{4*9&?Juv;4Km*>%s3Ppx<kVBnRY`IFIH$Z5D
zwdd7`ksCuXJ{6C-4WBuUHicxLOvKm@vbeM`7%dTh$PS&)6_c@nIs7)(S1SzzMO2fS
zIIb=hu1?S_b~q?xS+!p`d&-4=ep|~+xC)lK&AEOBA8l>LY<TV)%Y&cd6rx<F%PG}*
zhCjGpYrg92KiyX;y}B^Y?toI=VWZIq-;up@WI?A}<B6$8HC@Rf&-LDWERpdGS4NV&
zCJPn=?D{IosV$N&4@T}m-^<p|8=1kCvJx^bKwq{VFU~<O9a^!q9+JM4w-cKFm)TZ7
z?b^MkEW)|-P|Z{R)Ui97htVYo;WQKlXon=M_g3gL1u>x<Y6VzMycv4N5u$9vS?WU)
zNXV6+*6Bu#PLVB^Q${2qd+ebT85;AV8V|Vi|1JH@rmHahTGfL_+E+7li2MuB)k~yr
zZmg<p7$gVnz_YvIAc=}inbE1GLAEqiN@7@f*!g~-=nAWx5oc~F5r4-?=1XA*Xsoh6
z*{{Bq`NdklUPNC(yaFeaBtP_t9$xMak|4h4Ht{;{K7(qdqRdy2AvGtCCbk{+$H73~
zlDY406gtI$efui>{dN<dp|>zfb`&8IUDBHyZMJ#2I(p&P{J&mAUja#EGh#`?kkZq)
zir1EBpGyZGnf~_FhOcC(k-Pt(mgRn>*o*i0tlE?-ZzVrKUpm5(#hc{d7{#Ge2#1Hm
zwoyf`+$=64w1+?~CF^en>55lzk#FoH=68^0#iNMZn6W?f?5{;`z4*YHruXhvi~DSi
z;kS=Z29CTLsTgf|DVIOrr1j~I(`bU$TiosXt6xn#jityf)gFzH-sct1H|x?>(s{X8
zBx*0ucrL!}Dz~dVoI67@w9{eTgTo4Q*?my~eMQuAdhJ%Z{%X1&+3DV?Z|rhgjh`;9
zc5Ui7jppc~O}NF<YhXC~ohMb~(MCCR&;>--aPg~io$Z}O)i#GZ5z4N$Z~SWG`U*Q}
z_X`@ZAtT$*=pUg@<Ln_hLffq4W<{-}(n!k^<!-O-P{XT|ZS2Q{aT<7}?^8i8JQsBa
z&=@D^GD_0#pk$nUL~mc9Sgq-uM4cwYjAt6x_#WicOj@CjWZX6seb56KOSV21L-U#8
zGp0TrX-rhp{H*d-n4Cp^E-WLVotF?pcyRFgm0&+nt10Ga^m?CG%=Ac`%p}U=x;MLa
zLBol{rCusN%a#$MUNf{Wei!WHh%0f^k^udqpx9#1)p5!|kp=ZICDlgl)rq$@!Yf11
z{{vCtU?ZZ7R~ylZqK~>h-d3LPI?JQoGZNC2W7Ymx$a&(wTSsW|{X?e=`uzFjE#4$V
z46ZMB1*Tkmt2(~I8NN+~=;?s_ms?C!&~%o#JWfR*9bHvotadf!1s#<p(DVb&_K&vx
z{H8cLA>P92={=7l(LZ~r)cs~O#V=Sv^mVv|v3-(7Y5||T;P8u!O|h)#WP>wCjR8Tg
znQ*C`Lt|B(Udo~+$#AcYWkywc+fLWP@>I|HM2b24r|5Gbnn3(BaQo@xqE>XhC9Z*6
zd#3Mx-pmK~YC+?Jdxh5LKKJ>ur@S{XGvBTY-Uq>%c=IV8M>`Yo?iRy9f5f8D!!kZ{
zSC&nBZLa>e%RoFm!^!~8(Z?FJ9CmNUbJ{c1!@{;liad<qNygMD6QCQZGre~idO<Vp
zNykQ&4Oh_EXrJZEMc?ZdQBEhr7Cx^$jX)>SLlwREK}YRsc$*(OedHkKbD8R<;x$)3
zD5pe@J`X}lZ&1-ZSx;m!WDJoGZTl)Fpu$PZmbAR3UO%<^Y7C3HC9#*->03T$Z5U%1
zdLxRZ`BbmNVw=l#TZbVRUMo17#1R_HyqzQV;R8pQo(jR>hzamns7~r(-KD@|lwV(6
z)ctR%(3WNPslaV9Bj8E-yIA=MYEY<G|0^o!dd#aN6)dz(gGMhk6^{lT1}>29`O&6i
znxU8{=w#pGOe0`!<qA=CTy#paJ}q8ox1j?Acf8$2gC-tKAGhmFe^SWHBRh}-xci!|
zQV4o#WBK5jfrm;4U*az>OR3AxOL*_QmzY|3LqMcsJ8#n~9s)@ZNheXov~$e1;B855
zZmd`H=``(mGZ^XC+kKm)R#OAhTL!6kk0dSj@T;IFu!3w*3S#T0`C9aDFRhHG&JtZ#
znCH~Yx(kt&v=8y&k=KbkBqPL);E0u*Jk^OX=;9}ixPmc0N7$;>HK<>J+r(|%k;f78
z;@}Q+0qt|m#UVrrO>rz!<bizkrj-&`bBXT`VL7pTFR^8Jit)DJ@4^mY0yk1)Ri77K
z<fA#Tbv>7j9LHL?ztu))g-&Ojyzut=xB<9r>EFUEjALkd(LCa<WA7IS+p_m*(s{Ib
z-p3Vpzq{SnW|3NBGCi*x;c}mvHz87J2hqn<js@x(P57x`Slh;fE^tH!)ttHa?+{X^
zET}h%iYKZj_xkg9-nIVvM#b%T6NamRL@5m5ZPQxiTtU+k^j_G~bRLc!wxvYN?Cwu|
z0|iIoA)uP$N%lmu0N?pI-V(<6&`Fnw5$Yr|O@S_|xZZ9>NeDqovR%S*PgAm1WtQ1z
z13%83_-{N(7zzleMehR)DA3`^gyuGh7vj{4EOz8gl@QLVIQpB{2@Hnkb>AX1&svg2
zJJhWMiWUUjlwcT@gy9S(WfVWs^*r?Tt#e>*m#14uWqi|K{&vgE$AV7gNw2?!HEP~+
zLYwycfvLlRRX$C1L_e^=VhIwF1AA$0#UadO<d@{PXyQ2HKJ{*>J=ezCT$L4_UlEO&
zgM^paQ3)6tlZYVP_1lx_r(J8Y?^ZS5{{2PBpzQ9>0!|CU&~KU<g1+bFgX(#<`tIN^
z`a>{neB_y0s?KzOu!0nvK96o9Hrml{Q7Z#fZ}8Uwl|$%@jAxkYbek+c;x@LQ&VA}^
zgqw!hP{(Tr;nlIsPsv)jNe^ETqtO8Ib5P-k1L@coatZAmmMI<!yW02kN7jk1aya0P
zCS}U^yoemn?Gkm!KE6GH0SpHYM!N$AI{8s9Bd_8NHrE#emuCk}q@G`T$~XO<A+Rjc
zZQ{w3umc!sA7N)3U`6-Iq7k$`(^q1XGhe*xwVLbYS0}caKSI%#YeNUlt?n3P|5bGT
zu(w-SyaV7~`;v*G4RlkS7#9hlJXRO{H8PD=tmY#GoqUlH6k>1nY+m&<=mgy&&<Y&;
zh|(u?<%46-I0Xrx@}oYqWU2WxvW+TVGfnF9z#BB2t>I67ldW}fhoLCo{F5%c_dq{|
zpxr1SPbEY$2a8s}PA0^*fvYSk6+nB)sx9kP7_ZiLUc^En5ti(y1`KEIns#@s(B%Dh
zxA#_#wPv#CG!XH%(5HZCEcK9hsi#|HokX8{%a9L{l8znsdB%-SL~sy9!H=R6b$98R
zi!Aq}RJ&*dp><&Y!$`jqZ&(E#hZMO-Ge~zB&mXkyXZ4H<j2J=X#8iVE*H+pDLP?7m
z*_HWGN=j|pvtt?6CoUBpBt)o<LA0Py(3P4E-!Mc@g>7o4nu}7t>h)r$8$%7b`vdYD
z#G*^xCBtWhCi5LaMl|>^20}Ppg_3kk!%0X1T?L}F@jpK1X0U!e5K48*_~OMV(SlFr
zUDj{Mdm#c3JbQbOB;+1xtd8tPXYH-=;~jxrWxI}!VrWhZ)~OHdsb$XtMX3YDHe7_~
zzbJ9rbmpt*7`5kE8P<s`-uGB&WdX;aXs9)FkS%cF><(nqAR`^CYOqeYgHDvwsN!dm
ze!bwGy5+BoB;N5tGNzD!WI;?A-kLl3Dd~_hl>|d}41YSQXk#TEg4s&Gc1xwDTrikY
z)uEhE^#)>h2)h(p5$B;mOjUZ@LF{leO0f5E>--FK2}57(PxHy91u3SC#mW(M9^3iL
z_H3ZtKKrZfaE9AXRon=C1LcJc^&QeZ+rKOiI>G+U^oMN$<Ii>ew!Hsj713RYAB1}U
z3iwo>U#E4)PcaxfgcH1x)>T8NfAeJl3gCj_3U-WWC@G;B!pGjP&RzabPMMro=QpVh
zUmDA->;0@C#_3)gB_@WX6$vR%aI9IhW*B9eH?nEwT3=lPUVf`?65$lOG1`)Dh$Pad
zQso|<NT05tk)FAakvBr9A|44eo5^1_$8htj_g}ar6wIP<V(lrPm1aKt{*Kq?Ql(hv
ztg@>$Q%vL~zE;MZ4*TjB^Tgz5;Ps}L5(^y=4OR@=X^tbH%x2Q6%`teAo%F&vVXpkC
zk_IoXg1Ia1GgU0?E^O=VNIJNNze~7qxyGhq;_al)QtoW65T*PsSub8=j4==z*XIFW
z`;S_ndjRx`e`9}+QwW5#22Z5t%5X`MbEj+EvF6u2ewvWwfCTEoT?hdZr?{3Ucy-tr
z;z+hL)Juf>R){s|&_80mEpC6Tjv-K(tgkXK5Hdo-qH9y7Joh!9NMzY`i7vpgA#Mh(
zZSk&8FXbM6(2mD&2DgS&N8Qw`Be>at!L5yKC&4b{f#xIJ)B$lv?i#W~Gc-r#bsC-O
z&+etz7>5jeB5$bA4wCueILKCP&g<V89XVtMLPVELSBh<LPvp<mx~$JP)41l0(0D?A
zAwYsyfYEdAIUSPFPe{dL$5;<g_<able@c#4IvPg@Np9V`0G|<}BH_;Es2>xb?4_sH
zZ_H>R#m#NoT@2dK<oQQ^#T~R?B9ev_yb+0EL@u$O)GzrfZ=ubCFttrH%k(mN%8Qja
z`CB5Dv<??(uea!bijnXc;ktl}wvsLF^#sgZRT&@KEi|n&b8KpPFsc&L?Y&4UWB7C2
z;LEWuBbJ|IFYTi`Ngp7x&@u!Lr}=EycCY8AxCb!S{3}yT_0N0~ZW@quL>~l7fv&UP
zR9Z$dUWfM|hyw8}m`X?iH`*lg!VZV7oCTDWv<WOn3l79Pabqt%J;)$+|1R8r@g6*(
z{n$Bx`BjlpoV;Cw3`;p=)Tu^Oi#&sq-o4nq@nyK$t{N;%3`5&Eo3`Fef7P^~?%J`5
zg2hgRi|(ZI(N$}~C79t<_dC09PfQ4w`VFg9wk0o3?LqZ3cVS~%>`mQn@W=dgH)bAj
zUmsIPMQ+wzx!l#3V?~1mtx@Y^Ch46wMpeGFad@LG1z|>dEg)rUn?fU+Di#kRK#Vi^
zH2H{eb_c#x(UdF<{R(OfQg`a%b`^)xE2u5{6I~k7%{n4MFOC7T6J*L@F#9^TuKTp)
zeJbu_p6jE!(lcLGJg}cy<!I=W?A=GD=;4=Zu#F#4z6SvwFf^Wi2JNHw3B?bX<a%@3
zt^I)Qr0g{V6IXC=EKVGLF6HkF&T`?3-{oECssWKUNhDWL#;1;h_EEo#9^@_TI_xUT
zPt^$r$7MKNwtxu0U}>ty4sPdQojcsN=sAJvRF$sFL_hXuoGq79lJe?-M^~?FaA@bH
z{l(&x%!fLBz2&GiXzLc&!H7hw;U>9lmR|8J90Wzyp-^=1yp_TQE^d04`T_%+b-hI1
zN+u+Qii0ab$g&s?dh-+bB$|tB7?uuMR2mPA$Nu~UhW1pKM1d!%U+P`1qIQ)+E|f!~
z{IalW4Yu24fpZ6BSU;s3R+^tEZQHKH!XN;T!>_-D&X&uhF3SUt4C|xMa(k@IxgSXS
zum@MrX+#-s5s49q7VI#tA`mt!&VZ#=`!1n}J~$ufZObXt%Iaez40}rM7*W-bYLH&g
zsdO5C`KaQFN#v6|_Fr~zhX5d?K=3{mjRL4OJPw`wGx9>?Sx6$z(^u6@1{;wDxU7(Z
zn>x80@+N2uh5Q<|hD3=rbt)Ip&c>hB_QUZwy!q45Y*fAFOGMy;z)AJP<-KS;v}Lgw
z;P<D=lJwC?2N4>?&aUB3O&ZJW+U1FiCk@K*QU^4d1W3?S4I-E`pnHR(3RI7yjeYuA
zab}$;P1pK_vmqR}{~5oSOvd^QqiQ6`Lt<79HaSjoHbgdH1ls2~Lu?G15Yk6sBPlk=
z$xe|@xO;KOj=vFto6}<8w-w5RL?;S+??qex7A<y;dtWfxB1e;^UwS31=ClTRUjG9*
zgQ5Bu4R$$^9oB<HJG-A9MPE-5w!>hpKWLST!6>gtU7V2AP;t29L@ta!!Mlyrm8IK<
zYK42Ls?Nv+bjxt%KpxoAzAs7&m+YjJj4snJ3sLi2yRc*avDh3`LDHT`Y4p8{*F}o-
zd2I}OiIb{xqWULI8(zpH@_O%J&@))zs0u_>_M;uu3&tsMiz;HDT8A}Q##X&42K`C?
z)G|Y+sahGsRRN9@#|JTB(i=w4O5m6iAo(M^xlhu~mNOB>Y=rlZhzJHiCP~G!^Y*_c
z2HDT#<?$Nhuq48Lhj<uM3Rgyw9=#IX&iQ0WP+Qy(f^HqY92ja&)kQ8+d7rwoq<z3f
z!gmPTlV@g+Z1?uwM}i<OEGSE&EiveUICw9;aAz-P#`pQ!RXxZa1R~j%f{XLp_#f{f
zJ$UHYp($uXh)2hY2ObeI82R$*$2yQi)N!TQr2lRmWMz-RW0uW)!aRx%iRp-F^(KA|
z5A>-(3y^`|EsOIRBg6J1t5JOv7Dy~<h-NGis%1dz9h|(bjGk_N0f@xlW7mme$;PVQ
zF|+@TPG3@b#b-dXdwOrsLtGT{?&T5_0XUTpF{b-4&uiy|7z$b9m9pA*%8&<axyH23
zEXyJaVgE&E^x+hdcV(A2d|*z21O6E4Yj81<OhC?p5_$?Z2w9?Ui@;;q9dfb=Yu`gy
z`_?e}?I?~fyeo+xwR7SFh|`LM-c6nf=x)jtSfb2+p}k7@3;lY*A=~b~SucZOI)Cyj
zhAboRvi4-s!khz&r}v-<DxBYcaD0#@9yb`&V}tarWd<4gn6N}9NZUQcixom2juqZL
zz~mk(0dvy&R-hRp<eS&zAQ!5&`idHj_!nS_ERtii7nxduZ(mY9hCU{}_<w+ONP-vm
z?5{d|P+<=xK2dGsU`wz=*DS<TB+%^GaaiJuyAiivK`5S9CA&?4A0p`4V~;1$hl@tu
zeK@5-40GDEQg1WnL7eb`<#?Pqx>`q;cp%;3WQd*uF`vdg^fCVk8QQ@BasCJyx?}f8
z$j~I_|34v1f%NI)Qzr7hr^48(v5^p)*15w@<HThvG{S6g&CblSO$Be6Y=G9_xHTy9
zZTH?WO*0Y@?Cc6`v;AADPn;SIhZ*bBZ5k4tv%fF@KbS`}9r{a0&=Y|&J^ip|n`z!@
zi~qr7Y^NIXPLMZPaq17>b{go{^gmzh?5kj!tRLUL0EUo-$jD)5T4B(7XZwxurJrkj
z0UlVecZR^uwEDT&A4K2TreS>P=NcOzK#`GXrQMlU7{|D?GllV`pKCM&OZ_6oa{DFV
zPSpFu&OeVD4(1O#|L!pUVdvjO_m6h|sa_C;^+!AZu1o)D=ie;RA7|rti~h&i_}#Jk
z<81tKHh$-a-z4S#Q)lCS(AL!=+|2qxin%OYz+np+dPL+#S<zZ`5l8Y2lvDX{ZcJ_l
zl#+-EYvx$K&hjuSa)+Bj1YitEjE((tUTb!iS)Zg3>_r(|22cZVLaSBZd|nIihWBas
zQ_{XcsTcy@4eg8)^YYAJo2i}#2&?2^ZNzi9(D_v8vHEh+0|R`8oK5`!h%zt5Sj+$J
z7J~J^RZOYtz8~-<s*L)N2Cgzl2izxG5eeV;*$aTLC0g|n<45}ve3tWM+Gay-aKq3C
zP$|oyg@y&F`irugcs*EwAnzaZ+jKT8^gaUQYx#+{V`Oh9-dJ?NRneWGi)0h60M78e
zs=DL~1yTnpyZ{%%ps|{vhG2K&BCh*F@z%+O4u?7;o95M3C<(g<fxD=6UXS-<UJKQj
zI=Jg6h<V;$dDZzV1%N+h8r)>GDgzF06>y4rtT2jQ{f+&8MrC2=A*CwQatY&-L{+DI
zI4&xnv%sygx7_d1YCYV~Qy`bg$)Z(Ng2<Iia?Xk;qaoKMW4KAAEAt*A>H{(tVqKEq
zj_VWL=`VF{u6G$IMbyPb-e4=7jhOsmrkkU`GB<1oB?t%Pmo*l2Jc^$3+Z=(<&9xmr
zNg}}Hrr{aDrC07I5hg8KAGhiITl&Z=LuY_$RP2|g-Tz4EhS@$EzNFe&nNol>(AX5M
z=LEZrG=AEau>Mbhw-M<Qz)yRNHb#pKprBwr(pX71jpfNgHCNI7#V)TRTb@f~f62&q
zYAk5#q?s&?wcb&)@_>7yJC)br_ByZEmz_ZWIip^1KHKQKy>$2CYP<qUdlfB)Lhu(D
z-@Uv|EXayL3rjMgUgUZYPDH<xdevw~Bw*r@n=``Hfdt;A&Gl)28+QgkioU)eKXY8C
z%1l%y+5~E}%3}qiWYSn(2MEu+7hE*wyH6)33QQjrcBv5k`eMrwyKo5-0KDep=yyD*
z89c_+62XbQzL9-RifM7il?Y~GI#G8$!C0whb>^1H?mheNzMU+5ThE2xa54N$dZ&`*
zgJy^M(Ppj((el%pm(xt(N^UyY#v_qbR5r5JWASTWsFw9T&fJcSGv|9w;gc{eA{U#Q
z<5htO7&HFz^UxfYxPy(t8m7bk)VkgzbUJt}Ofhi-SGkJT?0Xt@t4I5e{O&q_cdAhp
za1&w1q0xtai-M9OyC7c<)x>fUpVXLJg7dE=7OENY9vG9s=Hb<y3<-x`S5#l>=aEZ5
zVxck8D6b9Im}=NoRn^(IV;OQ=z`;Tqn|!;APk3H-)+;s~9exzvI}2o_SfSjMlP8ib
zk$O9xqZKn8qNZ~X>XQ!YoV|_|y#j7qX-so4(`|9W4lb`Kbaq{D(lC)ElV{VZsydI<
z0Q-u0pfC3OIY5mmi<3P$`bVfn@}-f))MtPTcG{!yM6WL6Y(y_T5w067yztCY_iewg
zy<rw9$Mcyj%p^o~AAEL@7Z`Y05;1`5_MrIEjCoq`Jb*MSp+s7DO=EHpkVb~x#VDr<
zz@Nc2ZlU~Y$sNMy0FwDTzrwQBqLhfABfrAp^Cf=H4p|+w>wuJ$w)a(>>?|N;*##BB
zPOX(|+p$gbL1B=cGXKJiQEf4>AA!s;#(m*%OTL&E7xUSZnqYx3bzD}bOC|pS`E?za
zo;#*UCBds&lbN<8f4(->2a1;@-bMCPs8`{O$aM9O0Q%{uqLN9P5%mLdCjiuvBIU$R
zEE;tH+BO`|Zcb^8T)QCODI-6vu^@ZyoX?RPLm4%mJ8*eC6{O~TTL}m;R5d$GkZ~G|
zt4QO(BV8Q}{VFqI(8$`>d4t8fN$&4G;u8H%ukI56*8|c=T>BgBf}7o#QkONKSOj0<
zy9e;HRQYDm=IK-BsYW*LaIz=z%@d8?0N4g~^yYla7<Xz2!$$W)T9eIMjsyKh9FIen
zzDBPOa2Q%HOpO`rBY=H7dN;FCRg2RtacY&zYgz6n>BGOAa=JWM1wA}hK1U(t&*s@I
zal4W~&I4$;v3)$yYt3*U_38N90>I?4Z>%&IHJQ^vjX15-m&@60I{{A=&A>*5<S;a$
z-#_7~(j!C>Xf%duixg+cWX)_}as^>-Z{ow6F7v7d5!9vtTjcBmfT3mgL;`OVXdpW%
zrfhc2gQELs0Khbx7XjR{?=C|?eB-wrGPo#Nzy-X@I-HJC;**ArI(=KDN*gg{h^Zw?
zNdU%2F4!x*)+)Wu=URWI9#hMohQq3Wo(C#x8cS=Pi-nY8OYca`vqM%GTnkjms5BSa
zoVz>>*OsQU!*Jcq?UuIZV-;gy=M-<aLoonD)MOV<wFD>i0Ep!tBkd`#jWzF3Wm5(D
zDS?^y41{stpb#egUEwK{+WP=`XnGSBo)sLP_nE5sXTf7~%zr<j5KPzy__VvI-FqvN
zG%~3*n;I6N`jMhU1Wa{XLr$1(?8PT!0EPCbi7l~6e~@;`P#e&iG3FQ1eG7;r|3lWC
z^|`nbj@yF)FQDXfX4E=|b-{eog{HJ;qu>PDAs_|>BKBoO(5|-(iu^>(vptYni=vJB
z)a<qo^Ac%s+7Bu)BnLcz3xHW>v9g=RFddBtOcV4J>{<~->e22@p0nZojoLDKao<$B
z?ukSQIS2GpT=loZUpVcZz`gC(7H8+=ZuEU@|NUdj02X`>U_qwj8Kmh0Vm6`FCUms}
ztPADd!;bY0qI8~1`iy5IjZ%S!*og)DgN&Z4jVvlDh){HGEaZ+RpG`A)iRXl&sbDYv
zsdrE{x#@kqI>2the+TE~S78N}Q1hcRxz0%H2#T_Ns;ATtl-Rsz5h#R8s7olE>jdNx
zg++3<DnOfD%(P~VMw4>`UMa;Pz;cW1)S=2H=v2|=vFN{5R1Z}?f?sjnMuJR#i*SCV
z&*I5IZLzhoR%2WO^Wth|n^VY8#eL%iS9a*rbLpB`!7eed2SbR-Tfikec%J85Mssp*
zC-v<;M81_r+&I8x${ZViSrn{Tr{FmgjMU<X=1>%6S$`EQ1X<IJ8y*8xEVo(13xh{5
zqc)eLHV?6c+m<wRvCqDAp6o>GZ4P2fiLDPSj0zuXQSv#zdz&>D4?%EyRZ<$v=pV%&
z2dJ12g1$EAD@Tfox>h?E&9gaIpU3Fc0ZQ8dRJl%$o3`ijfE^gia%e$9hxl!y9jUEB
z;f!;Abem@?0uF0szMk>`2Pa#n$O({MW#BW6%#D6|r6dgP8<6@P!Nn1arE1*`vaZdI
zMZ!3I#N|q#cr)5G%>3UjI0#3#(`$1*?2;WA#WL_8H4?2_>7cp8)o?&rSMy1e`L^oa
zCK}121cg;VBDR*+*=*P7&WcZCmEEYy4Pf)In!j;;`PxjMzqCSXbAajRuOjzj9RMsO
z>f{og1+`7F@-advcS>W{Nw-|I%48-irQhi)Zz`N;<z?x^Dw$On7?7=G5d|Qcl<c$&
z&Z0Zvdga(|W9ia}FQ$rkz9%=<2F04q6PNpX{kQ^pN`0CwHCGB^nq4Svi^x<W4nG3j
zeOY&UZXo(LvG3|+k=O8eKVvkMFlN;z7G8%2B^OEr=C5*U<=o%TAa&k8LjeGJ7ET)Z
zE(iqkupNNfDxijjaMAkBaBTb}uXa@pqxeIdd*(Vt06Xu!f2uaVwI2h1fv`Br7t#KC
zoayxF`YX)<)&C`;e=fo(6|hT^U4<SNyeur^;Q*_&18HeAPhB<y(3)Kp)1xnzZ0@Ot
z#xwwh-Ur}h03F@<GOU&v<F&bA2S)3J74;cgy$-0$lk=i1VzdzHzPa01Y;@fc!2NS0
zjqHW<$u^QmKm!QW0C2odLJ91fEa*WoNe3r6)zDS#`HBn2sm&~|9K5^NmTE=Y^%S1+
zL-2Yet$fOV20%X?(h%|s@=42|0zH-E9PHm*nh{LtpI53H%{~I$xP=gn+FUb<v9bS9
z1Gr9=!PnVMn(S|!Cw1`ZkQ#w$I4mjq2KgF2g_(MQ&C(!s^hjHrS}GZdnGeZ~OTppw
zO8F1y<)J#%SIokLdDqjGCZ)vB_MS<ZwSs?#^L@#-Xubh{%5a-yW)0F)AjL_{^{P4<
zv3X;HO7eBF@5@9JP-Uc>0Mod4Hs)3ir~`o95{$(ia!b{jzw@9F$p8SPMBetkeU3x*
zN0Er6#rUb!2v`@<@%CJ<@QL7X>@Q%&dJU(c(vG4;sw%Ub9iPmfRv=?3vv%+nE}=p3
z&0W*rUVV9;`^+0U9F##waK)zhxGNvG;J3$yNOgCLb9{fJ5tNDhGk)pcuB+)45ztad
z9(xmd9WajSIeGzsLEX)HT?YJ0j!E;cZ=8CR!^!2(#MnNQmxvo5gJ|!L+I6PXT3Ma9
zA==fEElLiNwGl#<Nc=oISQqHjd^MM}_ITvJhjT(ukI)Ax0pI=uF#}g+uvj$OLC-#F
zPHrl=Hqk$?1J%=r_vr#4aHwY7&kd#%37#WXd1)ZkaHwmkg1>idwl45CiD=$8&aNF8
z==5u#daw-&7lZ05lzG64gmDdwsb3Q&W##8zuP?M44n{bi=#bY5Q2_@9$AwsAuKvvM
zi(rG?uh$b#o#$==H2%{T)93!GBY#Erhv=y3xEhT?38+(rmQE?8KWO8kJa0hdmWS(C
zg4B6xK#C)EPxA|_??^mJy}+LwDQH$SBqI$wseeu)H+u&U+7vNG%=-N3kXKS@?Sz!s
zka#UgTnsyX9J&)km_iAcWlpQ{^!lV%5oyXRSA$we&O8Iz{5xDZINx^!3LNkEH);`G
zMY@_GWe;GQ$l<K~`j?^VI%SH>b-oD{iy_qBq*pONJ&#}tp*z%F=-?psp!r(z$UF#1
z1&Gn03s}}(*xvpAvKa7{UfBvN+9xRTJo+z6Xwa4DPrEA!n@wov+1@w<{j9d6=>2Ut
z$0OG@{C6A*!>Fqu%G*iH_}#zxvMW-MQF<bA8yxYsOM{<D!RQf)EE24%IUfIf=eH|3
zG--z|(35}wxwaVbqDPyUU;93Sas=$J(dx+#eO|UxqyH@Ugd}g1q|kNy4hsfDkboG1
zH=LQ<yS8j!2}&4zi+=q_joXtn+Xq*4g8ixF8XZJOi0@4EpuKR|!ebd2y^M-C^(S?@
zIcY{WQDP_b*<?R3^)oTeKmTImYc2cwn)ZyGzZRR!J!5;8LyzpLID_v!>>zu#G8)&x
z!*liTifp=LY%%8vUf-EnOj>~K%izb?!Y1J53>4^copMVph{}!}T)+BlKHN!kRb$6x
z#fJmc2V`S+Kvi%5{v}`(vvVQ-|B6xU5#`DLgm?!b$6!_*ywh-;ari2H|LQnD0Gv@d
z<(lBXA$mOyqE3`ssdi9|+qjAi5t`QzhIgnK-o_~6<RAxOibjSQF7tMx!8?D5{0~|F
z5cwA`q5cCWf8ewow<7WOD7^kTEh4<IoV_&GM$5Wo(Q+qV(@TTmL_LyJ41WuS5AfcL
zq>2tZC`f?l9*M*LwT$mCf)YUyq=a9`D6fZM>-9i}_E2`!qsx#u7f8U37;}sUWr#Ar
z1ORO_gGPUh{w#AyCt9Dp>lkG|$P}zFg;Yqp9>x@t0AasjGlusM`~2Ky4Wf#qBK$Er
zQyf70)KO*cM$Z)>JN%*zlYhhI6GkV9B2wq~sW)JIrM03Otm%zvQ{K-hzb`^O3bwt1
zGnux1P1%T@xhg}1k#w7bE@g2=8v)z%bc%E^#mbk43m8)zhbd$)?}^x^CrRU>6jg!9
z*A#;y(HBA48YOraZqsRa%Ft3Q?nC2ej1W)p9867iQ1TJ_)QC}MgVI!`s*K^&=p4d|
z^c8cpcy6+N<BgCUf2CS57mDr^Jk)`y=a?GKqLX|pHbWx#7m9DtYsYXBpiodLM0S`3
zeQKbCcq_2N2c4tqm?fSF&DGyIV)S1CIz(pyO<r>P2cYu;${YkkIn(;ab}m#D1)9t>
zo)1K)iTF#Hn(eA+1V-gG(1G=DpvgVo7<P1;L=(f*PVvvUw;#tt#M13kso#rE6Gfzb
z3^m1PMf9l=daghf{%1AMy%@94BOBMf<d4B@$i~HB_H{dFWsE7bV2ZZTrsC~W5I|Ao
zajK3U^iRaU!oEjz^!qSKj_g|&cEM^^iIJ=$+7Ci`A%-Xzoez*T4Z@m4SgvECuc@8`
z)@0n+Zi+7V$j@(pG5##1<%V$);E$-5Ypq1+HMGdWqxT|Hh-b@TOmPjSU{>pD**?V#
zj8&!dxe$HZSSV=O&A46k=z{{`utftF(CaaA@UlQ!!wJwiV9TYQ0diDTe$v4lA1Y`9
zl~CJgbW>aucCdV=jSz$T>p{Hi*tYW$>3eDLg6X8)FpdF76ULFpMW@4-_qvQYw|%CV
z%}2$d8m02pcK=qK9#K&4wBs19FGx+n1v^85u`guR{Yb@P;PV2^#qMZw5Jlh)#GA3p
zqQqEiG?Y*EDgQYHgv5Ui;TLEAk%WIl;*UD|MT<Y`=pTLipKSsIuAAuG#3P^F)WQSo
z82a{RaUM>nSL=(ZUP@6LZ7cW3;)g*ok0?qS{vtB{$@|4d%^mz7RiA=Zs&_*PO?q&I
zJ1mP7hy^kHGynJR`|odY8&&g}XW{z#c$%6Ol>??ztEB56BWcc@wsE8=zV+@tpY4Va
z3l$4a+Ue|>3eJ{3+V_|A+U)rIbGU?l<!B%0729kqy)Sb%Ao*v=-^>v!;wIz!MiY8~
z2_7bN6CoL)>6uSTyo~M(V%S;n=q@mv>nS(;7QrCK>CW&9r+~NZXMSPFNq$jg`)q72
zN_B6jDYlJojqN9D^Dgbf3|D-6+WmiaF3yQF@p^N5s{XkG$ss?zI#MzkYH)==2c4tj
zzW6Tk*u%ry%ZT)zVW+C!xN^oQivN-$f7@?qmipZ&hPpyd?dvt`5(^wHr~WGQ_6Ea$
z`bE9uKBcrcJfq5AONSTY*|>aes~P&q&4n-8Zr8g(0Kj0|b$fGPrN?5x(RGTkL+1?s
zxr0MBz(@~c4+vgC615KwJR|=4SQoHIbIr~AE)_o&b$od7Tq{3uyUA>(z6OM?(ex-I
z-3(K8`)Nr6!LU+bA4inX=8XDM+2v*;ebRq5MJ;8Qj!(~_pG%3h$NpyWTYgCZIso<W
zGk5;ys0biLJdV|RlnrS*`N}@zZKkK4^2Ex}j}BG!!n<ZOzcqH;{q7i|-oe@57Rmp`
zWAuu+utPRw89G1k6OUhc`>k@u?|B$y%hY#9gXsNnl*&JOsD2J`eNM%H{fEU^kp>x9
z*L>oUE1Z`fR3>Uy$}hhjeg0#O;<Gp(>%!C=7=F08dte9q7N&fE$%8lr$Y6pZlLyBY
z80f-y7+p12fVqxd=?{mTjYZE>^`CrIP$o=M8-4wsP&OJ1c7Ppp_}LZMTL_!lYoSp}
z>4B8M&L3;6AwMZ*p;P1kBbQ4BJ2=%+`kQW00TRAC7c!TA0R}pjZQ;pLbSA+SMHKjb
zjgZ$`*7zrRC|(QG2;6S_rv(@a8+Je-e*La2lCw+D9+*hdr44y2dgI3$Pb(C!CAsqQ
z|7i5GJ(r=3gi8b}4VA|r7V-U(vYfuw8~;xf|Az^1{xHEGCiuey;{O+!U{0bNfEf4)
zEU6U0XW{G40KhTwQh3V8iL@|!?KlijN#en92l>3AXMQEnPL~`(PKnMwsr$<rpu?}n
zy5u3}kEF{yEG(}5GD!2#LGYE0l|Kgpg{`@isE@POYf5zh@us%yrpJMkVxtU&66B^q
z#Yj6-`7=F#*+S1&A3$Mu?PiFA@@I9Mg84l<aiYr~)C$RB6{3!G&c*WIQSca1>H)Yo
zkzG5WQf6#d$8&vTf|*}VXbNHDP(urq&o__J{Awm4J`OtBT<yy4Fz7B_hqH^<OTL&Z
zATo36{tv^WxCaa)kMdbMywlR%fF2>DvCh-y>!vK6y3>Q&dj!vqBjY{kO@MO?gaaf`
zADWl-K&v`JE|+(N-tNrScaO#ZNltwX<7H0*bW{&qq7p-+Qd6X((AHyKvf)7rRweKR
z0G^RTZ;_jDpLzNmX!-p{d>4LtrKm8l7hjU1y-(?b;lU?wJF5k2Y$MzFnGJd<0fvLt
zsh$w~(*C=-AKKti2AMt9h0EW1+yL|w6jRymm<-)*cDDn}WNr#j5{}(YgTY+xJ%sx3
zw3+H#VsN3Ym2*zPeG__~5(1#p)aC1_4yC;CbQy}7VceoyA`+aGGcb8YNXN}zsUum&
zsZ=i!^1{VTu{&O3vmf(r@Gib7@@}7Wb-eJ@IeEDuZG@oI>#XT}A1xZ4g<i?hH#Y9A
zYn?d};!Hqvw#%{RjtyJ^371(4ex&ll81y`w!Zm4*qj>NvVNE?BPt|>VGXss=McjMI
zqo^8^p_46F5<t0o;DJ0JJN**ZkQ;RgQD?Z4_U3I;XFz(I>O*%VeLyTph@NWlATl!r
zAITrVD=WVy`jwK6{tMg3W439kZ*;RQH*#EbUPb?TcIa^X^x?<iN5F2ovbISxcT-Bs
z|9rNdxDAf=%6e(fM*(Mlx4~I!58D>LVBOhX{oKH;>C(|ustt<)CaIOXi+H4_X~n1N
zJZG%|m`ucNNlpj2NN06GS@}@BhDOt+hfKovcS5%+AGkX|&XHD(Eg_m)eo*UL?y&FV
zSGdX7paOzLemxoLLDXE!hm8{kNu<vRQGWW*;sUt6;qH~sQwjUDvsfOz<5+zlSvtJB
z;jC^N6VrUeUf^Z~DeDE`A^IU^TB+Y}5~Jy`6*fn^^j8^Dh!!`-qt?&3-xb4(%9(sx
zQbNet@~BS;dioAf<^h<$q`xXSGXUw)+XMIwxrsTXA6n;1Q?BuN*L;fe9XjZ$S+&{_
zgoV0+zt%N*ggAH8Y{-6%!GFevMC?%E{PVXy?$@fyqbPGUpdE3$bo*3oGZ#gurorEz
zM8EysX;=_vCEGl(@rIypx7qj9a-qGG4uBu$>uJ6VK(r1v<y9w}$MX?FD?IUEMYCid
zXDnoR2;|%*qgNzCK@|zdSkdN&hRGtpHA!{jgf{>K;p3d&cen>?I6j8aL2m@4cH#Kj
z25&4BH9^PqW(NI7Twm*R|1NNURbKL{&O3rqvp&~jRTXFNPmv;xe<3cWI-ahnz@$JC
z5edo8I(NOW{N<*%8~aZ@b>7Uw-&h><I+9cKZBpI3Wt_oE+iRvlzmY=sOoMd(PoLBD
zp?J~O7|-QT-5$R!o^R9$kINxd)i4T4Yco$In(GF1KE9f^og~uyv;=_uBdGClmh>7x
zca%Wq?r&27C@w5;{Zh0!^Qh=A=uvqb0fFg1F$%@EaVe!Kk!4RlZT_~FWNl-yh{Wzr
zBmz0iAGfk0Ouf;VSR&N5Tst)zwJE&X!sR*bXX$40bYT|o)%0cQ>mXA^Pi+ca-5gv1
zub0{GJPtKIJ@7!6jpButItn#yZsAGSI+~UyPo5%=?5@cTp(98o3K60ywyJnk)%EX;
zUDpd{zBIBiR6HuSF&((2b&+zJ%(~Rv&Kvsa<p&9ozHFrli78>rMD8u**f<3HTQ(m)
zE#<P=a64K#BKlrOt3c3l>f6h={BFNpK@?Nr`08X=nu{2+`UWaX0Wb5sWJ`G-xH+o}
zn7SH560sFWyS?LQ8pycR&x9{8SLruABXqKyfA}+6h^9+iP*1LOm8RE4QF-2r38MP=
z8h~D<2p)9|3m}%~kXD1GbQ@DXQ-H=fNm!c|h&)!xr-rYc!|hNhhtNcx;S#`{n>~fJ
zZQtSQLmA5v7m<$OD0oI$4`6Bb0S0|Kb1_7hHNP8|3y(x(PI2LrRZc6j`*O7|F_&wV
zoVSHw8s`EC|6<L;{!&f_(dBD**~b8mDcS?wEe%{&hT}@F$Z=92MFnZyw>Dop>F%p9
zTQqz;2w<$*PCp^{0e0Odcl>&1YujDQn@dHIt{+J?rmNH`<6v$MAwWW;_ujQ#&)^vl
z!h{QDb%3bK=y@P_;t*NAS)c^~WP5ZsSJYhKMCguqtxb<j4uoo*K-w7ULv@Tkv~?^!
z+yENp25Wt)id$0C=hA0wyuY*z1CFmYas>WoUQ+X3KrSTR#%_xY&b%!>PT482ut&ht
z-p5~!K&!yy4(+0;+7H8yx`B6~0FmbryG|;}*EqfAr@5~=0Icy2l<BxfJJ?Z#`_Orw
z$sZJ%#OKccyo}hvGk_$`q_FQ@Em|Ak66zw3TKN)JqE@uhc(%QbD-z&)X`khV?%25Z
z(P%^u-rP-$hd?e5jx8fFDx^E#6g0InV&5!l*DbHFvklDKtaJZ%N*{%L@V#|rHatY{
z4!|tM2pP(|FR$*GIER#hts3#gLK|$svXeCc`Q$yBLuKHrAG>t7Ph`12xD@GgY2}wl
zRbC0e<0#j%!Sp(A^;2zyy(U6*+#ksUDI_k?Ex(s8IlK4p0abXeo-Gvm_*mtB<y#$k
zcwDLE4y8^@D+jGz+5R4fs8#KQih`5*vl-pW+T%kKbSw46RwYuJPfY7;^C6k-jPgus
zGt7?cs5bTa<r)zkU^Q<%k~n{Qsy|^l2*B++-Cug<Es(cfs9ppw2Jw-ik7G-wAU}-_
zSE*<vLDj05_j2Ze*&}b#aT~Xkd<?rd_4aV1zR%%MC(b<%0b(0VV@)dD7s=E*@AW)_
z_RPkUP=mG51H?Pba~~R9@>^z9F}M1jvBQn9@fg)Q>iw4>-6S{=C4e(4wN%p1X7P3w
zqyvklR&@^*MzwM;k;tj@T%w0i+V!VYBlmhDdyYipj{5G3gW3(p_m#{Fk-}-3m-n@o
z)z{btep;YdEf`ICYn(VzmL7Gz%}gULt{|0)`NpuEZdiePwN@t&Bgee2TK=&Pzyr1>
zzk$Bt?bZi-N5rNNYTHpnQI+|!FS|f3U%vP!M6~0UsD=J<;zV$L)e&Bu7j_O4h2CVb
zQM(xK7j5YPlfQ1c{6wZrC5d0&(hI{<(Ye^uO^S9DVM|M6CSo)xZ$=Wva<maueDKV@
z#E%I6wRP3b2vNfPJuiagKVXY<f3fdI?Ip%>BL=vtrPGtK(qu7%scQh#Bt*1dCMVJ)
z^dR2I+C}%y)<7z3-^3+HQ3Ds&BkZEyLbZs&Oly#3H=3;MR>&u6w#}XMO&l*9x5JH*
zJgC7)W08MYeXtMdpU$`7LX(u2>hd&$DOCO0EJN#T`WSmBwIwuY|0qhgN~bn5>|PeF
zyvnDE6iporH!s=*g>;R5UwJcoN}j=W!gRzr9nY26Cjs&RhjBWs1~jiPC96HcLI-qp
zrGG&u;#0*_)etM^Q5V4+2n9n`jdVt1lBoz|T_))L`G6g9x#+uyhUjx-sMkWdJbhJt
z;w^{Jd&Il6oo`e#Nvz0U$WE=vDOYhvsvO2#_bru8$>e){0sHpE%B{V<lh0NzUhc$e
zMoXx7Y+YEO<9=^H^=_{Tm*EuigvBRGN~<D~DQ9i{K~TxI3oE>-u9ptdn-?9DkV2mE
zkRy0WyOG&Dus1nRkj@S_%Ec{V>8=OV^ARoHO~fu2zvy1pes6t9T8w%o9SRpWi8g1&
zHiNCaW}kM2KiRijpGhOtt~O>e=A2)yz9y8CN3J$*CWiEzwOj>Y{6kz9#8*x$A-HX`
z_pojkJe5o%7(o1jnKroq-n`)g3q`!PPo4cXZ~;xP_$5I^b#z$5?(i?TMRXH(!IrYj
z7ImT)IRnb_k=NWi%GZb@luGqc_?ZxX_MO^+v-ibqa1m^B+LtSGcMy}=is~IXt-IRR
zTmqGGK?NH?2U0}0j4fB|EM%lX?av)590aPHVFg^7CzA5v6ci0u8a)Nqk;9E^W4~I>
zz1#=plVUj*sl=zJVw!?$vFa0b<`bscyEgdTSW=S6C8W~1rtP4-j0DFagD6?7*RwA{
zY<j=VVWhw*yYtYR+P=~<Est{P{A4vZ-iTU-@(@4cy2uEtur30L>6%Xq^A-Gs+}d92
z^R1q}oVC3lB*!b^;gX~Pn10LgI1?Vp<g>ZBxuI*^?pd_H$Rav@4>3<!y2k7qS!ySI
z+4q!7XDx!_D#DiB(s>(AEuVhH5+ld!6A=zms+%p3KL<Cw2k*0fC-kB8eVsq~b>IG>
z0fhL6O00IHv~hdKQ{VvtnJAAGz;m~b2#q(|7Opg=dMg09cK2ju``NEhG8N-7nqUJZ
zfV<fj%T#O)hGJ*!;7OL=hfWz^d)mwS0(q*_)HBM0_*sVCb$#X{c8!rsL)tQfP?kZ6
z$ZX~~l!Go5kBGN=H_>tH9Viu_4%L`Oo+NsQw^=)LexkdmcMwW811$Sez1Bz5pj4D(
zqc}6QN!`$Y%4>7ct91c+Fa=WZEFntHe?Kb~454rtMNlupn{vG?`Pph6M{MelF23Gi
z6J7P8cKD+;cg|dcoJrL9MjycA>sDqT6|EU{Cj(>hUxY`4bZ4{R;LYhqH*YDzP2Q&6
zzaqsO1ak(6kbjxg=1$mz<n$SGDGJuDx9)51q5GNX-bXgnxJAdZF<ew6-WWuzM>};A
z*P*;wXK513w->6@mcKpqnXMD^>YfKriQm>QA<qs%X%Ym}ys!aY=%Zx4-~!Y2=@GRe
zLBywMO*H!)OO$l$^YW7=;XOKtlZ#Su1x+7^@+`ku3yyCP@4d4xpacJlx2`QvJX8vZ
z^^9@EEjUO|JriTXkJJtv?fk|r+RK>&1tckBu7h-TAu&dw8qDok5IXr#QTxWcn5Z;e
z_065}1^pT5vYJdabFa|O3slQBz5wJokc4}kCf}~kgAu0U`|9|43PQ+hHX`uYueyr{
zY42}iL6Pm;mL7Z<S9{|ZctKGN^?%s=%cv~7t$iF;1St`fQc@A5L_ujJ6ahgAQMyGb
zC8VUo01Hr3Ktcg2=@cnJK|!QDB&17P`ajooU3lL+p1q&(+xz|SjQ9THa3G%Nxz?J!
z<~)ufj2+zgw*RQJqGZO1FlpWJ(fSD_lc$*+Ben%KmI%ASPw!4!;0Y6b?7H<S`1(NX
z=}Q#)5|DfLvz~O4LsLEjH9x2Cl#0TniJ%MP*25!#2Vbv1DE`voJ<wjDfLP?`CJ@wx
z^)PdKI}8lI|Gr@fkqr0l-Wk`q;u*J-u6?_h4z@1@Mz6D=5rZRXiSv*PxEmmYMH!-T
z?U7g8telP;o45VIRN27MG*4SVT)$EGe(P6bXbB~3Wd1GNqMmk0le;&b?^Cbn0K$W;
zsd?=q>?Ohw$#9C@e(?I2<A~_@L%nk$`PVmGE9iRnfy;jRz%5I#>X+<DYW20igY=Kj
za*^ZVS0BYGXB6!O@$Udpj9?C9^A^v@J{=c^;=su9GjwNTLXlLO;5%Avm;Pn;eRe@I
z`P6fY8&8-R9BL_Xy4%#v?BY+_=RURE?Br_s=*#<&!s=0r`^Prk?k0C+WBOsv+GuGm
zINf!qY$~;L=FYT)<$DO>v$f?R9a&$~opDzGHKybdZlMLkKUrOFpJ-4KsvXrnrnpjb
z8YPK1x_(YF@#^s1s1CA*`IQ{O2L}&4<{~3o)_~_%+z8bTDdsPYxx7c{NPtl0HWmPU
zn0}48EqHqGgNcWxc^gZRoj#kgt^)}4yVNSCQe3(~@dGW!b;LOQ>``7&(4$$zpjEWk
z&iv$fV^8-@a1!kGPcRX6)m@R_jboBV_;XsAh+6mtt*9~-Q$pNbcg<iGQJ=weyNRcs
z``HM_!?Z4jOKw7#qB={hHoJe<QDCL{5OWWcK_tfdRp))$p|mH#P8^M#XUvqW8G|`1
zMrE>Fi%}@h?^*)W?@&)7j;2}i?kbIl56nY`sor5OFf}#1q>AtivO{66au2T#1rzUe
z;8NINcj?_6gyCiK+0uh?kNgkXIKf7oD^gX}`M|b2q|?7(t{&wq1M=r)skOn}>1Vfi
zhyb>L1f4p=V>f^Ql(a(5x5yaJf0F>8tCPf**52g)v9H0K&a&NI_{R*&2m8?!fn>co
zKo(<iL=(WvkE!bg^LAu+bY}|tC!VW1WLQ=GfypJRuV{<??&?X*mt@2kzTT0C4I9w&
zQx(Qz?4^fsDZKM*hjH?0py5@7a__o~B7p;+bzy8wfXH_E>z5_ET`Zs-JuyZfwPil3
zg6=^8Oa_&MjhJO4<sqo>K^VjH?Hnn0P>&*4nm{%TY2J4vO)E(L3iG(n-G%}6e^0vO
zC$8O7M|-EadS7#<ZBj7x<nI!Nd8cxyR0b3;XN49Y?}lh_SANE0H$S(>JL6ETt;3`h
zupO)c58lTgh$}R$N6JZ0e$A_rDB0SWA;<Ubo%OqOAJGwLW{hcnUse0n6un7FtzqCb
z*{V4uskmn+M00mZPY7lBF9kw4Es+l%mmJ@3fcx_R$=M1qL_6ZG>(IeEnOViS>&Tmf
zC!B#sfgZtHA5jorctuV&$>sfez@qpJbgYF+Mq&0)C$|MxW$~1hz(>@hKcNOt!sKJw
z&6{4bOw}JH8+wT`zr{IVjPLC;sNp_%+%l6kj5(E6-OE+DHoS4>XH`gKlWIB)=x<Oz
z<m&#|C$GbLEX{^u*YT6i4NrC#(yM!G8a|-PYO5$_vLbCcr1Pl%1x!y5qRLD^3@smQ
zuOk%y32W83T>%TALnh2?c|cqoIH<kBQVMtF!C1<D<jL=;EeZor10lg<#I6KPd!qNz
zs5I&P$H7lKR7CA^fE)IP5zsqkmm|*ZTlNu4JAqob%^Hd)H%<=9OWj^SZT$6fx!jbh
z)in)v@4Ev=fQ><XTcSd&GAsSyurTCnyThw8t6Clm$e+2*|LUOU<COiMT3kh}R;bhm
zHLpEl&FXsksjg^D!HTwB)-?<2Wgy=LN9+hAce{NiMO})|+Au17^AN)}@2N7am*e{}
z@TjGH8%egZQCk7Q<=PUfwI-$Orxz}Y$NF}ei^UQ<m7aj|qkYo#vxA>)G2!}EJ|CRy
zo1f%qz{E&vS3d5z$h^<$eeT$s2;<@o7(prrQU$(56Q}_ljCthWZ`nkpUZ4p@N5b3=
zh=D2qLP5;*L4@PQ<gL~8EfN^Gv(~r~tXfX){`20X!xdg3$_n1wR1pO#-_ShXHB&b6
z8HlNG6odWU13{34h@eRwWa)L*P%QF+9fCk~cW*SGnTkVAT0Ttn?dlo`6#91X^&@}B
zCus$*mG|_bVWf_{U$T1^bF4h5PKG;_2>yI!68-dsJ|fA}a9eE2j=B8aVvkg6(OmVB
z5*jOoysxKf?t{7K>X4V6s?)jsiK>6f%EM)8&OY5m7gHnHIA5U=dT`%k#fDR@mEg*4
zNPc9Z5W8AayLkJiniA0%z5O+QWiLrhT3&Af%&14*4m4A~3=Zlsb(((9H%ubEkaE`#
z$1>5ZYQT(9hBT80w}TZ7AStFRaBD&A@d!BNtnSnFF0L`CK71!aY5(rxEttwR=34EB
zQPmE})shZojqM%&C<wO^t+%8jxzdYHtxaCkUv2ye19$f!5E0U*V%u$d?U#wWKN>!E
zoit*iBl`R?_<`}amtm=UDZ4&uYv!&kLy7MU>2T3<@eD)O`vQ?VV4)1R@LzkFvWbQP
z?}^i>b3A?(3vNpQrO290gl;j5*N(3}X7c5lhOm%-9Rd!zoQU{PJrTMqI>M%3SP4(g
zGTalNZPnpzrchUT-2bLPM5#+;cjSP{Kx#cQ1L<YwF;S1x&6Mp%X0O-iZkb9+6M3pn
zt)ClK5uvL7mOP`FKSubzG)~}Ln({TTX3UQUUXRmy8A7e+4A2$Ss)ar>41CTwo2Lzh
zS6K}0>OkrnHC_9bTSHNfHA$IgHbJ7ik)`XHDjzTxh%|grF!v2r8BitX0W41Ny>z?v
z+SB~fLKuF`YPmNsa;+?J{x*<MUN;Ocu!weBI%Ib-^IP4gk=Ut}`T0zJ-fqe&MyA_C
zuLVnHeLIrRMn>P1h=ySHVcu-Hg)7kKdVIz^$9svYLVUKgU;u`Xpv?W^Gg46w54Xi_
z26ScAnsB+Xj=`h|3B-u7Xx;cp5T`BL5%Kazm}*P+#cSJz8m~)-Zy*g7aw6I{notJO
zY&q2O7za{|uI)1cs*XvMJM<Aya3+ew{7+iS@B6kdw}hdg=f^|gN?l(QOk}jbrOF)2
z{8u7Ws70SW!nm?)H-}8I6(LDm{L1%8%)@Oh55m<5$w#UOD7&zRi?|19=9u2D{_Ks(
zH{$ybE1&fW!koy5a}%d<jxYvMUy6NXFl;l@MQ3{SuyaZX{3!SM%^AYrLiAu~eL7iO
zY7oxN6NR48G->;3l$erxpOt@$lW5A5lO;j`%EGE0KO$JVqJT<@`gWo8{yV(FNuJ(8
zYiM3L&vP7?hm_&N{mwB@L9aJP7%`&Q(VBxuO@g^~w?w0|&?{H<1;uc$9Ovd~#eDS4
zeAy^bkb*%<Y90xuZk+ahn${0&%};>0@IhjM2@Jsp62X0sUPs|u34EffItFJj+B@*6
zb^BgHk{sUe>?u;I>z9XKR!KBwl#9CZRfV50e09W#eCFb9K3pa39Uk^GYsDm<zLadh
zLlll0#$h+IDlDOd-G84#kx1L}-3^2wR1UcijX=lsz}N48IKow(q`_Ov%XhNfJF-32
zhv*^>)Z3V!YDDt&g}B2KH{}zYQZb0xkD=tqUR+6Fk?NoYX8ZD&WMpFA;z2HLAq32r
z6{w60$SSWgkaGc6$Ok}?lJHiy^&}p?`C*<A%!b`NN4Gt=g6sEYs5&u{Z(bKiIDCIz
zu~*_au#^q9NBAE|ECI=0yxHEi|EWEC-9qlAy|@VsK4pVA%!!v#ls_X5lI*)vx~5xz
zisg_pQ0vWyx7{b8)`-C`ZiF7{@#8$57)Nir=<ACsw)60UX??V!kh5r=vUFy%KDwML
z!ID87;6bkIGai@@OUznQMVqgZE8p3ci4pDqYNdQ!-%uyXS1Zg0<o(!<o3`0=ln1kQ
zhjxEG_B*q^1eX9?Z0DAf#RT=8&;M%L(tT!s3nEC(0+Uw{sK-5I7d<6+f{YOyZrj)w
zn~Dlph0!}JaaS?GL0y2^vR3mk8mwZ}lX|USeofD`YkR^X2KAh8q~Cz}d#c4b%4t}C
z7eg2FpPD9-RQFwWUpSML2^;DAMEgg*3?`PBRx8ik*6KQ9J51qh<klHnPg3|%o^G^V
zW{MlW+jkSsn4i%?_1%Y|{8Y{HK%386IZi-fR1PsuM3+Wq%934LYtuX1{wPJVx(R4O
zx-v3OEtH4-QoJX9%G-5zLwpe=$8|iC0=Psgmu3`mXOesJ-Og=UJa!yj_70NiasVt6
z?jKaQCnOjDWuO66+O`w!I=Lpbwf-dPQF#iZI;jcsI{+rk{uktDW<ybmlyx`J1mOFC
zB`e7a%tmdtR6O6#gD=Mvn=84xca~U@ke<vja2?;qdx1MIO!i`Sb?o<b>odeF{*iia
z@6t!TOU|<-{j-EKjG0nUx={dv^walUWUkH`2ZP^9)+r8HEXHJuL2o4saS%CtYKPN0
zDVVOb>g8qb-n}ih+VBO)D7rBo6x)vZ-@@@31=Fs-)5CY!&1ATlp4a2hb-X*f?`}!E
zkO%9`RaKCp<X|qdQc;W%kb4rRjWe=?_yfsevxd2F#odd$;`Z9>RbXBxlI~&F-dB6^
z&Y!NAJj;|p=3gB%>S(MpP@4Si0}Smy5il6h^3!-N@hhzGubiqWcy1QyofPk#hez0a
z<er?k>ZLdFyJo$8Q{@SRCDSD{vl-nNX-G-A3nY7d%w2JZyc3TCn(gFvP2S01TXF{E
zD^jSw6k0235$AFuYDYQJD?^r_R9yy)QkLH;4;haB6W|Bj)}Rp#GW0MmXY&jb$U*Av
z3PDV)-s$yT%k9TeCSriS3Aa`XeG=M&8Z#`X<h$kz;Dn*lp2OP6yRC+QdfM-A#2*IN
zb(-nLWr2j<F1<g!1@8mx;mIQ4jNR+>z(stu!QUI8U{niCg2J8NBOLga#6Z1m0o~S}
zKc0s92}YI8K2s1cVN5>o;C?O_yYqb`TJ&Sw;=nsKrnUB}a*l+T%-!?!k4&i0C!6-$
z{T&<ScGdOwJLZ6D^~?8WM&f-J;BRvIO}`7cm<g5NzxVqu=4gN}`!4NXug8c|vB<OV
zAGxl#yF9VpYz101ewdgk;S*hg5j&Fs@1?)}_rF^woC-fQ<tE7;A-gB30FJ|4<wC`F
zmF^$DANyTikKkJ_b=NZjGlmj0&q!Q!*$#gHr~mQ$>tl>z|M|HR`~QdE+&*8)LJ+UQ
z$Pk!efdDA5G)e5=9hfwzFQ=AJ&oO1hz#2~7n(=!V|M_1G*5OkI1)J{f@alz4Tlm%*
zwQXGfag#7ViMa;<gNp0L^npZ#;!sO^3taHOTnLW?d`qVG3d>HsVmR@!;>>^O&HlS5
zNJ_&YaH_EPu+;B^|LaZJ$@6c%%D~Qx`JNaAzGM=Rs>;j+Dg8Kxs=z)dw|X~K=SQ@l
z1y-l6eCfCBC_gL%m`sM@+6eW}ne;y-@F3#Ra;$Udz4woe&7P8}_>cE2AyUr%XHt8U
zg7rHHv&-;EAu16%;tC)VCtTDQ{Ut$A2haZ&oWuv7OrB3p4nh@!@*5<=8Bh<R_OSlr
z-sksk9<E2Tc8niFZsqfKXM^H~zEv-jGDw7e2W1NF-@D5RL^n;*uZfozs(LPCHl)lN
zFXShaVtjtr@&51H`Yzi8c?;NP@lQ<45VVn_8>(4Chx2%Y&^vpGFvS127zLp@P><aG
z(__)aM=|Z%UWz;CUn%-ti_$JHJ!m4kUGTvqbUBV5{(>$`lZaz<)1kb9_$f4UoyPmN
z-a9vyu<~WW^1YSB;<f}y>f*06r<4X$PLHLR#$un%ZXn)VI$w5Rb0hoCb(}TB&FpGT
zyD{QgM24p0dim8rr(P|b2w;g($arKWp|CwRT)$(dZO$f*W<kOhe?6w;Zkp%0qU<rT
zHFVNL7_OY7$E?`se=}nprh$#sImneCc+zd`Vm}~&-5Od=tyn~sCo%I03$dv%|Lj!4
z%fnMxKBs2oGWt|{X{u=sS%w7+8)!HOeDqc!Y{&Ks;n;BPNouxC*~Yab&{B4jUo^cf
z7c8J0dn)Fs0`OFN-YIP7bn^vx>X}cJ(;Q0N3-^)J31>ABzv|}<PsMbMVdcN<Jc#|V
z*XkR2&_Z&n9qMDr68nxO`huFB<IrW1ldO3MTQRiVQ694!Z`gE*^R>0oyYW|o=n?gM
z;PB>}I}U6PB|U&{F|5@?bl@t@<1NFe;{nzDC-~W8lis4Ij_JTtC(crQcOT!Gtcwu~
zFA~xhvBBw%<zhoAj~G!@BUw!F)u&oj|KhLE6h`N5#Jq#38^<wce67RCPxu%rcP%o+
zJXeJ_5w=%X(YY-q!pANWwOV2%Rg!?<YeTgrUw@QMt2AaZJSGZnSZXLGg>}6~okDEe
zcs`W|gMD1S_Se@2StbUA)M~t^mHi?jwO6L4kvmZD-2O2?+WFzlHq4&IoI`uS9b)}^
z3|RRj?14>d6YRm;Cwjlj-w^vAl1p%!VE{}(L%2&a?=AB!-~&@e9B=)cWoK%Yg6mpR
zRha|UJa8U1@Kc*+CSDHOZVX@C$HT;F#lYRK-`hEQ2xp-UP&Z!x)$rme)Qo~|p0oEq
zEwMsR#fUCdJ$x(lniFAUUHKY_?g5&F7XJRIxwZ@BXkTJF(I%ut!zSo56>m?q+L2s6
z(w+^3+}?oYTAJPE>#_+&7*T}r`oMRMRR7nxc4%oK6^+`p-xS<X4r}FX{u7qLGR|MI
zB^y_H3Mw$m_Bucu_pL#o-QXS%K&2h<jr)~`ZbEdNhJqi^3wgVIJ2U?^1ArE(Y|ap3
z^<3e=g-CM#FNxoca_@VTa`pYf*=4UxV)&rbIMvT2OMDak_(()SfZ}t11TgG6Z5z!s
z_smcFFwqjrOMG<rROpmQK02^tSI+F!EF8|{DYe8M$n3Er&qnFdW>o_f`nSYj$#K_2
zH6p5mPg&Qwlv#9fw~VwX<@dc3!}KAF-L~D(=K&xArPE$;)+B{3zkjOmLV%~5HOMXJ
z_KdJ3F}})(>0Pl;gf0VP%-vV+ZkgF(23pP8o{2C$zW57uL*>pNtn)-rx*FvkXAH!*
z8BQ^tjGv@C-9V{X7`Oy0MkKaY%q7Qe41qsAApgSoIZMBC_$A4*{ODtz3D_RM#2CqO
zdWDPxi4ReZkBOJB{`BC}MSW{tDx@#S<3kgeO7~^l_i!%G*}gW#IiQuvQl|M|-=dgo
zJ|8iJR@#~FAHs0^2RoPXMvn!o#y;0Ij{f^i7&Ae%?_ePIH69<LG&K67cq4novCvPI
zEC%9y$KDu9G^gtm!E)Md+PCpxM&E;Of4|7{X79pH#_t{Ppv?r`wAxwDy_B3r!WoB-
zC^sC?<<!5s%cqAzgn?vH9&~A}WxR?$H9NmtERr?~Yqjk_q3qtffAcxsAFU_Iz3qvM
zL8Zom-)<Z@{0wpF#tbvd;c2DFm7WOrwRFX~7PoHI{XDW4)31)xTL1>kUUgl2)m1N1
ztxD@U0Gd3M2mN<`sZlt0G`OFKiPpchukOt^pWQR88^NsWO;dktA8N>@NnVokluU-@
z-F=D7Z0fO%q6-6EN@Mo8DDvvrGJ5V}`6SyMRE$0Y1!~k`HAXmN3fa`&JxW+xbQe4Z
z`qpt*_Kk;9+FIoviI2AFNxj@r9CWewHu1iLa0O{16nr`F-Gx&m^+Z<7Ys_HMe3j|3
zw`vfMzKD&d;j#Clz9$AWu2+OMe+@oxX<W(09{D?B5QHnR7}OqR-T#2bn=Q(bd83bk
z8?wK?eG{YG+1lZHeGux+RZ)?hc=tf!cDp<s3x2|4bd7dNDv3tUyGL_C2F~ez6Sf^X
zNC&M^64vOD!N#arz}BOG5u;|$LSbEnKh6(JLHX0>Vd<|gT-}&EG=!?`#-sJy?O)P>
zg_YlT0Nd_|<lzNY9^x?aXJ?B&VgSzk+R(@qDIz<uXX#@s>2h3~B<?*v=sg4HN2Cy*
zv?H{FM{5_bwVq1r9sFZSXiR&vK$LSQUIiK=nmIL~Mk6Ql<IrsFBON5>`^}26NDH(l
z4g5mfNiJRYqqh3gva)Y_tobLlTu;emARaUwS><X9Fj|@gBLI%5G;=|=24wku!CVg1
z53%-1auycxDFU@3&4XhPE3I|?e*5|L6290H-i9YEB^!L4ndBI6=v7GqM;7Y-+BRqU
zDCx4@6Gv>n<6{s#`8(fkirYcI@B`F-_=9S?14_#k=q^c|=eN+~2Y@pTXy;>=jFCk<
zmx0~6AGe=_y=&aSoxD9Y5_(HUcnj>?@3%Ogg!Ndy9C3vvnbphf0HSG2y6g&7JaW|S
zz1sI%*7qIE2j^Pwg%6Yx6%L6DaIdk)yyj*MD`x45DW1Yw&Nt@6W9^qE3lg+H!lfsQ
z0<@9BelLRK0ajn?X(w>bjiN0LHc(S11vD{@Dr&4Apg_bKkTsev>aHD=G$?{M4N|!*
z&EsC?gn8M-_RC_?%T(uI!pqM6>1CqmWnFq-eV|v24i)8|sTYi(mOCMUk!vDM!hR}*
z5oIqxv@0(S0Y5Ay#EpTTlURe%OCvXsA3Z=2P&BPQ4XT{mf=vh)w2_$OIn^~3x-fS)
zaSvd-B1e6Z@ZC%wmU8lOLEzrjANR|~kzd7%CNjXhpukTbeer?PIy7Lj?d+5ez|T<&
z)YD=4;$mim3|>0U;6)>Yw_5xHmL&7XlAe{KH<+Cqk-~J8oqQv<IvP91FlqV@a87-W
z`k%=%dXj*ZJlH`eQa<5!Tj;;?0eIFS1#^$1M5qmm4BG6biK2h1H6a)BZ3hZnv=p3i
z2lyIyfKSk#vbylgPE0?yR?2^f4QdYSqdl$DtMdW0(&+oXg1tgW9H9lK92zBq#SQ)t
zM;S^}n+&$ak;nVb77M>FgS$+6lxpzUk2mj^7^xkG5;se;_##s_;7R*HYX~cSrnrK$
z({mR0H4eSQUnWY1AYh1||Av8ki8+M!!+v^*KJ~qNxmXXg9@YV@<A3#O<bLvkd}AR&
zf()SRzHoPE174js>fUs#|HH0)g>+)%*-EEltRP;wSn=)D8a)D+1Ds>ngHAt*@eCH*
zo`D7_i9z~^X^$Ai^ceLl_Uiq|QKa-5&90*+5^I_uz0k+eDB9Iv0~hE-RrI!({w%$7
zkkIU({4#ikc1HZZOC?w9%Q{k6Fz&>ev98E%`-K>-)Y}hQ**nBAEer)-Zq+W|NsL_a
zDVDji7ypp{<}<f<c@o-@F#W-+eK?JV<30#p?r8hzA93ua_pFu9)>DZ-g$VisilA@u
zbBr^~_0n7b*Ihh}-4CzV$b7~>LCUH8hj2li7S6WxH3%29!&J!7Sy0Yty)%K1Zc-HV
zi+Ey16O}^lfvEv!5V?<_bl)_`!6G~#3P>*kuzpWkkL8PK_aJh8ZCHJwJOXw!P3H$3
zUg)0=59htV`atu!{`LMSxb0;cW0*002Lg&p+V;0*w~L~ZhoLej3*yxMB_+~YNiSfr
z-5-|7fKm@w;wtif@on;0O;hHFXm;T-l^Dow(w}5k0rs($J8!(rb@VY;8WHMUJ*U?n
zRp-8x-4FM1DkAs)SYa5Tju<E%+Ej=2Nl{fd5JG4tQUIY{t9he8GTcFWCvnaw80}_C
zi@&he9jgZ}h0;aX8G%6>h#5*X-%Cr3NFn6Mc0hCnqY8MBVu+1#>eb5ca@1j%+l0mS
z=Zwiw^hZTys97d}Wo5NR(SJ*vK8+kYJY(G9B@=<4P*2teSz-K>1+Qd-pi4aco6@q2
zIAw*{?V%{>qb|tS{!*8q5J8Rl^t$z=w!IS4d;kbq*z&Yi3QKN?E&*LEc#mCO9d>oN
zXbPrk(dwG(V0Gz#TAdME-Fww#wi+HL=#5Ru3U-+zS%%xV6Gy(K!qp)z!K;K=86<Nv
z`%OOgeHZ5m?WJPSf~%jgcF$cOx!Z+KGpxI<J^^B)_sSia)f0c1)q@e9EA{O!z1Oa4
z4ojgH@NDZk=yKxI5v(e<00hpx2obrU%WS)lW(u__baq=zaCSzjkH-QbCd<ak=8XbO
zOty%N$$~VIAs4xU%&gv@G@%gQI4U#0c0O!a3NhA0qC$p51vL)Bg#xsWzxe*sV!NcV
zZxrIq{gY!Ntn%|J1qpod>^Vo_{JOEuDrg8RoXR-iR71j99dyI5RSd5`%5r}?Ch#l!
z4B6JvwWkY~9T@k4`r>{<a}z66jP0l$aU;Zse*yrZzd@a?iPTCIq=4ziil?8$>N0Sv
zgWO}8-~jj$iIX_X;)JAyE5r=mM)Ie(twY{CY<0R*^u|kaX!$9vR-1N$Hdfh*%@@5}
z%?aXEFxzz=VtTWceb`W@-VTDMSx8`#0|VKVEEkea-<j^V9tu{czIM3#3X!eZUS>#x
zNwB`8+Tl6NjI}?gWHcR4Y9FEKhBU+iz9ZmFg_wKa6m%~kFFy{+LXci<);Aw0chR)!
zdu$*t4Sj9bf5CTgb_cpAo8oY<_<+pBWc@=F&SCKX*mKDJMUsU2p3?k={K8h@!ppki
zN8l^uuwS9F#rx#@vB$5%^v*y6jfv0E3m+rnE_iZ~KGlWqPsZpK?7x&kWnui+`Byj7
z-W?-dUWYLurU`4A^=>FHA1ndSNxb|t@rMFdiDJeeCIPbRE%LZco#!@h9q9U_^3E4;
zQA7169eW%}TQ3$*wNtzw22;q73mt-_k;EOi5gBXmf8kFOr%~o=no;Gb7wje;bf;j1
zBKp^P1SJce!#3d8aoiaI?i=HWs2~b)@Tx&Z_IXeoYfc=o1=CcBvd&iXfK`9Z^m}5N
zE>8M~k!H=G#|qsuX3(tHOVI)lnImZ9x1h3Pgo>uZ0Eul4*v!Xr=sQ1>7fC7HLfN*L
zIm#2;PTGrK{m7rPl$ZwG1sT206Ay~irZE>&ZKa#`{ZRrLzY&-jFnhs}Q1K%A&880T
z@UEvn&$fxkmL+}A21Ssz!x%Ca@b#}x1pHu_MxSOYjr}wpJ<x~Q$70JX&BC3wE{W5p
zC4Q>qJ2+#mail;@y~Cf{UxY+F@uu3;Z5_CifN&eSe6(Gf?r)+-0Y*iboav7u<}k-W
zTs8wQI9Su6BE^k&%tIO#xwlhog@T=Lfs0Cv{e3c2xEL}88}E!JK)h4vK2FcQ@o>oe
zkVl``*7|$}jB8Z4@D^iB5QCIp^4K)0s;-Pk!B`UsOm<$VigMmsfZ(4Tjqg|zc0!*k
zSSPrFe$gd9wGB3=Y>z!RTOl|%n28nDv2Ri3@S$&%DKH_D;}!tp8`OZch|L~fp_uzI
zZ58rVUz~&AH~<10suUR;&y?xPZSwWSe&`F)Eb$xLj}eQ5kn+?EWfY2cF09m7i7#cM
zu4?@Ls5FWS><&xDWMV~?&11ZdYW7Y)OnKV95U|jyHuTBaKYi$&$_g2BH@{EmG#C&;
zoa^Sp$ZI3W&BD=;^mqfM74z4!D8l4%hiG9oB3Jtb`2!x9CCUJ&ohP@xJ)~s{NP%F4
ztG9N5g67b?!zQ3AeRTCQP=))9HH~qbMj`GWZE_c;5^P$etT36r9t!jpZ>`dER}(MA
zMN44pholpV^KKf7Ul$K@U)r_@e?I8f43m$lD5vF~r+d(WKG0&F8PbQ<{WWck`O($5
zN*)X#fhB5dpszid)mX=f@!<dnI4m{R!ozfmNnXNbJJ%V9n)cE5KpVY;O`V2A@OY|e
zxrX#R=<%z6s2le2AGQmH|BsJHUr#+Pmn7-dg~H<2-_p?NuTc_m_m6_77;?uGrhdaT
znqwyZn0Q6`7@Wj{`3v$e^7v>mti9cdXAhTo`zPgu)311>9X!!a;eWC_Q00Mh9{emZ
z<+m1KM2f)WKieGMQn=%Uz*C-eybYc1n?IlKTrTA#hrYrBMj6IYAgCv6%UAk&d!xNh
zp<0A`h3jXicuY3nZrl=MvEt(txoZ0fhRfQ?`=f%)n^Sxt=o7;ReR3OB$3a(ba<pG5
z2_MDiLbDVyu71O&?$3j_#pK2T$e+`_SDvjXuXe=ZQa)u_tVYg@fJQdl70WzXF2+HX
zq=e2^>u=9iB_qL=zu_n{`iVX$Ct|2=SZC(i{?uxW`y!h^0!NT7yRC4jbOF9$#2ysR
zh<NGaCed=-)p$k=$hXEvu_?#6gz=c{as7*bU?|2gVsacl+4(1Lx1Yrx2nnAv|5qOn
zf1I#_eHMY-2{@5owt;Rv<!`50V}Y{`8_!2)`|97E?JOE8MJ;x#d*z$wP_GPIq+Mfy
z%XL2i%Tr_+6}*qMLHc&Bd<NV0GGXz5J`qRQGJV*xmLvy^I=uVseK6{9()Vu!swLKs
zg5eX)C=BtqtuEZ2sie2)zD|omL`h;W@QqK<sE-c>{P8m}=edE}><en>@qaHnEH!wn
z33v6H2#poUpo~SXzW+J9+vVE#`z|V+S9=7&bmd|rW#nKmR_Y1M?+XH?FXNZ19{HWi
zFnN6!gZm>H3@?RFg7sBP@u28<F<c?O7OqH;!mM5Kpouy#`DlXz^ihQ$qmQ*1eT15B
z*4x{^rRX38j~3UX-xN*IXeuP-+(tDlxb+&Asz~Cqida#eH@EzV&BdWh8zT=Ustq8@
zvLMRxs_96SYf*}pVn{(FXRdKo3SBgK$wTE%wb1pQeJ?$?pgab|$%};K1-WX5sn}@N
ziZZ$K6i!D>F|nXhwcK~um&pXszue>&q<?xy|5`F0Jz?QLo_qvzPZAGT9Gto(-6B?g
zA$0K!n4iJ5hYzPl4x?R%Ya(Hy!*-Abl||$|<7U*MsX(=WE&f`I8JIz3E)-Rnw||rl
zprvkkh7puT5ZK@o`j(M`vNgVqewd%#g$<xOu0>vWOT2m4)%V*qb7&PowzP5iG>9t~
zh-<ES4ieWuB(D6o2AJixk)xE&ARv$rJ(>S4B6di|4C8E+&w_~~p-v^#6=3>(9Ry?h
zfa-~kY-|neryH`LIxCp-!sHjU0=J*Xc}w%NV01SJbT_^xRC}bh!6r5btIsi@Tm(;A
z@12MsU)<4h0$V=zs)Qm9(yoOYL`xstp}2T8y&I#sgGh79A_^|3RJ^7=fSm7DEx0{v
z9I8p3A2xk=xf5fAt}2(gCf=2vI2&5)<Lqf3iW7lBOv<dHxSE9zviksJSECmrySGtr
z<gK8?l4N+9>=BAkIl1AAXKZe$)yX!en;jcQaKZ)U)eo;1V`_$_{($X_xsCJeZegI2
zTcDAu$4!t%7NS&Dt>V~;faZ)uZ%kP}7-|eDb}Kd$MZ#K^K@OOoQ9($PY|!9be`Ed?
zV@xN`IwB+o6lVsO-1-nDv>S<Q2u57tBmWEHs$XfKMDGLQ%3p&KSH}N}xSqV3!R99o
z;`(|BBd(<X6>$v}Zl|a^3gX&ggAvzz{}pkqVVrzqbO6Nl))q!w%aD2bkHu9L#Pv2Y
zh--c+64!O)CjVpG5=}d)39L!wXhh%U^A7<SksyXzqm|9tH*X-Uv=%~L^uYE5V@y%t
zA(Zgm{8x79$~`fTDo$<4se0AUXRv)}`>*Vf!rrCx=?ABUaiP7+ube|rc`?C%xl%mD
z{AK-mcy5n$z&5~c%sN=Nq4dl9#rR~o&UNr&tlL|#ZZ}ddWp12e;%x0|KtBVU?Tqus
z9SY=2=d;lr0NiW6`Du3gF4W42gyvmpqytvI@ZEg>grzn)zX3T4$rW_ztRKtlZ0e_W
zKg$T?4K=|cS8Ih`Te~5C^=*DE&M__GBDrh=sM)A|i^v2?JoH4R%|wS~49{%zB8?oa
zmSY9xg0B&>?BG1U2M%B9Q<W_V(WR|d0K`|pZx6eT`>iNP_c7ogP7;H$?v$7tW0!lM
z#`_x7X#^FDgOQ`TAf;b;p|QT4P5>FXAT=;@Iq@7hM#H8LY(x)Yes&3r+;xJnD|f#7
z`~mL`tthCjq@rkwu^;os$3YM)UpFrvqLC|<q+M;44)CV#%|Lb*@TP}N<GB@HF`Sm=
z3hMfR+1=}~yR2Lnq6>}0o7o(FSmE>GW#)iLhkVC5QjA#^Gy?rTO&y<BS8$<`ps_ym
zL;zpTPj5`(5l_V<%Vl9kGW9}O>s9w&Our~A1zeA7o?UHbn13byK@^uuD|2CO9EBD(
zS3GnP%w{ILy<4Spz=>zxC3er_UOf=Jg<J=FD!2}_i&spmzBZ5yIq+b976z{43rS<b
z^b7Ju;$B0X>(Iivj(TkPSF6zp^?vE3iPI+n?7a9E;-9dPaTgh)(zT=+oWHzw_&HHy
zKQ;FJdl--W1+pA@wP)h?Rj!FjKlg6W9{_wb^2qn8G0x7avQY;0A?QYfdn1i|byAnb
z%(R~jQ%scS0?*OeNZHA2n{)0R1Ljgr{up?Uth)<IqQg>dckgJR&xie2!+DMzOjPA`
zIs}}N(dF^gr%zaP50kk;VO)yTsvZ5VvKBbx#*jDleqpgx)l&wTpPd4S96%MP)Wgd(
zPEk~WyT3zC)W{(N<KF)jQ5}WWZB{g&%As(vZTNRQyvA=-`IHB&-R=A)0|biw+|An!
ztdOB5Z9e4c+c-xtKR}FGPb#170Wl`(G{kIe5QuU7@r(VDU?g64@5IRlEyhSR|IA@*
zd$z8_dm=5$LnDLDs~v0EX8kq)p93JXZn^LL1`_IaUJEObp`Y7{bJJTdTK=D%e`Vx?
z?ye<UX_^bs`alp$iE~kZeI_zV|Laid*VbeY;rfsB8(Xvc<L-HSE-)7#b+$V{-mqVJ
zpFgw0`5$o=Oa;4yC4arTkly%QZ*PB}r>JC~=}1EgTd}QOP}+~^<?**>5w*h+lg3%;
zXBh9i|8guaY~)8o<aLR#voSOHM3YA{5@z<UUhsOC_Rw>|F{CVto~leCfZmk9`+lG%
z8CNj#5zifXL_*w>bhTsST*v2g8c1SUj^OW;+_968z?58A!;JNM*pf1FBFiKAy~FU!
zZfueU<X&QKcj{$HzS>LU$5mzHFa1c|VHxm4GNu36%dh4dNtDDp;SptKJ3CDk)%6X-
z5`5JTvJzuo#V+ac`GY*yk(wMIc}Z3|%&YMCq26N`qRFxM_rz5U(zuupi$$?!u*niw
z2eJ@qkqP!+SF4bZzkvAwMbdg1+rm|nnllzT<?M1qm{*<M*(tSGO^aCbjo4K@PB~_@
zs>7a=z~R*{1uZ2E3(xUvb@%Tan?3F&Cd}y>!6IC5UO>#2O{>N!eQpOl!dlByAiSwX
zU1QiRmP$tM9OhMfN$MGG0CwZ^si{{7l;sYfSHUl@DteYN;<>C&$M{pen0QQd!QX^<
z^2_C&-7H_c)G8E&!dfy^I1Zo>@NB}%X?U4dQhrK_E|o|A0QObNo=eZ}1*@G4F(VK5
zln=nXis*1Xuu3fLYj3(k=oVL|qkdk}`OHNFB|nnBV{&_QXf4aXy!8&2e~$Tpm((%7
z!*+{QAx@p94=xAn!@TMqUXBm_+nfVgms}F@uLhh)tCHN|RYGw%!>25&M0E5h!Ehln
zUgMx9o{P=Z9fEQ;F)h{<ipKnJuRgwywyQ5^hXr3kN_xiECK;}$0rxSlqNj`z`(Y3t
z{Om@j6l+WX3wjm&@<QYeFERY3g`Tra96W7Q`=2q^9cT}X#Ve2t*p=!XVSOsaFkXcP
zvt0=Urm@1hZVzrm31}Gxv0z^nvt#NAAy?}C<4ug5JA+yFqix+M`IpH4C9?mwMK<M;
zoWwR3z+V>kFN^z^#r-Qt_$x^GD@gdS$F)DmCZXd~_bcGJ^gaiU>z*5xzI1I30$Kn;
zs3aievv<nzy#t-KZi8Z?g!o9|Ypumg5KON2T@U%s;<LHkOLi2G%c3mJm%o=3q$R3F
zQb61%{$Ggu4B6BE3R3^dtNzNX{{KF&@@j~cjR2DAS6oAk{N_Oon<9~@i{~`Zmn#b1
zd@tTcW$|eB8bKOYvT7}=J&-FJ|0n7EKK9Eo=2O3thZ4>^2Pp3ED`X}}AY|I#9=Izq
zQ1d@Xy?5fvQzmMOmYA<_JdcXIf?tq!;ltPmS@>efk^jgH@9;W6klgB*uH`Nt$2LZ}
zGmT5d+WF^xsBF{=hYkG)RT#;b^IplhgDD}=2gc+?q&+7dQ@&($J#V12*Y2+^{%ec>
zlH$J%1*RwnmAbzS1-4}Q|5ICBH1!K+sap~C$AghuKR!Pn21xN4p&kIV>)68}k~28f
z9utnIqkc6-j|d9TOin;L{nf@m-4V*VRB7~C-l(iw9||?-)vQjV2XB0-j~ebo6<kRs
z;(EsM*lI-%&Z$?F^x~MhG+GcR`?Guo{oL0-wH;f`uN$oN7hW@Pm$p&>7tm0|rVd^p
zgR2PY<|OKSHKkC^cJ!iGkwwmjdxQ~Z@9kcU*JL0SGc3oq1%xFBog6^hkJx`y#;%Bg
z4pTKgtjy|FG<9TJQ#Fa2Fp6ecldSt0B~GGtcv+JCu&BCo+Fx_&lf5)G>z&qqNL%cs
zQQGp^#byXx=KFF~S+*W}#*Zy+Q-1D}nYehIgaK;YY%=9|n0hlI!CanC^?3RBv*lXX
zohI8ehXD*}tl)T#kH|5jU6@}%rRII`CBM!&=ov7tQ|lczCI}eF|4DN7++kAuC48Q&
zl3L6gRoi_Rg~HxlR{DXd{l|FV&*e3&%03KuL;O=pEqre@=1!x48SXTK!;_80&B#@=
z^*&SKisa4uF(;-R-AU8iGOMlnTauNa@ZK!j8_1<3F;tfJ^Vpe<tNqnSZel9xZXD0@
zeOp5{C`FEyk!$iEpvG*N5caktMH#LNR1;5qCfP2(2TSU9&f+gMPh_xWBxo(lp-oDx
z`NA*T&=I#|r)2=8=Y<2<N<9lvz05^P!^_5S855SIu^&@sFNeDVP95RB*L|0s?Fv3m
zy<L9~!Lv83qG-R8UJ+OpW!}ODA;oy?3AQ$RX_G)a`~o?Z9n5mrlYDkql8$@l1L-iR
z?%#NGO&W6`B&79>$<{+c&-mds=IQ~>?K=o9lIC~@B?G16y|4wzs?jKD^Dlg<V@nyS
zK(Y@mICef{@xtEN+}&Z3gHK6)18z7%!X8s$K8*9Yq)TFX@StQX+(;?QUB|2egwLEq
zIn?2dm)N4n8Dikfc0)H1;^Q_pwU_M#I3(echv0AkK6b~5K?e>fYXoU}3TPQ~s$o7X
zF=NNnD-w0TU7E3CKO|Y7W7YuMlNi6fxj$L{+T8y$n+sD@KJNg6gzJjH&<M;hjtfcP
zFWFsE8zUnE;aJ?aU8e9=2*>MYF;n@u8>zw$uX7;5RMj1K9yQ2Jyyr|PL*|hi?mk<x
zwBD<+cS`BT8p~o;3H~9TU`|hcU;C9Y^C=-)%;U#*gEaTmx}2)X<Pu1F1@~&dFF<1s
z;0;#!3sM}rlz3k6^L-b*d%MX_0m%MqmqTn-*hk{y7no3$^;uNmb!U&icuMx$7;2IH
zC{8<^k)Y?Z%S-GJ)1m37n<+;|0GjywK6LxrK!2`E#+d8=Qun-n-0twtdra$P2uzpZ
z)`#Rje#f5j^P090(ivK`Em)nIi~4Fa`B=f6T@oldw=adUdXBwF#pvemKa_WQ6_q{!
zGjXq)oYJc-YKII#FVejOj;kDmv=Mv+j;l#>ALzDCE#_+PpZt;d4DJxAlAkgfKPZc0
z9AKLaFN*@z$C%Xr`Q?}oss||jr3GzYw$RYBsOzD6++YRO=O$?|C;G4d@FIab;VZAI
zM-A2BzWIoIc^dJgoTTA?oMru&FW!J*W2Mp-M;j;jKMrFpDtIo=taa`<>-kTXh>H=B
zQTc1Jf1T{#3)g=p#!v!2bjC5`5TB{?1Nn1myP0{mBd?bqfA&|@KHPs)q}PwEk$`AY
zZn^);zf(sftE=Y@TNx<n_bGqQ0|m=rt<J^s5_{L)j%l*^m+}pU)EOyO`s$>E8POz%
z?8~w!8yz{Sf31wze}$R!$=k_=oT1Q}P@QxD{<zJm`oH<ymj@1KR3(ik>lhPHawN#Z
zYpa5<{%^F=S{^z$5|UGF!oYC6<b*C6dUX})f3raJv1Ndv;_@;%t@;M^(%2;5|7($d
zm^k>`|EZJonjNf)Af`Ab6l`)zo52h6moQ*1%M>}3ET;&*QBGQA&;LdUQ4E0-{W!iq
z9PPjU8~=Xr?Z>F(c()z?ztfX(R&YJV<AIHTZ$ti>gGN7VKJQr;6@HcF)88D>zgSRW
zD7fRrWF`4~za9Pm^la=mC{utgn5#Ue-1YB19=AxK;)9eWMVy1|ADwNzARaIjy*wo9
z@HWgmIqB<xi^mPsw9Mj5I7A8lte9Toqy^Ts1Lqi1^qUrI`U#m_4eV}>eg_18i{Z#S
zdfC8;6NXUUje(7dl*%!tsvfA;3}yW2&7$0vG_wH?Hyn_oxP!z`SMF=Cxctvn*!QyR
z1+URpAl5V%2dD+(UVqVPA=|Q@gnT1Tf4ese@?{ThQq#<}r>su2ZTQg%enTkQrv<yM
z@6Db$S;+$srC}%s?z!{5=MwKHO5?*-^?-$J`b~W8mhJSrk#JxZ<zLDLPLBu(3z>mQ
zM;`?kOgn>^{&7pKl??!RQ_%m++y(ZL!lYjSwH^Tg?k93y9{QV8jp@QWp_5(ktje-Y
zyL5U6pe{~HdGFyB>~%Ccb{EK^KI+)k@*4x3%IIk4dP;i9saRQm0U%M38C0GB@Te>*
zrgE#7qJ3V1wD+GyVnWO@eUUKE$`aC5#pKa_ym+B8bLjZr%2**sUpfC1Qm0Ad$i;~g
zJ@=Ix*#N$OP*>@u765_O0ye)1wZ7|ZTL6&~ylX8ESVsrz;G=HvT~}|tIS=GeU+Dp3
zN=9rkCmdW?$dL@TjtBAoOrNNzHEv~o`ugz^g^{o;3@U#Gx)yrvZ`eA(_NK%{Z0|>_
zro{9F>_5-=yRoE%D04~=V9^tVdDk4kbaD!(2ju2xfa<lsq+;!V*E35mf(fYNhWqZ3
z3jpwX!e#QaOdjjEZJCzHyu{R)*X0eA+V2T(zva)qgAo5jK_6Z=qMfrT1W4vEpkU$3
zdHQ$9M1%V*y=GmBsd-?{5IidL)RcM!$be20(pyW~_Lj(rs4aBBc)96<rnR}f1uzUo
z(}_KeEsFw*g4*K+eRpNNr2qYqElBnO_acActf0-0&y|69HM3);fY-r%9x!OafcCTk
z5%?f|sVNkmI0CgHk{@E<7miR#%U~DfwMDWgZUcLyZ`VEuJu|v>SRfr9aI|0;_;+p{
zn+Mv9Fd+O%7iMzJhC$zOL^V|K8BJpwhnv0m0$&c~AfFKh%BuyUN#fglCPb6J6R^LP
z^cWgEI*Xj=hs2z7fIal)SC~iacyc^L2`Wt#JHH9KsWebhVE%^}9)YO@<wH4@M|i~|
zm5O4ESJC~DaTJ=wKSZ!{Nqpr;?TrtDTflomTfCC!6NQ+6jRA{=2e4S4HaIxFaa-!B
zLQHM16o~)LH+A75A<;BXihnxd1X&+l^v02T6TFvJt8dFWPGF500a3j>`;O-<Dg@3>
zPg9{{Hw?^HH;1j>gc+ZjDcy3nEoq@6_{E-KlAQF$JXikBP=z<yX;~&WJ5yaemvi*B
zhwxm&DT~hjS&neBHJVojJXJ<%fFNFJIhL8wwK_@ewRxb=wMl<b!3-Pju%Gi>lG^_`
z?)kC$6r)L`)c~4nAM3*yQ3~5wQ?%51(;;u64T^Cu5b#|N)4zM?n=)VmFh>B$!zYMw
zhebl&=X!;Q{R`GE)Dz<6Fs;S}xafS;V>ANN3%@X6usnxHdZHl435?shn(CLAt;EK2
z*@X7DQ#Akj?t<sCY4SnzpJ<HN3F3926y?A9Hq(0rIvT=VpzwmpAla2Y=M}tOusp!{
zB*|#l<k7&#(}uh*ZcoL4G0E33^Bz8o+CK~vx_!<1a?%W#{_o)k|09R|-!0L*jLxda
za5ow4bv<D8sh>uY$pZ$SfXKRp15Og_wMI|x>?C(n>lb6Ur1Ik>piYHjc-43kpwWiG
zY7ldC*#et|AFmI+%>`2${H6Os;ivJJ_&6N@)>Gg-W?mT_jX5w&U6}{q#g7ibf*B(K
z*>f6+*9y3fn+W3XW1~fGA5f|0!>G`uqh*OLdHQaPs)LooV9L!)fJvLKXSBIA=}maD
zYF6#gc&E;aGdPA?==-oN*_g75fVPm<OQ&o3YVCK=4c{Xd=kb@=EO45y6|4oW(3`_A
zZNJH^k{-|ST<L5p5qK6S+S@u`BqJQ{I6xe0a}prmje$)};Pf{kqtmIS@Vwz(ml=NW
zuWl7tD`zVOdpKtRFVZB+bxv-6eZEd`F5S7Q6tQa_vh8%{5A}a{YF%P$-PRW91;UoU
zdwhQj@+YIaC7S>!H=jn$t`XpkA8h*`J$|(2YURosx+>?8tjxVw6_O@#g@p`9xlHo{
zJ?aBUZmf@+<<2a)tphyN3qbO(cndUp!@#O&e5?2jDPT@;&ksB~DX{4__l3<kMAMiT
z%0KC09-AwKIpo1+3!@Fbd1KVtw+xK|i`{ss`Q|7HM)PRcbs#1_g^=d5Q`KeG&We6>
zscrna;PE|CEGaL!QwGo*%yqOAr&6QT-bA~vj*c_bj(jmX>CZrw_H?l7lYqG_)yDT=
zdxbPTyRBJ&4?aPCpmd@2n5ZRSC>dKg$ayp);pqf5@4~uBS=|nbe++dX=7=Xn#<3T;
zPq>wqKL%{Nfo)p$-gD@{+6ZCg$GN3`A0}r6shtm`CVWkufFW-*!OH#9>aL3{qf?yV
zch`CT{1#X1MiHR7i5&S(Z<S^=e+Hrrr-Ufy<dN?m_gXf(dVo7~0c3G<F^PTbc~igx
zHjXevdl47=@M>hqyY;!6VxR@6in#We<JuUP|LLzh_eQ?QEF$tW#lUID4a6+i3n5V8
zXk5|f_N^nIL;+8%7^q@$p6g2#B*bpm#TJ1<eKA7#(rJ@i_0b#7Crru%kxs*5y?`{{
zd3iH8tENil&?d6H#T(1tzc+oqxADFZ$myig^1E#=4hrQ(6Y95VTZt*zhX_zPl{B~)
z-hP`9>{EwTJ$9)w^6N{$(qmu+cxhIyY-=C-U-kSt+tsV&K1X2%9_yyNN!QGEiDw_X
zKqhT-eb)~kshbI|hyZ*Xqp{9~V47UBI#@u+)U>kkmgpcM(}U``g%gFNuZNQna8GuT
zgFfe0d!|wG>n`$L?~Kiif0bG>y8kk?0v^4E^|FKzV|P#gJAi=b?EX{9Za!q*X}&$h
zNS1(O;@9m=-_ERSN0~-D7ZItf?#;CibD2h8JYF2hQT(~8Q)SQ?w0uxSYg1V#^UP@X
zGZW6iLg|&nYI{%SColeK(9f>ghk(N-y(5dC_fEIW1iuN(hwQ`=!EF_zKrbZfu9=pC
zd!1-6&`XYU{9@zitK7Y0S{$e-@}-Ha)?uZQOg7bujbLIAllYlgAAeIV1d%T*p=`=j
z=v53TOnJ4TJ)!QNm+O7Eeq`5PdzC&|={vvP>ve8r*5%aPx}Nj7np^IZDR|LYmwMRK
zZ!H+9@m!FX;8`iI;=7xTteu7?<;y%kcXf8Jlgjh}Yw~_X;z%cZ)I^m|!{#dQS&2C0
z_@dksv#aj8<$8%F2QoOcA6a^*TUWALjPJO4rSg%F%H<Kd2*-}9!XEN;Y)oe-C27(6
z{HUxm-K~tXZ+mF>Oun0|<|+n!_N=Kpb8|H(XNvm>o34j7E2@ZcawvbE>CYM}QPZ~I
zubYX_8ai?H7{L%ru$%>=8!XLC{VsZ;#x2ppu<KA{$?kOZoY<na<Gz<<Hs@Ug795~u
z?tHe|%7P2=;(F8-W{w9Y^FJ~}tNd*AD`y97==gPF$814hZcPt%F`A%=*_6%6@0Bv%
z<|XK^kk-$~pVx?=*n#JAm%qa0pZPHre9F%Q<k?N@2|M<=+^a7cm+R;gSt^#eA9!bD
zvCReQzmF}9@>_1EltncuN%S-|rF^iDouO)vWz)XX>on557?4Nt;(KIi{^sh$ot!Cv
zj*0Hj`oOdKb7$=WRl6OqsX0UFmwE@plh5*D-nd2OjAGB)XsrLK8Lpbg?)h7<dPCQb
z)E9xdE?%+AkkFAD{9@{1de1m2RTl~tmO*VyC2u>PJgbe6V2KzZC)m$?koVg8Ylk2A
z99wk;TjG?)WGlVK<j`&yCGk{kQiG_IBcqCGwH&74hE&=}EfGy8+mbtf{k+1@9qelw
z<kL@<v4<GM5i`T*#s)cDH+Shx3xhhKI%&In_|e^KjE;R?&68Y?>SA`gOKV!*h|PYa
z5c6P<krA~S`1q(7af{cBIBX!WKwh`p!KUiCDcBGXfmF5hrB$i)NWFA+ThL88&hCpR
zi+^zEHmkgHP0y<4>_w!iKh^x@=Pr(alFeH^BQo^GDBYhS_!U^`Ap!Foy{_66AJ)Z?
zL+hanJWuc@?@r)<pE!F+0ZvqHh0fGKsiBSggR9z<2qek|Q$HF1HW0-GVliBJkeFQ?
z-JNThrVnWo_i}2KiB^wievLN4(jLY$1ygyYK%7P^Wz~_hKsmcyvMe?O)YHNF5Dg1@
z=VwI|=u_Q_Wn$)gdd2$P-CNC~&)hvP6gH&U+Xr6l3CWLW9$MjRy3};ZBR*Jdu35p=
zPKBoyP}iGvPQHkSojhH*>N%_`5K)5XqZKSA0I`F!t*%AWiHrp8p_YAyd3&5y`@TRL
z_Hxt)pfbH%$t-Ey(&yf}ehuzVXR3Ow*x<An6mRt9gnFHpvxvS&o5MlDHO-%k;4>B9
zX7`mV)(L*Fs5#SmZ>?Vp>!xmyKBeVPUJ;{PKS5TaZh(N)D93AgemQlzEOrM9;0HMi
zJhk*_QB`N9*8U*UV9%fKxim^Sb@r6mdlYjuM=5H_)?9TE9Ws;H^CSGzbXy5$pD-=U
z)`b36=Ni0rlCxHw>_bE|X@K{|D=ue*uqQTDiaoIm#rYET7{vvvvs1GUSE7s(Rl|2d
zVQ;XI^+jE>F5$j*ink2CNB>F6ZYLenL^lKsT55wyR~Mh2c3=PXk*+Nj&xMcA(|huA
z+bnP&o}C=N@zeN?_w`0!4}vfF32f1>jTdP{<hmerXs+c8CQPM!*?gl2kMizX=FHSj
zsSk_yX29}#I6vXe)3EIVan8T7&o|En>|jxtad^_Zk{*A{g__ida!~@h#SjjCN&1pZ
z__D7l#>`8shcn%|nK<O^%wmphikuSBO`^0>0g^tEp`a__TMKq;$UPqPEB|DacHG0>
ze`}#hNpN%b;&r=bUZM1a`^%udGBaWyR)rwKrEpQyf|cD(VJ(?ia5bXfnAgZ(>%-Pn
zy)20n8B%xnBAd`)e2YqmZt{rC_g4^$KD+}IQjU;b<Rt8v?^||QDDLc(YXT=pL;Cje
zNr}{B1ipk_{tgo3a=ao*weg1Dd~uBkqBC3~cna{#EfyS4Lt5_Bs!AU8fkSD{@NnDl
z4{ltJvSRb?HWR8+!EY2&$7Q5m?M^u3G&z(RTNG-)UeGB+);c2DJ3==3#dc3_wMlPT
zR92eyE%6jb21x_|R0h3&R+%k4_H0DPiN{v1LNAwOlGt(?s0}_vx#kLfT^y_&^p{}N
z5nI><YTTQn!-YMH=~onmZyYsc&pB%#BghztXX(9{fmVZSQ|+wGhYesk=AI2^zKD42
z>jfOvsMIu$WEq!qKt_5Zs08msrvk}l=E?=q$uSux`#tzR{MjoiJJdRZns54hn>s?|
z*(5NpBQjjrpKjP6;GL%|W_r*Ub}T<W!Mi<c@rFJkn)JS1<GW5i7iCp_rnPCV#$?-r
z^?{r6cT4Y_QO~ug0~}Kg0ZR7hR&Ae>nm}jCb)asg?+SGezZ8LDye?y~v<h8u4H3S0
z-nWGBrjs<b`Y$@D&?y#|y#+eEUSNj*Q4B6{DYo)v?~)YHb8unf3Vbi_<VL=(__TXD
zb~$zY29!Je(rCl$0;?<;MXPTK`Z0xeCcYw}FS`W$K6+xFUP{2X$V&_e^RCW<g=2JS
z7Ft+^nvn}!N1C3m8n1k$X#DuR!g#M@O&#E=R~9^nXuVV_o+TxX*cB=bQ6-<L$iytW
zh1#-|{{Dum@HkIB-H@DaCd@rv<_&bNYiN3JRkC8N{QxnC+6C$gc2i~80{CKHkD5})
zZmar<4aDGSBf5Big1++N+P70l@7>6ZD0@P^)s#IOBoju-$!N1Kec&>K-BPi-P<;g)
zq-F5(BetlA9f^Cp6o6mT;g)DHTWSJR#2eBoo&5D10eq8Q-VKhBXdc%bm;l_g2n|B3
zqFI@NRR_b~c?Hd?!de<!fU%$=S;{Zd^(0(f+246dE@JPQpJg8{a*N#){T1cB4o!Kl
z3j_P+wYB<v@21|wEfvnif6ovJo6BRmyY|_(_O*d6!3n~NH@en-OHRNw{va>2i=0m2
z_>1`VHIV+{%-B75ARLVMrA_(xlJqeNlKzh4BgB(qmG&2@(svO}X3F~RN-rGC>@bdC
zbr4PooWGS_+d8`1?w6V<QE8xFuQdWSAKRF?>kX_sr3IN>e|~s1od}@sH$xvw{=L#T
zA4XeW;Y04<Q}B(pJ$7-@*K*Anpk|F#5bwgZE<`#OmL~>?-bP)By^}j1Vbw`Z<WKal
zy{LDcuchg14Y?xNPbe}~y0w{^gJ@Xu$)*ASSb??+RB%kqX>sA*Vn#Y5MLrgC$oR#P
zyeUY@IvliV{N7ovkaCR?=0g#!AayWKF7VoBa>XMGwl|g$%dS^iGo!*a#j#BvF&FV@
z#ar_p+)Z!O_i*c?Po%gT*UKKP*vJz`a>jqt_#IHrzL{oAnmXh7v#;4=p&uYbXt~-D
z>q%ahwEE{RccM3eBjM7eL%S!WX87aG-plP=hXBCNRGXlc(1k-@{-_e;d#6eEx*idC
z9kbYDEk;o{qzRUO2z%dhpNAUFU}5~@ww?G(?Xhpl(`jXGfNFPd<il9Dg}!wl+8q#E
zYB!3iyu0T_n@eQb6US32<vQ0>N%6)R<QqCgPOwGX%4)wYW1vLC>7|Rw-4}vAEkh7<
zVsEgzUdO>*I=Wusu&)*#XD&)LwzoSEtN`PK#me=ZLKa|yk0^#rw2n^9+0H21%xvxu
z*1JX$?OBjnfor*IZLylWTo8zSdmj8{;p5E(_lvC?GFFnDkvV#f0nWcJ&j@L5O3#~)
zWaBI4q{uEFwrC<PvCR0eJ836%VJKpC6PVU!or={SQgD8KwfNKZFi3yHG&%%3O=I&S
z3a!7|*XV1=d2At^b8T6&tyfVAcdH_MC4TkYOX@l{H3NKJay>FkA@*<cdw999)roq7
zodlIKrB@ckBJ3Y8nHn~QLnciT_^W5eeWb%Mi_*2}ga+%Fi_k?+zt&8r%=E&oma(Ax
zjeXrpQ8z=Nj(9Z4V`Fc^<<nW0Xm=~L(bH;8K0W>IXY=6o_Z)CkZj*_7dW75q#)}?X
zG&|UbFV4TlbD0Ax<?n>&Qh2*3z;3f)<$6q8QC_!`9*k4RwtlkEa~@+o=~y9YR6OA?
zB{9w(U=QU);>hkPnr5gM4=Dh4{iEo3U#jn)UT1R25-bqBiO^lrxaiE9QPoh*+J!OB
z!4%3g9{`3%(YF2Z=yF=r!M0B&cipoTQBo=<cAE$ONuabYmL?Hpq(^&3H~G>N(kRdO
zny&)D=Zx8&g|}OqfF7MIU%ghkwN~0}w^wvBBcy8%+ef!Z5MUW@yz;6iWD=@CX6~UK
zGjHGS6j+VQ>S}L$>m^3(@Ws+dKYLpJ!#fAMwL%D=yO=-e(P){zYF~ADUzF|~ZWGhQ
zaHqTBl)iJ~rQ+ox2>cml-e?c0l8auJAkz)u`|6gnukblt53mtC1Y%8J@B)iljzVX(
z9mFDjEgv9IY%*W0oGTiM$nslk{yKLAD{i_6>cw80zJC)~#z=N37`z<b<poL2#<(MP
z3x_pR#-N%s_f2@@<;PdH!EG=W4=CLZZ*@o<#_Xy=`KA+^Zc-BrgR3<tzD_1dKAj)9
z965a1Y<Zx)!M#Qtsa*Bzhe?TLQD>^2)p|sMz1BgVCT73GSTy|$PWC?Xq#Gds<310h
zocuO|5KYiYIJZE7{!DFmFOYy#Ze>=vLjAq=#*!}+{`K{Ta-v<XZs66PdhG0eIYNdP
zNkk9j_ev(`ny>Jcl~x9aN}uoV>L#D$RoA&KeM^`Wmqj0r@a%Mjv@->voS(Mn`_97)
zh2~IUs8#qf{bJ~<^vLS?Wpp_L-rR&Z;z_2-QwO&mUULWbBZ1DKO^V4%ra;jJC%a0I
z!fFKdQ8{CHc=b7|_J&Wjwfj6LsLq!-_I~!{CUf+<i35~xPnlEve~i6#RFrGmH@pSG
zK*Ru4kT8%oNCgH^1Vl<YhLUcSlo~`-R0Nb1kS@s~hh{(&q+v)I38h=Q;XN+hd;8qq
zdw=hG{@Kg57YsAkoaYh0IF6!7WJ7(ER2ISD2J*DU=l26!#e!6x_sF-a81o?!J`XF5
zT=0=;-gn^avCUqpoNJh$a^_luCS;;??^-lduhI6vqabxEZaBsrubEDt&<=CLV*>e2
z0Xn0!Df@In_9Dh*N-uD)P2Id7_Nff6qwr6|oP>t=P8-tdJ^HR7NRfrbeTuv3(}dLN
z+KlfPe6_D4JClaX2AF8M)|Gy2*@pEnbyu%@x8!=Pi}6H26)T7nDK>z0#x@11PZ~RI
zw}zSaPU}#+>#rDPN9}&x*j>Zhfs^c%eVz@B_<4G!s~x%yy3O0OriGMfdDm7!le5!V
zi(l-2cfY0FKLM_#^P+y!X1i0xxCF)9v#{i5$ZrBK@duDj&)@yF1(96L9X9{^57xXc
z83~eoEw*^)XJVUsp~)UBt$i9MX%^IHg)s>H`ml~6F(ajA+(vBw$=T&d$yqDBON+sY
z+W7^-`c3k96Ui3GNa@+KyLADJDxC|%cj?tkE*sMEn!aA}EuvuQn%d+mP&U5Qdd)bv
z&GWm&(Hoyy1oS#EW^>(6vMgBAZMkb|s^W^wThew*v9+plk<=!1I<1v!_&CEx{-Yu%
zNl1_13n}!RYQeue;di!sK{^mcdfXI^d{A%9sp74<vWg*<FYw(&j~~GEJ#QuileoRQ
z4+}<L6fP|@%;-OIb<1PFo+dweVB}J^9ns%SNPBQvX7~=#DFHjBD=u;FHY-Yt&v{1E
z?ir2xo&IrPnno+&Rok15mfSWu`?6vCwh0Oq8eK^LLo?wwrS{=zm3Zvq6(3&1L4Gaw
zmY0~K3HQrZfhZYWd$ay|O2Kc(8uKgh!IwM*rSV=}ehzAv#Mo+<d|V7&Uyo^_)+aEX
zQFfIPkt?=g8WK+=m(pA*Rr+j|?fTknc9`P&Y-p`o+59;e9Qv0}h3zHEG+l$2#uzAC
zqs4Mq3y+f9SF&$vDmJq*{T`>$Ch6_HTAGk)`6bnycxUN?bU>48P?T+HJsL)U8Cpz~
zp=+Y0I$8)uflR$8iP;*cRY$kx1igYp6Fr;4E*`Bh-Lwu7OEI<U&LM#bfk?T^u_vCI
z;P$IvDtGBKQWV6Z)14Wr?&++4Nb8*jg!Tlz{mmNt!pLpbI73!9r}@k)kCE-~1FJ;4
z`3=RRCR57AU80%+;z5u1E}G2rUdUmUd1dmuCNDVkX#KmdghbY+B02x%kVJGPq-4E}
z%hLP>w;yOfs3tQS>}sdq592MBFR*>N+;fk*)FO_FJs4A-H?8Zj@VQ#|*#PHhm>|6S
z@JA)hyN>##@1--MW)UW|iZ&*fejWccGV(=<vb$n$qI@s!Uo~fQ+S)PG51r?b$53Gg
zX7alWZnOO}3!s&;F}2z5r<%lzm<JyZapw;<FK(hp+~{gW--%W9T#FE3AL^Y~x{8bT
z{ENeen9c-WNqDf8SR}P;W-0@#$1frGV{t*UR|i{ux_^CH=--(w1eA-ipFLFBZl&`y
z)`63lPKa*f^6#$3=i2<lywSzZd7}=WE;1>^S{8-f7%j=38qK}nmlwd))AxZCby?@g
z%;c(*Y-l!AsAndQ>l0Q}-`n4v_>^^xJ2>VAu51rlnu?0GfwF61$+jVEx4P#!S}{$W
zB*}*+iu3&`wt)Y%_;&F%Ld@JV+J}$2ZwHm96$U#cGXe|3b4*^w0uUTF{BElzlCNmV
z*J))tzJkuC-aPQlY8kN-3|%7w5NvxUDpDgJ?@6Ifen46F<YST5yblU{x1bpj>)d+N
zuYfx0#6#xOyFz{77nClJ0<qAB;+5tsx0wZB{XoyP?A%(q(2Q-UmopYf<&@FYn6(4?
zzDLgdVE^m7kW*Rt?!Dn_`lBj>MQC~Hvkz}uN5LYoqOPfUNU<}MIJ_S`WT#Buqeo1C
z=aY$e@#EEJrmnH{cIVHdSDt<t2=BtY!ZGZ8H?0_t^%6~L5A;TXIqiOviu&$PiE~ls
z&D>~uJ0^1zdBQ!ni+iOQzT^%W8YmWF(Ie<k1S}-CF9<K5g{JVN(8E@Mdm>g-{)$bC
zx)32?%Cg`7;8|GH8*ilMF97Or@w`L7vdJLDZM;g?l402-jLi4*q?#_%tTAAbr%U0S
zWOuOPg<b0I780k|jZ*4X`C$brj`Gr4-JJlN)E%U(mC<_5twMC})-tn6T0qu{3Uq2(
zlm0CzzW@RQHv>gN@Cy^MEa5jk(u(TShb%o?i?)h@um0@ZHM=;quN3tBb&I86&ig$=
z&`>U~DJ`Y7@h2u|So$uqvKwrY4lhx(qr2N~90vr>@H3~spWSnDz)59n4i<(ZB!cGg
zOY)H;-1zOq_|;CpIObuAw_2`KVH^{w`v|SIUEdlq7dt`eIE6br!TQ`pd*|Wxhb8`z
z<C+hy<*-f`Pj)-mW04N3-XSEsycxf@Gv9kvr(cL!_A*_4&pqZz3w}_bXtw{JHx#Gp
z4X^j9<Cw~-p=ZfxSYrVm*%=X2%8ny`)j-uJGCmGUNs2F_)il^QUosvet6fndr>OmP
z*$0a>jt<dDf^=)w*#p5mEw7EjarKfdU@<Ehz2m=Q@dFLO374Ukny21wDV)awuk|f7
zo9VFQU|Q28_%hbQbnNRtl@oa;R$9Gw9oXnA*s#yLwP97gk*ivnT+LmP#V;2IAXAS(
zz4vec$|44BMG5J0pkd;9lA2XfIjO{5r}g%W^SLq3?|=T?GxSrlozIaLb38zNv|8a7
zcvNBPp()JOKbn=X+6Aw3T!(wP5+0{xY%?9X{))IkjEs#4+1v+>S#=dGiF+PE11?1Y
zu|fq(=q?9>{V7uSQqTUNk?suSU<O$x)-@uRuvZ2xR(18<JxTAx03i<N@)Nx(pCivJ
z{tBdOwAYdTaWGZOM^7~cTAM2cSrrpKpC1yE;?9)?cve778>~*r-Jx>Ld1q)Ev=|c8
zEX^N;M~4L5umdjC?>cP{>)jKBQeakn+~HT;^o1)dLp7s>bqbaj(3%kjQOK#4P9ulZ
z4xnP_Ehy;)GCSl?_Gr+<nA!zx3>2x4@>@mSx%P1KIo;Y;&j_<>t~3(`uvo3ADYY5v
z84bL1IcN@wY_QMbKW^~;y4YSh{$@5;K%T;e{!#usvk>a>f*97&FXAnNl1$StpBesL
zC8LhrIXu(uV#ON#q-Az3*HewEwcWs;5^o|_gq<3_VRIK6dLF)6IL`FM@VN>nA<Cp3
zgSAjrbE%5cS{Js`tCoOX?Y$JW;RIBed6E<GtVQ20{q9YNk>1oY5trF^R1{8hYE%!m
zTG`QwnVE0Z-c{)vWS!BGC?%qX6KT^^aT>A!(ank81qiD&R<A^UaqOyE2B6b<;Tem8
zZ6r&1+xfGfC#*6IqLZRdXkl)*Q#rOTv=+941}cYpHri$23}-A31%<Xsks5O#clh&F
zgI~U$58>hj7iHFp_K}i(He?x$#`l*tqUSwQH+0SMCs)!3jD@0?FTB823_YO`>gU5%
z?!1+(zGcyYR1Vj_g)UwL(6Scub5hc^?)91B=BqVFD5Ibt?-2f3MOwjK^o@Xqx~0Mn
zvlKM|2=AN?w7<U)EO<k%X^O(BTE)<fb_|9x$u|gfcLdOg{_+ip$Dv8Gp707CI<IIM
zj8J*;=h>JgsUDFL^WFCPtgz3z{aSkGzYjT2+2r<*K)7)%=0Wc|ueB#!zEZhvnApti
zzVfKU@@{$!)9z2-EOk4s`R?4rXs0CviHX#{!Qe4rXb||S$A6mEeV2VRvdCxZWfnZ!
z*GX#>wSi1bTlQ#>4!Ol%vwtl9qzX}t{d)8y)PBE1*`4E>=JC%A2UAxQQ1jCur#-n=
z3=~@JYA8#N^vh;I%yMmzAo6MF_Nl45TyrfxIqALv*exVbgb=G8>Z82B&iVneAQJa%
zK=qnf>K#b?F3sI~^K3C)Z#!WyHZm3#P?kf7UQe4kJN-c^k!})EtqcIKMs(hB=0vn)
zR!M8H{4$|G-9H8nxoOlV``cbklg_SorIRVOfebuq!sCdN2^~|;`Rg&`4~s<_-xv&W
z${a@p^d9}KlsQkPO=ea=H-NGjM95#G*>W(D-6)s>CCFK{G)e^*HA6LBz50_QBa>ju
zi*2fk;hOa|AYanSpAf_uM9jMT&jl6Q0+z(7Sbd{Uy+S2L-gU7z%Sd`-ZVV~I`LjfY
zH)cSo^OC{JEhoN%9ZK-Ex8p8jf477XG*Mp>mdXvJW~-OWf8J7L74eskP9}Py<F)iw
z_P;6Bser~n1Qnf`YSRf2mA=PYlci3l&g2dN_J|XvYe=}((J->ipyx%EB<3N|%gbpK
ztq&(dJe9x1WQJ=RzT}F+=|~GqUDC9mawJeP``2f)mV@r^>uI}f<b#^$wb){f<^cg+
zTQ&-w(l`1veVS>)pJ!)wLi28hpp(gmVI8sBNjq6DD<QpnIl4vpl@HI`5iJE<4R8fP
zCLL`;t_sER```W6=)|>>E$>(nJyxB7*|R&MAAgQ(akqoanUKU9HG`S4>aGIOK?Ez#
zrtChIH6yaysu$O{H(NAlt8u7eO6pmuZMV?oLf#8!PxAv!zab_aX7-%a;23iZKofP9
zXb}D6%Z+Y+ZN<<Moj<)e4yYbnMMI)#|2%YqOhkd%+DRk@pa(Ht2<FU<m-ZoR|M`AZ
zRK;;UW5_MKyh!i{KUe6SenHfL*Pf^1MxkpF7K`6?Gw*<gU`^$TV!(8aLIi>;2<`;;
zC6!{MOzv6+<0u4L-7DFrHCvSxy8r&L*52yXXRzgqcT;pT7(N`Awk9fIOT$m@H2<un
zw>y$$)DuYVD(w<U-JAdMFz0Ozzh*3G-g0*)i*O35<+Hv+m_?oZ-Z+nGYLBi5Yg(5m
z9ZZXYz4}5}1N()XL*{K6#j*>OI;Qf2TWyQKKOcK&7<j6%P&gXkA!jC;3pEA8EoBe;
zl|N5ECXl#Oizl!u@r}izjwN7`8N(NPS%q2>vc5QDfhGxa=~R+-l{w|$tUtx$0=4n#
z9O!GD>pzS3$XI~hkgEUDi`|L%I|9Ve-2PNgUHFx9SYO2zX8XXqG+!8C;PP`4&-wN|
zK}Hl1F)1bSAbMy^y;^AVhX^$3<tAZTDG@QHCt7&Z&O+y7F)U4kuAInlqY+wa;#Dbg
zXw07<5>U#BlOSvr#GmV|a3KgdM?C*z&OdqRvSgD2H2U(y55CXS+j1X*qQO1d5&a=b
z`(H^%NB2LHj&Qz~6x;M@x2ZJlwKP1L<Rk`ATDugxrR*r{&8P-@m6kEzDn`y~&fiGS
z#(q60GkRU@yP0U0W^##&F})(`#BV@Tv8#8h&vqc|;F_I$Dd|mvzyD&6;-p9rxMxYn
z4lBDixVPmMe}W@NrO{E0S!h!GtF4*-0V>9`y4(+US;aPFtkM13Ej-bnl*_ddgk48m
zQEo7BM%S81Mytu}gr;n^XguZFRk(AA_5Cq=!rentn5a3UNm1|q>=*ZoLvR)KX^u{V
z7aEhc?g}uuMcbB6o*;E8QyVb9a?2GnQ?G1W7Ql%h`oV4vO>{cR=Q^I$1q~EUpBJ0o
zNNF<aXQVIxyZz)u&LCY3YhzLXZNgwZT{oYQKfT~(TntY5$BPRI83N;l@>dTM=ggp-
zu*ptTxadeFU(vBYmA^-a7bmvDM)(`89I&F@skLrA>$%e&h@qy4$=m9*8?SzBzoRfX
zcfSN6eMltszY~e&w0j&DftzNocQ|RM0w?)pPNJekI`E;cFfOX0(l+gfllXUUlpBdz
zxUTa;=QHhs6WYzwfg^C}tTe=$Tb-V$vncX+sYa~Ddve~Bs0c}tKU`G|`@QM=3)rt-
zh%}=I?mr_`Z+U5=oWze2cE*L3dBD*irW}{&6=L!)=nU3G1R(T+cM_&?^e3-HDuphA
zB6T(8o#87}QJSWNE0W2f)T%34cAkwy=>O<-%hKG7hVY2wcnf|op{!nHP$qi%w5qCA
z2D`=SS8?I1G|?e0q1=<}>Mq^Yvp(}0Af-@5Z52d1%4NhSG2wzpY#ncPCaow4O(uP+
zm3!djhOJ68Fp`2>P$tvh?uhRH)<E|frb1+BoBRU5{v!FhpV4c3L6iR0zC{!Hio-KX
zuB-3{G_|^(-*r3Dr1yT(VI`pc*W|~0V&2@i-t(&!SD}4<&UkvW3ybN}Gy`-#U>{pA
zs~ajR*$Tr_@clV`Ew1Y-iB^hHAW8t2#Rx?hOaJw*=G6k38cE5<GtnA>BBdK%(BiuT
znd)}o7{GO5hi$_7C)l~KtmY5Nw^*b^)T+3r^cO(Q>btnq5xdLxFxWOk14Syu)U6yP
zu}`~{DuoHN{4;XaQ>TJ=`nxEKD_JkHE74_HX8F^{{j{a_O-?oXTrz9aD{<OkGHd8W
zFVZO+luZK!$0be0#w)+wzszlv$iYPJ7;#ST=7`&n*m`v4{2HK<S8%h!^W`VCTWxuP
z$*mJIZJQ4&_3E4|$^>x>LFH!yFev4r>e^{<Z5<z_1F>)ehlh?^I1KAdfzVHrV-==K
zz7!UCVNi_E_&aB~eN<;XK-VQfb^A6`dLvJPXN-vRbnjae!9XUBz7N$P%$*N@-bNTq
zWo(-<ZU7!?c@1i*-ocC?$NJ=dQ-N`9RgC_#3>6k}^c+)C8{P*Pf5w3HPf?MdW#vcW
z#1+p0Tjk|shLdeE>np~=wN{;yD~MF(bm^UaCsvn+GHj0((p{R)sX4k76tjNN77Mvu
ztpcc5mJ}wPo=T1IhAcO6CB07ByFc}n_opiiR69MHw;9PSbzlUal4YgPp_%r3%C9^M
zN86@d@LRg`pHJx@Q43DEAN(Zl36X4Y5YBM?tZ)SAITKP@CmhT7cO0qIP!yc<9|yqt
zuDrJY(XQn4XdhzZCY9H^&#Bp?+Ws#msO~rT23RnJ-&Ne`SYNDOp4m!@deabRTb#DW
zQKm=8AaM_=bc?;aS>DA%e=O;H315<Q#!66_nZMp8*cy0sSmIZ$w!iN%TNP?k_x(y;
zHSQ&A<}sUA^Fe_}sQW8z=L?e)U7Lld!%ryRb0_#mAZmw#SfrY2-kFdV|Ahy^p!c(m
zZ8w{jJb!MP0G+(H95KiKb(D@oSO}3tHMh>Xm8KG=7+?46QB^lVRLZsKGVuyyPa6+~
z@<zBZ#owJy09)Myx}ytD@ByN$XNeU(l)FjdeVx%M9;eIe(>$jK!sl`eBKev=B2ssr
z9{*l-m9L`=Vi*s$E3@0$p4nRtQyW_Q(AB;f38VJ2?>LG2brQ-^4?vgmn(%pS<$^N7
z^A1^I@-HNpY1+bd?01>Ezk$}A=6(EppR<Iq<+85<z0BaO*G8khwyD<a=+NeB1*5aZ
z;|jM;CEvqSB2}2~DYfY1y|`qln}=5KBo2}XV|=^c+^=5IHs$+y6DVuumr~Lv+U<=?
zOqJ-$W^rh=pLt)kE9&A~lLiOfu4W@?lL-)9Dbw@IR^ZUmcwyER+hn5&z&fmoTql-*
zQIH=xAhv#U=6cUy&SH-($I>q{?-|+Z#1E94%5BykNx>s9^DOkyLP~Ht$wh7XSg+j@
zv)FHite18s;`N>cr3?oMUtYMS=*`qVL~VnI)~5F8*)pcB^a15Cs1?WDHa{xaaUFj5
zlDxXtW;uOsB2VFP=ONpSwLY^li&)Rqmo}&5Q^5moelh(?jJD(i3o47YzbLwumrJ`~
zW-9rET2gxLk7?sq&hh=}Xv-A=<3<C0EUZoYq|4%TBgJM82;JUIH^Y{co~ac<{D!>I
z1ou`6-JMNzTX-I3h7jxuAu5ATU^D8KvkM9Y>3nVnlyj7nNu@0&##F;nTMy2>RAc{}
z;h4C?oL0wJn9&&NcBVIx$EKiND%aW1w(mohU8=#={Fnf<-)&y%W83Wm*!$>;o_n3`
zGJ<r$m~R?%v$h7)(mqL^>=n+{*)piP-hyT0hkcH49tvhWwR>i(=RcYYZCXPkF4I)h
z@0Fp^FE+VcYwft#zA+)MFpCED;GK2z$9|0akC0nf==MLBk7~G#ozpPgX#WsuL9kzW
zm&8OGbg3txglC85QMhBEe2SZ;{+d5pz7N-xLp}9*yq-1Iy3k_Sw>HVm@3J2#Cq%o^
zrthf!MvR__XAB+j6q{O8+}Ta^9q<Weyr`krQ9XaJFmVNq?n}C_=tC@U$3yGmI@QK(
zgDMpBiIy_I_-g2?YP(I>nC*(NnexB^`-+5Ra{K0l*{*#)LOL(PE;Mc*=FEKK<{7`t
zKiX_07|1Z1KQmlcD3C5k&NFGp!ksOCjS6HBaw)nasOKTY*?2TVE3!P2IEQbV4&%@}
zKz*V&(eL{NS!n`C$0Z_4yC9G6h&o7l7Rh@w9^5L^(`o^gGoms<7Utoi5Z!J2aNzEv
zF84rP!AUhiNr0&Fy&ZR^cmyVEgDwl@E4Y%ncdop5lOlwt$B5i@TV$~^2y}b<GE8_@
zR!e_IM)0<(Gy=c^H_@;QU|GI$Qg&+g+eFvS@-P=3GW-Y)S1@1Nw73&D!?#%5(*=>#
z#}dgUA9z-1%uBUTK0BpnexxU5SmCB*yUmlOSGT$cy-TAh6iOey?L@uDb;%!(dOjuQ
z7A#TY$9}yz`=p1Bi-d|ROwG_dSuXus>gKR-<;3aj-NN}U()$B4VlN8Z(&Kx)QP8p}
zaA7H*YP3}{&`EgjwKE-9ip*j71yk~qC9Jv4ti^eD)m&YTzt-=5FOX5}XY7P5U2UIj
zqQ(kaR(t`CYJyCS(ex%>m#V2WJi%QD8pvB#%65%(0vww;S`WNdcR)ryZ!tkar1zhr
zGC#PP<bDab(sXyWK6MEci+t%uDOiFoVgmd460nD972NHGEX1_JSo%i_wz@@E5?{_e
zOLIvy$2QXCRj3KG6wl{1X3)1qAK9>N3|~zjNC~l%|D|f&o3-<@Izo?K)s2jQ-qYqu
zjCSrzyVmraYvpH3Uye+LG<k<Iy>d+&gG1A`J}v<#`^sE;mthcF{rE?0MUg~i);q6R
zsKmec3}dVhETa+LbJB&AOt&D)(My_0bVOIAd!{Jy^Res*m=$%{IpQvOKOC;*7cyx3
zsk*wPjx7z)G>tGR2eB)z(SCSI;g{-Irr`~y#;PmhZU)a(-zap53U{+}Ghc`}FZC*=
zGy;NT0|?SL0j6ybn_D}czJ>t4JLvZLAC}=(Ns)l~)=(Us7I<TH2k)C9$i>hzn?3C`
z8)wy!6u@h(<X35%JE;!5?3*gal}jf?TqlyEIxhq~YqQE=%TYHvn=ZU-J{kl5rs|96
zN<}(ZrHu-xy|ir;AIGboTAP~VwE5x<$it<Pc+W|yi_?K8;20-7lc95Cmo*4nv?$A=
z=+3AVA*3Yfxf`zoV~}bSw#6IG7v!czkgQZVtaV9p5w?<BH*YErDG`!#JAG)~@NKs!
zShD(bQnKGA5+`a6`Yvvnw`vGbN=m2rYcZY8`v{D=lS-1B;_ZgOS|<O1vNRjtqYRN@
z{k3L@C)tqfcd}L{qYjCtev>ti*{Zo`N%EGeE`Wh(?sEjBIg9mPrtS>AiK_|(Dj~~!
zE5fld9A&VRdD~vd>%CtpG##3@inz(t*h<&kp0&qgP8DhMGQw%61h*L8u5^hSB>D#A
z*&6lauYJ&7X|U>%b<>6ZivFwm&pU*1i$?15V1a&6(!>uWp_2>Yait8jDH5g2T4&@H
zP;yD%skBGW73);1J73>$o;)CSCBHmT%D*9G&EQ5~g~qX0DWzu}BsZq#)1q~k9nM(Z
zEmb_DD=0KLb-yihJgPQeXRp*Xljynl>bK>kXg`2A6W{le245!(79IZ3+?V0oko4FS
z^*+{-NsfVo?Pnc%585XQS3GCK&RnB#I3RE=tG}F)_KT1c0dx8bzih@f8=Ia}f<a1w
z{p6Zbp&e0PawL~n7to*zIoOiEP?w=-QwM)4l|jNHmdw4D;mft8qilrL;N=Ln$994=
zPW^^!D$sgk*&Rgh&DDD?epy$%aHgWKe(%Cn*sc2{THa$N_6t7-E1KK7&BXlLD=2__
zEjEm2_t8}I=&nEMu(Tzen097KW*0p^Ltywo!6(sG-y$AN+|va-mtm}hH&gNb?bN(m
zvD`UTB#3E>7b7lX?)soMaFGvm$0iSyhAYJMN*6AmjT#dvYG0o*+GZu(@(b0#OUJk6
zIrCzcGjBlmEs;_G9Y9(NZoigwuX~7x<tX<mzRDlumgZWHweym->o<AYKiAGr)~)Mu
z)rFnPtztUdXkD?=otsqfZSQ;_ci`opi4^`!C4HJRgBP!2!s}nwsS>VrW-DylhWYPF
zGxa5lS%<YRycO6G+<QHUaxtgb8`BsrUP|<rXiVfC+-PxWf)bTZJF|3&H2AfcwXQj$
zDX?$X0en0D^RB%3u$akdZQAIC*e5){>_WLypXOsk{ul$zxrfC1ZIC9g_b`YJpD*}w
zyx>+u4-LM%iU~y3Qcdx1VgeN`Pp&KSJ)_;KSmr0*j;A`$yZZ(=2ED@}?iJDj|Guec
zIZLKZ@7-MaZShg5n65cBVSO9k!uH>$gQ+V+gGn|XVw0k;nZ8*r_fM=nMejYKihByG
zz^kSD=Xm=a7VQWE-?n3(HG!~zZf%<UjmK2V?wbQrgG7+Z1SoBZuRYz?OQPS~K2ZL8
z?`h70FS*^w&9!&idVX|krP}>Ivi)DZ#JEb!OIt_9SS%UcXRZw&2s*Pj<~Xe8s^Pk<
zX!6t7>gsNdiND5PHQTV*dUvsO%51Fr+>Jyd!Rs4hclhVN?+@?#?%BRu!8bYL(cx!Q
z5iT%8k3gAm`2C2BcUa7WhND1?<K+TVgJps1cFY4hy!@1CGoV2R+cBq`!Qu3Eo$?_F
zcdu?X^R<qw>De%XsIJzo023%?3c^_H0{u`7A!$o<)4z9KE-rTf*<V$loz@7MB_#AX
zKA&Xxz?IW`@uEvp8CW{657|-->orbic9whx?{53&zusP+>Q&Ox5y9kj-lPf0<=V*H
z6q3{}Qe70Pl42P3>{4OFkUig(Ok@uxs8gNWedcK{?9?;nb?uY7;;U~{#^#(TnzQ}k
z&G}60Cy24SZ#q%=4Ckw|gO`OY2ub%!W-2c!ynk2VX04@!T1C7ZVS!wz)%Rf4$(=<w
zCQZ>`+9rVEp?#2d)(TEgUEwR^$6=>7uQyW9R4OP%WL?^mQZT>6IcqlPxn@)7W>gY$
zwP+(3J9VqPKS?NWny4Xp(JTTQt)gBWW|*TfMNb(sd-A!(p*ADMe`Z=SzmXTO82LKC
zGn7#(jFWg_EK@J)l%4vX@2Cr*bDGgTgmq>A8HO;4(W9%iAHJXZ*V5w6v)?hcZf{@O
z*WFTpbnR!QYMQMk4v1ByMTEIMH~n|en9i{~Q%;|$W*(lK9?7jEIV+6H{>ynVoP?C|
zRioFBBG>QfQ43R|@-_4(5!<YK<%qZAT0rC;pMB$;H`Vo_$8DkoOq6?F^&rjKVGnyw
zkfA6zs5fM#yR#H)YFmb;Bu%~|TITP$lBh_u&7c8QrjkpC^WZ8!WUW2bN7lk(Qvxaj
z{Fjsa!#$e`DD{+2=**<=`7@PIu!!x$k9#*?bDw$j3eoA*hs#CQf-V@o2bscE6D^2Z
z{N84tMp*XFFa4e~==b<Wq;llkVv}1!U3R#oiqsyU%fLtNe9I%?cL;ai=YR6-4kK;}
zFwqQ`t=Xa?`A=SS)%iEO>Gq1dlB#s0aGuSZM%_RvoFsP7TjX?QD~Gcd-mt+er{9-~
z_+2T33WaC?He+a%O{8;8irH8E99at}zx?@jQhW2QKMMU;J%AFamF!^4ge%SUsC@wp
zW*Gsmg}JWARWXu{-L5lir2C8TGUD<JAHj_kJ)=X;JRAKQu5R`>A_Hpat@+s>T`0wb
zRWT-q^Q}5`Goj$2E%@GfH7X|i#@DU)W!M&|_)Aj<9`?!8De{(PJ;67;S1n4fo51Xx
zJE!bIe^&6?Y#r9%Jf^2HW-!_Dfzm`p?Y!vtdxvtwa397rdPUZt_02g60I(_Hg$rNU
zL?ySV9rSvE>&1bd^w9?o8E{TsOv^d)>#7fuZJ#?^4w{(HDsM7(XThS(?bDkTDrPM;
zQ`U>K>!e-sEtR(pO!H3>NRK&MG`&^x;H$9_cK6=t=9QVv!yzp^EQO6bb6)w@bm%&!
z6M|t>QB3&hq2A3uX7yLcUJM8$Cd=dXZo6-c{c-+cd>-c}-yZyps)DWo_gD296N(-|
z&w%L7Zf=&0?5f8k?wzFm6&tqt`~^yHJJ5C&G(_K3CusLw%A=BosFITnUMI0nT8x4C
zPeze<5Cr5hWiPooIdY^?O}%cpSgj`_zwmur$Ew9XZrrizY~NW?Fav6AJ+#0gY81m5
zG^#)QuV{mxT`-Q~T3k5_;f=v2L($3<jJn3ueIA}8h!Bs-Ua&GyRK9XkYf9%C9;y(-
zEnP4o_Lag8nXnf$zfR4j+z>dlm#N<w;`e4V!X_=|)Iq@1Pfe_~wbJjp5coU^wf(b(
z-g|E9tNm~MtouUpYILW98vXf3WA`fs8|}ZO)g~REoSxJqkaslEnTNb%qxeV3?@ddB
zkOJ1%?@>1Lx5=O=`*2Or?}@aH^!{t7nRo>cQTKlKsKRw;ORX&G^dBUohi8DFM@OE@
zTK%Dh3Z=e*0xmJv(_CN6m*U_U3;R833x6gt(PHhToi!l+dn@zFS1*zscM{GSM^?Wg
zC$bLsGDBiLdt;!0-{n$sYodr#SLV~#cAf63KiQq!MY3Tl`Ac@4N8g_{d$e4uI~(J+
zQ_nkj;QGp{(VCe)|Ldsd%=t@5IA}%Iw-W<E$5>b+@cR-iOAP@{6zCaF#61?3jRNm9
zHjLQEN6<5iw78@wrdb%N2ep4~TS+@#r;$;a%&8ffnr4nKZ1avQT^e-_8ADcN_j&&v
zo*+qa7Wn>LT%AQ~v_M?H2#l0yzF*u)+qh`(?XmNh+J7(ZO%lYsY-n?#pV#nKT=ZXc
z3c}|^rQ;rX?~9J6$vM8GWB|TZ4!-ay8J1G2<_wJp1y=uv=#XyUsKVZ<;>lOpzDTZ|
zE>zCB+qYbOfa#HnJinTC8OX0uHu)b#5$~qamin0>l5{?~aPGl#uwncm<6QZYf*qKT
zEK}kO_-@z!BOrP^5DTK$wsCFP92AiD^>0)9_<lyuWhJ|*wA6q?(oWiY%(Adld_2}2
z#L_Z9aJX5*lv+&-&a%fcc8c2^+^33i=#2_+Uuc`k6Vp@Uk~}j`#R8R<H;S`h`T#AP
z_%^`gQMf~Y<EUWlDIp7DOi><M{N;cEl$;?s<NG~etTxAv>D-lGlKbabJ?10<wq5f_
zMGWXunnKUNX=Ob0U}pJXO#f;jaz&q1#PmjOm%acpT#-To1s&4~yeh!zu>ouGe;ti#
z9kR;so5N7THoDw8#whdT`yaLc76;kxUQMh@0BhN`Fki%y7Gko~YdY-(1oe_0sKVOD
zEqFI`<xO6RW>rF1b8Qvd^DtLIoaVndJZ6|<LYL5i<4?eDX7(o;04wk{*@Uj0j~^R1
zg8&RKWu^~Bf`79QZTwbJx#PWk13NwH_*_AULd^4;uB6nyu1f*rgSVHi0^NidBE4Oj
zuGL)}J7;tM>UFeDCZe^^YWdrC0>R#PU!hvU>}g@F`)$is9n8k)0LWa2EB1D`a(Bu>
zTsQz0nmLQ&EWih@hp5LKs52Bc_@!J%G7|{;;76`>#^lZ>09l{vS7M-X`S638Voj@&
zuTI|k0B%dua}T1b9P>03JhTK%pWHWJ5_XuzEw;1#sepRoE#HcRL10V+BRI92WW^4q
z0b;TYlPuh)Y_C5$`$6M<xrB~nN2`z*vcY-LU@39?s|BKx$UaHAnLB;4LLri`9u03g
zipVhw&DcO0Fzi9<YoHD>zW=v8!}DAYcwe9Q9kCQN*JABx795=6&i!;E*U@^bwh|!=
z2LNa3=b1!=YBqEN+*@w2-=2@qk)!FX)!h>)y7|FGFG%#>vA44;N);s6{=JnZhdh8g
z&?8)3%<8)_)%5^W3rk~wBAU9hlxC9z({v^>di-0|><izI<l0YLqFIcL4r$gSg*y8J
zqU}9p7`!ifV=dynZbn+UAqIP>JkJnU+a##?r5T)5tzFB=Kv@KJJk?d;0rD4oZqsFK
zRm8n}Myt(+n}nj%66`B6y^9582!3=%$kE)I2A_Ircy0Lvy0Vte>K#Y{gbTJ|yavD6
z&ZMG5W02{#A7-h66Z=$j(f272CPx+#6vMQd*G#G>H^1yD)rkPn5N|ST3Ze+HcRQ=Z
zD+PUklS?>&;`bS%N49f5?ZtMG*Y;a+8%KT<k8C1FEevLPG`waI?mi2?&)WG&Y)Oom
zT<8L+-s?R_7Y8lDf%R%ZQ*Scbs~D_OeE$weB5g;;;amyqo}8`St4>?kfLILe#7S*I
z3n>#s4BCX4y|;VuB}P(#j5gh%)gQ|nqWAhBG)&b@qhq`sU;KKI`7qL{Db`29ya?uk
zbr^@h2NN+UxMd?3>YfR@&y|=>BxX8Yrr0}B*M`#y5Z<#893HGXY%{)g*kJqq@~y^y
zc=*`;+e4-f9}lteRjze?Qy2a?|B$d=@x!rch-%4VoAHu=-)GA!oRng{y&j;V;YDUJ
z?o!D!q-qsXI_?aK@wJ)HXF`NgYh>K`d-{eq-J$iHpQYCH92L#tIzY^9y3N?^&t#5j
zOZp#aExqcbqR9R2fy2!}rP30ydvCmkW5O_sMz@$bs&Y@#+-HQb78fPA9P{K*Ve49y
zJ4>WuB4Updf{p2>`(jj=9rL1>3;bdNUd~A<u3U0@X;mcawnBQraB|&SMkYfbY{elE
zMJu@r$J?`sTbs#w03ZLI{cZ=<dajJ)P8Zd{t5FR_#SL$z0SIgT+j5F%6VJ%IHFjTl
z`g)voxmZ_vhY$D_uH2!U@28rvD2rDA>C<Dlu}tdDW8+>!DfkJec4gme%HZy5QSL0?
zx8LH*U*+n*|GB5OkMeE0`&HbyxI3P;aT*LRvl~>Mt{}h_%g`2C{5_Kce*Dw^?E`zL
z9n)nXAvsse6bdMl&Jht-(XNrZXqm4`SjCG1qO;46%BJPPBOmG*_b%F1ZuQU8dA++(
zK`FnRqcH~S-%ZT5`kKJuW4xXHz~Aj=doZtCVFyA9Y%gTYS}zrLc+Nn7UrQxqVT%>~
zrSJz$nMC7^9Dc3$!hx0AVK0j)MDOQ65V6h7G)(omYkduA-}V{@xhmIxm?G-bZK9zc
zcVq>(R-G-r5@b5HOyl2>nG^B4obyvF6Oc8nxiqlKZbIe#%TUMgHQ2gu<W6(Gy1}h|
zBdsn5<n5?#iV*qMN)S1R{%yki58E)ChWOgKE2R_Xmmu@d_s4dMt$fU4Jk3AJI0^$8
zENOKfR7+u-r(P<dUXD+`%>5ShM(Cen^UdWUu@5!@*MGcN4m}}S0R#r!Z0yrO%}!S6
zj*z)G`2-Tuqca=ziWoygP^5XzDR-;aC@=z<4X~=^=*={2Onq!|Tx{<v+xo3j9n~^5
zG8UR?69qqBaYi0se8gN@JX-!dv~;1CULw}YJvo?fg(GJinGDo^?jW^DBK?tQIZ}pK
zvbrb0$~Wdue+Dv%gV6*pW%J0#5bCoho7SjR<l2j6c%gtfaEi5ZEq2KAh|BTq2ap?i
zF?M3rlDTU#;CQK&lgLY!mvJ>?%jUcGx7h1HrrNwZ+19oMMvs^X7}G$%_ewS*<ewJj
zpJzFtV^9U+<q)MRIYI&CSr+o0+dXPn_E*atNy`G@acbj$DIGi%+y!qI%8DqtdtSR4
z)Ss|kgsXrIB5Ey%$k)urCp7|c#HgiqL72i|`mOk3{O7~nr+JL;AVCuv3aidU`3I$-
z+2WBb2%zI{um(|8wLGHj>Nt4&5Vtk2!}da56vQ&If^L`w)_pMUFt^;hw;50FSzqA?
zZHb=I_QF#&&XO<lv~67AY>*N(3`)tYOTi)I8k!(oo|V0%7>MG`%z1JK7&@-3Ard_O
zMD6-awF!)Rx(X&fs>^ph=BYlLY{!4O6>E~^;Kg=1(aA?8>@moaq})ZPJ_#0!{a45H
zZ^zdEutHbu?I%Cmo0vpS#b^x}*%DrrbRM&u$aIG68O&xGJjk`viYA*m3ts$Y=I3A)
z`&<s^K&m3`cIy3_=MjKg36&Nfw@z3l%7e(c@+%lcBr^QZ-pFgeAa8!(z!b`Lt7Rc-
z%}zEyH&qKkix-zdS*}}XW&kL`TFkak0k%!WJJm8MCKfUQvYwZfJ@-`F4c_dv4n$WG
zRST)kdNmmzd}bULpLFl}i-eFL@{x%*=#H^)$ONZdl)M#uj)l+s4Nr1N{Jsdkpdi;T
zBF3eATbX-hEoH4YwvMa~ILt+0rkC0W;MT9{ZIzYJn=x&DZ6)MJih7pXso-aEHpB+_
zreUb<vj}Xgnc04I=b7GfDEHNnXB4L*^P|DX#`9oEJb3DQernH95XjTpR(^iXH}gwT
zIGSoM9;OgKissY(4kMFB(Kfi1)0Fq}y_Ilp6$DdnN^3ic=)(lFc0BQh{YZ}d$XJwJ
z7>sRm#W^KwVD+-S_A2O-?O+(YDx=%PIqVdS3%JF%rTih`Og^7PaK37$rc!qndOULx
ztwGw%Ho`^s4k<%#WN7|oUH;=@{C{r&wERa7-@ce*wy`*Bnw}u8a&pp9nvU8YrdZTN
zbmFq=$qxPOcd8Nc&!sY#`Bg_PQW+0Y?U@$%<ad(qh8p)|>*5;nJ706a?44vYqD`{%
zY+s(^q;hLE{E9lrO)s?Kh_BGy49zbHB@SjqV|q?>NlEPP(I$%{^`;Hi%%X{<L~8B6
zxhikcGk5ngZmK}`Tr9nPUFGJa)7>yGc901zpN-<wxXg=jn(Dk(EkI|k8YjMQiB-SS
z!?g0Dykh2xP_58fG5Tuw_|t$;qq=|yIVmU}gA>(n=M%X!bHtbXBG@-w<Vc@-$gWIv
zjY7Awl6ADW!eLyY=dTOVM!KYZ?{A6${LkYp1XJuDHq$08VW&lA>AoaZUlF2tPriOV
zHM8R}^Pu?qY~8ZRxah~x+w(^$pA-IKH*04f9czjh&a~Hcp<wI4*U>rBWSTR-C8xe2
z+mRwKIj)oZf8C;gbKn2CS>j`4=Yb3i$Ae>a1X_YGj*5JmQa+|nLcztRVGGV=8ClsS
zOZ5VCrS{1TMBRvW>u-xYPWkHe4f`asSe$-|eX_ZFw73cU)dX);rOK~%m&dv{N8C4J
zbI@x~myg%aP7Fp30UaM<_JlE`rz2AoX8ic;m%#v-ch;9N!X&EZ%%P@->}e@+?T-@;
zs(I)PRPY$j@%s_lJwuGm$B@p`&$RWSY4XFEdp>An{+wB;+0NEHJ{$&IJ;L#pChXIP
zzrQ3Tmp`&@TDV|2loziOKiP?{WO14MMo3d(J=t4e(n#yT-FzD!lT`YxXNM!0`6<;q
zj))4#hVDkUTTKp%nYH6jRvJxyv+L>4+X-P;skPI-P8=w4RCqa_+0&$gj4aA;u+sH;
z<U#7o$9#vueynM-D~$3tVe@|<?|;JmuknEg^_dMJeQ+gsUyNj7NtEQxZ;BSy;ER&^
z!1L&*!DM@)uRd=xBQiJZ3k<blgM%Lq+++iEVAgr5*HaI57(}P3q&9#wdD!JBIzGae
z@3A!R;SJhm0*_Jc5kOX7&R@>Ys<XWMaw!rQzd8-Ibub;jWiJOdnf#nZ7z3rqLryxb
zqz-zE+K79S`AmgYJuO#??*vlN+)j3;(>cK*&Z5fnersjE7F6oGK51dxXwxSyN0~%t
zL_TFst7;2HqfPzXtZM1d%kwYEC#X5qo7A5bc1{lT71=StxSGyVC#%V}cpoitp*udA
zW391Q%wj!psr4Xmx&&jlo|qA{&F^)tFAPhj-l0(a#xvK8Lwqz7d(I>@vvn-tI_IIQ
z!|I=G=1LkKTE#0;3UuJjKsHeqz+?F7V3BU;zrD>rUGjfD&8JC@Q_R4Z>8NGXSMab|
z;N*uZbRT{~OHr)gL_KkuvbcSSF`B2L$o%=S=F17*QkCB4A=Smy&#06-S`5}Stw!=&
zjUctrzMJ@+)hH|atj@m<G#^#XHI#IPu>s8mYC@x3a5P5Xi{Vq@CKJ+3%^-*e1H#wK
zYeE3L%;br`<J5WnBkFutCG&~qZvsvG+a116HagO=Q63?rwU0GNndQ-t35AmfPb?dg
zk{@}qING?y-%%9@JyZvOTMhQ%I~pPX9+2I$MUK{LIlkhX^+Rjr8NcjIS8KG<=yh_#
zW~gWJ%&x>hB0Ra8i;FiUSPYbxG{vBZ>Lr(*wDTb4E0j7-oANB0<IXai8cevU2e&UR
zqSk7<J1b!y$_?8x`NJa+Vt!=B2m8^6O9rzbw+V5Y?tW?7*N!jqqvc*ieX=TpJ7Vg_
z8%?X|)4Dp>Z$;9{*KzLy^W~kv5QQ7v85&n^fKjKJpT8rBe8fUrJ5IUKQmuWjzu#+b
zhrox*ry~20^Z%bSH2++dkEdRk8wBd9rv|x!(pY}To4jW~N8}CZ)H=(qbhWrkH>da3
zhp8q9D#{MMXg|VR{fazfqEl8#{w=!uxO*nK(UCJepKPBOc3PD1l?9Sp&h!;>W*S*v
zpv+{`Db6v!`US{nYp3U7Ei=NAG9M)UXwL{$D#nUldRITo+-$uJcDp(ojg)CVdVf>(
zMefM&QlyFRC3!!aKTpV{rko=T(QUb0<5Y2<9OR8wuo<fI#;6onya^?@qGS@);Tx(A
z!|ZHsY*^x0DaQyyT--ury5W~~XU^T{UYXA~b+#Aw`f$?xaeyfgVJOSs%;Rq@iCTVq
zTY<cdZ_vVnrEtX}_p&y-^ZozZQU1UFpEucg#$KZ<$ts(2Q3CR&&5<nh0@iP$g<T?#
z2u4;&>`g4ABRwx=NVW>QE*fh=RE{u|;SHRimX7SYz1NwA?lJDEpK?;3eMG5DIKn9;
zuO5HvnepcGq-lxgmh;zKBj1gDyFt-l{>h<gAJY=AJ<Qh%_o71&2*Z$Pr%&Z1b)bMH
z*1)`{>rL|JK;n0VXr@-a>7{A;9C0puaIq><mR?$VGe`If+ayOb?%pRKB@i)lJ@O2@
zVvL9~#Pg-KOoQqtmGaFv@^*9ax8x5d2kWH#=}K^k|Lh+Bb1wetzaAXE&7D){dc6Cj
ztT?IT<%(rxUYoF0kL~pZ0_4XdgHKHx+b84s8vQB9*p(7)f3b+=@rY+Wz-Qgtw!h2#
zWb;KQD9p(%nlJyaLbf<F2CO_UV{>0b#XytK4`cXZ-|d1Jkb7~iY~D1@t`2OThiN$8
zy}p|)8&2E&D41as%1eQ-@yptgpDkeg*rn-v&N@+6|Eom&7l{@PC{{9_$T)Ae#Y>FF
z`!X(m>uQSNO?F)xi<xc8HaBS9T$vsPFIG5}=sI(LYJJ<oNE7pqPY;c~Q-~I-2Nwsg
zqlo`M;PwCe_IYa_hJiz}JJ-urpHtG0Aorj2z|lu;IJ?10F0&GFwZwO3Xtcgx0_yec
zW!=(8ElgRV{8rs}{OR~_=aX4zReF?kW@_<jhYy$kzXI}~Q!M@oCbhr3Zt)}LL0jC-
zMZ_=B|Jbyo-C8h$8!7dW5Gv=R26cuO&N?Cls;>OhC80SeeDhqFOm6_z$s!$aHpUZ4
z6MVMcH2LE0$U2)IT6BCu^S}Hy{a`+`@$st>NAZB#8pG-wy`rRjymT{(TFwn2?8Aju
zdY_Z`%S8*R!A!4xyky}q;~)(U&B!k0=)He`wu4Kgzsw~ZoSTFDSE;Y)=|u8boafdq
zcyl+DLnc<#)42Kx*_cj=!+2OIr^ekv%dU??g8u*Q6!Wm=7=Aw7?&a71x-H}<nQ6;=
zc4Qmidq;xIcc$C2|Kr0D$R3d^zQ&Wo=Y9+MKohgK<6_ZQkmkC(HhTk4;UU@U4Bz5#
z=beQWICpv=QhuoM?EMXghS6~S%LwGznXQY4x5pCcU8e=vi|K2DMX(2i*#s5a$-Qe|
z2@u352l4p+CW12Q)4l?WlifGwD%SK|;YVsLWiR}XxAM;i@qSHV{jFFj2CBKG>UWsV
z<UIrVLyALiU@yg`T7S)31d~()!odu97yQeH9=l_)FQ)pW(K`5A6kS0KZQS`mDbI!|
z&@l-!N|g{Br4#A)<)1$cY9wKn&#4gAtiCcAEF3_^?o6$dv^RIRH<FF(gG}i%xzV^6
z3f|Dqs05NztjrQM_^z{^NY*WL=DGjz>6=u`wm9#mNPbzkmDG9M|Lz<4pLc=h@>T2J
ztCL^zQ_qS#YTN&z96IWRJ!gqSisHN~(|n7u9*Zw;h^H1XT{C<NZ}UR=$+pA){yy{Q
z{zv42^XGaZYye=9Jp(1M))X$km~8cDSrz-I(kK`*Zp*>0Q{*(Q-p*dK48@!F-4M2O
zFyRNM<2o9ug$nIi5aRBIsRL7L<7E%Q{PzXEuq4e~!^*PWzb6v^``3{6=}Q81CkIAg
zM9qSShgya@cRmKJuO<skX)cenwIh$5y?)PaZH{}BYIE~#ACy<gkjB$|s~P?b65CHb
zbVh_({@YC32(?hKzv<K=)o&#31>ATFl9x(YxD!9us?-c73CTRhUydzN@z_Q@(xKY4
zGl}3aNpfGCOXx75;nKXS*&#QJhAISN={=+>bn_tMY)mHF;uE|jssGzS`EuC2Zx87g
z)iX8kD|i4e9JRa%t)Ab^Gvyg;MA3>235A)>$AdREfoBN^qsYb2)Wy1gUQ7^9O~2|7
ztU~N}CxIwJvaDZASYowiCCfd7K%2UG63%J@JIvlv@o^H{*E|HM?{xsX^hbbE;qltt
z-LhMvI?|_^IxGaBaWd37gaQi{i|)(_KFiKKi(}2xF4=d~p$PBUPsOcQt{K1}Y`WX3
zlzP15aRP8|Y(e2abpBgp+oOR3*>|Jp)K1ZPZXiG+fHyf6B<wP`hckFf;|-Yn9Wa4h
z3NDduMXU^wf%AFx3ITtHyx*}#?t<Ai-~C$9SQ-uJT3rpRl<2AQG&=-?0c|#Php3Ho
zAyj^!zxkZZkx_g{ED0Hbxn2MB(D_GIushUbZ+AzogyyD-!oLvZp9E8UXyR0AxQ_ix
zQ1X76$yQ2y-V`gQkL(vM8tXJs7SA_#neJ|OScXcKWtLd9W}W}!-w+L8orn+Xs-~4x
zOMNh_d;e`%q~rD%o}#>08RJtG6SBwo*S|meKRU?2NUAZma~A*10w5z02uUW5VLUc{
zP%q~5W3>zKwp<^6J#-*Xu!dw)sh{kW4l~ju-%wAz9>r@Gz-QjZTr>XTuYbRH?@u&v
zVjtGEtm=@i@{8ALeWaZE?&7WD??mFhF8m*V`Og=0lMGBkYH*7EQkzMsM3@0>he1>)
zw7lV@X*u?7z^bCCli|?Q8Q<klOIJ0%^6Be8F>T-eAL@_Ieik!C<_LF}x!55gQDf@?
zOfF)^oO;xryiD$mVn=O%6Zr?r{ExR#O{R|3;|6@-EjX(qo|7G>ej~QKE`hkAkdG`&
zQmMZtVEe7tc&<Kz*DS21vHH)OUbs@l`)E5*={jnrBQ;P2a-5z#FLdJv7R##qu9ye{
zRrED}y!OAQ^n;h<Cik4QKfu)X_KqCsKt>*Tj;#Wm?Z<oZUU=0kZ8e<r*B>QSNKAnE
z7-OUU=ko?B!si`1qte`+sOA2iCD9wTefhyu@|=)2DouFx+uydAKkxI`ziyHtkV*-R
z@Tus_)Gbrz(8!)qtCMovkBFdLVT!B>Ymay=ph=Q&*GAKf{?z}>B@i_gV;0U{HFzun
zu(qCHNIMuS-VKuJJ!LPsX`p!bsIcIFErP`#c%NidxRa$_X#DvVP0Vu`UmupKUD%M{
zeAkgI?6NKgw6UwDUnbfUM`0({V!F|vWev<XkUtvmu)t*jpnVyvUd2p`mg%J&-~ZKL
z1t#(?1h@)1M4q5`#<<+s^uaJZ$~^ngyGRzbCgS7|Q?oI{QM%TlNW-DZ=3};{?+{&T
zu(P>xiP-|wnq%*wUJ5~$8FrWh_yZxyqXeeQUIaiQX7iDb(#hr^V_CqlZg0i1Dl(!!
z|EYnQW%eiR9EpjRLl@bGLXha#zVYa1973ilXY1TZyBGfi>JO0)-S)xbyk;$Y_zNw<
z^CEchN2px#J3CMxChWVmS*~B<=JX4xI)}}`pW~BF*-dGai`d-u1X3<?kn<eXs>QK1
z_!b}!csx+C1~&jT{R7+ox?{@w!G1bEb7nZ!i5eX@9LB!DkmM1y@OFO|*|0B=`c0sG
z%0n=%niZxj$e%JLu=9b%tmS>SjRz;y+3N-Yu0S6majUW{*<)kT9%>QU{!%AI4IBm!
z>D^Ea6rpZ?M;S#lX|L$bd_7u7$TN{eL=-wvlhoRq=g-}1cSnq4q_eA?gMi+q;A37b
zhk}+#0zM6+8O`*cvns><=BwFHm<_24x9&7fd~L2@(#P;y!$^IwWGX3zD&IY+*wp@4
z<evZt=Us%v<43+!9D4SK{T;yEoAN776OIVBVV1_4c`Q5G+Q&U9OWO%WcKVb3WkusW
zeNx=uVLpi;+|P11<PBJ}w4oCa?XhYBEC&d5$p|5*^s~awO%4jWHz4aEMSM^U%H7ml
z``aCIq(-p1qil=BQhIya)Blm@|L%{yhe#sNNs%hPPk!+I<060z#k&`Z^L67^p!Spv
zWJDPQ9XdwQCqDarWz%G48jHUS!#}<A`|Lai%3a4kKgJxl_QUu`iFsiF01nJrL}Ngy
zKq!7_q__jXEW@g*5#58NE0oJj)BUgX_MkhqEt88Tr?%*hyC>=MXnE8ppU<jW&a^dV
zHq(&d-#5ryO_sNKB)b)1-t9C=HqUOsi^%{Cp+nRM4a&_}m6aQ7(0}OFP4T*nw?vyj
zQ-W#Tj9ih*pDCuP2qQlp%AwX^Q)DjC0v%&Lz_X^CGxQIPNZkwibMDv3#qAALxa+o>
za5`~tHNQ^UC#79rNB3Y5@8yNBBpS^uBB(Ufv#c|Nx8EK?ask6LYTcAQmefCq??3;=
z=nai|a`K)-T2CxAP425^j-yZ(+?Az@ToBTCGm!0rkKTa+j=X?|Q$30D49`Vw6bi3F
zKPv%ceDkh&k6dgmXBO0)F~gQU+0kK)%RyngRLf&;mQ^>A&Q4Ysf;Gx!rXu;k>Z51@
z)U=}Zrk_uN+sf2NVPV|_52}}q8%2E6lWR*qSuJNVso(7u&56=v=SCf4=id43M?8md
zY7FCE64JxQic)H5Z#JI9<5d5Ah>O|c>M_+F`AR7eIvPcUg=5jt)01boItV`I1Olo4
zlk<}f%W!E35cY(y<88}-MHhhAo}Js1Ll-;T!Ml`{Y&NuDnbG=l(*qA}{gO`hjx3e>
znx)ZJ7^A_6lW5sB!%r^?mIEnu0(0w94y?Xg0ySm-P2$&8=d?Nt8LE4Ct=bbLp9jdd
zf(-t&{SAYvzwi#J2WX#(v}EnWQ+!iZP7QdCn%tZI|ACo-v4P_DtPB-yLtk>U51icR
z0LxoRr!bR9-G7JbaM1FpzhcH9G4Wf>b&#rxIF%-xaf}qOnOaM>2H>ZeTI<<inng=2
z{p6Liw^J$F?4jzdhvw^w<SXiH#1=q1bP(9z{FNu#gLk#pZKqEwb>9}KGi*Z5(zKpF
zk<T5qSRV{UmLNGynz{`IO{f|}RkC73`vASAjG*|1dHL-yT!I-u`HK8kPJhae{j@JW
z@vehQ|6CTj$hrAV;u#BqkJm?9!lZ-KCOdk$=<)8QpWF<>&U|N_-miGasCD>+%c(P%
zp>_8l_%_e~W#;Kt<qsFNMr*Vvy92WWaw`Nei1du~{{Wr(rQy`wqZ>fpUPKjtnJ+f{
z{3>Fyysyy8Eaf1M66yv%i_Q;k>oG}q&TlD4yAfb3h-i1zow4k-aV;a}lxA0_4(ub0
z9B$G5Q{?|SIm|;GIRNO$v_2j#QruH2HEWH*tW<LZFrql%Bjo}sQPnX6_MPyeK@2f}
zwk)B;HrZ@pAxq$<0NtlyE}<25$Vnn<Abfk}I7#>nKM@pj;||}|x|DG$>fXe<$<F)G
z;D*6B=Nc`wm+J+QQqQScK*(`|8GHad@&tTuC=6{D(RAqX9Bl{<UdE<Wtb#s6m&a|{
z94jR7rwTct!W`5(L{~m6pqnY`JGEq9%z|1<8ZSW{GvPpOZ8D^{Hb8Dc3M<#u|HIy!
zM^pVpf8!;jq!bN=Zc654N`!7v$s7_9Dw&6j86sCj8kH$yW|52;%Unt3c}`@WD^rI0
z?V}9$@>$>S`aNqs|2+NE(!KZnKIiPS_da`g?R~a$X|(WTPVMfsNEdBx>M^sI2d8_M
z$Z<?nGioYEH92xaOXV+Ow&^@4S&!ydY64W!rp;hCNb+GnPO4s*92mQ5Etv>@S5pIc
zbp;Ftaj&Yq`d;T`cLBD>15#|$$}`5Nc@j>`m5mS`>+p<*Sj<%yzuilQ`?3fo=5_8^
zUv<|#_P9~QDQIIZ<$u_Cw*S^SgnkRvtZaJ9r75bKWtXD#LOxIVC+8mMxo-=u$z1=q
z;ia}|xDKP{r5wOgVl0PS8b?%g--8Ym=OxwOu%jVt2HjLq@3W^*fAnO|i_Pc8US23I
z^AM$f_A~UnCJXgJ^&>Tx9B25C^aFBYJO;8zrOuxFPH&F4TTWQHdBSRk<sE{$W!Igz
zN$<p}rO8%B)qIb+-h9No^RDfqTBm)SwDNJduIw)8K0`%u&&(Sx#VF9OG^-P=H_hnU
zQJK3-w^GR8s!-LmzM5H`XU<G#ZOTl{xt9m}6d(B7r&^UoknWF=OBsm$9@|sufBv)0
zSBiRR@g^3B7!8Yy6w}vt%ss;%V<S7qNZ(K};w;}$^yYbPP$#|G@H_&RFg1wVMBaP;
zEp0`I+L+<_z<#!tP|bRSHIEM(n~|3#&J2Esek)r``@gCoXV+n7VmxMog070o4c&1R
zoOjSqHV2qBwDI-{k_o-(U-!1Ne@eVk_%?3$@UhMTa&m_=At%>8;u~e*J*gSmTeCj9
z3z**n{D3LtWx-c>EutrOq-kld6fd>fe0^wLoxF+fq<2)_&GUd*?=6xDt5)u_NR-_&
zpa%Z)n82-H)?`Y2$z@PT@YZ@@J@NG3;|b`b%^4I#sSjV1mrBtJd7HgJvS(zX)KCI#
z3kkZNA-J%A><YS^z#vR<w*R$mh-COW`>BhRlESu=R1F?)uM`Kan(T1Dgo6YB^Nnx)
z_QA5pB2+Kub51`VpME@129-%UAK?arK1jsY$6vM*ii1<lW<uX!4p~>2-Evo_Kka)g
z%m}?@LX$sff29GGK*g6J2yIECUL)?{=UsN#8_I<Zh>&=~*xhvsLc;;Ijsw$AvA0}b
z_s;d+cP*i&(1V5zQL~$CIYedr_`bc2B3iN!5{VhdH<^q~L$vMyi{|r?zzEQDNjBdu
z>v|2?CF_)SLc1OA8E2B7k+vL72#B=%R!MHD*c5PU{g%UpnR7jG8_(0>);IP;nE+ck
znY@F9opv@1ccd(|yl8qY_Jui|O}-_<W!ClT77wL^gd0`6GKcFlDR`2~f5xq!OXI0R
z9t>GzV9N}+OvB&0hL^O=vcEpMJCW7`B6y2qVehe~og69^@uKo~O#P-H1anW8Jh*RI
z%nN4p3G#|ufSG!^p}TS?J;JWFT>><N51>qP<29nTRo6^GAZENdYP({TEdSc$O$TZ`
zFVG5yO?IfAda!FR91LTxCXC|MGpcy$ofu3tAK1`pYcW+g#8g5{Lw6i+CG`^F7)XF{
z!*A2LuYC9jJq3nhc?PCATEO_tu~Q2v|57F$n0XNl`8Um`(YOiPTx1bSfSN@CwBxsa
zwgwp&*lJjlr*DCqb|5<K309Bo_{Av78MrRzZGX#QK8wf^K5Ym_J_S=foqsaszNOY<
z|3D3QaDoQUAlFM_GrPblaRdNgb`$f3mG#%Y%-UQR902aKeJx3IMJ}C8<4uU!$ngeN
z8Gs2NG$awLm+-(*W>A^0_%Z3{vBAb_>8!5011&;xt}}4p+VaOk5JGQMHaN7}sxdmv
zlI<3>7GagIb|1%Se1dYe68;3%&JFrYJ~iceq9-<_lUaKM?Ixhbt~>N`ZZ<p*{`^%4
z18P{6Jv3>}bkoVL2`z6*)-AUlw!T=u<`)oMBE!5(eR8BZt(^FIuf@<?+aI${x2YP!
z6kGb0(#p1(XqAJja*}t8pj0|e#E(fLnMCmZBOjf86#sh-<;iwV5{`;)s{mP%6y0)$
z_)M+cexE^t$|ns_bm=FY|5^P>nNPgE4?rA_LMRZ6LYJVgbS@-7o=>;`DMC!@O*^)Q
zatXE&-M17<<hlB37jIy;5%`UekM!jPFXSmmOXJ8WbC6B`Nb!}NQ_JkacUjo|5EL^}
z;8hZZZpV&qh(rnvoF0i54mYI~(}k6X_^~o4Uq!(@^G*qX<-FA+t-vzDZKX$t)5W0k
zX^+)C2yC?QwBCl+9Ewx@O4((hG!K<uUFZTJ!DCCUEji@szKWqw!A04nW3uFuE>qXI
zQDRt1kC*=G$5mfEOZ^30uRD(^^Xini@8G;(lae!<Pkwc{B}2pNNkavfqrZUZLDwyN
zHrokN#8y&IJ4}ylbufGSB$Vdv7)c?BeAebSfI#m;?RsOXA3~c$Ddf>9wm}@Ga_44M
zR>er^Cjc>-2|TgRd+fqeq!ITHS!wD1PNIbd-6aspvLDKZikwb3;o(IT6akfMXgO6k
z$fG`wS0Ulxh?^T9s25NibcAqSH|@l+cH)WN<k^NX{loyKS_mv%{6aTZbAcIJgMRRT
zv-+w3!b$c`gzvQg=u;%i9X%dsl~>1qGCT6qb#F)%Tm*NsH}+mbrVf!obLuOj)=d6R
zLGqmkjhyXoD1LJlN`P4Ecfh*Z@)vut`^<vlsgP=<8gEz1_*$=$v2spmP9>Oi!`sap
zlR~&I^4S!nz4!(>`~kJn!;4`jtn&u(4jSPK&MmtRVfm`VOe_1+EiW9@MfM`H>y*G(
zkeZ$x8EK(6p9A}V-<Jnkh^&`uZppClAM}`>eWfbGZ(ok?*J+GrC!Vk|BX0R}{W#Vy
zybUx#)fR4nyQ>BGuv0{!J;r82)37+AP|X<^wE)fOrYjJ-y;K&Wc;#eR@7N{U8(~}<
zZ>;lZiGyt6QFOmuZ+&IECxmBEDg*)5G)f(!3NHN3E4|jQj(m6n@MjuByvf{#bsQOk
zNyW39k$+iQL+W_+#gefl!Ge<zzO_E{QlLT_WTYJx2T~1cB`y6})gXU*0A9}HPpdZu
z7U#3CgTJ(jp>OL)K<^&p%9%~7Po4sei-E+JCuBivH!~;noT_9^;ZdBr=uVT;dBg?m
zkzJ>Um(H=}Tz;1<0ARXFH-4JuSm<N#llI(J1U&!IjNZ`uvwm^v843n9FWS%VDcV@-
z8Mao`qPt{+?FE^h_&kyJ9S%R+j~Y8=-~I^)36tsPx7~a~BHkiB$Yca=+%h?~77AdC
zn9WQ1>!*h5eVC^KRkm~_B$6<G+r%J^O&$@31iPb*<FB8((HGyOMvt^)BtO{2ogX_x
zIyXh%`=UZW+mcp8#D^0+rKjisfxcO$FNLLu?0NQp$wgGI8Rn@5*GqP}1ngcdwSdJW
zIFW{y1PE5g=lz<SiGbX!<G9IowN#@g5Nn?q`d(`CwH||g50C}q(3?XfK?;1rb+5n2
z1@T<H2dZ8XDUPfH-2s#+LsD9kX(|iJX#j9a<^)0uoJYR0S`dIyh?J)B@O=Rp{Uq>^
z6K;jf%J#pg_|XQ<8KLS;vo`+u9w@!Yv1400thY&Fwv~U$8&*kHg9O6a&GbT@wbJWM
zLFug%y!OSZw<ZJRiGPOQ<1y*vZ8p#_-jEST4&sPDK+T~krxePRMU`co(30y$=GB*3
zf6og8#1r9lHWq11;i6@K5pS<pvu$qum9`|5y&2?d_s+jVexG!mU@`QwVH~KB_d$Vp
zReJMQN0C?@|1^I>@+>$~3K`~Ci$J84n_~AIb5#HEh@PERK78&v`!wU_iNQvbH?#eg
z_u}Coy{}QNil9rX30gQf_@pzfhYYMKAm)Yc<kC%+j(<!is32t3-+6r&4ggkqA<NN~
zMi#?h+_>Pm=jQ`Y@|qhcmqNbr6F4x2)_t!p3Q_pT7H*Hxi!4E1_g?zytnWvW>_CeG
z&V<v+4lbY4qZ<#8H-1)~o)}`!vY)h8&WPIrI0lsNjCX(=!&3A}10W0(5SQ7{7Ow-b
zR^^Xd$i~)`J`HCmgBu2Xx6Fs(c~a`&e)w14x=mJ(OS1*jXaOp#v;Ko`WSeP})O(26
z^>{sMJ6t)2j_KF_dwt+SLLa4<$}(fcniOS~5u6hdxd0{T1nrno>nJ56tLcrrXVd4h
zArYDfU4L|(*uuc0eb*8)7bC(bb<hI6w7d6AkwXA>w}6Q!c+fvnilyH}g)9aNFrKxA
z{OBq7f#}F|ts-twqq(af^L~qUJ*}2zgyeQXyRjGNli~i-QsIJr&(sSLCekJPnnjyo
zgz|#geXms=ji?+p*9{)S-E(lcK5W{Op4e-dowpGu!U(qJguwoc$*m0@EnkaUSGU&p
zt}P|Hl2d2JFa=l66hWuIf!K|ij!>Xh1ZrWV{#5b+hgvE}tbP`sSQ)g-*(Y?-JNw!$
z0h56GpD2_Fb<(F%qb(0~8>->+5tpEs;B)hsb%?iJJV3_WP`tipt^LFxceQb!B}rN6
zWQ68n!%Cmc02;F%E~}b`7C!sIZypzN)nTdfVWBB??6lgp%t!1vfl(;`5@Pa!^T-EV
zFsW@j;TlhGg$PN(mlx}OI8+%Sqw%dND(|}=Kd*`zV5+pNC@ySKZkY};ph3>cps9rU
z2y~>*DD-*rr{(1`<%yITMvQ2B%g?T2pUI=AlBeLd!U8b1-`7Mdc0tFl#{<K=P15uD
zdkfPNWkdKOxTN^1$FMOe;yl&v2E|bE4UKPYlLsu-%hp+seC7me+Y317$V-GV>^8g-
z{|U3(&x}O{^F(JP8e-(`yxK<fP))eEI!bO*s26&+dn5m+HgVbJKbzlttI3z=<GleF
z_Yng8>i#Y>0debZf<aIfbB;adhxUOha#?ocop6|p9ra1>fUvqJijt@@f!Xo|U8(f4
z%=jz7<h>9gVhHTtl%GzyFRQLzfdF`@abOpMQ9uIZJ-8=vIakL79wuwQYp@N7`|&3C
z{C=xc5SzfnZrRnvMk*MT>4BGzk`qajd-*QVu=;2^W+3G}?hbH;GVK&rh48y=zl@tx
z-DFQ|{cit-j1b(>X_Ca`hFpeM_K;RbSz<F}58w2xzZsB*6nc_j{lRY{X{{rvHBNre
z2cyu~=k^bU{PA3j7m$cL@M($a*!vGqViRyKb}5>)KLHyalqa@TPpimvk7I9>5$Lfh
zKzr0XWP3rphbr(yaT<zse1|r_Kb8HBMNZuscZeFPioVX@D^Uz*Nf+c{y&<M6^@vVT
zI`CHyKCHM30=w}-X%wZ7(TyenCvRz^TsnA)nL1mH!7iv5br97|GYLE&*z6*tr@U7;
z0}7`)AxdGPd?%}H#5p1QY203@KNSdXp<J|$7R4MK2~{c^FfF>k#F`XxBkh~gd58d^
ziExyA1RawXa+LXoZjUU6%EAMvU=Jm;Anjn3)UIbuuVy*T5!h6mc4VDB<ae~PzrTXe
zk|}SzabVK`<Z`~%q!>3Vf3_bHDYs5%pU(dZn);hzoZ_`UzO_%-5R$jryHDwISt!@8
z8##2=!@A3Y=<`l8Lx0@3?F{l543)vL`w3_ET>rXrgDn|~=$7A{)}KFm1oGy}g7-K?
z)?x#pU1Pcq`6qvYcwZvvb+&w?U9cgC){4^JHoZ?u{FAi9a;T}+^wR}^snQ*}7L434
zRi>R6EuR2zs1UeSENbV(9B1erhAEx3R!8fM`(q=a#Sw%&hw6m&=yS-Ozwe)ug<i>y
zcQcMJMnD&B5fEvf1Bb2;t0R1A{c*4YJs~7&S9>Sf40+^6LhY}u466ZKU0L(f<UO=+
zn&D^q1)4lu+Ly<qwA53Lb}Ed<&Zv@=7+C~wXePdco&OXCF<v?(UVbkQjodI{5?+9w
zk{oJE$&x(}K@#2_(?qwIC8g<zp4=ea!W3Gbg8;q;k*el5lACP!uQjr$rN-L+8gKPF
zdflY2CfY;Ky&NpUk`z96-GopYKJx8s+CjDK=VG*ZrW#`XXglNTt5_`qyWo)FFaKR8
zlxCG?bZpum>H{cDO0)U{s@EO8-m8@R2jpvtK)$zJ`B)@45y-b*RD}d+)wln*DA4NW
zgN(l&)t1ba`M-Gf2h&*d`9E^JGr(b37yAT)WkN4c#Qlcy{sa3SJJJMHhE`wbBY%Wl
z%aIKs7e`<zD3GHH6e{t}yovIL4!@mA$c0;NprEkyd_mB?O|g(nl7>X9Q8q`aD=KD2
zVBt43Pa<TwBlFIdg{nRR{9G47s)q6BCWd<*;5_is7#ba=kgi{2MWIwKQEB7_Zqk++
z%F-i^A!4q<zRcscj^Lh}Ltf3k4At5c!2$ZEF_nG#?J>t$n@WxV+&#4q^v#EBe!Q9Z
z`c&%R6M%NsGjymL=<E%w;;1n70{phb8|xwAp+I(D-)_j~%I4LmfeSVMA>ioh7H2rN
zE!&|Bn)-e>5YQ>vl>xUFWcBAaHbAlpe1Z>F{V~|Wfn#Mka6edF-`o(l2~Nbi4T>jC
zB6l6RF_nODR}itXKqg=-XIiVZKUDS|F3|jBC+nWg1II`umrY8q)ScRw=XH;a7SYjl
z9z%8wN*!u&X3tHi<vftoplDd;T}a?0L~pqgRpk1@DdVd%{UE-4nugQwOcbE*c;I{u
zcI__qDJO12unFzXN+Us`6i?I29_v7J<*08}Mj^Bn$=XHRDvkJY_jI!L0e-Y0v%W8z
z?|u&;a7!#Q9iJuWRfb|+{iXqubk|WnIPZsFk>1GXtZr2`UWz-Pc*1GyU&q6t^IiIQ
zY|a)-Htsp>@nfW9N@KU&fHrc&Ilmn%OGHC9=8&J*z%=Oy>}P)?W3^kRK~$L%Ak$Mx
zJyjMfg6<1|8la=TLTzx;b%K`JIag?1+ClXKkae=qIG7b;;~Q(<DVjoqhxa!jU{HGO
zi@i0eO6AU7LZ!K&8g~SroT{DPG6yHPfaYlbPo0uyTYu$UdR-H|U#9lv=$GSi17LCA
zxnb6!U$Svg4P#Gi5T6DYwRFHDlC-{|=Q~&xpNR{J4fe$Tf-L=as2eK=K-0<RKLeT0
z$Elmc*;@>WLz&!%t@ZJm$85nZ8j4iySIR}*Zoxz>H5#yddok1Un*o)97=+$*9pJJB
zA?0<Jd>)nHRNm5AV}A6p4rf)iVX!A}wY!?yJgoK;m>CW)vr+*}^Zv=x-C*>0CRkK2
zGxe9!CAW7}SWZrR#d1Qz;jm_zLCmj`9Wqyr0eHoi-_RSX81jfWpNmzgyzweVGq%4z
zsXgc>AQ8TU!&8K=5GOK%TAWMSk7~KJt@>YqeBIBctBk(}($5Mo>^M}j3`YUuxv~4|
zrNLoqr2sH^w?@Czw&qL@t@RFv#$YabmZf<Wu~I(Ny`~urMu!@o)MIaO1|)PMOZnFy
z{gtd=eZYer)yg~tv_Q0GPBuV#&u`qp(W^I}bIu{I)`_vSJ8TF{o3Wh0SlbPdYrema
z2_Z;|F7-sNZgOx2`(8%M#o5+q0E&K!p!`EmK=ssFZ1s0T?|o{aPqu#y@@HDw+Z$xs
zg}gSPFlFQ)eohoMg<~-5o7s=C{BuriByk+9=gt6>gF-;y{_?|kVRhIK#9)yh02_9k
zE?m_VVnzGGd6}q5+-8e|Mz`RgpA)YhHmu}0HO({*v22>3_svc%9!a?LZFjqQr{>Ec
zuJ%*xWK|F!%Se_FIbmk?+p_y6(d@yg=O@|X!65$>R-UzF$t-O&hVsN0;Fv0?Q$Shz
zs4j@xhXZT|fH90*6^U0)+YY7F3o|naS-v>$<fXfwePQBjh|Vxmmg<f|by3mn4O9<Q
zqhtfMr$*b)^!P{%yc==WrY$XLBgcvOrhS!IobU1@RN;hHd5}m|@bPshCzR0!I{p(J
z4dH-<U^V;Q=->gcd=KysiviqK8i4|4D3}I_1nukR(;1e?|A093+oo)tQrc*UsL~yC
zY-y>5KwJb!A$O+O!QGOu2_d%?bdLduhqtbUb@9RFyWzWI3Uq<YwQ?}dN(GUlhi;_j
zaS+{Xxw9fZxpo3zxQ8g%lw(<t3k6|-H+!(0i*LeKZPf0jQGd=XVEUN@&<~mCwWc6E
z%MSbJ4e~=94@bKOic5y4NR93Odi{9lV=e6&IW0MD5y9PyZ&^N{Gwh>arupk<0BWp6
zBpYxeRoM<R;aP1eEe*(p5gPLLu9iJ2(_I$G4o#y@ab)K6FEMim7Sp18tu((n%l^78
z^uLzTlKa;2J_2~TD)c0Iv{Lj_h`F5J!J(ox%mT0*1m1}TE1Rxm@5O6WU#{c)CHw1u
zWiqdn2d>I8ka0m(P$;)K{oWLUKAk982$)swf}Zv@H*s*boXfQAyKD{RFw#(swGaBi
zJh~}nuw)k~NMZtXM<L^<gy6N_#qc~k01@m|jP*U|fNMWG{;<hm#EE<pJu$XrInyk>
z0(XS!nSkj${yYiNx|Q(`a8JQLQ16BT<2+`WVfqv6H`B4e%^=~uK6U&CHSPf2L$zj9
zJqLQ1SOBX;azf-*nqj@{0qJLyNpK$}pF1Ei&On~$L~6wGiH)OgF!otgeFo5?rb@5!
z{HlE1A^rtbz2u2Vd2?`!R7vwfuEbl|_WQdtTd2ZzLkg7V_*u=Hb18kJ007<(S&FB?
z=?6YgR4iedaYj^R696Rn>et7lQfcJww_PDwOrb*xSoa%0!?(KKMD`uV;=Nl|<&_PL
z*9IH7X#kgt+5jX${N1+0jX7L1$8kf|Ov81>C|-8O$Op)h{^Z5ZJU27x1KvI71bx{0
zB~dnGr|hLsUpEY;6D$xkvL3Sk3u9FPr?Y-e>4PjxB{5mflN~Z~Ec;F#TVl3_9f4iC
ztB3<TSQ4ZffttQUVf8zGuCMWGP%!8TB3TSpE*mI1ui2kJ3N@Mgwk|FQtG+Q^<W@PL
zg)*;?9K64iYX`*|wu_ms5!SxuB6LzSv)yAUc^kR|ags5?T{bLxCst+5rEJOkU+x$p
z&}96_C#5sIFzZeUY$e+Z50hc6|2GfMzk!l`Es_1@kf6?b_30S~3)hsGfPl{HpN1Ag
z8*W?)mP0q6x0Zcr<otHp<LHtJMevuzVR+M1WAU$F95c6Qn%Hrt{}42**qymFWlHH>
z;Wu5BG%aqrX<D*_t40szg9(@&fKGmQLc}dTy3x+=f%>K4(f5z|@5G9n5pHOq{Y%vS
zZ=LFs)uP691(GKA&ptC>x3t*=e-$B$nm^Pn899|Rr7=;cQT^D0edpGsiFY?7dKZ-v
zrF`x{<Isco_YuFoynb=-7Jsp@l0-_t2<+%TV<*$Ivu$EWQPUMB6~=`xzcGE89z0bS
zG)24L?$(M&=V!(<SNKXH%HS-kD#spLvhZ-gj<X65R%Ym<*KO1uO5Db=z<s!>ZPy&E
z_ZiCFpxBo6w~qt9w!pk6c|?C~?{(ocm40O|wpNoDPdjK^9AltBz$0egp7t5*nT}=+
z=U5SC=fxA9au3n*=dB-`sP^0v=UT?z%XSnSd9Z;|OjMmc!@S$wEUS#!aLY|*1qHY(
z$dhj3W83Rym?|5=R4K{+tbEsgyM+ME*6l|e8D$PQrW`lPh*cLaFD_a>9LA35uv&TQ
z#ovDSU%w5eR<#c;EcN}W0)rf32oUePkieBaB;uY@D4Fo*_Elhx%j}eO$-7nUFCUdq
zUvRM*4<7m&Wz1t!Wy+J;W2?|zL$vkSG3LcbU_{(U@&+T3iIVvVE6TtCO1r<Yw3wm+
zx<2CVU*dMMfyRfiB8l|-<Nh={TSK&Udxeq6^<z0IhlhkYd}=xOK2CkLqw<FZ&vJEn
z_pOs@DuX>q(uB$e@96!&a%>N+sfYOkV2rEyJd(KeW5xU;cqP>TbYWdf%y=-k(T-GR
zwlqLMck^CDxp#L2|MrTR=b0xvM@#Fj#AEB=dITOXbkk5o<)i#8mmlxbvN%-jV~H1a
zE#@#0jU8ecBT~1Qf7uc%P7=^)H|T~x4KfVnb0S>Qc*EzXh{n>}L4XymhKF5yPV*CZ
zCJ2vb+_Q*$=U+RqdE5EFqKmsrek6Ik!NUV@iy#CjN<wsXATf){@UfScFB=GR_$P_e
z?V`NjVYb5fv$ZGT8AtLm2=xAQ9~`3hwc@%fT=R=vmWl*nAoF@vIN^;Zc9o3`IDBIm
z-5vXqlAUd-yYZQEaO=k^MB*e}_w)rXMO;0XQ|)mU8Quhdr#hrCN~sLZYJYYPgLdwL
zIrT6%ujY+r6MXhdI*pMkH_;v{3Oe4w_E*hNp2cYaaqr;s5fFC|QM=9}6$XI{`)9~b
zxBp54DhzGLn*=74Xsa#SGt2dE$$xb>)J-`95qIuGM08uyB-k5dxBVBXkVFnt7~Ds9
zXQucQP~j*(6?!*{cOqu4q~Bk*#LQ&OKm`om1m-2nz0yb|HEsS49;zTC+6=8MEo%F-
zr&3!P@g^aD8WW1B;8un)Q~oOhchj*)fC?v9u$UU|!4efp5EXv(o@)Qh_zkGA18))-
zA)>AG%FdG1EqRG^d7i#1)Od!+jG$~mhKv?Bt%waaE$NXF^)UNW*tFp%cnrzNPU(eC
z32|%pBMa83xH@JZO!xhY>F&mrEKN5SO?TkqsrGEfpD<kp{OKkUot@t_ZP>KG=OqXl
zkH@P*YbV0#HnO|dE>ZU^qAp!6ON$~OqV9ft>Yl>H%u}}nW42?7x;h}JviRHxk_P(5
zZ;=^zkkWq;MtjHp-Una?e|%>6NG3jxnBfV{{<4KQ+t%$LC4q`?7J|}Rc5G@t#)dB`
zhQB9!*tKJovmMRWK6s0B=f@e%lcA2(E$r{Ph3OztEvmb6R1B%B<q|x&_3o0W6&*mk
z_E3*{nAs2Q+PoH8RNF}h+&Sk(tqEhs_xCUsSHf&-v@>#DA!-+K61t0`CXIHjTJKc5
zM$*^&rp?B9ivxH|(mWY_ckh4nw^9xCm^h=_;~_YKmMKz@iBNXRzNK9|gLaL!l%++=
z`o=foH!^sO`xt;t6@Bw0h1ufPyR%E0K?0bGeuW5I5{E7+vLvL)IQG%pnfX}8m2EtU
z&yqA$cZ-lB<2fwx`fq3rO_v9zOSfXW`zT*6O&3b)g0yLhSXxe7JHmA3@TdC>D;)$T
zfJfr-Q2r(A(gJnY<4*t-nJP|iiMpE*bqiQdwKH4>5zeB(r>+eV_q+((+!AMB;$1$(
zyZGA+yqiQ+Syp$&UTXeKqd{ky%h?C{BJVoxo7xko^HJi<^R~q(shjc=arF>!=MS0X
z`ca!$IyGqReeLGqeNO~0-A#5+VJ4=0Lv;$sahoP)a6peEXDapj89Z{39i-fem~Dfo
z#Nb~H2%<7CP#OPp1*Tx!=YA2$WI9i#1DRmGxrsWP)%-5bEVmdcEy>%NS*N%(AIhU^
zgIX`^mU}ITh9Obo6OdSUJRt&wrb>0F(Jn}hquF}SKZWY*s9w^9HsynEGpKZvQ{OCg
z+4`Q3n46%<80}yCPeIX1H?4;5`tY9*aN3j~9~p_%xvhNf$Kge$ph9|GoFVJTz)>#O
z4kK*u<^?mCQXF1~IWo8)66xDNmKllcz!ARAm+J1hMZN2W^ls=Qb-D?EA2G8dES4J<
zj8ig9ZnD~%T`>!*jw49|V&1{vX}Wo^+2tik2u1qq#lD@cbKPC?belI`T&yG|A{!mt
zvBt$<fu+=ND%t~yyYLho4E>P-N71~w=;FGaGZ$l}-DY`Gdwx2OD`DLKo;x}}vc8$K
z_~&KNEiBue1hyy1sW4=LtElGREFg);9$-NZZUv*?J|IK#G7{HaUp*<aI=kdgY@!NW
z)L0MJl$*?+{A)zx#JAnOn#wRBPWYl;II*c5uEe<?L4s7K^4>wtR*-s9Onm)<(ZBXF
zyoh1rLfsa|jRfpM8NTYrrhR6L+x5@pc$)J6xdnb<(#-IxcSYEWhv~>QIq0%iP$;iB
zWI4t~K@4%rF5Tfj+8C#_6Mk6xoAlc6zwZDJ{JWcUbB|nrsVn7WDlNX7vz%NX)V!d0
zZc~F7xYzZe>^>*#CgO>aHm+hHE*}&?KgNDOj;|QfqtM7ic-t)ovoo+8ci}ZOm7z;<
zq%}m}&`qtg4@gAZvf#_aJ-5R9)^fQ%)nUol`gd<7*20b`Xt`&c1b_Q3N*i2-pc{&T
z2%2@=?g@a(Cr?18*r%!<UxS;XhGmEr-qBd}S3zA#0*}j|7?fi|_N@Nvup9Aj>F)6y
zidyVV#K;27Nz5&{oro;_cJ2cj-f(CT(X9cQqy13N=?MtJ$ah&59%TEIkZ=pzjV*W@
zptQW8MDt8nWzS~^_~YC5OGm4T-qL;IqhG)10X80lx2Jko94DrfxQ6>%P*!>o(j2Em
ze@%TQ&e@mw7;AvIi1usA#l^?S_CJTwiUcBC&UVAAi{8?oo+cYL|G56P%wNi2F^@=W
ztnYu#3o(RM`VgTs2<nCzX$q~)A}xC>_8y!$6Ge&5RfA$`(=DmXTrCM|pvAVU;k~-n
zXkhwJtmYINn<GAAE&TuPzdJRN@EfN0hs>$Sph-57zP`SBlYUeYpdL%M3m7$E0FPf3
z`O<<0w)Gy&Q?>G|%EGje@yHC@IyjhNKDz<9`{tFW2_hX`^7)(T>HiWPM~zsAWPPC8
zsWmuha;Q?3=&yTa_$VF#PP32|^%dO-6vR#-4UqL&onk?cfaFP_SF)Zd^Z0x6d?*qb
zHD{3}bv9WH@?JvffpnV;a`k?192;rV1h}|yDqxP7GG%~G-XXEz{@|vuL~YC$b|@i^
zCu@%8fT^=RcDy|j=p(k3w99fqAh6Vk+KdOIPGRnlSbmBDWUv&3^>?BIuLzASJBEj-
z$q|bFq52BIe7{#5d1L-1cP)zcLNfH!#85LcAYr{A^D2oWBDvu|Z9HJ+cky(y2E0Zg
zLF;&9hDDTM8*1+0vfIR6J;VGbeNU_kWWWUr;i}?@vtg$q;{kI01mfj>I*@<o9WT3L
zd9geP@YM(>(PsglG2Q$%`@P+D4}#l;LyR-qH32Fr8<7`t7-O+v*&c<W&#2`EujV5n
zCy4F5N#D~YKl^Zl$da1f0Cbdav4c&1k&B9M0RAkHCyX%BdRmXX8AR+pLiD`E4k|Fh
zp<#qMgyqmB1u|;M1>g2iklDfY1PGzrO^1>yIFY7KXaMU0kUx@5k1C@R5!?xi?oCl8
zr1|WW8C6SIGi^XveXZ6Witih!F|Umpc~E^mYF&5$(*u=S<{j@g*g`W~?a?oJ-E?q|
z%so_pi88Z&8OA>VQ|CVr1nm>df%Pj9`1RyEKMmWi?Wh{KXBX&C%nVo+MWTjr^Ov;!
z3DIEpgLJutALx;^L^9z({Po3e&<|mG!Hbc9EE%Yv6KfKoJAeKTl}yWM>F1&sQ%zb8
zh%5$iv`S%(wtOl&&=e;1!Y3at)1+w|wyN`d&!PYo9-7AoIh;4B#84qxfyoxyj&#Be
zqTDh7u49Gotr&Fb0R3i+Iwt9*Vge6JtuuSg1&x$lQK5X$1S)Mpb>%_EJ)fXj?!Xq|
zQK!z2PYR@yYoWBIG27uyUp2(v`}La;+7e*Cx@zZxZvL{h1>A#7LOIkrMu0Z%p7x{H
zgL<||<6w?s8qAlWI5w0KHeX3aa49srhUhFh1Nz^oV{pj!3eLMX83+3Q^Xa?47PaYb
zsxOU0H+RJDb;4>B^RYjY6^Upxzqf<afar?LCYsAc_Ot>fMv?7R><fiH7H26w*!G6F
zN}t+Y>I~_OT~rNM?89lSEhV`)0f*}YB{6x!JW{2jVZOLAVLMroG|tS{YzO-{tdL&f
z%{vO^o!doZ*Wvh*-DN8ys&!HzIb5yXo_|N9+D%^`Fn>qxld{c0^Fp1>1c;|MB}?&A
z_ED?%Cte@X@W?An_JF1!ACbS3^w5udaT%a4&%$SnpNfjMsl(|&yeHSDUlc4LeZF4u
z1yg`h9gfH^qG10)vx{C&b-1vZZ~bS>Y8i7_ToJVB<v_qjb?DZo000iT(GtM7={EII
zpQn8m)&;5R!@hPUA8!6emn+ysCiDYnvp1C!(xBPBR<C=NG*PI&^{RK-FTFdmZdM;?
zbZt^9zm8>IAl<Y}i)ZA>?pVzn4K}40S}}a1r=!~0G~T916#=5=WYnf=RQ;=(V&Fy1
zrt~rSH&h=TI*!iUv6@8i;~GJa$xYZ<@Glc}MilW3-!CY~gS7pouwBfPxy2zzXvi`b
z0&`@y_rfW)OM(xy<E9N`&`=}DuAkMS1kyk(aEn$z3+m<siCM)c8VA{cBYPn$&fiRH
zV*!^?7NK;NLUOBK9Rq^m>hw<zHePYoEw~o(d7&2opI#*N1kss<ufrit6-wg(pxM1b
z^eTW49}3YCv>sHEBi&OPkZd<@5I+YQCxM>&2SR=n5;I|nW`g&1Mggt+8ntV%l;ms+
zb!bXA3yU@ZaJ8>!NorqH%FDseS!w`k*L?|bSL|UF0xA8X{^wfY`GW&ThR%$=60#oD
z$jEws=tH3loEf4Ya2kteMmst)9&IxQEwPUX<wa0Iy>=HuylklQ6|{&vpYZ6%4OHI`
z@Rw(7#`SM)Ey{->5u$GLX6P^#br_A=d|f;Mn1yGoGGkG7mr*&u7Y&~qkD&$-tl5@o
z^;Q=SsX^Ne4phL7u-}pWP_jc^8uwKVB?5$6eY|0hrJzXthQ9=$h>-RSn$``n{op|Y
zJvPK5vjKV?G#3k&^8fMCh%uT^UVw6B;^Zu)Qw*2bTG<z6<U-#@RT?CshPo;#zEV`t
zcY3JvqtY%gNa~gq^h?|aSmehWp0cVYh4)bVW%NMS@NhHanRxTWg90CU!7kYI3?lt_
zFlW}h{PW*>7UNW|zkpZW=NB7O41M5uJc&BdD2@XtMhZYbLnh_sGZRDRP%L_vwt+{#
zYJUc_?i@M*am>3RU|?{+>#15ei6Qu<eH_G=<BOSI73{2sr5Np--C$sV#lf>6zV2s1
zY4fn1vL~h_J8%eU4)xozv!aRsr=`!+Xv20x8QRR#{)GA%=~uEj5ywRVqEAf`$*LUt
z3>yFb=B~EW*LrAb`vtci<#;16_yh`BeK%&b=qcM?N*mq<y*POC1R)Fm<8}2K`oL_d
zVrT|eGA?}d_nYrfSX2zVYLp6Sz<rQl8a3`{41!d%x>fDDpxgslkwI;*eh1xB0Z}tT
zUWD4SjK#Yc;7r<z)Nn_5z#$wmEx+X5PmB8jYWURcUyHno151S5sDJ@<TY4@aO4)%|
z69p5XAO-;vz?jzQOT_WeI-%x?(HdEq&@xVH%!XpZj2-$BcqqyWo<xn@B4?|N1Zl}S
zIiP=)Ku<#T4Sjca(vb!Pbw@4Th9U<WlTY&%xF-WJu5RpfdBjQH^;xhfWd|Gwgh=WE
zh&K>j|0M$T2$?s&8$h{G?Y<uO=Srvf2qjh-nXWOILS_$S%<}nhXYbV5{N83w`eG9e
zcVyr5IGt34ePSC0{rCw`L2NUVBLYKwROJA&q#u0-$bAV&Q3}uO<kH|kfK;RPYCoZA
zK=_3egUB;bU3zlewHw8Ex9$sv{RuK?9d1Yr)3xt-=T?<!td@K+<OB&X#WP*q>3g!w
zbx>2~d)U}7>cTB=NX33_)OfA;L=~DV{1CCSQ-@OaPS`p=D-!Y@e7?@8JB)q-#EZ3a
zVET=wW!o2hGGNQw@YxgjPP!s=mL@&WrfXRbM~D_>FVO-IQbvO3cIiYVR63s62vBCR
z3$OJqH3d3MG=|ccLEXgG(IcfbVKnweP^4o2_Lap!)Z?!5O9<ScB4zoBy)-v00<yCJ
zKQhu)L{=oUNXgZ1Pdo|`duD)bzV@$*lG|eorH`c(sB@yG!?>63^SDjB4&}~PuG!B?
z`z#r{G4MwC=#0W9J#K+w#;4E`fCjL|`d5A@n|(cntTmD2vS5O|(v>{a>;fqv?Rd1*
zfzsM#?!sM9wA81a!bQkt<auy8pk*B_RU6Ywy0}yqv{XzbG1y4kdcoj&3l|$i4x#e{
zS{7`v3(!erg-(18xxRp6l@*;G6|2AU*|vw?!{Ky!blwq^B?fH&vs8pFgjyo|$f0S#
zj*73d8hZ<Bf!7#y^_xKrXi@QnMIcn;NdkP`aKeY>><QH1M9`JAzXhs`za#vbMY)J>
zQ^T52RP2OsInMXT-hxZ#g;e@$vtK=6UXQjsmOT@xc>gTudu!w_mj$se%aXdI<VR9y
zVd6qia?^;bkJxf0ge3(bEXM3A-HOEoW1&{M#8TNWq6eX;p_Gf*$nd(}A!L9JMIyYq
zc2OnA5x?!HqcW?o3;@t6Mmqg<AawRsH+#Q5=qnig%(T~duYRJgj<osKVGd`mjq`oW
z=Q{P3(Rt8begWfFM?BKtfgra4kFZB=r#5t>0}%CutnPktnt_|<)`!P?xIhfNlCBm@
zK4Cmr;(4*D91vjTL7z}7T}~;VgWKDJ_r{+xf<q(~X;0%Cb+Tpt)m?B0sLsmDdVfJ@
z<<y!FijxqKkPADmxxf+7oI>~w$>b$2BC^}e2?$I9*jWA8yh8v8*f5X_Zcr;-l%i2%
zsOk+q35iL1VH<hWeGIxmyMqy1clZh9#C;C~sP0=$qZfdI{Io&<<gBV;(@p~LZ(++`
z_YHJOR(@7d&_E_AR&>JV;zVzy7}OH%r!5I<bTwj?4|eiRl8Wa>tyG}QtTYZGx#5=K
zlB~hjH_*pP-~B16pnk$x5a&TwWD`69i`H$cu+Or3F3g`caP9>jC_WHQTh{P-6PN4d
zf%4h~8OJ>UlfZi`XA`^tg9G@zjF9!9p7pV@3bD}R{OvnRt-~5E!GqrtFJ^h~6}0VC
zw3`@Y1ytCNrl{Sh^LJ6C1;}_Z7qkhUuUvHsfKb9GRMP>sCdW5kuR6o*tDlD4pk(9U
zynyJppxCWq3D7qwaHzzS$Mq^%b5ZBrWJeJ3`?8f*;7|D(QyfONq@H{^^2Nv0{Q`AC
zKy-3QDfImTO!dt87o?dJQ47J;UyCVG<Or7_H&0dYk<=+*?Wbl+S1jeg&sk0s9%KU^
zs5k;L%y{tHSnpY+b2v1!TLNZK0VYDBL4km?h9MMu@|b=WW>tFOuD64}bofd!R9R~S
zIQ+mHIY8%TO*9_PRL^nDx&iR&(BmS|c8*5e5`4Ex{hZe~Oa1t%N@uSWLs89DF!HM+
zWp*o~zIA@l1$-3}w;aagUi@mdNEBa#TL_K)an$}CD12xqUD$f0m0%UytU5aq+F%t0
z$3vtRtOE$R1+^08$}cn$I)jW~1qx<?rS!Y|31IE;3N4G5d*OWAa7O&rH4+;^>Nx75
zMHX)$2Xk<SJ9q&zLg@hGg_@T9xpC;v&+h(-i**J3@8h1)$hC@T@6=fsXf3gV1n?g<
zt&s^o%KYzj)1!=A!{wUq6*_<WC|Wh8J*?>RG14)Jtjs?L71~;IV-GVH4;r9QC5ma#
z#su)5MP@_IllujZU^Z-Cf9K$$+q0G!%!V~$3-SH^UNGJn{(zQE3rgGxyAdgb6SqoV
zxu8he{;d@G+rM<lrz^Z(L}y&)0wx(ObYl2xjbO)MJhyD_^UKfz7+RpOxR{s4ffyLV
zO~~A=!4!`D_TUPp1+N62E;m`L^<p8E7`_e0i+1twUqUrV>;;l5=p|&-y2I;k?+-nm
z4zab3=kHlBZuKAv$;|DlSoim}ce7!<N`dZ;3%>yJiHiY!^<J2li6fynd-~w!nNo0m
z=%_OnqX5K|5R6*0srK(v?n+4W)m}5_U3v{G%HVILIGjd{Ko9QXd}QaHe8k!%D;M32
zyIHWJ#69cm7TC2C8FJGf)YmRku6oEuqBvLP`vMKo|MC5$98j*Vw@&yhFa&sHpj@3o
z7UKF(|EMQf6Vk=kVD2C;3d!qM7DKxF>0Q$Hjl_0>N-wZ+*W`{bfr4>H+Zt~6rVn_5
zsnLEx4&goA?-FHL+<R?$?s(jrbZJKG=b{k0B}Ou}hB0dS4wAzB-mM9g`tS-UYIc;*
ztv(S8TX~&%jG$i;MwM7{M=5g@b&&^0T7d%`Dx@R`7h%7d?#I!XLYZ^nC+8GU(0zd{
zKvT$B!@T!=#d39^&6(Qe8s19+yvY5ag?Nz{=j9#!<32{Tvxen>t|#lWvB@HeWidPV
z7ew~e>&Bu=+B)jzc$mHyHw5*}K&=;+2l;>{Vg99)ry%Zi%R+F0?G_|xf!eFkR0LLT
z!=h4zSXXs>r_KNP=o%ar9wB}jBI=7oue-9e+a4;SmZue#t$lG>^PeqgA_ib~Vd_}z
zb6RbAS5!qv3YV-AZI4`BL1A*Aq&nh(yV<#qb7uDy9QeI;j@<0;HL=Uqb2l0$&fV41
z!Hq=f)TsK7#YHZxo^jvkP!)aE^_KpR0S}Uw@67z{uq}wK(!|6HLs@|9sm%lzRL0Uz
z#WgW1TIII2_dW@F-4oBDTF&Kv#|u3uAFgGh?{KN_g{CVxhEfi(RJC5k&J`5>n1D^l
z4SJ)d?4Ii5T{<c#xG>RW%YqZk2B|oPQ#jG3Y5e^Q8wr#zS{`*@l3cQP8Hj~Nv6d@=
zmCh*n(&C!6?AMF$c0+-bjz{z&i+O^D+2By5aLRDCO?8ijMFK>?T$AvX;VApT%2UyD
zU=<E`hNr^61X~mnKDGa^-_)`gmRdHdlOhE6qeuq`AHuj_L$hdcue`K^)Ii2VK&uT@
z_jFb~2Jj!)p`P@Lg;eb_g|G$}i)6}5tZdl?{_-+6IxAeJ)E89N0-X>aU}E5%;$A);
zh@F2Ie$FZ^K3g}r?q>cuas4ZKN-T^l_fmFHl1EVQrY7N}AY$xd+=jik@nmkW`4ifc
z@&^^<xAJYnl63Bn<abU<b>3*~e4KbclZ!cd8l{pr_TaIgxpBJd<-Q(+CBtPUqCbM_
z*#(c5ms!}~oGdRNpdlhAqr_mH)~tm;HeJ9b=vL%^VE+66#V;sdNMaMB*ld4nTlneU
zuN<~g4)&3U9bbCK!ZQXem<GjyP?c@y#S0(89~D_nT5^-hLHk#XvFqU4yrA|NpIX8}
zt`HO3e{f1?TJeO$DQp6>uR;?U{up|ZFa+;_Nj=`c@HFlK8QBD1$fpU3KkHqX*0vp#
zgJ;syQ%~W4Oh&1f4sR;Bxh7Q({~MQ!K&S?Gy*a`W))G-|Ae)fh9~L-I(1iiWD0gjy
z#f7kVoFrTvF&Q-_<shq5B%n(1-X^~Xo|vx5mt0Q0KO~rD$23%H%7twu*bOFfvI);k
zCPMfxLJO|~5=w?0-bygWAu<@__zPMOg4trQ!JCPkMXX*p^Diyo&tP{*i0wC4_YqJK
z<|uCg@ABLE@jJnMaa}Mgj<&)yyw%MwqyW~z`!*%noM6IVY}e*799u{5DJ(+?23L3$
z>wv#-oD164@QTzk_`3xA7YzHi?!HMe{@55Mn5t)tJ{SJS=;<L?_7JTo6#-?%m|>=t
zOvzd+=X4O}lzIF6E&|fZ%fT3Sb?;*c$N@jI+X+I`T0urII`$5Xy}iTVZ^dS41;aP%
zy7-R}jB!X0Z6EpW=PRBFW`l2HI!mPpC^B{u<e2LL>E1t#v}|rKLV#2h_b;CzoYf?Z
z-Ku8$bj282>A+MjFEx}22YCQWfo5HY_lhSlG%zb|N&S8JV+8xc5T&tiE)r6m*bf7Y
zew(d%AO6B6R6y?AJDO<mKL)OChc{9DUfUpw|IIFI82D0r@c4=m$fWnf&OUhn@r`A~
zfDjG~T3o2~jukwGk;eiJ*Z3X8f4~Xr08jKQlC!U*ZErl#_NK@~LjB^&0XmW=y$^5X
z`5lv_0-f-%Nt|GeTnvms8e$+#Fk27}A&`NZkd1YO8%qtkja52|fP%o@*iv9z0}V{^
zK7vQQxiAgMH}<rx)Kr^b9p$5SWuG?SYXw{nylQsS%4H~Q2gRC_G;tE24QUx+Tis7I
zAHd%w*gqWFKaNaJ{C&cC!m<@&BD3&^MNhv5<?`BaN*aFvvQ2?7(+aN=XM&l+Bbn47
zecL`=As{VQ1lG{6y7MI?0XfJhdqLYrT^HL&IQsUDWD`nvO$}CV1`fX2<XQiPV2s>g
z7=x74=*G$?Vqk?aPX)LLC{nQ#7&rS6?d=t^6zl?|;^(->gs%(N614(FEKR%4tQ@0#
z8^~pxohm26Acx=>)!lSgZRHc%FsrAm5fAZKj?;nV@g)yi%MehVQqLI}ohruYG5*3a
zmtYmQ?;9H8e+*pf0+Z#NU9Sa6=jCH)H2|SpvYhJ(&RzJ?p(n7j;yg_R8;c8pQ<UZ&
zkLrGc#Q~$Qh0%%hGM5j4CO!^oDq4;+b`{#@0Bz^S`3Uukd=f0L=u0G_#E?<4$N-D*
z8Bh^!#w|3f+cDe(vxPbKgKATBWm8#4FkgAlZBd5;AK?=O&6gjJjc8jO$72FsP{{_0
ze0R7(sD^=7Z($t;YDd`#4Z0wB)i+*(GY&?SI0uTg_EXbN{7LEEg>{Q<H1;Q)6pj>e
zLPBpUf&3`|r=5xRn>azpAef4SAo_9by!ZoP4q3uX#m@7PA6O}U3P2DkmdAvo9S1p%
z*1Z^ekANJYMXtk!f2Mdy_$no2lJ3opUbz{hY;Xw3Yuv(DJ8Ow<f%2qzo$zht6EU#D
zXvxoS2`GZq1c{l>*P~o1OSC|$ngkYaaRQoHg1W8F@?u^&hKm%)WqSYB6oNr=cftC&
zF(xaYcnU0iCergT0p0yrVTh8<W_|*yV}myU*=1fNiQwyc+-(rQZ3p9C5z?HP7H!|?
z9;YFEO5rX6L(xoZyl{+g1gBha7=o%ukB|jjPQc<=K50K8SRCvo3D9t3a38@U$ZAdl
zBZ#IXY+Fg&UqIWj`f5FbcVUZ&avZ+<CKQ_q+#Xf$c~>sdcCE4GRsQ`tm53kY*9m@&
z_{c>ZW+k!ls_!b_;b|zeiKV~;`pQ{DD-9ijEe6e_B>{_a377X?X-^QcDfmbnu)#XD
z&8zJ2ib?#N@-j8{f1S!&;#z>uxCE87`_g`GR?+f0r@R4BNt!CgDYc}+9_=edl*|%%
z)zU7V;0#IGY14DGjRsxk+7!<lL4xi9Z-+Z?sfN!)zgVe9cM$=H>}7e%L?F1x&b5B{
zUG#Tr+uX(Uu_~!M_Mi3%P!URH@HY_3hrc(hSw=fJjW2=K869^S(i4Vv?3FKC)%NeW
zq_a}mJ|K51vgWWP9-BwQr%Gx?(0o=<-MonDczRTm&@)0e7*jzEGd!Tw$bd%~3@d!v
zUrb{-T59U3m>xB)pt6>5&KQ08O?a`P1)gZShr-b$5|dec@Xc*Vq^hJxF?1ztB%Cw$
zDXjO5cWoq|_+YNW8>f{7T<klFL|(=wNjmJf%7pj!dC@F_h3!|A+>K|~<06s0{j%!K
zaHiovCqsLXUF~}SsN&IcEfFjHC|v%eJ>JK-QxGDkiEnu`l&Y2vE%E|d6mzOn2wiim
z)(d0<W!~6<Zv)B9AA7ZJ<8jyTKo1i6^r)b&M5q_TA3IkbxNOtaMh<*u(hoTtZT)b6
z<#7A%wkeluat>13R|%e7g9|~1VvM;e2mT(A9Z!h*oa1~wXQuVlwQ{&9P<`ai1$zA7
zz)G=0FmVG-Msk7>Pay0;AcNK0X1M1tzdq;JIHpChuwP{bb86WjRT-tMjL%=h7l=;Y
z5?N1o4f~Aes^pdJ(EQo$8sKfcQg5~gRlSfQJXa|tl);#3tZKsd%){B_`5JzBJfGcf
zl-j5hofLab6YinWdj`eX8=nvKtV8#Wr+;xK4``{DQ@XP900eijI{0!k8{V?R_X1sG
zADtt%?~t5qGK{|lxE=CIYIc8(6|_BtEZ)r7!}z;efJ~EEgL5SI9R`s298pf!jqp-t
zOSQ?u-&b@-(W2A*V^T98-V`<^gCHdKpQ9OcWB&|ImGc2=xAV6By;?agLgv6y!3O!M
zSlQzlTeuM9DJo~j+!8UX9kDI9&j)M^ltp>TuYAE**df~U=DYD<fQ$fE<-E|mAm~jH
zzx`bX0Il!K2uIcfyX`I$nN)8N;y$>_nf3ul@<5=H;Ji2BL?jW0J%=W^`svW@xByy}
zh~?yqj5v(g7o5+TKGz;}Iilu1qRDb9RUq@|gJ;tzd=;}DKE3QU$kdPwy}nSV8Rc|n
zJ2K3g>cpJ|tqF4uL2J$bn4lw=Xg}=9pD!N;>y#@G(oM~-7@rq(#4x8o1Ujmj-m!-p
zx|EC*?N%@j*?w4$ho!CYBfR-4fH<=%*E#~{lmMpn^Z+y_8_7y*IPqbu3Gjpk=d-_i
zVOn@eKM*=?AZK91@59q9Yl*}TAfjo_ZJc=5I72@y_{C{mKDtpIQL1(YrQBnIXObdP
zxbX;xoPa2mt22m`rk?-)`Mj=31d`O6i6jsEtN`rg0=>~rr|Ukrzrta}dLYJ9cd*AZ
z_HZHdYfN=qZ{NYT<DmMkROmHO5r3_G=JRCsHE1`4BqedBq;vw4J!Oxkz#}ELAGFAO
zHsQNa;d5rwBwRtq&jC2QhWrm>LcPJa@_n+$m{#9Kq&W32*ADsO(cc9^-cc$)Nw*KW
zz08118P%94$JkqbZ23L9r5)~R-!{yuY_L)W|NBUwemRB#0Xhkd`^weYgP?t3P6JWl
zm=HAjDA<`Zb@D&vhAc^ilp!y?xjjXJVdW3Ufo?I=^uxBlXWI5;kKW-T1A1kwq*n;C
z!Kb%1;k|>54|%34?~BPM-g&iJHygk5u!km=xXa}9)dTDHOkvqucq1EEVXk#Z#O1Zs
z@RT_2HGF!nHaQlk-wM5PX+O_)3%EMou@{}n+u~Aw^Vg5}hvNvGN8r%h3UnG7_i+T^
z&CL(IF2&gIS4r(VHeN~Bh4YXcZfn#^BzfFt^>MH3ZFKFt^ve}owg%@3Y^D$wmxa$6
z=!}2=>4J3vbKG|Tkn0DwR0EU2wKBBmPOZypR*}io!$iS-=uM7m=hgYmiR9-g`663H
z#^e+4z{DCZx!wT)p~I>Zqec^pjS<10Sk?T*)Xu@gQrp^twm9^PwOc%&QRgOG$yvlM
zC>WzBc{K!2r(wVR1|>;7!=!|yG8;+)fKi2^$=UnEha&5tmVp6XqQ(7c745pZFaB_z
zx4P$*{js?|xdOG%-@TDgZaq!66}s!yT*j<IuNn<trU|AbNailrg%#*T>Z=^ONbf9S
z_KLgZN4G$ZAP4m7XpZ2T%z)}99&{m7vBL)+yvOFPs{(kIH^3?xKZ>T8fKIcHdZq@d
z^EaUnXo~D`-}&^Of<3>k9RWb<{k&d+kRnV+UJW#SllO=G<x>MSnF_B;ExvbQfO^o)
z@FLWZfP6#ARjgj$MTcWoK0Fw0EO$(OsVlO=fWRE0gPkil%lr~g2-XsX!N+0L8KO0u
z4OBbZMy{2gJx#w|^oMVGr}?)-1g8;9zyY|TUu`!#fys}e-o6~{MzL3*{%o50*v~Vd
z{Gs^e)wM0H!lO?f-`J1txoj2g$IH>po0vF1;D(yFSntz#COGj2kh2Hd`-ZdQ@`BQ*
zQS%)mxSBacvpwjs<5*bxJ1V)s%^wt3ZbAygX`_wxIUeB2Cz&>kqWvcJMM@dGZUVY3
zx~651zj7d-q;nXnY`;w<tFra}2;M0>KOF;PWq$2YAqby+1SMCbplYVp3JP*PcMFx=
zu$$pfp;ZZY;cYpGe#N<6pLVUB+IBc(m7Gi@2{`UPY$|uwCD3Yt0ELXVGM59xppe1d
zXsVO>nCd2(6*kKnoFr@mTbF<!{sw)7v)!YOV}gZt&{x>jyg@%I@Se<tgnG`jmTe|)
zzkP7dD!dLD$*=jSOXSUvJNOiW-3YFOeM~v{_&uJ`ufZV~>#IzrJ}d2DpU&VVo9?fB
ze%nmFV;1G!B$aXse>ortxZmJ>eN$ZhJeS!rA)W%5WF|Ujb^CtJ_{tbH3yM)&no|+@
zgP)PU{<&e|b^^c~UEw}5>UAp(CTy@JNKRFe3_()Q4kaPFnH%;sC=U9Btx)U1Lhw}o
zU@1ZDZzFg&)R)PfMUV~^T&YB2kds;VxdeZ!gWX|6_a4p*>S=s??`4DO=@lD@JqH9m
zSLE-5zXR0BxFGF$#{kw{4&6$HS_{)r(lsS;W921)T$KJ)_CdN9m*o=M9<4l1SC7cE
zAVYmqncM%b)iF;9mv_H|dD5?(=Q%V_CQ}W3#*mkXdF~3JvWIP)Gap;A?@ExDR1D1W
z#TP>CUf7uvMzVvpL8Wf63rAPk1rWxS@ylS8FXcYMJb})du9Z6Wg)B@*$ex!V`ga0$
z=1xOC9E<eOd`vCYqSOwq2|Yq*-Eb-6eAm067HuIy&zP7EsT|L?PJA}N<iG?RlWXjt
zM?c00&bf><rUnmx>Gk+z0i_xNV|sn}$G0fRI>81iGWEW&B`>Ik&EOdR%R$JDfG|co
zU*UfqoC@zs;gr!Qr&=xP_CIKj6b5IJ=yNrW_@@pqX(PO~@!$o5tuY2IWT56(wc|j8
zB5jJ>il@9lg)4nBB}j?8Yl1JEqiP@m_QlH=b1SHs1aD2c_?kfao8VnFseBQf(yJyX
z{--^F*oE@h8N9Fu(w%=!1w8^d#@{1EQtg?RoPN!Lm7I!<J=UBe0&+OQhRo_jk3-!0
zRp^8m{&7Y|`4N25BI`<mNPq$01{9mCJW~&W+FHf)f1w0ys^I8Bz^1dXAwNqM2KF|1
zTsS%8imyi~G49Ag8w$m@!O3bMiV&3=!D@WTStR1D|6RhT<fDOOBW4faxjbu$e!_-)
z_m8L(gG(~*HEqE^b1+zp8tiBDNp}3tU%Y~M8A~ZP<*$m?zy>P-wIqMa$l{afI~>!U
z0#dJBwd4%8r|elFD)R==eP#U5|5{6tukUP_@H}PAi2v<B0V9aY{}Zs~I`f}^Ew|JD
zco_c#?C+`c?*v;eD*sNf<;Kdt6AbU@{%0aCH#?O6Up5il3;ZrE^aA*2F8=e!mh8_(
zpZK3Y2HN7EKZfTs{tI|4J5X`|0v=0h1EO930v^k3&wl}r<!%Y)Uqo@a9{s-?Q4H2g
z=y6)NnG7!MjULM8h_%PTec6~2KmM09n}1Js!d-fF&?C>|&GXH8TQ)D-wJ0C<)r2sb
zrvmSc8oe!aU>^iArCE&$LG7zUxZNK?z)>iL!=Bk$<m!EYLS#9n!XLgjb+jW0q*g(2
zQ<7Ou=frB&JWg0K1RYS&YsSl(;w<6Qs_3hYN4E~Ls<qr8Tp13fs_cGL`{M^#Jppj6
zC}hh+wQ(f|4>oZ1N|JCy1X<4F&u&_BjUC+tc~@exF+j;w`#=ATKZn0rFb`k()rtx+
z6_gNsb)jxLj2FSi(8H&e2C5UX#P*+d2<%zTR4BF|ECAnso*lzOO06Y|fRL@HY*5K<
zXOSY$s#646fd43H>qBTl#p0yvj17vZ=|2e`T1%7#DWtLPC;|LA{K*V{*FHeK8Z?}Y
z4-n5$d+(&mEce7HB>-#99C*Pyiz+WD>3u0dni)!B5UNDrSQZ{FoUjNu5#hni8v-40
zB3^IBdnM3fFMLhG`q~&TO@S36f>taPaKVSHu9&rgUoLc>+4cF?M4Hdx=PT1A#3(7*
zS1M;p@a|<+z_R79LzQS`FOTC-6Q9EtkBHT|V)YjjlR^m-FQxJ=yruldi~oaK|MBZh
z2!f7@B_+tO;PdLjtX3D$WfWNc*e)pJD95T$5M)zU@X`D;|G~J+-?tP4J%nTfmq6?J
zA8T5qG5P~*3T%+|vu{Q9(9-*stQai}{CM3ZewP1O6W%lbp+oRSUz}ks^mMQ%%{P$$
zUvbjvvV}ML;tX*(5Y$b#YlBxO*3#nUckv%XFK57i485E^{xS5TP@zBmY3Svu?4O2S
zI0ffL=AVXMe%!2Xpsm4)ZbUb!t3$)&OVHKy67<m9wA=Xg**&M@d*o3agbG|MJkSQ$
z2~yX={dA+5^;rjv(%W4wK{Z0Zisg;rwqHYS`OvPHL1U(=-sM$lQ_R;H6BB5J8-4Bl
z_Vc#^DYp+{vW=2U54T$W*n}s(^M^kPoaftv{qfrgRP9tr(ms>k<*k;FP{j;4M)M9q
z_w<B3(%rA~X>U&G-rNDWdcMt<pnqC>Vj%;$E1mhr^brM9xbIJ6s$*x)u7jDqp`*=h
zzfx0L%N-|@FMW7GQ)DuiIS@w<*H)XS!>!o)(D6{f<oQ5svg{Fr=YyN*=unN*XzPfh
zfVBT%PT_CwsKUy-xibMOS0LCEKm~1hYo}#C9REBwd_B!sDQw;Ja^3uAY4Yq~`CRZS
zfGZA`83S6+NZ>ZwU&@qSYHicXcbUtjE$51!&z@;;m~1Uzupi{M>yX?u_U?<^><n~Q
zm962Ap4j!p?Re0xUbnoU#PdNFpBl2QvVYgcJ8qwVauUj$%&qEx<}^!(&bj%hngZ@=
ztf*unb?h)TF6FNu)Z3F`3qbmUmJEvmRo%ea>D?pETu=!{@qC5lJC5LYnliEYKo%Ei
zj0;nl=*-|~;$sJn+3!9%B`MSOZ@c*~71Ced^li)=fj?`N)n0AQ3-aaFbwLQ|jI}SQ
zZ;4!QwK|VLlljo?xn(-=?R0JYX6U^r*P!holMnZ2wr%|KAV8(&m)HgPrPz(D^<N(!
z`~10Ony&C(6!u}(!MCIN*TyPOyV@v}i~For*$su-fK`q@O&{yLXuF|5T$C``JM>-Y
zt;vo7_*}@EQck`Rnp>*j?xX6H@%}KV=@3ze{_2R)>C02w_y9~S$C24_kNpA;UBT(n
z2Kzd+=~a0&i9KTA1VB}%X_W5}m<rvQ#nJ*;`Mp73_{`^~3Xg3zwJlaVrFr^r6~pfj
zl=81>?M1$S+;<6j{a*1eIKv!Rv(Z0m@G2MiH_zZUK*WCy+B!&cIuwUDUto~`V3r1A
z(NWx%*Rrbmnv-zlK|Z?CRpZxppEQ9q(ZYJ8_w{V__JfJ2$na~nsbm5(sZ7V2(>JGF
zZ#LMStU3Dm+at3bu!Kj)eoaxzvC(E6XljGXodR^7C&TW!H_*8B&eX*jtBT|6Q)KTO
znC|{bchiu2jOvyM6dIW}=!HEzv1w<D=BjN^i2D(~FvpLeI&G6%F7#5}cKJ}~ToYXS
zSpcO*^`AX|3X-aAh8s;OHGOIv`I#E3vwNm&Z&4g3`^JB6s9rqJXF}Jd>xal>OZCRO
z*UzQLa(>U`__oKVN3Gi;d<X?UuGm|2muxVMO*}PF6J?`N<NtXqGU4~H8CNgm3(;RQ
z+)Q9g(yl?L`|CG28>;i#T~^TdC<tSd*E?r?VcbRBQN#?fmJOibdV9?U8nrzRnSOpt
zsaa4vw7ozCilruczsgWd7(oR>JE)A^R(JO(Y@E39>&Npgt_*g)y}v;D)Y#26K*^hM
zQ{At2=v=-T8nx>MRzPpq^4m5%HEzs7yDsO1{z2j&n;UyH>?}f0XM#5mbXuh#>AQG_
zgFyO@e{nrFdui(bWAClQs!ZFrQ3tURlM)aW6(j{gI@AG1K|s1eLX?zl5Cf4ENhK8l
zr9)!TN`rz(OLuq2f_>f0_r4!%IeQ<UzhnQik3E0AzT?cZp68A$&+9zLi#Er7V#<5U
z#Dv!BeX>Y8OCL$|+aaZc`vN~3wd#EMrpS{E*T#sG10D;&@pf8)2^y~4qLqU{WE$k%
zCADyNucLOV{$i72<cME#d;fU1BzNyqpW!_Yyxbi-4b-;hG(el1_e$v~j+G{QQXB^p
z0S^I|ZOOC<Ce_+@y}i4uzWLiX5IrW(YSs2^j)v7jCKJ&$Q>%R>bcg`H-kHZ_`B7Ha
z>niuQsM~Zt1a?oKj-WXksATo7^!rYSh^NLqDLIVuTkuLz*5D8_8DH$dfln7--e0<_
z&Zcla+$Y9)(opjd3`N<9sb+Jjnig2)Y_5&pkbNSMH<2zLaz$TqpCZHOi03rU+%T}i
zOYP7ONBYY<5l`Nqd!0^5jlYOn`t*CJ+qD>GBg-#?N(7K~D;i*jT{iX6Z)fP&5u+h5
zqf~wP!o!McW+K}^4e!L6KV_9N-r8PHX-WknPObG<M?w>2V<EvzGWmt=<DV1Y!z8U{
z71Slb<}6^cjP!fCfkd?;Gvz}snj5kmWYlbH)v5y`byA=WpCgWydn~Yp?cSo!>M*bI
zDJU4?OgVstj`-lKEkNtl4TM#5bxwVbASUM885kfh<@ysUMUBblNpVLWO3pk4Bk_#Y
zWwoY1h!<GhA68$AdrIrWt=(b6XrKfv{v|0c*zp-~S;-^y_0#<s-B6|}fwT7oP3Ae4
zHMEXT<xLaMHip3DA8S6ch-`A{y>65~PE5;KxaLQvKxn<WsDj3Yf1{oTdq>FyyI8W9
z(&(th+^gV$Qz??4BgAN&I655Xzx8gePagpyUgOu;)yOS7+cHLtW7<yh8iz=k_Rn&}
zDYg5JXQntACtRdmw%gi7HX5{DRBcNryN+IMH_AxPJ#wzK!89THYTsG>1Yh_GDp1N9
zsp6DaTsIFmqLYS3A1_i}E>sy~)LgxV=0cy=d119tq{Nnbtc0C07orCrG`IGA<zNaN
zyPU-puEJUGqZQ;B;7fn=@XG<Zr!*Tq`)F@RyL7KLXS>Nt32^3bE~d~Jj}P4-^519>
z!pg3+xL{k3mhdYp5)~o4a~F)zg8~gB7$C2*PmE@kN*x9(V-n790r$w-i>>}P|Fu!p
ztvOb@HTzH4K2KJYXPNI>rk$7b>X-bEusY9rmViviX~D&hQkhF!Ke{%#42{!bp{Qo!
z!jzcVta3Ll+<NRjCFV7oj_5t5VKhDIUX;h`G%B9dbH`C|vz7&JlV5TTw!KbEN+>ZX
zPf7FF(iWQj+Bk}9)P(<VMW^5{2tFL1Eq(PO`I2nxQ<u$Q7h>>P6@U5;82G;RV@%0U
zCz(LjiVQ|BF+Q8LFi?{9<cJ`5n9JrWb+>Q9i@33k^`b3@RuGS<aORxaL{p0^J^O^n
zQG_r96#CYVMRV1R&w&&=HrjaLUZ`=_gFRC3%I1>!KS#h6HY18Wrp9!m$?|*l^!^ZM
zJ?yfc?ZVf)z?Nu<h8AhCBVs268#c>sF(#cP=9@@WxNqRi*?fPC0fcqv^T*^<#_rcR
z%qN*TG3NJEPm@nxh5N7I&e*)nRkZr;zQe|UCMYaF-L`D-g+)rbs1k=y8wGc(8%nzy
zU0*<7YUe_|6z@SJUy3OmKkq)L{sI9L>E?pe7Dvm<<Po6Uo=s}d4PsLM(bX}_vsb8(
zF;d%oFgDa=Y{leV=<Ta*=Qk$wg$}WWU8yH|a^36*@GMV`&>ALV?lHES7yq&EDH)6f
zCJBr4|3M#dj$_GV`~}?x({%NTi5LC+ORYM+FK-YZ7!J45s7Rf)i~=>3=aX6O$s|f1
zDb%W6UCFR`myLgZYEJmtCv2w4R(w)w<wYEGS%Wnin`<`2@LsgvyjUnX4sK}w?D<GP
z#)z*O5UKlgWNiw9uFu(stOn;Y&fAu6WD|Ax?~KMLBp(F9DIyTsF`WK6i}|W0Rg}bg
zweJ^iEqq26ip>5Zc+^x%X~O($bbBxNjq)(&YQOA<VP8!&Hgim4?MHC($gEj+(Oh5(
zQPD~nYk6iiS80Q5pW&Onj|{UOXM-UhVMr`k`k9mk@m-dX;M(frLoD|Tt{Q&2`}O&9
z(H1tztUZ(!Gw^CA*`^?&;Nv1sVeA;(X3CaL?6ivzWG+@}>wn$o0%4|XSmaE}qo*kP
zy-gi;#^vSNzBYWzgwf#{MVCdzvwMU4NRs*zlFK7<GAu`bG~3K_K;TY97H&86S|^lm
z3TEV|-LeZNjj4^<PFjv9Wm|vpF5u~Nr@c<(%RJ4%*eV;;G`BRyhkHt(IQq}C*@Xv6
z{gwJ&aLjV}n<U%j)`$tt`8D4Z($Vbyl)uTAcm<4gD##9GOPal)!$??`9j$*ZxbXJj
zO7HXM5Dl9yq_H~o9_sym;|TTg!4z`I+@UvyqLGGMsUSJhOitSyyrS;-Opz#0SG`^b
zyVycwbN#lOsX~vwt`}S`@(&&<Ukr{kPUKA&CN(_nOUTZ}^f4x(K;R*8U1Xp3HT0n)
zO9CVFy0hP6_GDyQiiT&(U@my+63vY=P2D9=Z<yjnqn$aBTB*_<!7juJ8GjR6;=Hd1
zU6W65j@FamTo&Y_++cQr<I_wQZj~5&!WLwERO2et>;?nXkDS`gMHg-<(+bgns%v;M
zpWs#x@=&R@yyH(%-lvw$ZNh(BAC~|0vbs?#WPl|iWa^ExoirPtX+CL)kUY7elj5=(
z>~esO|A%u~!xLoMmtuYD%%o<mWF{4jJmY#LEUhRfU|023=xD;hctzhRi<0~F)dI!j
zNs2GL7cju$UlAUEWv;WbdxR8QuEEwguer6TsS_$T?YY-Mbl*zR5y*>m+{cZIs-f-k
zpsA4S*Sw7II*flY?#Llgh~{+MovyF+<xT$H_w9^@2N?<l#}`vvoU+wb>hucUjFY6;
zii~G!K>}a@R7V|1^6x@iLcNTAbwT_K&DJc9uVLN6rNj|V88K=rIW#xPS8k73M*2EV
zsCPop1`G-_%Pw2XZv})&hF52v3-Q;dzMh<2cxww$+{YVe?%qA+jY|C@2nIX@d4%Vn
zqm<iUPwwt<H-WQ7EhE}#AXf8pM55xm!@SUvYI1*Fgujn}5NyHdN(C!kF0^pxOd;>3
zy_X@to3<1fGe}4xMfr;YJ$<2QLZ~5B4-Rp4TFO>Llc_zt)(R0i@6Ylseh7$qW58#%
z4y8Wq{nBh$SnhnRgb8dKc{~Nihr7@H;n>GGpAMgO2jGQx26LzB(O6?shxVL?dp~KE
z?xfNc5*UJ@MK7e(##Ww^!(kSZtuSQY{(N+7{y!;Gq<0D(U%%HC(dppjyC{2sw6!Hi
z4CM1#>@!<(SR+hF6jO({XZc{#u#K%O>WP4!s+CdgFOlt|M<*OM_U?27FbREzu<s@p
z+pY}YA4q<_6<V|!`mjk}@fn18WsDOUkYRf}U}YV#^34hFe(v*3Nmqdgy9CZ{OBdBp
zT_jD{j2oBbC+`k4Pm9JmiG2Cz{&J@y0>XGkqx7naw3@!_lu|6b121pz{_OAC>=)XM
zj!wefv-=!@LbB2p=}w5Pl2{T_3hi@17N7*&{tv-FTOzw*7~c(aW{Ub(nX_8o^*g@{
z9SN{vW|eD7(UKxsEPW`o&N%|zRE0^0dE5zl^nt|*Ilp{qF=2$`T-8yjlvTLx)O}T4
zooZ~U7ErBt3YNkU<7*0=+zW}$or)`P7PX%T=MP-C+m^0J?c}mjLX&l!kJ{UqYc^2g
zh|tEvaS{dNZ_zaU2_FhqznQwBN=|87xdvDjww&}5V!o%g$7S-hM=lk57(y&%*h#iJ
zEuUg9pIj`@WkChB3U?t1dF{Kz*lGcK0?&M%CUmGRh|rOhuNe9@FWS$YR$+=$TSU~j
zT2-$u^oHp8Y<p5$2>%qh)g5m)LL}8RD3}c%)eTfTlKWqqJP{aADRM^S61;2WFTaEN
z?S$m%wj*^blx9c@ArNHc%9lIengKhKExGz4<p?DP4G%w!_Te=cxVqd;E^Xg?NVBz`
zb?@^&wE{IW7dAqFt3^hrUP!)y@U7zH6QU)#yG?04zXbMwKrZFH8}|-IF}bdsigVeh
zaG|pKA=$jaMwAXb2l`G;1`Z9OT_=;lUHVMt9bU@c7-T&H*}RQdcmC%H(9d)8mlEW0
z1k!<Na{U_SY$C%gp`|Il>iD^_-7^Z(sakuEQGV`X^e8ALJyd?*=)QJiYgw00MRFSD
zi#t!W<u-`QyL1#MNE|-99Sm~+@aRnB`gt36o7%7LPp!j;)y1cH*7M^8^f%41&Cc!F
zzK}y@#tRApY3ZR2{kR*?iDx0p5G^cO@dCeb2CQ#Hylh5;q;<z@C7tVg9MabtuI;0i
zTMRjIrS%F=V!TEzF@9qC=#iFh5#B$%d}&X)4>hIJ+QJJId_uh|<Dx0j>#2Zo-yZ|B
zRU{Kg&d9@u$5&uUsWT(fq?0!bSQC(y(0N9pK%cujky=7*Rbhc%Von7_I+$}wKydA%
zjZ32~h|uYLj0uY89R`XV^|Yr-@@526jRvXu=1RAPt%H9ViQ70&7`wLVLhpmUng1df
z&-D9CYk{azeV)zq2#AM$1#>I3R*T_RT%PKr&0FEP3bH;VPxc-0;TZ~vPAg&i_(TBo
zVLTvNJBP}6AGV5)n;%F}JDmV2lvUm98uQ!kT6NCrlVmMw7PlS2hg!8H8%;nqW(I;8
z3f2{?HpP$rh`CVlV-E5znqk9@))$#D5ly8a)ZF|_0$VHv)5gPbo6ibx@eUFvUEykG
zk4VphJQNSA*39OLxjnhss`bPcYJ8LeMVP^yi+ZX0pwpEGX<<1qJjgDGU8&{PsUI8y
zN+2usq;$5oj|ITb*n=btb2AhY$OY^z2j{Gc$kZ+GKt^%eYP~<bUWx1t(Y?WR*%OBU
zYkd<<dvbBlJ;V?yXvVaObr^$F1=UJFs5|@yveDELiH2;`?pfy`IQCpY8up_rkXYSn
zsr2D!dmWp`_VKX*=&`t4H7nDlJBaIjGH$;L>30@Ngc)PvyAM!I@PM85V7d07^PvRL
zFPL>ZEeG<>aD*wk2=QIuY`!}>?Xu-ylvYN%80Ry;1SgI>At_5m0@xM2BV-dkM^xQy
zqK`U<cSePy_CN;lsmFJ^3HP-(_CT4VHk_{l^d!r-u6IXbfb^<VFIZh6w;YlE2skV;
zHDz3ex86r>DIEr{EnoL=z)!3k3Y2p42}KP`sl(5M@_w4vFh{#=Z7#B&XiKAeBEW+V
z_()5NX47m()k{eh{j1;?gM2s@$a&0lC7_ly+iBA+PYud5&vZb*u|C&oJbvxSWC!eY
zT9^HJ8R9-D08u(i`=q@N7oW>VKl1oxBS?V}py9Dlp(4(dr0Xo%2bhvP=qt2oo0!w*
z(eIQ-l&JeKrZ)$|Q*D&sDX}ckiKGsLJ@t_c=PgUUTM{1uQNBUq_-#EIozD@dSdyM`
zTekoCV*yZhl22n&bTd8W!`ZAv%dU0h8%F~Nf8)p|d@xy^;JVRh>k88(az16K6oHHY
zwM|Ew<u@icLv)c&GOB|Dhw<ZaB#sLCT+0_y+-)M)cU}7&0s2~kh_cO0FuS)D3_eSc
zx^<h|AUf<igMo@HupO5k3s5*$X^CO|YHoW|T#0r5_^DGVRUh*F7h<RF-g9KjtN>c3
z7S2?Q$1eGjs5__IA)4)nyruFR4HdWuQ<Ngd*tIPV!{PUY8g+_gV%ir(Pk`dHPX<j4
zAS-zsoT;i#LmxoLybg19q>i<|^JF>23i@*Llh3WwEo7{v<At18Yc$J+HfMd;VlHM&
z@eBr3nusq}Apdlq?tOLGO=EdXisr8i-~z{SMI6sO^`Dg%YV{A-)__h@<zQI3zOH@*
z$|}biL87V@sv=5HMdOJ#^y%^%QP?K`oL#yhTFf_IMTZY^0MmRZ>&ia|S)X1d2yhY%
zvb*dJQT!sG*+CKMBhXe)&qvA0dFs$onDx&GW+0BK5N-~MiC8T)pS6*^fDKFGN?$8f
z`^6XCo}s{|8C<K|pmved`jQ!}<W;+Ta~YPcz?KydJoW4bUYf|j8!_D=E5rjDf*SIX
zP(|n~Snf)Q$9}J+*;=XK27aD|SYu!WGb=gFCD-)kCl7?n{YP9@ZZ!3xE|+b0ymmo2
zEYWl`E&yyhPAG)2^I3ho?oj!SDCbl<DNS8J5p6K8Y&A?78zTvvqy|-zm*A|J__wIs
zd#%?83Wq7Y&TV?ZWkBH~CvRv(7hGTcT^ef!oJ$o@jaDr>NwHLQ;nOJ}=$r^5dk-Hf
zRM!SdW#Tn9dLyH737Ch7XaxCxOn->>@%R|~CjP@952}VWaY{k?mwRm_I>7*AQF=_Q
z6m;WM;-$b=mAz8)7O*I(0uSa*Gz{(}+k_ew9S687iA5GGz#uAG=Cqj>VYR_lm%SYd
z>a9LNanc~?{ZFDRUyqK?=#)drTWJm~`{6!RmX50#;2LjAP;BZTqUwZKlh1VH^Zyaw
zTY$$D3b#ll7eIyLTZd*gaqPrp4T8f@1nm6OOc?T}y~brE?If*Vw?V_qBE#e>?|9<u
z+IUI@)PXb<4Rh+1TEf*l7=qugz4dFs!at0UU7YR~a$Zc(ly4HmE;LjUUp*|=o@Lw*
z)s6UrScrMKM5Mv4)xIDq;-I3)ts>^67^u2@lf^FZa}ZI^0OD)BC2pmc9Wn8<;o*Rc
z7UokE>Qu}~-wo=0N2J2Z#r|i&-FHKxdDfF2G2heu?|@2RKU{yG32lytJ;k{V<rkD1
zT92#Pxuz7ZF~mkXujLwIzl+-I_I%h}d3&LDtgaJc2=TjVAVtVG1xIhK;Z{2=kfp^Q
z{2M2H2u{<3UYA`OxV-M=Qc-DIQW_6pU*BWMt)GBQ^)^d%;5>U6bn7-Q5OeoDB4^Y7
zGswg5bFedZO@zg-v5}nL_MvWS+}%g_puG43h4&vDJP<F;9VUwAhnEt7iSH`xUd|5;
zMpurdpy{EmVkpFejO+vR^<p$-@O`ImxDh?zPBIV*W(qYcr|5@g_=05lNovRRAW~#>
zKGxJhq3RUM)|xmO_}a|wS6v8Z+7mEmJ6s!18(jVdtQTVgCDxc}<i|+qMUS@d(Pv~D
z?i>aN16<XTe_`IsmzY7nq5??4+EY6QplFan)294!v$80g=m7{c$2(){Cy8c}xT<^w
z`#@lK(p~*`@i)7tB@QOIEq}tL(L)DH>k^9#UI?s}Px$8Gk{nC(N7=X)V3<6C`Kqv2
zB7OcF=q2|fmD(?7P_^K}r6i)ofjL!hq={;=3!!I2h-Mb1I`b@<Zd|qLveQpYD?b&Z
zqg-jh_49u8yLQQv0~CpNGoSbAK;W0c8&Kmg351>UnM2m70KEbQC7RXt#>&gpNZwtX
zDeX_Tm^4WayE7#>xH=x*@DI0|6Aad{e-NF0a~kv%^Q$c8--CQLN%t@ma!nRaeIvQ@
zmjyVHtCU|UAOsr78|=QNLmNz2qnGOZAx}CYG(5_FVdU81-O6@Mc^rD#u|uM6?sFUE
z<xpXza9L}JJjExc7}}~}#g%>yFPVW(<b^05DU#>ZxM%OO3BlqMfwsi~<Z#t|3>+(v
z>qLPU1Y=3{;`z{wHUr;Qf#No%hNhAC)1H_Cc^1RPj7TgKVvU0Qh{t0v6qXi(6pz8w
zH@sZI&r~4m(}DVsN68VP{E(n)uTX<_<WWPTow6x#jiOvuWZ%1q$}6C#$u;Kp$2+2N
z-HLFcP<{}O^R_08?_lIxO9U@lHg^ve_I8?A3&_0cK<UXBx^o}u9pq}ib?XHw$Bs-`
zK=57;CSP+)IxW06RA8w!y6Tbi=+o&#@Cu5kTlGv<rveoK5!S%tLNd4!&+^#DCOE>m
zeOUGZ|GRG&>XyIyxkynERVNE$@|6S)Wod87K%`})tCg63Qf)706AC3KYa(LgErwY?
zeZT%pN<~*cvb?-B)nG!<0LP?lJA^F=LRV9sG2<<DEkU4`)ZVk{?Ny+Dt;&=g(5cCA
z_Kg`v$vR~e>}b7J3}!IEyeb*U_Vm?Ql)31F2dk$9DPpTqzxh&tCY-1B<WG_~I2AJL
zP})~{u$tmHL`0wtQCb>siZ*xlL3M!X@n2TYrn%DnrQI~xOJ7|Jfb;fovzMkra41Bl
z51MkUxu=<d>{5ukX!EJssU)dL!9{WTcEm$??3O(=S`)?&+obo`lPId)vD7lKO-zqG
z8#nRrieR+!a`;5yCkWv;Kh%Ul$HQuBRCUyYBbf}jB~NJ?i+Q^D@)pZxc!W}@R3M|a
ziC+}4J)ry7dA`04sBL-6D=5cUdV|)T;1mEIY^N~8=lH%3xPYp<F9f;SN>dEM`PEw}
zSg(92#q0&5?-kHrk+*zVCCST}^hw5Y@+arLkh^Z8G(y2kr!H#=*MFlMPRbpN_pS|n
z<pGM)x@r6n@)XEhl2jA_)pq*<O6$5Z0fH?1qhGRF|1PKUY9}76%+dR233!)2y*x6>
zi4&6la;^T301SY)J1(m*I1Z=3gd0EpH=Ke;KNesfxZ!{c!2a*w<NU$(C!+01njU=c
zfAStGh_?OE>UWJ_<z0S#^QbIn?ih?a*s0(iRJ@?(?{bzaDN=B_F8HYh2jl;L^NAmr
zQCogZBjW#rC_CyIfM@?*>i;g~ze|mG0{=bL|2-SqfB65i0a^LO6;xHrL#8*enxPL=
zu*F8t|G_XyV8F!6PG!&P;(6y{?t_*>ef2BW3q>XTi&B_=FL@p4AjE5?<9404|Ffuj
z$UqLs+TjljL|t<ACU!b6>Pw-@>p1=(p_C&jj-`nNxA4CWRcb`?D9nx*>hOs6_H!_%
zR^9b2|4`o@IMiWTO}&Tq_^w^&q;LLtO|H7$TBXJ;Q0wHc&oclCg}94?{?GA^@S78@
z6P16=eye1Qxx^LL$Q*0ho-3z_<4OKLL(+#3_+}=MAFl^{=6hmsiqEN|B}JF^Ijk*@
zkxs^vO*|m2yY`E3aJ>ZlSo$7Vj??(Ao?OvYlV)DYakgpX$HEq$b4-X{`GZ?lzoa_|
z6@|1=4rx|h!G|1tJ80z8(LcKy+#NKuz-L_+VjTNG{I?%r^X0^B6<PExi)7whL(7nu
z+aHthYzkW2C{nb)y^MDF+>BAmuGEr_?1Gz)x(_6~g%Xci)HY}R_Hvp^IWq`uSpNB-
zc2JCo#v(H^`L`b-87D|ReFPiJF`Ql!<Y_k?Quy1Cn3i(k?OM&DnpmwlTT@mybn<_&
ztDUuu7rQoR#(yky?PI1q^p*3*Z?C67o&D9Wm7atjVM5fY5=<O*UNN}@xMmHVR#z2-
z{Ct)&K!JCwUGME#;u*%+E*A90@C;99TQc!#=E<nzH5p-39|<!<IMMh;aSr6I(f{AR
zhueRVL9iaN?3lk%8w`(_$dF^I*Ngq_N9<FZxt_nWl=>qqV@ZQO=#9(J?>}N|w9Izu
z8|y8L+E$AfO4gs1fBTY9{ai1z9a=O`9d7Ilawj1@_CMHFT9(t~QiW?NKNkFL)cF<E
zESrD(OT^UIWcZhU49D;kSZgpTXlQ5s_9H4b7YA(z%Ia=e_^fxnOZzF45j=JfZ|Be!
z8e9X@M_Tww{}<Co!d+O$IQ#ujWY7>*M*q3A@*k7GZ85@Hmz#37+N5q-Y^IicNn0|f
z{_QOZo6BzLi<COT{Ea8S;(FTDL5qOj{*cM|mcFUD>dz53vQ_Dur`H;YmnNWmRG*?*
zH1adkZ~-b_DAT!@M1{W!`;UG=bE!X96(aCQ^6p>n`RbffLG63w61Bkn@jd-yoxDL~
zLV;h3U3cq1=_b)3T+nH)6skf>*``nadtLtXho~%={WVx$x3kA{1w9U~8C%odIab9L
z7d){p1;R$@)nPGfQZbixD_@2{`9!Dj!<{n5PDtX?pvbc&r(yc4r~<U^NnHS5ijvXD
zfo1C(cu##KtOFj(`{M^3bSS(G=Y@w|-Q8ZHr&m8Ld=<zJ5xT0HB>QnwmI2U^%z%VY
zEQKaNf@p|Ia5l*a#k7R#PP0S31eKe0T2t@N0+v@IeLeRV*^MY3ShyieD*T~ggX=wy
z$Gbf!Pp_peBs!ROUAU^LlMLru_=YR=Z5bqaN8dvQZg(GGPz`MH^#?)!CmrmN1gSL_
zeolDteX$5vJ~ftxTrGFtbSRc7OU-Z!J=Y2K(zuqoUAEro@?@;BXW`qxD&VWUf+>2m
zjys(Vh>MKkGOhyZEgMaq){ODkaGi0hq1DI~p%{%G!0menYdZ4?LUp?ps-mHeUzfXG
zgb*lCNy~MQcD)E$QfTn7AfVgRpb1lc{SdR9Vv5lK4@#U-k0uQ?WfZIBi!e`{0Dw%L
zQm~}*$+TM#nnDUxG`yDgiO4UX*?Z))TE~@O%$s{}9>mug8w^9?+Q)VGzH+F)Q$PXV
z3yL)td2OaT8W(ccTGMp;K{Mu(#GLXRhOr+Y#_m9xF9cnhB&CO+_jt}hkLDQkmpl;C
zvNh5qcT~Ib<=Oh!+oAV6bFG@v?}3<!2|tv4XMqL1oPYl72|)X3pk5sUwkbh9#8W@9
zqXu7CPgv_ZU|=QaZ?@p`n_dhpj>#tSFx*_dHoBQ-)WMUbz8rUAVin3t{m}bme9XvN
zupo2ndR1@)lr}w&O6=(Yj(G+GI*PA$Ap6muQ76bqyj_O@1X&rB!TL~RlOK?BfwFyo
zp>zLswLP)5uJ}8=->DW&7pKly)R(vlsC4ZusJePX+a0riJ|Bu?e!US?ZHq%t$?>{D
z%yS+=G*MITPMwQ_vR+`HJw@XK1OA!f-DUuHDKpOQBu!Gu^q{yGau$`fwC#OQ3-oQw
z*ZH>H3r!kZAUZPWp!e5kprXbLC{f3AJ45BNf<iy+j!`5Oa6}%_MP7Mwn`aq1w{{)S
zfs${DFS3B8EElCTJ_(IY^}Ei1d8i0s)6rfVjq^1vT#a3s1lX^x)(CWwJtG7hnuR*A
z;`?x|$LQD}6(##+mE(#vO&#cxbJwvyXg8+F=F67G>r~@g4DQ0^6j6$SzJwY4j@aIZ
z=DJFBCCq_la5O~SbRz(XwF2*`e2EMSsHZ3lK8C6C<iJ{q6**)3pl-?HBLtAgz#C=L
z07X7s;IPb+r3pQ%ck8$J>W+0UzY=G8K~${U5(c9wMf=M-0B%}l8j%BgQVK10S+(d*
zA=LAQ${{U+S)coE0XX<lgf)Op<py-!*b4zY@ou-2n+N-XECq!|buC}p;4s|!_<(*W
zSTW|#Q$bftMCpUM7SI<nMzBWG9`=F9@p?ivy>S1T!45dkZ1=*<%DLf4^t;pgxI<Hh
zcWrrE^J(auhN^L<{6NM)rH@dXU**ntW!nHtAi>uI%wzU=F0uGLSx%piP4sm5bl}u_
z=;0ynZWC@zFo*Q0X6+YlXRiA)&S7~frls14AAuIe+CC1!`SXC&fEl@p%FLw0;lMUu
zHcQm32G_3al?H`eRV@$ZsiJjWRa%22ABnoHNBGu;e~k=4c+{+7P%$pY+PVXVYsBve
z4F<^eAx`~J|IR1(!mXNfu`p{v4<ld=?;NDv^wH2l<;y$RSx?Q4cl)Pxne&#QX(PCJ
zgsi>In=ZJ#I+-`)E4w)KT@WTqcYA_?_3>XkX`2qZToEvfl@StJk=|c-&Xeg?s1@xK
zVE{#IAfTr_wKIEufcV2fEIU(RxFZq1`zU@21j4Vsze~nvFpw~JE$Up<SioRAe>gcL
za!E~mx-%?O(0mAUQjX<jV7N(axb;u%k4b7-@x7JASKC%ERv`2Q#z=mz;b|aYzu#Tx
zR6jZlT~-Dk{o!z;EHqLlvWx!7jka&G$0w>{vePYYV^J@XRNa;#6UJrI_kVBHWMRx(
zV}jcmcu6EUl!$DwQu?d?c#@)JOBEJo?mormNj94|y7&_Pp+^?!M%&FCyk6}W?wVdx
z2kj9`Xl{BVaMu0_-MB~q8?gjwI<IPXfpt9%rjRj<3{1GTa*{S&57Z7!=acavVji5+
zZ!3#qXMY&*HWF;G!YyY!KLV1~K$ool*gO|kC<){vc?gp+!`PgKW}92tT}#6j6)d3}
ziE=56*I5Jt-@c;4BsQZS*$CsHxe@{nnHd38&YLKG;JB(MbC%Mb8WusYGzMwme?&sV
zi!T|9u(_#{EUfa%CS#zu<GbCc7UY}Gw?LI&W`hQaE4R@QUfNgo8o6&2?7DwhPPC>a
zF98y${2*hZ^aKr5rp*AmLe?J6XTuT^b362^l?5Xu3Lvy$5RBeBBU*Cr%TqoY1@<B=
zy@CBOoZ~fdGC<v35+3_5h}tRrhSc7-4AGldhouP$goqxL?wm$GgT`^@T3QhRAxH!R
zpP;7u#<=E|2V-*1JDIs(bTvbe9P^2QhBDlYVwY>x*@u$_G^Ls*r{!`N6%6Z0@~pO&
zUrEP2;Z&;>z6r+)aD}z*fCI@{xQgHkUjeDZW<F5ar(~V5Iv=@3g(h0kgzkTj${kGS
zwH!GLcw`SE0zLYR6;KdAtz4JP3tAZJ^`msbx$$05`5>)%I4GmS!+zeM+WC|Z5}Q#f
zXBem0tYYD&NUkcDVzH37E6g{I$&)TIY&_D#1hszDM&$~VvFxhqktx)-rmWU0Fdt3Z
zFg8g1^y8KY`6XX2Qnl(e0IsYf`8bOV^2P2ed~}0<(fkq%US@-Y`J!9AA%52GFV{1@
z>~gfglu9mlnAEVA#9?zTj%=dpVGTGs<@{-oEjx-KR4DPQkw+bOt3fq$X!)WzO(e-?
zx+@Ku<IU7{F|p=tiPf((7K7WWr-V^EyR)Oi7JR}u?=*s?u-VS^gn&0l7#fhYCV*(K
zz{8%!6>O>$NqgJP;gM)xmMjaq1PE`6>gO%4TaQ;I3-Fv8mtL5ypdtJGo``ZV1X`&c
zgK6pQlCrc<L{Ot;fM<0B^QW>-=VLcw>Qmy(nQ{XH4AGiJ&S~Ib6s@uNUrKoK4N{|x
zwz`XFc(h~=Z1jKrY}PB(#nL$yoM7YO<jm>vQKRM=1O?3IJn>W%YIb16PC11WHRyhb
zWgla84a|gAB)<tv+gA6EbQig7j42kyw|B#+z&mKnlx<>$BPiQ0eAxUz^R?;%X3Qq@
zqmU`wGk0)$h4g^EFcE2q>QKJ__E{eUQS@WCnV$sivsg{sL#lKpy+qnaO(S8`K@YVk
z6UP4xScf4WO%F*vOlRXrd$#GuQg#tDF)Y3S(!7X`El2ngp19*w)n~p#7)nNw%8qx~
z!E6A0UtfDH7vJe$zDa0ezf!`TvZWg1V5MnQ+iFv@=hmVM3$)*P%wSyMWXj7L19i{{
z^e&VPW>J4bRN@HDJT~a<k<@4pR>4kMqX@*RrD|6Vz>KJAoLBDNaeyYVPVY0aWWET>
zgydX<`akHiX^@c7YC=b{Q2L=dH1?~_lhT}|S^=(w5;a`CPI1O76*Ao5<3X`tD`W=a
zkt*5YZ=;Cz$&0c~W<rFkVC~tx*84VDJ+GNLMV|BH=URZRZG+_Qe47#h#Jj^2il)59
z;N3N}6pyPf2mI1X4GBTUK(-|l&#b!MJom6Bnx>4W1?%u2evPZccF@PHSV1T$nE^J;
zxdVU=WrkNt+yTd~s0|P?#vus+EQu&a`cU&V*Xz2|1%Hb7>Zp4trKh_~A2S%lXbmP{
zGgq&Op#lzn!*-+9gqPxIwH^e02J$>=%MOiq#L5uXYeJuV6ao^4tZ^YtQWT_WV-otb
zp17cx`W$z_AVA<hDX}DO2Kdp6iLNhAj$gk85E!98G;)Kz6f?=DQ~u|3V(LEV^|SgE
zfHm|J_mdaZ!=#cZD7}?o$X!H89V_uygRk^3jx8WmsmfKc_m{};I>3e`?%!#_Poyql
ziYI-Ys9f@wx@&lwidNkiza86Wnjwvqxovs`+!lf=yz9ShT?{FTlr{o8Jmy?{!Uq^|
zYbx09vb=lbgU}&Qwx0&zFOwuJ#=D?DA`mitYz2+gK%QWGr{==;Wbr!(sW{1SkDmaM
zp%$=TpwG&f|0b`fE^p;Ll@DPYax|ro4pFQcVV**#*9XG!g81Fv;eXUEXD$&puMY#t
z#T|{GCjU^nKl~qbRfvD*BZP+wF^K;-R{OxH-%srH3p5?g{i#t#oaFPn4hjHXAj~GR
z8{0S(@FgF3v0(b?zlf|-n{zaM{4jQ&4=;=Zuo=`fhFyv+aODNT73=NqSR8oCjU8(d
zKa$aPkh(ARbd&gBfft*kdtv4<aa=S5O(DSZl?ZLw@B6<ExD}-b=yi8E_X)NcL~$^c
zT=dbthNwd~IE*S--CB79_D#Ox3cbJUYbxj}oA6^g<gL@I&PZrXLu$<Avpm_(1^OM&
zwe#krtV^pT`4A5KPW5m*+z-a*o{~asv-e8fkBLk|^7aCu(xz-p2F&kLj8#ot<ADg=
z7fD-sOs;J$Otf8L0mt*s5>UM5D3acK7~M`6VXKW)gv!z&?%`msY)nv0;hxT)_juU|
zLEdo`4Wjmc%zXAwg9Po-RQ{Krqt(`GevOEcRPKul<M`mu;`1kRhCiM)8<QW@s;dh&
zq!F7)tvv7G^TH__yWA}VYJ?qqSKIE;<t^(9U%0`IhK}!S0j~l0kUN1je-5%MBlWS)
zXuX!#;A7%iO(#|$Q;mk$*{`L_?`;^8Ib8x^Bnsj^r6sL<s!Xo&fYVT6dL}>NxFywP
z10E}3+}dKUimn1QweG+L4ZZdj+3ME13%TYWldCKBD-DrSs>lOy>fT`Wvg(rs)pIRh
zTzj&Y<ZUcWXR{7LVqUHHW}_4M6#4+8z~Ivj;o`G9ZlP$>uod;%Yp<^oQRu2TuMQ?8
zX8=7wIhbojrolyo-vM8!Zq@M%GFt08KpG|fy4`}0NWSP<>{LhO?Vs$JWI3G~WK1K5
z9fH)i4F+oaAYfzdvUu|#_R{sM*!5*!qIfw`mb@lNu(Y{`^&iG;RuXRy<lZqbdYwG6
z(Qugx&_V4dYmD{W2N7@)Wg1hj(9m-I<i+6B>I=y}{BxD$(4dEr_t^aedgLF|W&wLF
z#uCnH)S8N_5gt+Bv^|+gDfp``ko`WFXn`yoxNprC?+=@qdx@#Y#aS@AcdA+b44e{N
z9OlRhuZx`a(}6797XXEfbAoM|ahvSwpbKleFy4}qBPOW-Q*YNLK3cuY^w?i>g;J0b
zdL5FRjNkl`IPSVHa_hWg&PLOZ0tHQjCGI5~&!DkPmC7l8jF+6x#20^YSxCRF<x~7W
zzX6%1m=EfZ4WvPcu0UOfy+s!W_#S>|h*A}(X*dn!>g(zFW(O)`lF)!%uGTkL9!k2f
zhu6)RB75Oo>9ZKOKACjp?k9yz(d_Sni}xmEPzh!1T8N{=EFUs~Sr`u$)c{E4X*1Bu
zJJ}m4Dz#bI3UF~W<>XmHv6mOX?V)c}<b?q1;c--oG=uNXUV9{Tsa+enxe$kDrf6>n
zv`y0Uov;Fo--txO3HJQ~7UmfCUn9LVDEeK@N*%|qG@VfI=_M2N?qAE7$*p4$|4K?X
zk+1YoBjFjikW#oE>XsM)W?IDu6~B2o9nfisfNuzD=W#|HYnTG%B!u!OjR9kD?q`-U
zC2Qe|b{b&|)DtLC>~-wvM;A$pcFaH%Djc!q&s;}o4JlaO%31+%ma-OcSgDG+h*f7y
zM=ON2n)x>dAFrCMZo<EAcc)sf(EMgH5x-t5-Byjs&0igq3@S|=)LX{4DHr35AGt=B
z)dv@-<%(sO>3g5BQO=iBBo}f8pfWeF)!0c2uPT<Sn8+F#3VZ_&Q4uib>G$R2{$jxo
z?1``)oadfnWeuEawenN$3+N1@Zz3s|>H**24qKhZZ3jt=d!f_1PxrI1U-j)SOG2d4
zU#fc?&(&VOaEO0#tf+6sb6?PDGt9GlJ`KxC>Dw<1+2Xn6qRq9w0hqvL&jKol*}Rt-
zzeI9`<u~D_!$KtR4MME0tA6*2B9-|as3d+RxKi=j;lPrFyD1mk&=IhQUgT`=a@1H#
zI4h_uB;3XGepzIgXM~r&T}+REax!UWzNl)Jx{?(t+4o$?`+8D;Iz#Sb@=QOWsPAcN
zK;U_G6*!R5Axs$x^v{jV_b2P(N18BC<qHgdt86Mm_>sBn`(p7YZe`BHMbh?JWxUy*
z^GoW`10}a{ML{-@I=o$3Xpl;fG-ydyX97H!85Fm~vQ0YUecJF#Rr`;ITt+codA&T|
z%DGn2&s67Wiq=^~vI=}YH(yImHc=$;G5PWo0pixst%!&%M+mw}h_J^}dGYoQ-hK-j
zpYuXJJb0QzM?=^NHC$Y5C#;q%VxB8MoTBJ^w>y~mcYDe+%{?jF%R+*xZyI767R+Wp
z*987<4}yH5z9ns(qAp;_a*A7B&3t6;w;Od`lyg=Sjw(E@SEQbg_N6>Ow8Y^#eCF4+
zI0>RoSIV4E9TENCO}g!(msdh@k5u6cYAL&1j)H~><9-iXMYuS4dx0MK|IlBbsn@6N
z<DwIge{-cG(YVi4%<8w7^YVfwZM0Ylow&u%I(}m{r>`Ery%vI$R~fQ+HedV*Q*OU~
zqUCh9%x||&T@G(u&Ej2}{So&1-hFBpp|`5P{Rq~g)6}X|SfvNGUJaQZ<5Fsv-|i!~
z(27>DZb=g_Nwmbuci5(DNoCj)<KIET{Ss+#abzdrX0`APWpU0yFLKU{xioB198N!*
zWh<EDl4Rl;aXPfyn}77f9tiMMR&GW3{r&&{Ypw4q^t9J^mz6&kP<hjk+azb!KAI5p
zyXSjlDlaXq_pEQgkOZfEX}Er}!f(CAr!%fcgs6|D6bbROCtitNVI4E}HBP18dVS^U
zTGy&Z^esx8adMq<XO|?lEgqYR<zdI7@RrRfp+)WWye$pGmfL(4GV^sEBa0{TvPX29
zB^cn#Z(f)b#9uq^1CDcLl`E4~Qf1lEMN&NyZfwIw#_#HmG5(qGvB-2;=O30a#EW3-
z$gW*mTd{pQoBz?7nCNu)M6G9cdFFY?33tM;&z~OkK6^$^`2L@XZwa&ynV%RV2z+!-
z?wTuU(Bq40Cz%KdX#TopLeRDAnF5pa=Xjb*nM9dVlB18$J$>pKbSCDg%JH+S+z<Q;
zmcwc$!UU$fuFUfprlk*eUYQ<TCaM#v%cxb>U#t=8{Y820`wfENjvVentMwD=ZS`jR
zj7S{{o%)j#jql)<Rgh)yCK$gz@nB#Ee+qSEkE?#r)>&d6Y_jd$!T9&>uk$xUR(|Fy
zv)MPgzaLk@$EUzS5_Tat<*+8phHr5!)q85PnO+z=&nW$HqRQ;1)F~3`K0o?~6hD^*
z8+7^Qqw)w>T`znGr%s}@+09!1*~LbYf9s3*%_X)61@&f4H#4*+M*MCzXYBvQ0`78s
zOUSWAbYZ{&@2jH>30gLK>R0BN#r+dqRy0iAkK`LRoHta)CwWGnbovi(0lR<>d?h_g
zk)9&O{FBeaoE0~Q=UGW}H|_hy%A4p4I4^op%8&5k>(`+(_uU6R&H7UZegX76*-G=I
zZ^Z>JL79=<=Np5cR=!WAuv#9*)7#tr!262AI#PMsQ#gZP@HejxCT^6(uY9Kv^6eJN
zOMhuq;Z1Q&F><5X_S@6?XJ1Wa@YVqBydh&$n!fQMKYlGAQoT{SBJZND=8|WLHa*F*
z{m~9mthy%tzZ#^Y<mj7vN9^#SsDGWC=@U=rqt)-CE^@FJ5o@6oFbL0Cd!Jpt_e6!I
zrG`I#0~!`+<S9H$lKu^TyiBjwvu-0maj0-sORU9iT2i)e1gmdpDfy4j2td!c`3C<^
z(98A2$S3qi>u;Sq?KD@!mtI?WY`nB(=$nW4Rq}SfRf+3(UJ~wnR0AAmUS<P9ytRwE
z?AlHyNaC`bscEIVDbSU5$8ioDYNN~I{||<O{yJ>fvG^f;NWsBlN(J^-jvF1%X1|}P
zJ7TyQ?4m5Cw)$Gg`=fuw5yR6*@Gk%?5WNgM%P)5bkNldDmu045J;KL7MU5lTt~(Pu
z5aKWs>c;u}pTK)3nWNI;?%$oqe>^qWO6WXU`Ma2!7W*z%>LMqJZrd*^LetmqJ{~R8
zCAe4brL*v()9-g&^IWhJj%z>Gn+<B)R87ev*DDna7ZacVLu~Q><o8y8nRXRCV=P=j
zC}>sBs?&R8t8K}V>~x-S{)3;TS&{R2-U_)b*GAYM8BbMwn)MO)Vkgf?ZynbB>?BOr
zTI@}>Ege&naSAT{Z$@V>2+oU1l`5VJ?oslr;uo5^LQ{@CLgsZAU3c0`o3Cj2uRC-7
z!yOojMb8*h#iLq3N*+{fdRp}~jgaZxWS%TL%-i9PU`inT`LE0U?;q$r;R)$~Rtw^}
zEm!9~lIvr;XQLgQyQgiupXe8E(B<4K*oe#d&8GxyOcnJ-4{w}PNSRt5=X|0np$bX-
zgzDbH&;Ky=PG~O;2kpq-vLL*Tb6p~=%^hl)eG<t^IrCOlC|xgKXS{`YrGORhUqrTQ
z;T?RGhL;DfdOPf+@d5tXMwbe)+FU+`q-w#Xa&NlQn8=?jXy|Yv7@xd-`%$UzsMEwg
zN%&S{bx-a3^X}4bLfGZ6yXT3!&RKVEZt0dE{)aUlu>wn8aQf^zzSA*3z~pdG#Bs$d
zvVulK1zWKmCwZDo(Rocur*wZ&(y%YR;IB^$5QPPLa9uAN@Awu=6OdsWN@gc%uuHz#
zPIK5RI=<NjQ@$<#@QkDM{_s`oA~hQLyMArYW<Mry_HD3>Vcj7{#WGR<@>0d&ZytAO
z#fa9tSn>Y#?FEWQYwhjjv~z)?v<b!&`=!pxGrbpZ2>W)A=9p>6>`U7opgFTO>dkhr
zj=ZYC?Anx|i9PYKiT8B_{%3Z=6>YffWVrL%M2W(kGdMZHLSxm~`Xa4lAM1-qCoFB+
zSWKXLm-pcKIj}A)e%6w>R^z1tESHLD0?ylIp(jzUB~v+m%}wzRtlKJNG*tZ*0snN2
z!X^&S#)%xdKXFN=bpvO;Q(+f;$y|52GQi({G`9@*#<dcbQt-3O6G=N2wOv&Xz`rY>
zLmebM31+O3Wq^3+U&CU)+JArI<`*Mew*Bv|{nX$Qs^RV8snqrH&QRwGSSbMqMx2#W
zAlS{hMDNxz#`#c*AcpK^%`KYi`~G|7|NcYF@$ap&@ekXW&?f$8{i_+H$ThlJ+dx@8
z<zS^=ipmNuuN>k2ldQ1JwmP`+eQ^zeN3X_i0>bz+XM*L=derVfXQFkfriK5AKP_E$
z$4~7KP!VR!Z|u#mBOu&V6lR%c_|HrH=WQ<S0bAdgMY{6^4Vd5s`+N9!T@N0p)FU|<
z8;$#4xgLilymalQ5l)UFe^i<%<@(8RK$T*?^5MPQ{(tfZAJQIa+j>8l9wB5n51&LI
z;)Ca<Z=HcpBD|I^vh$PVuE8fAykohSfN+3)8uvtiDlzh*w~P&Oj$%2i94B#M^ob8I
zAE_z4)#5$4^K-%Ry2AB3PFO{{^Xf_O`J-%x;m>-C{YPKc<l~J<J`%nK=H^mn($2b*
z>#0L<9=KP0B%`hqm@7YfJWBt+cO5Nw6d|0iTSw2U?u-_MNeEAtBpemdKO+nqM2x%C
z;Eey>pno?A9oK(v(7!im+YbI~;!tw%uMhgy2W{u{{{{*F1_|5Y+`m!W|LZ|Q0X=XP
z7kR_{Kj&#cpCCA4yF*TmO$szg*?@h%725dPK<kfk!vmHI!Yt#7ac%!eE&f;R>-rfY
zSfZ(Z)ty27(UF%B`wLb^Vkh+3fLuNY0D0PrV@<24$LjM$y-$$<IMXEz2oci&6|peF
zV(cmG03o%5#uQ%SNa#ifRZd&y2%LZJ8U_*nlN*bivtdBm$p+NKpG}#1vk_Dsc)snw
zwOy15k^XX`mEBcw&p^5&vm<^d*MkfaxBN8Ru33*BWCWizp8zihKccbvY|5~j4aD|J
z7Uh6%!3@x)VR;W#>RI^|X6(!q!C5Up-2Wp~6(?WHJSj~?!#zQ!VSASfQ1iRHrQe#a
zbv==C>^orwrIIqlW%Smc=axF*Kzi<<KY4cMCQ!IGg&b_uo4a+-&QykhHHy5;#U1E$
zMq3dE26zBRx7<8@fN$H>3*fm>Ke1}@i1?ls&j1M*XL{|<^aUPdFALd|X?a=l&OMao
zfzRH&VUO@gpqR0~mEITaG+hcHE(%~5cL&H;N?)8+e*FGQJ^2^!EpQVUS96priiJ(!
zD44{MNfS*AJv9X=a=*UNy7HE0v+lxSQ`?d^-CY^&CV5Q*r9R5{{=g)Dt<SK*IX7pd
z1tIYo^n6XdbBUt@V85}?jW<?iXTKa`UD(q5Ch7|Wr`!PQ(CI6E!O4f|54tJL!YKP`
z0<U=-pb4M6sxv~kGgYI12B-6?L>>RUgb}E!w6PlS&#ko-38EQ?70;p-&r^)t_N=hi
z1ys}yc*TMyC>F>Tgz(~wO4-W@v++{~AqfpOP->YV6YYzp!QMv@&)GLE`EGk%BaXkU
zv1g#Vir0yN9ZidyJ8Lbxa}^fc(CZPOa!9}EO$(GTo`{${4UmUi#Oo4K32;#qSD7Vo
zE8YfhPvWf4+Gy>3kfj*B)_8Ino#Bt}N+8-MMMPVPaJjKg#$LiQe_)s{W#hrNg(u-U
zH*S*b+L__Ia=`j4>w60*`@4$9N`{}>hRgv1cXy|DnDB9U&VdY(oo5O~a9bo$b&<!Z
zn7E!t&$)-5^H!8tVBtH$xM}}*fAdD?m;=ExIAPCH-{6nvkqZm!)a|<LhI2upGSHKx
z(3ALSLygjJB59K}fEf&9It2)M;rLqxH2gB)eiZJh+pamzop)-#r6^jY4``8U0Xsx>
z9FV2X=vYkh4`~pTz`nVa4d30ltSC{r4ad~@F(w?RK+!CT9u<ooHB}jKNwyEo*8sFN
zyHl)FN-GI&dZzX~uD-<4GpRxFR?Ww2b#|W6Gv6cN3^a@DNHyykiveHj2_PnRp`ipk
z^iRoN9s&$JOHINA->&hVhf{FlWh@S811$8(k_<s~oJuGVdC~%QI8IrsBgZVQ>SAaq
zLimC?-g?t*!eP+(VJK8>ko0AZxhw|9oWaYDJc2R`@tHSx!ZW1OM0KYgOjNxgzoelB
z%&1-floi^XW8F$$DPzcatfufeiWd*0GjbPJPk@R_b?4m@4unEUMtx(eKPTHOQ7u4I
zrUAeB3eXC_zW4cH(z)NNAn{X*z}YxK`(RHp;HtQRrt}Ak%3RYx`m254mvBzfBjGA2
zBnB{PW$kS2xuzeachjehgRRs)pHz(Iyz1YH0FKM+)1hEnPE2Cy=L{>{Ga-E)KI#I|
z8)h8K<pt-o?Zfu+++;<95@mqC4R`K+l}sNT$ng-5(3|iU18m@X*9_yG&knc^$BL3|
z`-4WeDO;rk7UdEF)Hh1T+UsGieBjBw8l2N3994=+QB?e@J3EcSueVqD2U_7D)UH5$
z#s)Z_$SEd`bos&bS9`u%<F5?Zf&l(USvzUPT+{cXm5D(s^PZ0YF;XMsYNv9(ZbG%*
zg+n2exIGrvSE%-Dx62rZzD@5KoEp(vaJ)<ZvLTcN_C*98&C9T@QV6WMMS8g%Z!q)%
za@cdbehTiqy4B~R$RoT1$|DkA_)YCc;RG7}v(le^mBbJnh_|EgAKqet#U1$LlM~M3
zilD{4w7s~U$Dpg#t<F+oI*3;S7uH%qw!`#3lqGi<y2`Wg1|t7!L1Ft>I|tJ@6sC68
z|B!)j11qn;;&u<$7*Zh<cK+WV{{O^MGZ^aB0mM(|YY1E0Zi-7u+p0j^FrFozw-rDc
zTq9$k>WSkE_8<Mb>9E61|Ly*U*g2LYUbA0Vf?*CM6{ZcEVqPFaQL!@HpYfdSdelX5
z@V5yjLv|iC@Hk>%rWXD>V}#RRmxoBJPBaG^lU<9+Hf3xTR3`w*Y6GJl40(36ga+UU
zn>c2C@G!i&$W4gP1ns<J!kx%C9GfmHk3g6}1PSLnEYL<xdG@#J))9xDQ#hBXK(fvZ
zNIX|%RokQ~&d!Maftm0<`AZk=0DRJU;Vi<PmwNx7U|zZ(YZy$7t}_e3+NfFkLsRiS
z_~AF)YPiwo9|U`laUpMAz*(yqA!Ky9-YTAQ;8@hv(Ro@dOkmz7%wCRb-gdd3gWo^-
z)&TbaRwDcM*E_OP4oEGE8x-==L$6$aBwPgBQpB_u{~3?bGYAcFdBX04Q#4%~$s4oQ
zscY|A$^v#uK5dxzZ~I1ZRyax%B416?Cphm6A^$b7Fb5xVGUKcU2lO1c92V5AGSc+n
z?9J9_Zn9tPcATFm_X4y{)FZX4XT0&-x2^=}ogNw7!nIY%qutJJYqda}g?q@<UUoL_
zqGeP}_5!xQOC|-2%}V*UW=xuej*s{bD=Ke0V3?IILwu$n$gU;vhT~&(#pJTl|6myu
zbn4)A_rqkk2MFTmay8XeDi+QKhVa=;y}>TcXTE)N(2J6bW$jeyHf`eNyVs3R4y6Bl
z`V&v9T38CDMK3P5;vAWHNQ)x8&9YB95DRV0<ZR5<WKbGV*a4OIyXmeJph0Es&N)95
zU#y^NOoEvGwPC`*$tv|r<UsnteJ%An>^T@{;Ufp(Cq75D;>0ZR^ynuH;3uXaaSP#u
zNvRvmw1_LWA7Z=*6jnFSFkXh}U}d$=MZ;6*+l64gep7rv61^nkB4%0laC6zckbdp;
z+QPPamq$=rx1L3!P!so5M3ERoX_v!vC;Ql|;$i|cgV-MmdV0lNas!l@#B;8ec$j_Z
z<0t?*5P%~n9yNjP!hY-Mv|$)z@$z*Xpl$p1$qR`rnZd+emRIii%Q?kJF{u^Uy!RG;
z9Owowek2NysMhrr^j$A07aQRqE%Zo`^<lBnse6CZA5^v?+-LhM^tl=?n*-T-Sss`q
zw_`L122PBu&l@A&%)(G{`(QPNAH{8W*k$^U5L&+4Tbs4>c1d4_l`$%n;ab|5?A|m6
z$rjB@llOW9#k-GqA-%u*u<Ba(T|t1c9439Z7rZYN-e6%gn=!%-%J&B#I0Sx(L?4ZE
zEQ61f4y4Q~Csj=_de*duX}30~w@#ysOK79_$l++g<CS!mwM*TVFy?w*R6Ok6yD#L6
zvoPNMZiEN73s8cec=1GkeCM48fB^i2$|Xt%l$N8Z{EX216ssG<LNsw<=C^7-h!d>t
zK~jshAhZ>_<@&1g(r-pB9|ODvK;uN6>S6I8ZMQkR5>+x8Vx=Naqp^sR559yyPuRlr
zht8L$2POsN-sKIZAKz1>uycvwxbFd5A$P{z2PfZSM>a4zAU;hWcm{l8#DmL$b|2oJ
z90Mem+C|tUL-JWaDA|CFAndyByMpJpth3b=(jT?4;8?Gt^cb|Ucrmv$DcacgGMr0c
zFl1l9Oqb=hhNzAE9pH9KXRk&SKO%tY(+f1%SUjCD^Gr?Q89ncAyeIp@ljlmSC3k+0
zv;lhZiVVjR+AdO74Jz2o8k4BI13-u2gHFm=Qv#)^)A{ca#lU7=(h76??G7JwZEAiq
za%B+FsRMokN+%a>!w->&in<o&^52@w3%LHV2MvlLZJ&J`g$$a%S2d|~y9`f8<EDtL
z#d_AwkAzX76j2JJkpJ&JMS_1iTi9jO>9@zbn1cf^-_OFsK-^*i|Bj!b1l&Y{ZqhqS
zhkIW`j)97osrH$5Me7C7jbj#s<6dMws!3pc2odzBfgTHDPj0UZnsjyDbeYqis3yI+
zzB62aSZ5F-99bXkwhKEq<`%NWhws*V3&5m{0wP0gd!{P7IZ?OC<P0pl$lkov>;d)=
zUsPt>+I^u=_IK~B$kin;YQC!=9GeDubmrE*ZhoL|RStP<%jG^=06E1<ygm+EZ$5=K
z$;3kD%GB>B@aV3hY~au@NHR^}Sm;2M$+~4%mPhbd43)xQWT5&dy#XDXbtT79aZLkc
z<BN(P?>U6ndFr-=tEK~LZk-lx-<e^-s$4+zA^eL_&4Ap8nxD4-BAnL&-QG44N>L;e
z8NzF206sVWZlf4j1Eeynlxz6_q-Im9wxs_DKXJ!f0I>oiJ%YA9s-j3PjvGdvR7BN}
zSF&CPmKhkbx-usu?aDWpevq*K7@qTc0SgmAuatFbXV}UPF^GP&qMS}0%*vibPA}t7
zmX{Gk3`pZ46K$!K7H|{H0_R03%wL90yLxn72$nzp(g$xq_!a^fop(*gVRsEsM+fus
zIjDj_B^nC4blOnUYuiI9Y66y~6(k)_i8`Vi3xk8lg4N7#IvkPjF~KAfT(cy^iEz$H
zpF<w-U{KP|6g289@_-{G^{p02{-KOaGfX{EZ(ujgim)2!3@}HtfQLe-9dXB`!R+R+
zL)3Mfs^0>})QP6Kmq(ye2!rRL>WOy;s^}uUV3Mz_I*<|B8#wcib`!fJnu@L6B)PII
z3DTFTpW!A90Z{0YI`CN6bEgye@EAD}k8qO-y9&OpV~C8{`fZROFc%3RlBJ+K!Gv>%
zHrk7@s`-marr)-NpG^hSFr1VS!zls5u|${k@tHOpeU11t=vZpUQ92+Uo(j-_4paYd
zSGOr%9#uM!2sN@ebBFRlzRd>(AgeSVTEoqi%LvA}cfKTKClydCSg?I$goEak-_G*W
z{k;YJ>7;xrPCGDx-b^U`!~mQUujRt!g2S|e$L~u;rF@J_!NeuQWmTu0?!bFAAH>DE
z{U-7|_qq{P+9>BFuP#x6Tp%hGO`Kz<MPikGdEHo(ZQn|WCcJW>Y&zemR-ZSJ?y<Ky
z4ez-7qFArKxD^K<6@L+}%KbP1!y5q;#%EM&Pxs{TV$#UxzE>MY;pE~|bHp|CSPt-i
z_d_tE7AnR#@1>^jTACLR#twDn@(7uyRTpu5lr!>CM0`!#l1Ia9%=~EDF$`=$n>YE{
zmp7tP%qEZ^CGubuolONIED0t?;dprce?(_SJw_)z$~r!6IC66|vaev;$<*5QZx;e%
zrP?4TwH?#2J_Ken&_AH^+j8~jx)dzWk?MorDUgG+=)1Q(Yv+PVbD}r@ldrzDp(U|l
z+if#5D5_h@gZ%i}w%h3CwW7VRdR!0>D$OzLH}6oNR2@h+sAw78`4osbQXrTd&2(=)
zveR=gARBqui0+hwAPBcs!03KJc)p%4*veV|$c^`UjTc-U36Lnm*Kj_r?r$VT;w%Za
zkEn4D9KcpXV98G(a-ZCpDMrPkg{dUYPSal)sZRriLNqL#;m_rT?Br}vc7Dv=Qe=NG
zN`~bO350SyklB3B;25W-(A&d%YUfKr74#a|7y<H=syl11!X&7~dWYIXr|yq^R7X)v
zX5AO$2UPd$ECmZcwr}Zy|D04uu9hAaOyb^a)_?9i(lB9g15Ua<J$cqnoE)Gdk@e;4
zD33sWIuuGylpdx90ai6cJdozNI_Gu=1=-|R?n&@(%#^TMjWs$RY5X7jL|AzQqI80Q
zI3OH6AFvZ`52gQYA-Unv{ouFjY%Fe1y{a)de!l{*%q4(~yBNV;%ANQ3LoTvkUrHWK
z@N8|&_nU4GnWp=L^Vpexd=l>7$^d2(mjzD0RXcfM7KbK$!BF8STUhooN>yA^OAX|*
z3-=nfw_dsX<tY+F)PZh7+f7PQ=LKM!;Okzk-OfPBkYrUq48?-qz6}QQH{Twf@4s9}
z>ooI(#>>=sKD6zb_tDC;GbvkZDYMW(6{rYb%!O;Jm>-TeNMWf8*C?A-3#a1P_$RVm
zBbmk%81M($&56ow2Pj2@m{m!UgiWFvCVe{?Ahd=83Tm9rb*g26`FgXb!)&J=fqN7T
zM%R*OSJ2K7L|7fwar`twcVv^g%(U~vb-s4<Lw`H#o*-3|7l_oW`x|E-TqEUTYMi`|
zaKEK8ym;_#EdZAO@4MT$T`C>Td;s)s!9`RcH*nuN-OW!vL7@x;Xi1Uy-nPFufOa<x
zGzsCP&%jA{yJn-i^Eq(RMPN78OS3t4-mS6o$nIL_c?(b)rCvjP2W-w8D-*hqmTlB+
zQ9-Xys>B_+BQ>YRCNn6d54uF%v#HkOJe1SF)8Rw!i3Q3+qvpGqa0=JC+lA{+xTByf
zO@>-w*6)J+kidrtaFd}s>P8wO=w#0i>NjlA33+RpJW36vxZTsOI~D;of5eeTA~aRr
z`B9DAMfzE5@ERMZftdsJK$+7|aC$9ZSCa~bN}8630rh`hu?PB>UO<r;-&tW`Rivhn
z@Q@D&Jq!jW7c5xv^I{xX0O=#h^oF%Rn6O*U>!$AB+FZ`?|K#O3bS3fy$b2Ut>-cuh
zMX3lX@gSxb2Toiha<cHA24Wd1N3or!10oJ}hy)t;Sc>mlvM3*9AU|X`v~ElOZjPmF
zGVt9jsf6f72hlNro;s)3+FxfPixH;?QU@vgzu5ckc&_*M|LV}7c@j}lr(~0iWUG@x
zAv0SES&6L7=+-dH43UcLy^@tv$QB`Gl#!h++wZ!ZbMCMAr*r@Dy?=lGevke;$EWvt
zjq7!d=k<JsnK(iyk1*oZa|o~R2CvT<sX-_RTSGCtR#L=VuNFA+o{h+vdP9N|9{{8v
zJ0a9F!f;a56(o#~1#xQ8L7D4f;RfUX>H)o?-*aa`9~cBd^e=-Jga-(^Jo|Haka!L&
zj<*Zn(H{uDvFU$S_R3-58UWKun#bis3$tEKPEW4Gyk8AGA~d=!o9HB0<fA~^qyb|N
z$8|tG-~$_i8JHM@MqEQX7=s%Z$gh!T#ofm=!KG`J1oU$9%GuNlGYio#&GL-z@<XQ=
z`rSJp&-{)oYG{2obBuo3<{|(Yvy{zF&xD$wP^wMh`ik$5hH(%0*hDy%yzy8nu7zp6
z#jvm+BBY$5*IN!&26O35>~TVxOFTL@bJTrb?g1`>J!sqR6t>Wt0Q8!Rl!5p^;u2dv
zZv$VZgj(Yb3W@#FQgBqS!+fBB-dn_ZV<uZFXpOxA3Z@U=^s)+FB9|FdQ?#{AXzq}#
zN-__qy(oGms1ZzDBv_n$^6s9>Cv*-#H<K>_R4N(Uvc2+qILz<C>&N^#vCa!!3}ql$
zdIu)fh3bCg(wQ$L*@FG|(YaoCS!T6ca8rEqjE3RBf(GY$lOb9$XGf%4;|B3y37yA<
zrCwso6JiXL`*eQo0|)g<%*f(cLj;WRbbc#II(_42=v1g3T@WVz{WuQ3{4L0aLqY&a
zST8de(?SkHRukk`{+!lpu(?Vg$0suB{q4WVa9>O<T@m{>s16Nu0>>R3JZ{jgzL0k5
z6dK~&)OHhL9ebfFK60~2%A}8Wz~PRjO2-EwXiZpc(fV$~XI+ieAP8gd;g((R6uIgc
z`Gw;kD1{k-|CRISxEF%RcqvLNKQXsDEh)}fd4D&d#08I`fG!lp{mF5JY5lAW)=ZT{
zdzM#TuNHWgdIUfeO(!?0!$mboeHX}OdzE<7e4j6nSoh1O$;F@;+*m+_AX3Lsx?#ut
z%Js!#yNF+~zs>><6nZAU1BAa%GQo+W+@F$W2r_0~WehIDrQUmjogf_qlu15d`4d*$
zKC-`;)<oM-ve!+`paa~ivtf8kDrWF)#5EOH*LH8dN`S4KqMkn<S5z7y=REFj0ZE{I
z%gUR}$m}r|KH_NjYQ#w*Qp~lmG!2tzkUIX<(OhZ?`}g?$*417fKEvuPEYULNbHsL$
zTbscBvsT~ZFz4mGIH;2{yHlejPlp~&5!3??&MoZ?W>`185oLocfUQ4!;;w@RfCH$v
z2d|!DOl&^^^G4CbN1Xh8sG%--yh&Kj=ARydjMiA&i1YWxJ?m$90k-u}%w=g!2i@Vi
zqF0c<Z(hib0NLa3F1OxkLM?>z@Uhi)3=;<Q&=1pa;0&=9>D6Z#CAG&*r4eA(4<Jj>
zkoT6Ax&y|?K=0aeWZ)niUOu98eL##viPrfRG|t=_pCOzf@Hu_hNVfG>s1G4<p2Hg0
zOJ4rzd<bHFf$&N|(r0f-LE6RcJsmTWd#&muFv{Gmt_ua2vCp@*l}`3Ipq2aY9`Xpl
zSAV7~Z=iUz*xWnc+b^rXlZBEYN&};zhKHV6JMULE)l@5zOu?&<5csGG9g`yroOl=x
zP%2cbO1)@zV(!*7^POZ%7@Z_=wibQ>ilaS{|9}+Bmp|Ooz#sgY*P``bynBBAhlcMl
zSx7!Qw1wj!iePJ6y*ssHh&V?hzUMQc8&qsQZgv#_J;ws~5VkBhRa>$4vgRoZ;n>u@
z0$Vxp9h<iZrY7>c=tthR*BwcRs01@x!r-)F$hlkn^Bu8;i>R=AvZK%EzgRwnr4dk(
z<Utjrp<s@KC>6FBh)4f?`&1~emCAw27rwm)tW6EgtArv8Y_HoaP=kMXb06W{frA@{
zJ_c@HDE1)wg4NZ(xd5yng{xm=O4|ZOkxMY{e?I3Lh%S}^E0fsFxk57;aZVG@%4*Un
zJ_0`Z=ACt2gumaH1DAhr66lA^Kx9o1pf@dC5TlIyKYjzp7EbSx%vWLUlg}??UOB&c
za=E>Z=fyKIS65*;EZaJUO9?}cJMA^>G#++QVw<tJHq1r7Ld*PCP-1;C5TB5)v-`l2
zyRSg7>cL=+#FnG(DRW>x=Onm^CI-ygVK8q;)OQlhn>dz8o=Gm%V(QuSW&iWEOC=r)
z+j29e*CFBx(;={66&*Zfe|f7PPR&PDzlG`~hT_Fmj>HCa`knHdb%ZL&DXHJYh%NSK
zIebve+J!*AF2%=lN@le>-NilWzgbrk&r-6DnUJb6+GRofvpcx-IkhkCrv@G#zixoh
z@37d=y^j=7t7Jh(awn5)Vv@vz4Z)Qj@A-}q&ISlV5No6ShKL5E0(;mlI=GU7|CVbw
zzs>$>)4GYT)t-=XD`;AOYM(YIn?f^5it46#0(`2<j%LE7%io3Zn@@xUlsOVB4~zA|
zn_-y~8FTI02DMc+Zav+`8N7?b)R}M|Lxgc2j)G^mYl-)J2={ZyL|Q1sx{TG!aAkdu
z%roH_WpWv3LRrwJkCA<^EA>xNh7rIvaF;9C3>y1|hcE;%>$|mgC*9k${S5Q|V8%Du
zRD_?m@83@6kWT<k?!S+*>k8P9`2yfANh|Mg%3xRQ;jJ{pHxGa}KcD`U5xrE+pdb6j
zZ<g*OY4tk<sYAJyHckpaqfxBq7a@7KI>a#CdriQ{Zlq#_FA7M!0bj(jKRBAC@ey`W
zWLg<X@1hih%l=0+{<X%{4Y0y)uIE<AjQC7mu-QhP*NB<yQ*bKNFq&wuBU6B*ilk8x
zuevfcM(p(im}JT5p(<*q|La2iUstz*u-RZwpM&$66yhL$9t_l}^3dQl5S>`6k#s(y
zC4ea+HZT61qCb7mpHoD(@#ijDIXQpsA}|y`4c*EiEC2H>y7}`g`m?<K(_H{Yg`XbX
zO0wstyYSOp_$vzg=`Q?q7ydjTKpOt1yYPSMF63yr2C}J^<`gXs&s3W`??hFXstA$1
zHkMW$J&1%$i*2p`+IX<+FmF&tj@4oDoI|Ta{sBN>YnR1-Zwg_6^vMh0cJD-{Xskt6
zf87e$ErjDZm^`y-Xn;Uj?oQcPV$w{!2Z64fQCnB%zFga(${|M+4E4<t<Q+#=?N6RQ
z`$PnxP(Y(s=Hm^3<}v^)A&yjIq5}xS86c-*>QHl4G}X8<{zXraMrWrdm-d8thjo{4
zX@umK#}km<|0l{xX-MhB9I7TZfZhU7{<eyScU;(VJu_IeG>97Sod8lB`p|mtw2xO#
z!LXS{t3_D};MNNfrB?Qp^U_mu$F4otSKT*%PUM%xD?_wrN@xD}Jq{EM2y9+}&R3t=
ziQY1(#WuW~%7(%<pG&<lluifw^1~kup=?m@_1;qKXM}FC6)}3Ag3^f1_goJ^q2a(y
zM`BJ63K<9<I(_^ckx29xNSSOq<ZrMLd26djVTW$R+{DY~0!}E`)|MbLA3LolVzT!E
zp6dmu*kRVy0tEO<KzX2H?nCHcp;$}fv0jj!Y99ICwoemkT&Qzr0`>Nq_qK(F1~H`!
z-Xh*~qP>vf9B0lYrI76VM{3FSfM90`h#BQt?$KFdM4^T%L8y{yEp*Zpz4a<OI{K+&
ziQh+XAa(|D9*Y(74Ml+d8btIOR*#JGX}IQ=94~Y{W-(a+R8;Jj>-!*g?(643TzhL=
zWfAcd5Oss_cU^f%2}#_bSRM!DomZjQ<Hk8209mqqM+rKl;4ZRknKbV(|08#KX<iQi
zmgl+(Y*rQnP2s5fb2B5%OFcLVbvM^;YT_0WkJN+Akf5Z|Z-l?^J}b`ccqF{DXnt|T
zg#kc$HS7U7X}#_=+IK;ZO%!npRePO&L9j?IU<($Vo7rwLg<C0L(c|1F_`)||e!Bx1
zykE*W2njyO`<21=vhu3`o$z-^0biQY6)bW24x0oq<UEu;W<&iBqbPW#Wz)jQbGqYI
zqFP2LC>jiW&(o$A9?op&%y_(L+F-*w&1>=E?ScKlJ>r&>t1SUq2yQI*J>w-bVQ<_F
z`ApG=vSS^f0mF(lAhp<V$m|~nRM+n1DV^oO5JNyuc?0r;bG^s)udbY%Cl|9L3OhbF
z%sfkrIsdv>c-=1rT_C+tGx36S29fs=@_?iX|I}>A&zZ0_hx7`<=;ZiA1INWz9aR~2
zW1R>kG33{L#g3RY<S97EWAfim*dQznlw3R3dhKM<o<dv@2}#*FX1gDWxj0va9ZNA{
zTge}G1D~1Y{#+7p>GnR%#`~NPz5ADu{j<Wk9<uQaXJckY1fhfv9SUUt!w+s5jXnRY
zbGG-fFTg-A7Y+ux{zm73QiCs`l-y-oG_SR=w47nv@y<qRs^ex6+rQ`}#MRf_+W*9z
zb$_I|c>KomzpHPiCy%07_h7q{pxaG1iZg-KO4~O|oWD;kSj^15QM&Oy!=XvPHJkEB
z_K7qrnwN`AMHn_!yyS5(n2uN+@}D&nH5mbAMpVfRHAxPs)>P_0)<N%^nb?48S^kI5
zd0@jF*#F(-OI-}hc4fn;$k_@YFR<bczmMlMo{p@4q3)3&x};75gdIM-|3kyogULm!
z7wVP!H+{csaa%UeXxuoBS~zN{hZ>dhx6mFvMEQg&IHn5(-Hu^_^2~vUHQ9g}JVp&N
zif%HZ02LAd*svbd{>Xb~H=xWCHvP|@(*K@7JCqB}LSRT{P9e<jK}4FJ28apjfgs>L
zD0{AGTy{3>@>LWt22`S2TX~+{9^NbRi6HKg)^0xtfC0@NSbCxUhLR&5T-KTg*N&vq
z6~~i+%(gQFSG;1-DPrwgN#j-7sIljCVxcja1K>J3-awUGGy4lx(I`U$gpLEYYBxxf
z22hgq=q!Vge*@sY$2tMk>KPy!yDK12=!~`MoOe%hie&1F?hgvZP&RZA3=f&3xa8>T
zZN4Dx!T1$)m})q^p~}G}>9u<Y;u6cOq4~kQ>R5;47+6ZHfL$XfV3h{{Lqj8!{6IXq
zBZSt$m@JvtGq=C)E6}5xE&WnQM9ay*ci2k>o5$o7M`W_UHx!)+G0lzZq*j<Y#BrJr
z)DSuq$;(h#Z{;9E-M}ICV?Aynac;Cc25*X>CX5Xe88^>{Ha@pYIGxHw9&mfh9%|HT
zZCbdw_Loj53HyNf-VtiG%MZW+Zs=qZo0ZCDR~?&30*$`0x|7>Zg_g&(@&mHMG%gAC
zVuir?o0P#}7*76FL4cWK!WyMPp|aNn2RD1{iPV^z&%SPG*XLtuBJ4J>T=$vwfAGI^
zkqEalX8NpWH*A`xc-`Dh1X}T(`43hBc2prdZl|ok0cU_628gI8AK|z-1rAh3w^FpO
z+>$$Dz$&wq2X$-T+}qi8{qeje^-<~CjN<isuy<h>(1Bc9>e65P$m$ztqi{0IZMd`7
z^2{TU9%k%`(?wtumg9wjPl#_5ozZ2_tUJNj{JbZ(aHfPyl7hB}zc}L2?@mkh!Hs2V
zY^%2lB-dZLfYjZ>uY|KKl5?Ht?PmAfO9!vh4M}CZ@d6(TJ2cX*V^S+Y&f@_Gjq<eH
zQFo)#5YW4Q-hI~AO~fcS-m+4*@#qJCgN^F8ZPr}?ny-qO1+?#FY&J9<J*;&<Stes9
zALftT=`7Dk)T<g4Hj$4AYOm07W3Qg`H{grU6airAiaaj}orGxp_CZl9p1-GEKg&Xy
zjpW#n*9M*Jd`}{qc5FKAedYDT7p>d2OglnvBo@~6B_eQ(YG}tRD54~zCg69t;W?@=
zqfBREF{|=#d>^hWXR+~cvpB~s5rCFBW|Lh=P6Bchj%?VxtruI0078Ec&<+LNniNx5
znk4Tz)@G&X*u$pN&nyJ@%3v%Mq^KB*3US#gG?Y5zr7K4pNk@GnxN*B`Bgu%1-@B<R
zx9Q$G!eJik4lkZ1KkOVH=q3Vo@fyM|$3V47lA;>gJ}8mFN0v<+dK|hy+`|{d1)nzz
zvWB1i@`Tsn+*lL5esqho@vfl|Zav8;T6=BKUK_G#2352(m6q07(To|mA+Xi4IJ$qi
z7{JH7ep$b{k6EtrVKif@rA8<{Glzpp|MJ<FoFcU#m)!+TcpG8WXgV#a&;?KCWfMRS
zxDVn%qFK@G{1o8_akaG!uPf8vC#5Q+#^i0gZ_q;vqClv}l~%!;VD&vay*W!EwHG^v
z03sjrjKSz$Vy2RcZ8o^xoRreLbml?Z&_Kqarvi`@eb}yRd8eg=cTX+QlYBCIVTd9!
z(B5|{HAXyn8unz?u_jVv+o%{w3oNUBSj|SU#+8m@X1Dgq)~iZdEFUJ)e4W@)lQ{7<
z+*XM4UOpHAMn2T=e6Ws%l08+tJU4<mQz6w5%!-H3(s(er)>)nbE5k5oWo@ZMHlY>Y
zJr}ixP@%|bZ?8@LS!tHvdBK2bo-gjO=qlBBjkKz!Gf6yutwv&)EWgs&Fu{N@;ZPR_
zZPJ+czi%R}>-XYdA?YHgBP*oALMApnUS8<+358CJJ+E(X85F4Lm<6>nS|lQJez$ix
z2VN467B7&9v-g=(>wo_4+37N9;`?AR2Tqs=^nT}e3Aix1Cv6q%i$s0p>`9fS<MZA-
zMmbHq@<_V}^FTW*bMWmTp|(E24(2Sp*fVtSYARD(P*9j4taeZ1nc6QPGt8i~IG|b?
zF?jW20vUFqoj-c=9l$7e;x4nzQ=fkM<m4Smip@RQ-7ANnp07v{8|$nPv^<xf|AW#x
zo9TFCyYt=VHrWL?)~d^p?v;P(`s%DDj?gI>p_fNgO{18KBd`l_{T|cpE`X+z8PEU-
zOc$5StPDHErHs8VHR@Q4BcMk)V3Mag)8C*qZrTQVoJs&NO<P`=!BPDg)D-70^DLoC
zPPo>9rZ-L$SpvyXCLWmx8&-|lo$zk*X0X%~I6h{~+5b8u#U~dbd_3&b;c9iod=&l!
zP_pgsaI$*HSr5KYb`wD_hp*|vvq(AY=+^`^IJthKAodmNeQ&e%1q4i7B`eTx6tlNp
zhh`3|8%dP$Y@fAMwB)h~8w_?{IEW8ged{6ujg_QeP*^2uF7k<NnT|48=@bK3rZU^E
zf+gk+yp8I#rmB0=uR}LRi#O;&Uk7Vn4K^a8wylI3-!^R!SrZtDebxyU$EW*!BfsO!
zD2k-&wcqF+9#H5_I0P@lc&3qnxP+H!`*r(JKefLn@%YGr<Kw$Zs_pk$sZ^${7a*}*
z0%x*(z1{NCq8{)*wfJc<GUYK}3DlL#rF96>w(mb9W7TMn`-^(CU-m9DM|I#IBH42r
zVNLuOC)N$r(*${9A6ylT)+wofP;gsxzByBQV(N0jsZ>MgLeb*fVgvLdfmKH{D>?Jt
z09ILbdJOaw=~&(>=4-huPK`CDIA^_hp;3lvM~+Jqk4MJUt_ONUt0KlC9Ns}}Y~R1b
z2zp$6eLGMnI<aD1yT0;zWqQI=h1ezzmQIYY>}q`WP4O%;$IU50-H#f1Vu~Z)e*`aX
zC+d1d33}x0I4JoB^yJhuTl+0I+(a@Asy#psFQLK4fo=#@RWG@cQ#1E}mu5?%LT%vR
zs@3274?|b~U2OL75nGxO=#TQJxVl&-wzPwVSvh$dOQzLqPjDFq0nx$QtS`81@0a!h
zyh`y~T?ZWn4DkQ@fGXHD9i@VF3VuhVlc%7smlYe%dbj3H*+RO*hpM!^1LOV+0PLUe
zl%59Py5xdEh`w}12e|ZStZlNVct7|Ls6$I>y4pRD!A0n@--nz878WsOC+}>nKB2%8
zl>u6^)|x1cdYa-)8`uhHWYw?=pc#phuLY<xEwuYGc*vEYzA>a4`lio{Gwx#U%f4x=
zmJ_FW=I?ypU%*dbp-zpVxEoJ^v$`geJ8mBN7#pM-rKVF+eIx@UL<6@ZQ;EsEbU~Cn
zDh|&mIS%PDcy3^DJaT(8t%7B;Lh)~);0Wm*`{I6d*^u#$Dqt6b62(fO!L9r1W@ja(
zPw$}**8y`@0u9Nk@^XR~jN_s^kuCw$4+9^usUEr7`r-q8QwHq~+;<niY6)=)!C}`4
zr=QO+TLt!`sHD%hrI1nfvoEjjpfbI*tX#;^=%M<Hj5Mr$f$*IvGp!_z!W$>$;XCi>
zauQK-$9Il$JJz(uTdPtOqDnGjV-m0^JqU6gsF_9#8^{YoJE_$7=k^F?x&Ydrd5X%^
zcW0Z6mXxmz!<wPPW(^6ov0{{=6@Vbg)AOLaKE_Ob;|p$zaw2>hHLOLZpgHY>$HD9=
z@_F$8RI@j$NqlkHMy+9*Hq@TES;g?vzc%em7X*JQX~=xcPc2th&VDIQF(ydMhplLd
z>MkAVik^KeDpDk%&}a1dl`Faa)_5+kI!=mwgxv#HClECISU(96vv0~&qulu(53kEB
zM~7S$9lIwNK=!*}f~dPLc3Gtdn}UcTx6ROps~j7$-bwD!&sgYI%><d%H*iQk(oss)
z;e6<y7b#R}7pNhdn82^N2vLA`)HT^$j%qiQ_mm86i2=1_^5CeRS~y!@Vhfet$4sM^
z+)n)$3Kuy)_)l}a?0Y&>daUq)jZhmzB|&PvoYEP)Fuc&>JROwR-f^5}lr!wQkS*!=
zh|qu)%?%MESW(@rU?m>^lF8aW4w<YKia~`gK$~N3t?&H)%g{9L3mTD8QOQcF5U`!5
zNP)OdSiiQHq9IYpXga)e7=wFsBgspQsKah;6=c2&xIee?g<B|&t8Sc-F?{!+&s}T*
zCJ#2nLkp&m8dh_Zq4bmJnrz-Z^|$w21>um^-b>J9=mj#h{Kzaa18KA-i%-zsOAlkJ
zQD$+KN$dj;cm-w91->^SrO^JHRZ9_cYUs5>V;+64-#jKGElgu<z(!_eC=+pOkd<cz
zPty1)p?#aoct5xuTjU+&dmxV33k?~Z4rOqLY-eGvZ$n+y@)IX8OVFZ$YStiX>lD>;
zlhw@Va2N=kCv{teP95eb%L1>`pweV<%-5yL?O0rOW^{|W$Lr&9j$lLY9yodX@{>>(
zu<T>7>m)-<4!g)4hyaYPP!Y33X{MLzn5hss&EoPc%O{bYCW1n;G6B(Q?m>{}@~a%b
zCDPx`nvnQwPKUDtI|#Z$wB^P-uRyT`uD7KI#SW)2Pn|0IIgOJT<9Eh~qQksua+yKd
zkLAF5A}&2^JFH*o__l-z<o4jLtN{z8Gkv@G)&|`Foji|c&$dkTP=G%uX2|niDSiT*
z*@xf_EFjD^kO=G)gwBCoY)P34{2AxQa;3{KRfQ^smp(JvcA(wbCOj&h2;OV%Q%tYx
z&5VAbFPjxI;*Ye<*wCZnKbMOKHJl@UI;}~IylnL-C(QpgfR%acbX2-KwK{W{E3=5k
z(sDSsp9d0{rnqvVQyajA4ilQpDP1?)^`PVZX$GeNmn)=z#r^WTz$uSE5>K4tm%qXn
zvppFIhx4|%3mWl#sn$uYeFUw^9a<{wMJ~&x)DPhu_`(*yg50Y&W6*`C4f9CNATa*+
zpisf#@Ut0>QFZd_tgIZ0fuHIVd3EOMgIU-6ST;xN7u2(hupA07NQmS0uLhG7%+V4U
z6fG^IUL7GKgrue0L?-3E7{o*s9>&%O{c*0JR&kG!!DCB}mtx^5nVZ#eSKN}s4kPMk
z(i&i@zJ0WEQ9oe@si8Cw|1|&|<e^jjq1bk(F_)M?d&hEW_LBq5cKe=qVpk&o-3!47
z_@1OGML3oUaJp&iMDnxhkE)8qeT8j!IuQ(Wc)OngvsCdTq@N7e4wikcW`eyve$lbm
zZe4viAm}w~PPCd+y{S-c%fCEsiLkuNfX>6xoWiN33owwB@yMf)EvzbWk!AyCzX7D0
z+>ij!=WI-S)5{5u{T|aC9i6t)^8|fuP5suny0wlI)il3s*gDk2E@{+~MwJNC#7Tll
zpUMX)?Shp%l)P6jU+h27hY79VuOaw@&^QZz&2I(_2?LOT_5r<D0m#-qba$ygstq!+
zj;6o>pmvSXg+CM%en>@Rbo&Xi<REd~85wiFOAZjRcVRmRw5A+JZb~DAC)Q1BOko$9
zs`q6=D$9B>?eMu!D(p@~ZVT=!b&+SB&Sq?mb}?#-#ufyKYy%r}P%s*DjkNH8*3>}*
zD!*{X$HSSad|a7W#(?t10mvYd-)|s%pFBPK_Fb`%ho=KCyy=2V5)t3&YO{0ukWtQ#
zW}Vghn)eJi;p>Y{g4VdkgJ-22sHji4EesnCWW}`B948i88h5}_5cDwyG-(=`Nbjj;
zk~X|6$6&N=_E;>gChZWv1?PkL_I!g)+hc!z++P;<=gj?FmOuC0|04z^G6fu5pl6sI
z=@3&<(+^xPl`WZpseDa{AI887dILRE``8jwPnZbXy;CfN%<&i`KMj|mFPHl9LN}A%
z+xrL7fFbTQWLhYaWt%Zj<c6N$()fgAI_x_yo=x7`4pHaKu(`_~^qllSx!toCygseA
zddpt$e8mhF)j^M4{aefRZLO;9`ngLptEpqrZ{y8@^ZUB5Z_MhHqWI$-SEhWwAZONy
z(};rw#@LYgy-$8*uIjT=+Cl96T7b@FPpG=R@uKBP55<tR6r}65N>q$>8j(Q{i}7iC
zWu@RWDb;_5qUt_uJchp8BVXajKjRt*xrLId5FVlA;j2ZD!3lT)uEz=uG1e{eT!OW1
z%H7V>2Tr+$Pdc1&jR0~T{u;r-#1D|p&`4KJTH-X3vf4&GYsfxcfFRL(gHH46jxc#S
zSCysR&XDXW!NLb5x~ttb6+!jj5;%!d-6jLs3c3f$Zzx1Ekvn+gd|R*SF=Fy&3{&*p
z!!InlUN$a#F204748B;Af<ynpH93lfs1VQfn|I60M{E^^U}*-L5oyGL4J4zeU%+M1
zvpAw9!?3Kma60P+6q7QNx6%vp&Q~X9e>9%eGM`&iaSrAaN!~z0OF<?L)e7k|R7Agn
zgG_x)Zw8W%J3Ao$>V$~4`V55gNv2{P8JFjMUqME6H&O&p>&9KKhm;`tugE8Z91|JI
ziJA4%+&~Ox+LDElKINBgg98;KgHSl2H!4r2T3~Hg(88!j!eUMYW$Pap48&zlmFUV1
zn1n9&dGXZTManRRx?Xh1k<OsS^mHJ5Xj1;J4*hC_J|av2TJOGL0<WVYkYK0kC|y<N
zcS9}YBQ$m+?}b`QgBQC|ZWQDYa%d==Mw%e#Ij#|88mO#tFDAmC6|0qgve_q6d!d%+
zi%t>{20RjRp?HLi&n@ZabGM2h2NLE@*tvt)<XjXtGEWuRVNONw!6|AMW>G*~bRogc
z=Dh+(>NiWUh^ypJHR{SGKh9~jl)<F^gVy<}L7MU?<scK2EL*jjR)V_&a}db~k+}Nx
zR$)TlmAY(Xpy+Kmh~@d)tQy!qJc!HzcM>`Cp1CY9ar9~rZjdbw$n<&-^a+nRSh65w
z3*>E~D2~9L^uU0j!YnXi_OBWNldW1=2Q=Uc)#ok@17o6rCL}0kPqi#T9+ie=)Za~n
zMLhct3bxB@gBK|0?Ngbi_&6?RgF|e<^F68yY01JCS%!~IG)a%78`w;m5(XU{vhT2-
zUiNmguJ+@$_eF~*IBu4c2;>5uLT(@<Ni2i<ygHVo#~_%@4_-HHF^}r!Mp6k2NeX8U
zFNaEggQ#V><`2Hvj4fEhLeDhaHw#MpWsJd^3fI0N%mk|NVk^~K*wh)iAE5m)tXfj>
z+QpDtyw|HSLwfBwJ=8j18`0(<-9Eij(M+eO9(c`z0#{4JRc-|HJqtHn%?W^Bk3A=N
zu-AEP30vaT-i?`ml#PNFVy^e%DOz3}$=Fu|>4BE|vskt0jVv3!QF%Db^w+L3U#>pK
zj13MoV#^EIxh)iNd{IU5tzmNR*8%3rju0prWt%F6KEbT%SdzIDo2<}qX6)pi+M=9R
zXNv5on=(sHoW&8mpBgY(fR1_7V&?lcFd_74fkjU%4y)4jPA;XA=CS`%1<6$fN;P{&
z1-kE_?1UCmwoeT1J>=Q2Mvy**LYChl)ewovkqM$Jq#^wh9vEd7(cx61QpFKelKs^-
zu&c{hVjd~TpUNY1P>4ahUk0|RBkfT0(Mopz>W5j~elpY#c`GH@@+(72DoA7y)~+kS
zd`!U;x>X@Nto1$Xs+YJIkP8fV8&o?UK$0P0607YBG8~QG?YW1Tj+lN(-7REe`t8X4
zrz#i~y($Q-f~)KN9r_@XCJ1D9enN?}9XsWOxq&ciWEX5Q$N-OZs(JFa+jW_Kjc{I^
z+x3F?wM0fyWoYqR9>S!;Y&SyO5f>31N_?}Ot_DkBJL;q=*NT9pr5Bm0u`S$*?fqqG
z;omk50b{KNa6eLOw!%AM?1d<|(R@5v5q3qcpCx=IV=79`VLZo%^3O~rRI+A@mgkGM
z-UibrczYW|yG1LzQ1t?*3H8BRV*4D~MkP5oQ|yC`2}{#emMnBuZ)T-RH2<-gB0`j0
zNa5A%CLUL0aUFvsQugQP1|YKzl@rXaj^?xIjzgTkE1+@wGsgSppGofGeRmbcXBG-a
z{PpIHInfp?jB>CxD|YdfbQr)HL4N-ZA+HNN0t7QYC@RYa5>}izqpQIp4ALby7qU>M
zfZn$dx@sjU96#7|rv6=`xTeRm!2$E2$U33{LzI$lxd8AgF6M-uVx&t!4r7U(!@GxO
zfxP!ulavA**Vq4(BYn0lbV{q~3JnS2>e_W#knxH0PEcQc=!!?4iBuE2?MusZ7|T*_
zF(Pw)&nOMHvIw+P-^$EyhAQxV^4s;6a#;OOgih}j+w<%;pd^oeD_UNHtf0tKFsSrB
zfi?!U5%exVoPn4HFGPDW9=L0zL%|v?FHIp8C8&6Y5mb@eZ?GwUIv)|iQ+?krLa5p=
z9bBiCFd0bxUCQ{hhBMVu4S}sMt^avgDTL0A@3Pw0=dIw#(&~3x^G<eztCKlsLy}?T
z;EYbfD)VlEYr6bFkLTXUJgr|Y>Fi~-)%5JJb{6VimRdS`rT^H&rN>p^6wC(mSrz%|
zq^A)k1?|^=C?w--6(artY`ErV6&9+fr<s|Dz82#-kB*WK;QGb@dluZv2gddS_>6jx
zo%BTM6pe8fcL)BDSDyKqTqjqIk@bc|>7fK3_p9|b^-@eFV`U7F^A6SXVf_(vl#Iii
zg)^OFK(WTPRp9f@^tDeBDGCy}yShrfJ=HP{tlx*&U?Ll2N)BG?e*2(J{nt3CKE788
zjWehSU?DeC5k%Xm2h0?SRxc~ZEr+AG*FbSm<J!mxN)D<|=0#2tYTx#|7kU^5O7*sb
zO~8)sPB%)<x9j-g*g0f^f^G~NCm5vm0;@!&Gt>9h>vBL=w}J5-7i8?yz^QwKX;AFc
zE}xYq&zjK+HZg%}P|cD~_v1YD3XvVb`TWW6VX*$>sl18linZ(b&~DHsc@Wm$ew|2K
zvD+~r67`!8%xX}G=0n;<|Cb}P&4Hha5VQ%Z4PyXyvVJ7X)ANr2AgY)aF{E{5XSSTE
zJ{C-N)v`Fkstr7rVE20pFQPIyI7R{tm@B+bg|6+}AT}d`ycR8I#lqY209sK3Wv-TJ
z896w?qrd}WD}fWeS*@WZ_wo#!^M?y3qq(eam_<`oc-uysum#49HXtdeg*_19T6}S*
zDMhv+mIc2eC)kNKZYs0-J<|z0hNO5nhGB2CY?kl&$jmv|oT9VW`ADnu^50wlGhbRh
z*zz4z<a22*sXg!b=!7uK&7dMPAWIm74tAEJ*eCP3%VH(lE|x}I+EGU8E%1T25dS20
zE74m}SQ$ofnWe;IaXV2Z3AwQc(#1+>&W7=GXc!On1nbnuMyqMnh|UioOe<>pDfO-W
zkU+0CD3p>h>M<`Vj=(~HO97r$6I^T`<nTm27h2v8UMPn^bBhG$_JBWlS?rLD+1lnt
zaW2DR-)Re=z{Gt0<e?A#-_B*w(F1cIV@76_{+pGA+7q9wCUrIqS$2+XO<^3>RsN;N
zTpXZkkkMi@Ua27D$)0}3`Fxo(#i(gkqY9QrkO997B9vhP)qKSiUAa&TqYtSU$H1M|
z_dNooEv)@yw0J9s)hGTgHd`*2uAl4in43c=$a#dNPUULC*O8Z+izWI-v#FU28cRVz
zU!*JFM27@~UNkASRpWVKN0eKD6O|vbVLu>ys5bF-m5oBN{2*ldlt!Zf@}jB<WioR8
zw&QQNWk#jG0c^?z{=4_x*wF$Ubmz9zut+rwGt0h*M9;Ym`peGXi}_M5NL7kObGYBB
z|5e5)E<Q7}5uEoO<^@wK<pnH#4-6`3MIG&^Wy`fv-z&(JAk&oT^rOikmb*&%(fYad
ziO0gp*~jTEzb}v85%bFYmvS_RNM%;D_9qKL4#l9ktnmsq+ki&nUSF35J!DTMZ;TlW
zgUl-w$|NWR!T~o}QbJxfwQ>6FBLVI`IRIa~Y!(yU<EJ%tR>?@XA(z*=M1w*lTJ3R#
z*=$$Krw;4fInPRk&+OIvm4xT3Q=oJ`^Rc;5#qq{TpRh5lrtKTONvlTIc&1Qh0H6M<
z12D2h6n6BPpbMAo`MF9}{^q^+czRTRhFj)j6L{1e7E+<~Jph6m7T|2Z0d&$CYEqUm
zhPyU6C^#AQCxe=a2&K^wxZ5SlR(oSw&iWkx`fw#{{CMUKlVFA75k~TVEx~bU8_;7@
zpRNMrK2=*4XS(sUZdB3WuoW~B;<ynQq|4e~@5KkT6bB27G7IPJhg2`HeDMEckJ0hi
z1^IG_Z9_P~#ZTM!>Ct+J`l_==kLpcNV68qwyb%~*E{wj-f{H_QQInQ!o85RhQ-%$$
zHUo()>EZ+{EehuXgpk#SoVdtRl?DGv6QDY8t0jMfX+s=CNzo@>2Squ9=L1o5+K-L1
zp-w6kMMPS`x{HITCDh(dcrS+ce~;drk9daDW;q#oGf0&)zJd-1^DZ`>S!t;ohUAF#
zDi%R4p=>Gz!|zc?DnWoLNK0YwP<auTyAD=`n6(Spv@+0+fyeWh{O0&b`EZQ>Q;*kl
zxz{JEc;t|t<$(b+<nHUv80QF(`K^;-d7)$ks2cRFTaI3WXkP8KG9X?}5ui<9HuWK-
zL04`pBf3LjEZqP{M8CV<(lVDLOGkKK{+s5v5hx2)tk0+)K|bsnt)RuEo{^W4PoCb&
zbZmnYS5ZFHAV7L~18c!)3eMy)iD*OS2>sYn-zWB~J{rG1@*G&TkC!8@36xRZ)K{{7
zzZeCDf*Tf?8WnQ2-tVMH3&IzFSEplx3WO56NwOz58W}|)v=>6>?cwAO3gMtcV^KbI
z{N(M;<w2j+Ez23NR|hTTA;uD+RBqQ1+!j?=3fKvCd2Pr0C&Yj<ii^^t;pipMz%h|f
zTarLDVRj^V{2UIqTMN*Qbp&>va{8uW^DH-!4fIy>0F)souhpv5l~cE8t2Xu5ILcAI
zHz25Iz*t76IJ)NpRKk?68O<$+1YO;8L|6RujU<<5`YpSLqgrlMaNPM0d-mDodudNK
z1}#6Bj-<b|Qa|`V>tJO;h`)w0p_5FxxO%KIgK>o{!?p~nwow$OK*xet=!{%8=+p^f
z;>As)-c%S94KZI<*#EEQ_5^X>5;0o;gz8Os2U7f=;mNkjqDO}JD&g7YVRX#FXXvI0
zY!gG)l-OrJtOCI&B*6$74YC51at$57>4nquz%iye?TF+f(X1xw+DW>#o0uMLH#$3>
zqrTojVU&?v*HS~3hovW3#w8f=aS$TYjC-ap^1fE2i0;W&UrxS0OtpN@r#!<B5SU1M
zS5jgvb`^F<+ge^yw8VR5aORU+=?K}bc?#Z7Wup+H9@%H#_r!0l6i;Y(t-6AxM2Oqp
zD7b(}Qk6W>`U&f|?bEd6+=bjIlw{E+O&9dM)0?CIzy<;2yrOwD0!tn+(iX6@X<9$E
zy^^x9JEUCSSak%JPrZV4c{jLcwGoDpjPeHmnqQ$V_Phqnd(R)tZPudLQUIgIw*QQt
zGt~8W1@tfkgq>p%3-EU_8(RW|frn>$7kKSQq4fFABDtw((C4;?4QtkXg3cnV4czl-
z!k`OI`W)pIQcbMb7)bPqWM!?U&voTr*#`-U>C(cp-TveU>r-@!9*Q{FNT}IvMfr#i
z;LKHO3xdaXTK%~Z_=YX@8v0xVPe%uJmhF)oj$GCH*oO%KwhQ=x`{?G7w?{xKgDkqm
zZ9J!Y`pr(cZkk;~sua^v{5c@dfel0Hg9rpOrG6@Xl{~O%G*Z1d0vVzUfw!4j*vK{<
z&?OnrE>#MD)T+}q-=?r}2yQbJ`~0tu<pl5>^RZhEh(6&LSG97e-WE2jr82SAgiwHl
z6@MibKKSG;-)i|e*XbR&@xfufoDcB<`7fz;Cn!r`Q=gY_3crO3g%x7&p|mnqZBbGn
zU#E4CY(Q9ebY5|UWyWKxW7qtGHpx`F_n0I5KL&b5vARW?sZ7=BekXo|F08>9(~v)V
z%PvSkm#LTpyV-+-Lo2Yzx;YnsQHc0nN6CJ_Ly0<cVAzx&utDBbyDQAHmb7|5vMhxe
zuzmzdOAk*ypaNsekPFmQy5nZx5Zw5rGIZ5}xbn?(%#b4ycRaV+EshK)3mw+8Q(r#-
z2{f+VZqY&NtWa`Pmv3sjg#h?$C^sSzQA|!0ts^>ArUn62oC7sM$#RvKdd!okhRR*O
zwJu@W7L&ccCuMWz6Q*)tw!rZh!ZstfvUufGLx>O3puvqgGtl6p6wfy$hXgRw^%*|&
zLi_*{QLz*x@51QAZ`_`S_fuYC`(pFH)yEQ;BY7Pt|IOK(c>h)8TEPvNvDKL}ylzO>
zFp~6J7)FB074S^#>3N9r3|H9Sz<0F1p`-4>>dx&ET+V8%r?YuF%K5pPPpK~h1Iy*c
z*rZ*z{*6q$X$tX#5BND8+3JhQY$Ah*^LFp{sEuLlM`&zk=SzlT@P6Oc(2?e32Qc1N
zz3RO>w2O2ihtZw+Rfhe1Dz6e<2&TT;p&vzii-LMPTXm}?q#=XJ?o=oX|AjOZ+LDuL
zpW2I^2#_)l*dja({(}YJP6D3A$*8fdci4yQ-OxKh;fPs9qynd`h`%8Z!De)C9&~wN
z3k~aWCjdi~VuJs_8_#Cgu^8OSlw$7COS)Fw(t?S6=i#$&*NL`m(un#jOYvRlqq@k4
ziPsEENBBSJ-a=|vr2S=VJA2Wa>f@C+O&+lLd8XNcD*8cCU;LCxHN2r=0pz&#AbG%k
zd=c{C$50su-%!b3jH-5x%1CcL>7>^k11Dwf>x%&bJ7TCTI(Xp3Ha$2&U-Ah8OM1X@
zGjm3Cbzvm|`N%-xHN43JxhPXFzOK-2pm7s{0)rKHvS7dvG-(0dKlcE^lyI3k)Q{3c
zz7Ad(a9ENQ<5cE|q;Z@|(Ba_x)-(ow;2sQ14sx7t4o{*h8k8Q^gor;4PPm{)q98rc
z`B2Y?g(bySBBP*kpv`4?^#Z_sH?<2q)Rdg2fBRVQA9dO5s;9Yy$u}t6jI?tAS)boH
z8X6mhSKto_)PxxbY!!5v9zy9K+}irV&!mr;{uQ9bb+;?ksX~sF3!F3!tP>gK=*mb0
z^6sgE9GtSe_TVj*&C~e++Dv=3j`H_VEmju|qv*g+NT+IGnW^YS<sxJb58~z)08-+v
zD1TYj7llDZgKoFSx7Kf#?U-xI8JAwlm<Ph_-=+bv_}{IghAIa1TgvSg0Nv;VVM87|
z;h0Lpz8ZuLWw$Ctn<!pkCihr^03$%t&mal0`z^GJbAXUXgi}zD^xA+<Gyik^_o$AB
zK>5d0u5suEY}rjKT}StJG6lV;Do-~+y9mEBQ9cIDt2VuL)0}PFJNb9A>Mw^OMa04C
z{Q|PkdyXb3j@?27Gc&t)wHd^Da*}}~-`qQ<{cm3wCoIwkcd%XPA_^55+-rWsZfFHs
zx2M#N2T=WhlJiAmRLz1Lo(Pa0+y@O3<y<w>F=s;*Z0<!DQVmJbI{}5|C4lz$Li13@
zo6p#q1DU3xsC*T2mdjZPom47<wg5Q1gyC>$ECw4(k4%Qe`sgWkLB^b&na%bZ;B-q6
z!F<gaTx1rZjPu`~0&b5Tc|Kb&mUZxUy@Ng>s$1)~3@{Ix8ln;=SOY_za|N(~5UgXP
zsXf?Xefs1nq0@UUeqVr^A1CuUo{P~PGNEy??=)cT(hWZ$v?<}&XJc$Sr7T3S#S5yq
z{g&ZqXWip`CSwNu(*AcZde&_Ru#4T<aKWr0OU>C`SgGWO3@UaFI$PvJ5}*6p=T~m&
zR;jQs%b*qPyu33gV_**@Wkm&j)cq<*7^q(}hG-i%F7|6-5;@9bA}9F-7v$R&Td%#|
zGMn>;*6`5|quWsn!_dEK7G<JX7iahH?2m=TIUjRyI01M{Pcfhk5c7kJqW0;KClF(u
zZL?~gg($5$$f(lX{0F3_qsA(vX1<1@hWZ@4)&!jMfm#&kLP^(VL+-s4PFt00_E2+0
zT3xqvNqSsP=eK@bjY`2IYTzK82MmTugw-@m9tt%&NWSjENRP7gD|QZJvV>e;i@tl^
z_WqaOZYwDC9V9oj{Dh5Mu$0;IHa7xpN#&@6Oz4-M07cpIHtjL<(Ah)PH~osDKfeKT
z)4YYJ>GLcL6M-p?(VtsmM5p5H;Hajo7~clk9eI8)+1&4=-&6`+QxPr>*^h%5Wn?VP
z^2bUUd*ik4|G@)h&_p83h`k*u$8PD#TV+LOyZs^?;nTH#m1j|1akH^!wm6Q8|0T$Y
zP*!+sL;vT@mq(TwfM(B#{#supx;Qt*_M#gOj4Ee22w?3n?IUChCE?RcM>XS8VFcao
zR0y3Z0x1U<(UfknFWCS(PmD}FzoB}D(<H#1M5=R*>qjUNzHh%kNr*~DmvIQ~a&@I_
zU+<p170Ph)q07>27m6Wa;J)nVw+Lr;nr_H*D87h5jy%+FP(dO$^(=)c6_={gx39b!
zHE=*s%9;TT)B@TLXkeW&)7a0xVG=!mv_XEU0F8YYP-50{RTvCP)t=8su$;zvHg`LZ
zocA?wdedM&d@L@0U$DjrHoj5Vf|FNTVt5ap4;^N4`2>fvH2ZjM?{p=HNEWKiffWnM
ziyG>ns*x`8uE_?Ox+DCsl4n|bMQPB%q^@`gMZqpK&ShqS&J*C-*$Rud*Fe4eH}V7z
zLrduc8J#W9#;+y}UgasQ9H5BgULB9&Hcf_YYTeVcd-XQ`sEP?WUjX-Emyer-+NFDp
zE8utL!GZl^YaS$@t5jaWlvf?C3@G5^$PIK8GrAqo8uSM>W|Xeqp?6Vyb%0|_;J5M)
z?4VEdUP!l`?=hH{s+|Jil)P5RMLm=B_Q*V|G!_)JM+4y99-arut*N!F-q}GSl(NPQ
zlA`1pJ9{3VwPy5G)p6o068NM8(O@S6?-SvegXMSrzCp5-kC9qJ=P+d0ye~JmnV|<m
z>x)RsL3BhKjrdsJDDKUaE`AOL|J=*j;hi28?rt6_{X-bx`5*`Ky!S30)yfWQ3ys_^
z45ZT_(C*y|C3f0|<#JwsMIKXJV(V#YiC|N*jqfGF=JO{T6E?5wZ>|OlyCd9~ji~)q
zwQod!HhK(_X1f8IEum((p=9k{D>Dk~N6<O(AZWT`WPYVWA&CV7AJzuHGZ^`nSuE8c
z`2wcmCLjb=>V>sQZ}j{zpeSi=Et@fEFa~nk%(fTCk5#{%aOZM70F>4Re%FmcVnGn^
z7^8%zWkyp}SVOwJQE8xbM&%Hsb65PpOn(Z@!2?E;jkR54`uSC=novHrFsVLpZCD(p
zpd})J!@V(__7*blzP_`KD@6aWGG@`s!1>h4_Evhh?)Uz7OAXr~cT9uoVcR4G%h8&<
z!h@n;y+gDs_bVkB%jgX<GA$USlgGd=>F_A>b)y`#y1%@9yr{fW?e=(W3jtw|ORoz0
zJZc|R<Prr;><`r?GQ5_9xF5PDc2dk)f(S$je65yzQ;S6In%}?8VNnp%To`EZhn=I)
zjna>>`~<z0J7C6d6j!ku+4xcQg$P(&ilRNsY6gbJa|{26{bBFEqHdJ|g*{ligV5Xb
zhZOjzu~sH35e9G6pg?nwoT=Va!svN8@2z*siyh1Q86{RW*F*o5fWuInGA63049bxT
zV?o#ir}3+CU8F=B)XgO+Hk<?Vg)JSYLyxm?6u)oD1}ud>&wP}4J@)I?a1KwlLYZkV
z)J7_Ua_yn8C|JPHkq{J=CVk*Bm7A2z!)QY}4f}Sg%hDNbQH-w@<YVjkFTcHSN*&5O
ze}2G)A|RR@k_dOw3vF%8DIP{I;R1SKFwK$^J>M`Z{Uz4&dsK(M`kYx`-IE?!L;c1j
zEmc$IkvGUK&MqIQN`=*t@+$S&-ob*7#F2D+p)15;-?%;27Xn+SM_H>`xm~kqt~B|6
z!2~cnNW+t7OanOjH|*xS>Z``Nh8ts9=zKirZu0V@!iR&^)c!%&Adk733kb`vU~h_%
z)N`edztHI4CO>xivqWBpm`O79#`6r?n;s)FD<A=ga&uebJoH)4Od(Bw1myBAK))%Q
zw(7Sx(7qW+{-h#&P__js+k1?H!j89$8+Ob>EU_JhzI@}4Lkd14O-kEn!67}m#YbU(
zTn3!HS3MC906)oD*{D%>&h>D)gjC;t-{KBKW80)682pF1(WZ1Lk>zyNqu#5APtvo<
zO*UAIB;(yjibsxhg~8b(uS~=vN3W}1;1=ev4m5aMZU)YsZj{>fdM}Q)meqLM7BBXO
z7`a*lJt>~_sOX&fDqa@x0E4$`0D94@=4<v!%4ru2BY!S5m3LcQj%*)DNt-r>{e|TX
z<p{}!B-v%2)9|nCTSiiarW)sM7XdXPN=e@%ra|#2`m-938J3t2Gu5u1Iox1Mr@>sU
zU}RS%F<d#Z#B3LO;IjKJmA8Kg=U&B#pyjAyuAU2^Vw1$xWrqM`GIZrowLzePq;Ymq
z2tsmtet_h(!+AT|F299v-qvk#HCX-=6a_9-(>s6Ogc?XNt(08#ZD5gkzXBSSbp4ik
zOHCu041EBD>tUPg`{%{l?>`QWq-$GMClT%lc$1ybW6Ap^((cgegKY8?K4}R~@=}P|
zE`bY6zwme3vtJ<l<V|riGVI_)Ms-2fWq_b)ZqFPPct<8w4H&O=zxk3Hcct?%S{@JM
zi3RzJbxEV#3n1Tc7i~xc&<-aagy``u@W7WvO59?#t)c}u3FGKcMr|pe@AiUTmV`<z
zbW4VC6f5_Kze^~Mzji&k7pm00@$z}BtE0M=7jTEO*nOW5;qQ@>>ypN~7vR`*p#F9P
z4o9AQi1<c!a=N(KAaMuEsF3vf<uvfA_-@U-j}hI4+ChLwN;SHu-lNr&d2=6$TGsZy
z2e5v9UuY8BpRTjkNi)1g2iR0Cxpnhi8XZRq?2fqp(fa$2{qho@vDYPYNabUYx^diY
z(1&mY1;gZP*4L?W>hlthzY!*q4@0232eK~heYwQ${+Aqe#tg4hAR)4%LIfeQCpcPc
z=xw36-ODx|nqT+V<-UR=hXD<T%{%P8!0L5kS}9$4$9pnEjbYf9mw(uneeUpCT%)b4
z)%u{plIBW@?-w|RbP&aXf)ITjMK0JtijcK$wjTjT7O*j@C#MFq0C$VnTLBxB5HRp$
zy+Co#o<LR3+-zD(b*LbD`9$1Xoj`(>*%twi{QgLDb;1+SFH~0=eZL_01NApNc@Ez<
z*ggBs*O>}<FKoF^waP{=l4(OzANyWgh$IZcI{<Eo!nya>{<1!Evt&YHrTY30w@9m1
z<ghr%_iiOF4wZFD3*3&7JsAU<7<TL_glhaT6j#RVaI=SW0oh!XxO&r7;2F3F4hf?u
z6oN_t+q0smSvi#z*rCsHXV{fc)&_{(_Mx$ZgdCvj%Z-!nzF)WsCKXVM4c+~jsT3D2
znq%JE``Fe&wfibGA~R#XOJFGV^(9!ICgsUf?zU<&e>hxOy;osjJcmy)6C8cH8d>RT
zPYJLka|_6U$WuZeh<a>1p_~Gv9eW7EA}L9tWP_42^bEvX#(S-CT@NkLd>gehh$bWQ
z9OQvMT79+-TRnMG)X*}}-udyMcqBD;Rmm$et@eOnl;x$s1o(Mb65bse1%Mu*M~Qar
zx-Gv#5BN&=!0Le#55pet{#SzpP2T|yXvQii$vS;goC8~=pTrYYw!shggO$Wc+e3JF
zd1Y*)sE9iXZ4}Ty+_c}Wnz$-l`5{4M)Zp4|w)!pKBiV_)r-|<xS9R<FKA{{=xGh*w
zM%bFtM=J=^4!Urnw^dhqE}#)3FHvGNl;GO}=?~;78=n7?@@e($Vsl_frl>LL3LRLS
z3GVTNgxkVul7{&Xyc|Htnusf7uM2BT1!U3hLu<*c&wM1f;g8tfe{O?z&g$FAo5PR{
zawLnaBQrh#&5e}G*2E!!=0>PIv6eFvTukUEfz`DFX{y@~t81mngfQWm*k~NKHUH}B
zw~TB@9uEyo2XOaKz>rYK50K_=SvPH*R`NT)x&p9vo$@YNUGa}tVRcEs>e}MgOzbw_
zenk-`fS=y3ZM86icmWPc&WmYS>h3Tk97>Hu7XwalXcvz=^z6~<r3`Q_SY0cvIndIQ
zUr9%nLeS#+x#50pI51H^H{8z+xB8U$dBXiX;lS_s|NDefn%k{C(p!hMLE{6X0L^QB
zs{Vt+V1^9<U&Yq6t#%p23&4#SU`(P1-wRt@H(Hw!9Szq)F8K5JvJ2Z*`zLi&;Om^@
z-n`EZIFrCNc)w94YqhroBVLXE=)#7``PF}q4F>n4AnH?2IEU|Eo$QI7=ED5X#jfkL
zT&LKigw-W=zX1~NKC*GR(Kz4GYKP^fcs0DXv1CH$GWel0EW8SE>4UkCtKAIn9h_9*
z&m%rb5vvQ1k#TTosIkCY_iG5|%P*RR$Znd>tdhSWPc$2K3h;>>e46`6nfGE3r8b`L
z3H~H#(V?vO?0~1r%X36ueWjk^`W$`Mk59&QSARbqP;&5m`=>pm>_Ol_s%(c#+|Q4b
zDA$ALE4+%<ueqyI8h<%;ktJ|5b<<bEoCGfqjNU)7o}mZt+Jx`bmNwB%`d%31?Ypf-
z90j~W6c+~5?ek28pbTE7cwZ68WAUFCBH3cIq>>BqpZ{~nfd3|P8>Ush6gTn_pIs*l
zpFPa`agG@3!h2uY27cxAZo>i+H`NKhx^`RIR^l3lm;5LTzYO2jyt8^0!(?<&K?_G%
zSG+38XsmIyp@wnleb;PD@=|wT#OLaC2Pl5{baD9bt~)C!j#YP?+yw6`%Izgc+70m^
zcmd(-wBU*+dLcQYw~hDw57YV|mjQg?e^@(L$l)O>yBTXqab4g??}VQsuH=#~{4474
z_viF<+en+^LH6qoOx5*`z5K+%5|7Nq&1NDjyyeHy8i9RnZyx&t+IJIPqPty~l(B8a
zw*y@DN}xzsTaoAC>E3?Lq?-V)NP?RgGbVs$6Y+|Fst2w73vqh5V&G=tCekaW;TiwH
zBL~lLf)5!OOd?sm;*ipE6XA1H=J@f`{|PmS8zB^~Xne0k^Wzmf8zCR~n)yd;bTeF1
za9D}ReTEiD+#l>RZ_7=)y(C|vTLWMCwW?Z+I3#FXs)1CrAz!bnoG@6hr}x3*C)ZR)
z|9A-p8i!wBcOM{btTnF5@RXOoT1F6JIJ|FUBmAOn_C}bC<TK?}V2`qUo*;Q&csw=k
z#V^GZeI(C+wG6jfGQ3Ed*N2Cg;~RThyr(4Itq!NrY<cGz(rvoE?>hYAboGoikt}y(
zCZhcueDbsj2M~VMTsr)svro%uH3SlVFoOh6gXFEG1d_L1<H`?EY$JcyrgJ1O)eI|W
zZf@fsshXZRq|o3LHLsd_NF2sVDcFjl#}fsKDLHuH6%+U+d!KXo>Qf#rcEZH?h*xw3
zt_|YN7T~8cF4FXW-%cElOS`Szi2oX{-w%CUOxs?QZnwIZa7B`kO$0G+UzziK7^)jI
zr0YlA6V_<vhnqJvi35Vi2q%om%R`o=3%ia2A96%Gf)terbIt@miM+EU-Eei8@b_~<
zygy`!_lxUpg{OSoUSmcwRpQri^EBJp(f#;h|5yZo{k9XgXz_kKQjuo(mU!?|l>UL{
z;~C5@9b?KroVn`gR7tB8kiM4|+M0?a()u<$Ll7SSey!S0(kq7H3S;v`@*h8B4&Lg0
zvxW)j72fbx|3GgD&$x}F;-&qX^a^b6jgO1t{P7tE=+fLiKf0Opk`%ba{#-06<{ln$
z7&%kY*NzF0T=MD>JiRI;ERV!E!A=ZCNAE^issCz64_CPeJidQzkPL}&f=kp-z$J{g
z%TJQJ3R>_KE5>eN65|9DoB+RMe!P3*hr{Hbcpq0b83#L9f<JEcz9V_Q>oVMGq4p0_
zFdsam1K)U0S1eIXQ9Kf-@#7NH8j`&Wr*$t(*IeuUf$hXMz9I*PV0y1@0f|#gws8_k
zc?yMJ&XJtsAJ4#$c(2z`C3#!0XurXC9R4LKj5sBcDd@OO?PwAsop|wi9IK9pgWSZI
z>hp)K=wx}p+Jn@n)HT5`W+DrMBvOoXK43XIXk|o5!-9?5XP|oPt3&E0)G^_Byg4#J
zMtn3LiT~{4Mq0zzs@o0;|LDex-+-YS=aeQH5m#zlqiJg=UJwTaCr1l*m6cRC>B5fO
z1+S%kF+lj>kFWI~A_73xfhS~hMVNqESy}zB_QF%D4m|oqvKL&xLgv&cd2E#T#~1s@
zZe`qWxJBJ0MmPR=i82B+SCZF+L&H@PW|vo5JMYIUvM_TSm}pNrU@y_u1ZwF0c!f1Q
zKJJ(F!=zvGkJzS_UG-0JQ9Lm)5M2<P{zl$jm{!9uQ{U&deba$MPZf5YJH>J<S^8+~
z?hxaOB+skT0rl?vC!}sj#Y*}A%ICrP^m*(dt{#3Z=2M)f>UT%a)-I)t3?%A@J53)i
z%Bpp4YRUO-U*KGnQQHpxGm-x|>wAmINUd{3kghUGE+4+(2rsOdC+|LyEd2fVAw*@0
ze*Ti=Ghs1E0_~(Bm88TTlAl~%VfxqqlWX_tE_|uihnrWnlB}gv0eE;xZtjl`<f}fI
z0IzQ!u0TnLFb8#aV1F@O@E|pVWE;Jp&{%t6v%e(iPn9soYC-ouk<2mNxbz%+N~`2q
z2a=y$Bk*KaBcV3K_c&4tzvv0ug_6cPtpE<Z?(0rU^a~rzA3RPTT7rc<to|9Wrf+4%
z7AH-CkFll(2tQ4u164x%mV8e3dX)=!)lX)fPc@t?)$}EKHMk$-MR@uTCc6C3xRh#s
zhQf9gKVUuLX@Et;c-k*GKWeVcJ(}joVHxwQzL<Uh)XNXv*Zw1-@I75TD(Zj7th-HI
z5xv*~P2oWWpp?CP(D}+jJtg3ZIa*Hc8^1pU+%=pwfTD2``2_z)-Ry3c^+j#H^i*Gs
zJ3t%HBkbVtg8|=u1)Vv1)vSNd%zLt>?=H*7J{g3Wr6X>6K4p0><s%2@uU8I_0_Kr%
z9xCR6BEv><mCy)XWD3=96PvPA0hamvCwUH6KMGLC&@Or^JE#9w)a&st@iLooL1LHm
z8RX>f<}F=82LaovnGFz0Q!PK;W^n^3M#_;vxap(wSP6|Ger5)s$C>h=q?~RjHs8`B
zv4K87q=!W=#P%^@d#~F7)X@;t()9rdqG|~3D-zQW%Y+Tu=W~hP=V3U{4K2%5YX9C6
z2M=CDac)F%W<Xn~!2IV8Vgg5p>$k5fAlsW#pmf_IoSQazSzuTz)$oy|j7Hfx-9}~A
zYQb>XX=vAtGoAwu3%?R}+h>JZ&}r5JtuKDi6{OX|0HEsl+~skuWxsHs3}_r-;c}dO
z#!#&d1f_<)4y|43s)gf$Edaj?EwKjXRoNZ4)i1j5vj6riq}I^@s_Vh}J4RV@0IJIk
zK=V)ODA?1Z?ELR}L1+6F73|it?{o)Q|Jr)GyVo(y)!j))PMebC;7gmLFHXJ_Wjg(8
zOweA>=&YpqS;^c}ZX(Zw1wZm{TF0HBRV6s#ZSMH2uf{}sB6JAA+Vo+`CFv2!DzNL@
zap)S%0B9JW(Y&5b<gxrOTUxlw_WC2VumvDn8f2EJa5FUVc-b*61#~G&T2zbX8dtW-
z(=9*)S^tRt&3G0<o-GyusQC!YUd#207nibMt=+KqemLNgC|}kT-s$*zDpq#wK3wJ8
zvg&^o?z|#jA%=5tGDeJ}FnVu?rZ&K*JoMdO7CFs}El-Fw^3K)TmGY!Rb9Zru{mu?Y
zpJUTq8o6Uz+GbKOK1u3#c4!<c6I)7q*d!M8dM>YMK99Xne=&+lBu@3+5Mr38vr?Qm
z96&l<Ghk4;Z+Wq3Ik@BwO<Eg_cZBkxZw}C~pi;T#UqJZLs^FN4Tu)pt0_yiWJ!-tW
zujtJ~1UFYa6+yLl&ut1wPI`$Q;l+m7dwNl^$i(sP6!0{p?;gZ{VuwF=M##4=0Obwm
z{L=R#oeRj+U1F)%wl!|>URMvc-d~cseWen5wPuWI(x_Hsu7={|oQ$)fdqcI^R_CrB
zM^NO8uc0RC*jnuX<iKHOsm1HtN(>ew{?JV`^Budlv;?=VjQ}KVt#Pws(WFq#Gpqbk
zi`^XqMQ{_(-ax?Ua>FZp3l6s^rELM|*XX7EuY<Ol9kpHg)!ej`4{0x-r5?BVbQyfG
zv(qyJZC5q|B4KL_f*5%ZM<jGjy?<dkMt2={TJ3M2K`Np(#?CAa+L4*emK-|f+lQdN
zmqS~x)KrrDq#UyHYELKI7O**VA7lt|f;Uc>z5%#=1Fjq8GVvXilsiJ_p#MN~!5+F^
z40oufRda1mF@sG#5&z0DS$QoDVR^`|o86Q9?_7|pC>bz$=va>%O$p5&GZC5{ER<ck
zC_cu++56;?tf;l~`A-_qPa(T6xvzCl`+NM=TCt7&-4YLtSK3T^jpQ8EtsUUUZid^c
zIn7+WZRU%jJw3wRbuECG5xNoi{EFPZzJ8QYDRG{-FWn0!j?2?#yl0c(?6Kekn+_mg
z9_cQQ9+;kUzau*NMETP*uAU$lq1>0EnlG2J16%){n%IW^u0#*}1-}i9tb^AsTxH2I
zwE3G0Ah$<wHhq0KK=)r<yT||a;$zu;Z&<%Mflih6xcO<@@%^W)9RkV?83q=>2HP3}
zW^|;x@<7#q#y-CR6oJo}j(EDlSriU1pu7y|fL1Qk%4A&Tx64Xd&}dT@nz(=7+G+8`
zzPbMHj&!cMed}la=#=Lt%zbRf*PJV`GrV9{)a4^Y(P3wr2>2o8J;#8tLTz|(N~Wl5
zueDdjSm=fx5AN!dwx=F<&3JY3MwoQw9x4@HzL}z$<2e{AT0?VPn2+9{s;p|i-cc@}
zJ<la|%bJ~X8U{N~sa82VhtX}PXS{!5zxK|rt`9rS6&9T4z*s!8+cvE7J(|lY&R%P5
z+w$nPa@(Lgy+SwM=6xsw09t#-lCZ88ZE44}qhw`754KvxC*63JZ%C`!ez5NSRD{E%
zEsM4x2$`7=j>t-{&YkRkAfD)L_V1j@73OQewV`bOW(4}-!oj9xzvzX7P^DmOU-7*U
zg|pS>Hsj{aHq*c`!Fp(?d3$fiK8~xkhAAv34q9l<a=HChAiBTbu;^3|&;McXz2mX)
z-uU5?LMq896iUe+$;xaPMZ+dDG9n4ttGg+)BBQLx&fZ%Ch3qX98JXFe=Q`%CzR&ae
z{r-JkpFh6eS0DF%pL3mSo$FlZI`3;)%$XthV?mP>uA9cFi_*P2P}>wEiR-T20sSD9
zP1SJ)#!QHnw!cVLM|+S(@YqN6#3D@qM|g6m1NJuon%OTBXNc^7?1(*u&;(4?)jF6N
zQk#SurUg2>IXr2laOvBZrZNux44k=t6a0N_%RvW+hLj)$1%&H%a0mfu>A!krDcmpP
z*?-{&BhEE=Vp;Zn)O|GGgt7lwViVlVwAr)`Xz<3eq6b7S>IjM<C0-LMaR)B<eTN+>
zaN*4lF@Fx{{ns9OY0rBu&v^Flh_aswR{4;8BTXpSG=h&lta4I^VL4*%y`#l)vY)BP
z8S!rGZv{U!-DCt6(Uxtw1kK$nk8vEtB#kAZ0&q*K9DKs5aX!uR(YkxmvWK}`k9hHO
z#QR;sXuJ)|yppKTM1&W3=Pd1#@GB>;Ab@dBPkd?_>(=;D7m6;IrcZV>6Kec<5W0ll
z)5G=Wo9Z$uu)UIhTB#T6v-{pxIoUkrD`9Azb(Rne&26dF#b+tAps)&_?y<g1p5R4y
zTX#d$Xfx!1$mz-3JHe~LR$&vt1yOpv-_LlPy3S8ubsY0bnS0JoET1=v;+g4`JFc3g
z^Tmr@tsgj<ui7i4XF&H*`Bw$h*M7I@qUun7gz2}UzF9CUyCV=mIvXh2oAZ4noE!oG
zz=W2=DIY%PAM>8bED?(LV*f4QM!x8tot-gLY&eTK|41`C!_VQ@St8qO3iSCOz8g`!
z^UUiX$RFN8Gb745a50rTV4(FRAcV&_n9iezMFm-z0qhrfV#`t7)_HmHR2|vl_VJ+U
zawc<t+enkWY9Dp2qG#+ro?r;RV=jP{F~^trcEq&&sa#&DtX~@M4sh{z40Yo4;T&#L
z&A*^#(UHD~ymrb!Kc7^sJ{|w)w~W1KD#GsO7AKK)k#dpMXZU?((vxsb8gXQqn&(Ss
zZ0+%TXZ`*35ii?dHFVW*h_)(TROpZw`^oy@itABAnrAX>)vK8^g}<=ri094-P%n2{
zG}Ef${w&Kr)?`c-FIxVYvda_--H9@ePGQB{g>`T?kRGZU;9Ct)sys#Q!!qp6h4k!y
zw;$Uh)85Yf>i}2X=Tlc@=+E(t_S{`?Q*j_~+0|^q5+|P?9dIHlKLP6B^Rk7`iHyc|
z!XBIqt)~_u%;6cBLWVKywv#yuzB*Zv$x9kiMvD--Mcr{~`b4eRmN5U3)?qL{c7EE{
z`gh=nz5evKz&SjQYG3@*7ABKtLdyZ;S6KgR%#2o4s&lZ6$+S;skgc7LUPz_Gu(_D>
zNj*0g6&vk{AU~ns5+ZXlN%Q#6>wRtQYTvZiwY~40#xL?7^6W0j&bW~()Uv-U6kb$L
zGq;d`@G<jw`?w{e$&`9(l(Lez`_P5LNIS(?e9fZU*V|MJJjdr2=Y@hg^Ia*t`Wm&V
zc`ui|Yd`4=Q5@ec{8Z*^Pp)-kEkYDvWTpSiXr;<_F|G|BUU@Z8->940V~2vfT`?jb
zr@99opr%+M+23tAFo438PU$-9bD8Qk(eRd%Kwa3uSz0C6ZXC2qcXoan!T<UB%|OyU
zxm;7tj?VKB;L$-^rUNR|e84%`T2_YMdQC>$ts^Tt%9^@2xdV_2y6W?TrgM8Izh9PC
zDELbM=AOn>vM(7;Ln9#qx~X5ZdwkY$<`a!#hJ+U5xf5-wxFMxP$2BHnlq@>b7K9~f
z>=c!APU*!k`kK0kUMRm(lNSgmJPK4OLbif$P^WZwFFJS8`f&EQqpI8(Y41FMN6U7-
zCtFAI_Ej<~v9rsPnB@EKp`K2Fv{R}nFv$teUiaFvjgvnXS^RxfX>#hiw$>qIPfG^~
z2I5-CxYQ?GWZli%3=fqUO&CKs9yd21dN9vG(JJin^2F8U(*xOOX?ho_uO^Mt>m29?
zzje9z_>?Z)H9%d9Hp-ONYQ{kiBzr?$*_iaW-6$Bl#_UyDGL-Z8J=487pvuyW9?6p@
zFlK*>o-}LorAwv4{KEGShA=-6-&Mjn^>@S7`LYx2VMN_yL8C3d;T$$Ro6u-fGHfX}
zucKIMou}UNxzZ>;D<Uc-$9{pD%fS&|@i9J_$h+gY<g1DA9IiQ#Fl76RJz=2n)B_db
z9?$MVZN<+`CgtxaK8$K4Sd}fFUTnT<r@iYViX2E1zUed+470Y7m56_*PFS=xE;DnV
zpQ!D%{)S5T6}u>P4wP6PX0N8E!akw<K-xf`F>T0oHD7i}3d5S-6l!~Z8i1UN-wOvo
zupX#<fU&*M^IGM(j!r@7%@SjK>{`lu-romvI%e+E3FA8RfJK@W;Sr`#w*ByR=V^&D
z96(R3dZ{^8rDuN@nFoyr@Mn~eAsvc6HCk+)V4UPIsGu}$0lSw}+VX61c>aYJbo3Zz
za^W3<GvH865S<5~_7(rkP6*r-73e>TwCPZpEJa40^synFR({E-*ZC|_%Bt%9ZLLS{
z!w_vM4!O`)*{cePqK0slOgwJszI&ZHq-+9;Uh3qflXe}R+&6FS&U4{Q{GiFgdm59`
z8>)*BD@Z-Z9X%oqG;f38>p5$0N-s4qEt6!z;kTEX&=;TO_@CO~rPLIt%C+>-bez?m
zDzG-bzDqjz&!dy~I!C0ZEdeH6$thBspxMkdQH;XOvp`d(GWi^5Qu)o?edxWf%b6~V
zd$YV88eFW*%E^9HjlR=v_w5$)V%VuvhYJ?6AnNi_ADrcS@Ljig)=^w!7;3ApJnHF6
zYX#JBm9~=_Cetm)3S}D(CYbxAO&U=o9O?M;gjc>SCmvoO74e^U6U*J_=711gH5%^p
z=H&%XT@V$Tq9`l!acGCP0(v|3<Dmtt3fDDpWg)%r!IHENVG}sFDcF@-);mqTj=QI^
zpBMYSh88`^z`3BX_BSu){ZhKGOgf8ec8o^cwZAqBr=+-XIE3Y8vOYLG8rOj@%aU5}
z`|26d`-8|`JE(;ForvmS6=q<Z&xNhDd^X*5I$3U%wPTo-<(NF_<mab5QKy3(SJJZQ
zujVQ0(sRb(C0}y?r*jRHCO?b)EKNrDjwQQSduq^_6*+tiVVOodYtHM=H+|hhu5rBp
z))FHf7JulCGLLco&=)$-_bsGn6J}eer#m39i%D+d>Uzh~mocOptWA(n!>UKn*DmWm
zp6J<_tckEh_2beSV&)Dzc<NT!8b1`SXr=TrFmlm}`06)dW_kq54NA6Q9`8I>e&9nj
zzU=1$rQA-kKxk<9He2oyQ|Cw}a8yxUNh6AVL}gcj=JH%)<y?-!Qk#ZBR<qZYGtnec
zx5b-LJ27*3|4g3a94=NiUm`1?mtK^<3~y&*AFw2q>r?o+-<M>uP#Cs?ZHxIFE2o{b
zN`#U?XSUx_`UmrzB-g$tyr<J}DzO-4*Ry)o?DmoCflT-9mon)A+)+78&^#k5X5bOY
zmA-tNh7s2!7JW9C(PiicsT|OH!8#(}y9v6k@)?O-cKV$mWEybTX<;;QrbblgyBr(c
z;RJVW#cwq{j^cNPDp==ChcYTTAJ$0-kG)XoB1`b?%ARSgd@p}_$a_mTbq)^VgugS_
zARqi}F}(ntTTk_MRL?wDrT|hmWPo3$OxX*6wxNr()uPWoa-_DlG6VLExi=jsWz*YQ
zl-uakcb+C@ZUM!cDWzgj(E;6I_78>fU$y2r^veau^ji;`2e&$ykIut(t#V_i6u^fQ
zTb%5btBhPv>@O8AKPVjYR7V5aDon*hWZr31w~Fg`A^MV?K{{#QNfI12MLgu8>qp*F
zYN965JnQ4*GB2TQj6m@E+d<#;6a?JTC3f_C`pzW2Z}c$?c<PYbFi*~f@BKFOP!1oD
zolwc1Ilhr;YZo3K3NHrVPQP2H=#$u~Fn1tnb%(eBJOZtczzWDm!7Q<~=bv*9t{k^!
zTMXdUReXp1BZf|4Q@ipCk_3XX=eKgFo;5z4Y@|kg6ZG3e9G_jg!mp~G8}AWb(65TC
zTe?^>@NL-m`BaHfXY#N<ykk?SNA`Qh)23X<r6BJqyw#}e){Xs0sMZ9n4`SvD#$F|q
zt0oH@U1M>oH;Szcn1xs0*-9@kEOggs$cgJ>OgNtqXS(<w{gfSST$SYR@Tx>SmGwan
zakfjM%%f(yANSw|gOzqfl*Zd6-7L0<%2CQJ;V?OtEWIsqPDEkz36r%3riIG*b&y)|
zjuZ?h>9`~eFTMmg=da{8w^pVDFqFK>ObJy(TbXVB!1X~#Iu?0H17Z=Usg&UQSmS%M
zDwVd13KF>(jgI!&3ag4ao}n=6$*8$+_1&n!u)>jXhg$0-hxeAJDj&~;FbZ#TUMj&O
z8QnWH3~nQeeMdS(G*$;09TS|)XHja$&gYqmxwc4B0sd`uc)w;|8Xgr%P2B>mG<pZa
z9K~36DlRvS!+Ug@0-J_zK$cCRjP3gbDV-se40emwvOpU1a!!RD?PoE=tlfM><JLpN
zN(*qtD#+`%bGzu~;2<^1H_~g(G@bXJld&wEPw%zc#VipKf;z~J_fxWw__6*0^U9o?
zm#uJ?WlsXM52XyNr;e}^cKHmkKLUET&uIHm8YRZX%%)|gwcaSAP?`Mn9apE(jCeap
zm3GCn(&e`1?X6kD@A92z^YwSRRtc@?<XB<>E1}nNBz|;c^m>YY8(9K@4!ysIYWX(O
zl$~M%$3$zHf6@GWksLUlSK*O;(1?QToA;zdQ@9=U021t?2t<aP_qyPRg{$$^yc&bD
z-OTT=rLv#9nIx^WFi7?*y2C^z-gU^_A-&#DtOolc4s!#d@jY@K`KyU)%nS0qr9TUR
z(V~9OPRBpennVbPqWjHn6T8+{+Xmh*jtI_EX|e=w5AmNqeb+J0^OC8Y!tb+SpN-8a
z89OuDV=J`(PXupXbar0u{<r>SieXWajlAx7zjd;LVNdgmKJ8&5INhR69-J(_Fqjsg
zY;Tmi@LUhlM(p|A-(%xvs@}dI08EA;x#MC&N#8|TA{B;D3QMaR<Nyq0Y_e%ELRcRs
z?|>KhG0A|g*Zkh>lYw3Hee}lMP$@x6HU@>+wt&!buT#^()g-TRhb2lozAqUIf&JYx
zGTFF{a^0$R-z*%jf{u~3tesXkP9)C$OQns(m3Hmpw#lF0n<8AxlQ?N@W!H)<x)(bp
z<E_Y~NKtgFAT%j@sT{W{4}hUNm^CECkBqj23DD9wwQAFcti;VhDc}2Ohem>Ix>0Lh
znPoUwXp~>)>yu`I{$h%{DOCp&QB$f|vo@(<9}Sv?A9%5+I0Y^z7+))cQ+Y;9iL1s$
zs7j8r3N;;WeI<{9UH`mGvrJhRi(@_{^`knzkDHM+7ENSPr)n5Xl?Az*oxqOb4nTx|
ze530d2R{xciPWNdw{@-c+aZV~eyds3E10`{&bUCQbwKKBeTOxAcodsFdbvZnhd%x6
z6xDW*Q*H-71k7x6ddupg22qGW%(0}~=bC%{sw2KbQoO^W^G(d?<==qkmcnHIVdp_o
zvHtli=m#3xiVz@u6;_fV{bRHa@0t92&)G95OufCMM(#;uQcC1eYKC5fm%WJ%4xTO7
zkFm`bnP4DPhYU#3)2lL-0L4&0PSwKL&-n00Qt-@_C9Vj0OaEKn%&XH+&AT1g0h1`k
zQzl-3<1{$&)=o~6j+>r5MJ$q|am#(<mwDUddGfqgTzc^!ex9&s4q#sBKhEms65qSx
z(+$ti%di!p;4Wszhn+a;u%D^ZI-Yn=__?F{6z$;zqAqTuf>r=clgra{7A*(t^ri|p
zY^p2B1=kP316Lah2b;ITa@y1HU+D)sSaKR&S!jb;7`l>8j@>D$JlhJG)=mY6b}YI!
zkfX6G^{4Aq-Djk@!)H{=fgoDr?ey3%@xEum#RUMf?9g4G;@EILwR>PtW#^dYt|YhY
z{-bdQWyBJPWrmJva8*oaCo$Do+^CwWixq%Kn<)(6pL_P@IV5@hbJx4ASqB6Y+wpv{
zub8TyK*&b5@t}tO9PI;KPC4<}C&OScJSGJIMRApk=0aV^A$GzNs;^0>&*G|T8!nkO
zNB9{E<8Y3$c3P7x^H<tjG4%yKm}h9E66xBpFq?jUAOAbQVd4Cv>UR)qkHzws>PO;*
zX23obcylNFRxhAW3<>s~C1NV|PI3N>#}6;qbydWAQB2m)AoAV$L@q*HV|so{k|y@Z
zH}jtWa1%3C04eK=IdN;;ga(N6ncPUkN4H<Sjj#qIg^N)~N`=aT#nysv&8p>02iU@`
zreBhEACavyuR4pTF#jxkpHEs5rvlsL{vpW44)lCKA?z0Gv_Az9^uIB3fp*X#uvUdd
z2b9N8W!5|IlH{s$GF`q)aIlM8IQ8{N>ta&fMH~A?NHnv|9a&yg5EqB7o4y8UNS?eJ
z?+^~nOd(!O!AVZKO9id)rc$v@WQncFsM+spQhvSu^=A~f`&C)O<(Gt9yMsF|F0n*Y
zB*-BUx`P=-@1NrxzAM@bP9VM`EmP27UdN~0Y?)KwM%GOuwS!#j`?AHVqTeqAV1R!{
zHvU)N?qh0odG!y!RTtadjbRap(ycRxs|HfcL^A6r*T*H-X4%xp{EoRXrSUxB7+=pa
zINF_BoZ|QRg#Y33<#N62%LmTa>2KN3Wzm^GEpu_LU`6M|p9cqXLu_{lw>P=p{Q%cF
z3KK^k92xByhkjn=q0-V2u-rxIMS;=F-F_4i3$&eZ^to3*GG_)iPen$W^DPz7-?ZwP
zifL(F$D+@)Yw$oqfn$9Se%wO8Qb7zH;FkcYI>#Im{~3N~qZ|=rWodstLLoVsUvxNS
z_yWU8AJ(W!TxfjJN;>UHp|G-r)(cK&RmSx@6RjXob-IrZ6YbgcTPH1r`e1*`u_C`B
zGh)v3*H|*h&IZqjMjrOfTs*;UbFmlUVRCXw<ECo65(R`-n%XVIdS3ocZ8D5<y!AsP
zHkEdr{rsviw%Qk64J7Of_j6^PZtj<gC`#Tp$@-Dg)XitH6=3Ol?FH=ojGew2X~?Aw
z!L_9MXxtP&Q@wuNMb3&Fr^-QT3YkSt_b=T)-9nyn^z$JxKepnq{Jmo<na-QivPZZe
zb2I$&6b6-td;jq&6v7#`O<Z`h+zc%WgQzGqTpIbi7rT1l(ji-A+U<m5mxb<ka<1W`
zj*L0-6h)tDPyZ8ixbkeRlp;Dk*$DMyGN(Kn)CJv%8ZK9RjExwnIo5auoi@^3TspA*
zd9umq>&%=HT?bAseHAu6!v3_P@@KmK)`5^*tV@2+r6<gy`S7zaU2byKA@AvdCI1*9
z-8_rPKW;ikZ{su^g&W6~7AN>C2uhQj8&7Q9pP-#&k1FH=n&`Ba*Yc*4grBLtcjJ+D
zzsX6uP^d#a?$^2ZPbM2csy<{zo%mvSF=b=K6VWCRu=c$_#W{Z2o+LYeGN)tCVSr}$
z1uE^|kClau5Kz)NR#Wxd(M5plk9ExG9D`;90<p$2JbF|v;ilRGfT7U)ed!5r?%T|`
zg%N_Q*5?%HeP~_JpzhUUJ^j6CL-}ob)Q5XO!;raU|8nQ2H-G_{M8ZYzs<Ftr7!R*|
zkC=?E!`CMa8K_mKc5@v5L;%3i1fs-h(PCGVGlgqUE4UE2RhBn)CklOihggrP;01NZ
zO*Y?Wu3FdlNu=#VNVdnDkq6keeY$N&;X^n-v&*S`&wj4rx(t*#*gI|=$Id_bMM5}P
zneuoGDVN<Vsn?$m72(OWwQAG!x(WqB%;=&+Ki7OuI{Wg%6<P?T598tL5nkX3wf{yU
zLiet1WD14SF?0{Y+itkxKk%J*R$<7=Fl&xwAhF1UyZMm-$Qp@><_l!DICPwCrAe6Y
z-|mx<+s#@e&DULXbCfbMpT&53$#>EYH@q~zWZ=_m`q+rAw|ldoo$=)TMY!4=)ewt0
z@tItwS^b7R<J;SI3vSn}3tDM)T8%WJVS34X9t-x?28Uq!_{sUxL*s;+GaqA)oD8^o
zE^qgi-GqY_jy!J@TNziQ3yv!|zG`L{Blkqrbu4%^YaXLLu>GM~4qWnqpZLXakmlHh
zsL1&pt|*JOTkXQOaJwu(FLInxZI4{A0lX}-l-jyhD&H>EYOvzGZ`}@^y=dc9p>h2D
zlzmb+ARf@<`!5gsoju{x)Zc!LL3Wt;99Pokdiz$mT1}|P6rQ7jeXEstdjPsq4n$!@
zQCsJycVp=@Pt3}H$aVBf@l5O{D0(V7`cuMzB4uu|tc<aLwy8u|F^#)8KA2U{-GxK0
zxzX=2gGU*D*=K1n;xRuHD!o=Ze8qJB1s%Oujn)vj<JvcFEd(fRu8!hSkpV|z)cV=x
zXetFbxvK43wW&8G9MQq)Us*g^3q{Oac{&Nbc{@APF3s|CYlQhEJht@>vKvCl%wyZs
z8%&|waXDiu;GHSW_XV9u6!0?#gt>~W<#S4VT9%**e3oG=SJ0@Nq0(gdCxph`Kj+9^
z_lb{~(%;0jQ0|s>F@^pF3wfFMD_z@fRea96xji0EldKQgDs3B5VjO(R1{=y)?%6z_
znuy``(Q{pD)aP|?BquZ`%wkKg<YYGD50g(_;q<x7cekedE@)*nNjbjM-q<XZMSlRu
zvyso$$slz=w-$@G-{pN9HlOmb`=0~sy3hKt1n)gr7P(l7id9e=Z#G#`CErcS=ju>U
zcS_6z3yIQHO2_b}G2IWz_C7wLZL@g%D)|dn(xKtYP1M=Js#>L>m#g;Xtn@JzVIy%L
zb_b4u^qCXOqV#12LxJ68Hj}sQa&vP5QX(0U)VRt-PJbQSRuyojUujQGiLFqZFG}~4
zFMcU_5!n6Hhb34gn(1Z1jo@ysk(;wm6`E^<rczZh#NHwsP=Q-tnWd*93pc8t3gUWo
z(>&3geafbiFR0>!#}PBvlk+?#Xtv+P%)w29nx5ICZDEI>wlvMCv&6*V99su4e(EP>
zP@<5_vb@V79f7ym^o6&ls`%nj^hO!@l^7k?nZ>h$#fw!(+BfI2Xuf35fdgTR;Go>X
zz~o^$JmtZzTn!1Yvek6rh~Md31*dB15^D$80a@89e}cH#i#pA=vQ_P()=SVGquiyk
z^M_Tahrf-T&`lBiBWU6~-h2{<d(r)KFti+HT3CX`!ZYEpAj0QbpkQNd+pyTht(Uvx
z&Z@!Q`=PFrTaKEPajBxnDAKHKN305UBBN8_IG4uL;(<8pxC5;rL(tEmEk#gLuFm4S
zX^Ad^EH>rW)L^k2Q0=&`R^j`me2muClz*wc$f)VDM)1L%`b(~jrL!#FMJBW57DWTS
zb3>;58csrgFhnd%`Mp2?3{Dp;JIE}fGKjog?v4JVbEnen^{QBiPV3I@aa0jjy{SO2
z!Nt;Ws+dE=t5-^zY@T6xrCi6}za`Y&R6T$2mJi;1*_(q4{E~Lr%=^OZ^-4oW=7%yg
zeJV+nNuM=hMqdl1Tek(Vd#VNo!$n_`Lq0&~B?IJJdXv7}S9F8dIfhVZZU<U4FtW{W
zsmzyRdtPwMa6`4B{9616N*U_0x-`q#vdY!tGAjsOo_dvC&X*yS_IU0!0i5^w8SC@!
zQ`oierav7$@Z}*@y&KEy8;?LgGTr>4=jZqR{<Zy&3fF-K-&;b^?X&kgx6hP|ZI>p?
zA_bR1&yCj=D7zhcJJ-X-U#s*Ks?iBJQ_4J6b@8O0ogj2+D2+YyFja)8wBmXYx866F
z3_&f;v|E5J!4~h4UhMY7#xIN|{YRzpRO+A-^>=?RDSew><twQy)4xuNWM$65v3U4B
zH14e*C<n$bD){>mpVm`sfymyaTdvG@SJL;JUYF0<Y|D6W*idQqtieEs+924X^<2T8
znxJ7@9h+q3CEj8IpX2Vj%3(sLg2l3#rfNllg59QM)0#d5+Vh7?Z1^=e$&-p$!Z~xa
zCz`kB3GkI&j1KirZ9JamNX9{D@mSt*fDarI%ivNFa=lZy&FXM(I#HFe)e+G`POfc1
zIrLD}d9gG)nSdDe0gY|sncc$`Xx|X1BIMdJ?te%i>E0~|f-m@-7iQk=;s2E~TAr=B
zAgkGa`XfbxbrG|RuG?5MV{9_J^#{V3GokVLX}`Ge$TCWjC5V&1>KibYPF$}rOd6`x
zfBt?d2O5MZ4m~4G>Hu&rxE{p0nqb#9;L?4m%kgshTt_~tMjCHF5lkG_QVAD^Jr$^Y
z>I%!PH<cc8Z6|S<!_8UFQYM_G>w8(bOa9d}Lf51Ph<fE8H;nsl*RMZV@QbUPblyYM
z4%c9=nY=d0KK+AsRwufh4piph@%ikY5A{C5@s{eE9?XK;{4IcH6XI3bB=<_?i^X%{
zUalBL9aCr%dk=xEejGs;##2LEs)>EBL(mmnG_;|_TJLtEF46tp7UuBwEb^RD5Lz7O
z6`M7^W9PCI8|JjVM0#0oxEfU~3&a~9F~P7qj=I(xy}`fstIT_)f}sK`&uYH<-e9@|
zKsJ+SW@VF8&QKnGe(Ls1c6aW9^f6!F{84uWe+H_QaeKH!WierM6z4l}?!2!0D3@N(
zSKoks)cdt)a`t6(AZj|5GOV<e+%(nLO8z!*m{qSgI)hO&JDuc4U_>kIT5@e7dY=9!
z%;%+&66~LS)8=_SJjU57lgCQD@S8iQMw?~)gk@~@c(|~@?3kXIoQ__Ajt754CaY#d
z+&DT~=d+dDEo!%Tu}>?eMGlT*@TmEW5Lyicl%{Hc4yKz@%JS=sWbk;mbF4#eSt*$>
z5rl;*^HYS_$&Y`zcynjx4v!hy@P+3;Q65HKRjxiIDv-!J()7;!eRZp8vE6)n<OIEU
z;^0X`>c^NGg8h7e=Y0mE7`;i#Nc*@Hl-Evq`yCFu=C(HAITgA)#(Xl$vxMUL0RJ)O
z6yTHH>4lWpkjgLs&iwX1`pm3jl&gib?8sz`Q(kCMsuLutqFWHocYQ1~n-hZPyx@1T
zg*c`Db2-Q2?dpz9lXAleh=nZLbHo|m4UR*nz!q~62-PGM^dNCPc!5JVKTfj@I=baY
z37iv33XpH^NWRR;HtGMmr**`>ak@1K-JF<tipo5)&DP)TqFakv7ly{&u!LLjZ_SJ6
z;vY`T0*Y}$@i80&IAIS4IOJ@3X(Pp&s5ZzU@*-vPO@p5y#{S}yL4^cc1<Q{FCrS6s
z;t4e)XA2vmjV4=k|8JwnXS+SkTS<tcBCdiq25Y%wK+_`F-?BWn?r+y|$4)#?2Doje
zGT*w=PKztlv|ka1;cY<3Qa*h)Urosp+CJ#d)?9Cf2E3VC>NYdsO;*A>(n$sOKCf2;
z5S81MLj~?ry;5h%=wVkjw?s8&G6JDhVC{k3XZ=GLjy^s7@`y#9%y>I5QQS?NM1~y5
zTqQ%(U*WP(+~%cApk!aiD9n>lqbxhPDDdmX`10%aD^2M5<C!GG30Bw+uA?;NU$E!-
zbf<9YB7CIkr+pH&oKYEX9kYU`J};sk>ny=XnOY$mpA4X8xf%)<{F)W4Wxlt{#=?`5
z1%Jt@YicMXi!V{_o;ADOj4BmDovx@*&#_zm-hzko1j;J7PTLA>RNNjS!_0(qLH_uT
zC8{enlwD$l`1Xmt{{xm_AEyT6JjaU)R}->Z^X~+-wx_$#*%9*T+2&+gW<rwU(`1Gf
zPVKS%346!8^IT}<aKoW&<f*h&<K$MvpC`01Dek?Tjkm0aSTQq-cgh7%b&z+^k4z5r
z508Op<E<DSX?oMyAA35Klie*Z=QjQJ2a@1VM0Q;l)vDR4E$#~?5Y;OA@8qW-eJ&&b
zn;U&z*__0d2P5a<Asxlowk9EL1BKGP=YaBf(mm`t{(f|q;vd7VU;){0Jvb((xT@G{
zGbTW#)M(I0woP3`4+1dcaFn_(s<0*338N)W!17U<G%mf$4?u3bVF~f_%>;)bqZG7F
zHbZ*j)k=LTWNmy_`tn$U5nmk6&)+hBp)=UO+ZR!TU4K!_;RluJT8>af0VjHNafpxy
z$&?1+7OxrgY#M^P*JiPU>fV7iJ^sxQG1_0f;fr0M2v0#=fkf7STdpggfdf`}O8QBh
zTG~W7vFf6fXfkSAG7sgRe#X!zakab_TH4oEIXG+xZG|7i@V<}Y1*U4x)*5?JEe;s<
z!@~DB+b~%;sEB&XD|q2JK>2-3h79Rx$jx>4kemJWM%Dm<vx%uGVAsweVM;++c*?zo
zsS8^qi}C`ArxD+~ZnLi@306x6fH2WAwAdZ7|Gvf4`U)!K6YzZ>cFGb-mIox;@-JEx
zD&Rs&7j4q<_Z+8Si8sBOwm`|P$n6j@d_qit{|&JSTavrr5e-FoW!_(N!&k)Nfua6H
z@dekdLfeKB1C3z8`KJEY-_cLdUj6Q=MAu&*VGs(}<?t=Hrfann2Z1hA3dW%=GK8VX
z;Z3;8^Ah*JRtUe+f&KtoPaZ~m{p+JbH5e+3KK|b~N{39~lTm7sxWArb;_9I+#O}Yw
zfhoL@3%W%YBeL})txLf6h`-^z(04dKzye7GGqbUSLs&%w0DI}b7FX5$hrPAv0mYH7
z_r8Bk(Fs$~J33oD-y$V6g#}C@$NTHdRwzAOdO4WlREa{uUp(KAQwtB@xW0uYjQlml
z16Vb^0f)uozo!rlfoDLbwAjTCNLfe_sl=b6{(jX6A8`X?DWAA*o#1}q40`7>(sb*8
zyyZfoXlIN63+)THSQ=JQzDwHtA4cK&FoLN`AKUrY`Y@(IEX7lq8u7pE91O=1RybEJ
z`}f=VLn<)&izAeKu}(j0)}Z_jyxFmz-?y}FfMp$Gg`ZSY62<)wtx^M$@1EygVHc)h
zXS-u*ecZ^W4b;3Yw!7728Iyw_^#&3G;_^%L?Am)St<#DrBFydINu_J_-nOGi=||88
z2g{J&)4y2#p4JS`xWl~t3~TM*FWF#`x5)UqpO@$k#ASU<wbU$FXPgu~6HGZ?Zn+A|
z`TuU-HIV#%$E&dZ5-tRCn~|!YS9L`T7qn3P?zPi@gAmekbR#Wv+}0-y+6@v(JP!XY
z_3`$}l?80N*ws%reGP3M#N)Pd-0P1rvSEU3v4zzFNbzgHNnAWJufIgR9Y+p53+l$Z
zKfLCD|6ZIQdz1O+{5N|QlkC<#G`-{x=rG+OWS7$ZaM-ua>2dcKzUvYmw(3c1j<fnM
z6d9sbdH8_Ry1Kta_k$-Dj;@d{$<Cpf*h@vz#IsmY_b)nwXb64ch>S?-otd^^qsE1y
z(EqQIXq&v+<^P8x`|JIZ!FiCt-ZhLl%xB>1v$G0YBL~FfD%2C2^VNejjlukaFMhqK
zuu9`!Lfv%rqFkWIdRH#3@9iz<z2W&aus+%(5RUe>`t=Vt#c&$E`Iio-T@xp57JSi6
zmTX0e7!+XqzFhw=cdEdcBmKQ(*Gb&6k4b`8@;W4MvGyUBqPT-jP|7i`k0%MT_X?1I
zyz<4r^#_YY^a`}+9qHOYA$wt=JnLw72e#b#ZmQ#bQ62bNvHb3Jef&%K|A`Dc^!QI?
z|5@38R)*!NaPj|hj9AVT0>S_GvaS61Z!cR9t5$aE|H6;WfCVw;zwl%IAhhaL|BD#^
z;>UkM{?_gGzwl!#KmI$HZJura3qQ7QF2Ilf|Aimv5t0MzSOBm<loa?c6SBsLwb1kQ
ze@^m0C%In|prZfFH*VgW`APhkQyngi&mO)y{{75y{mA?vyk(gwK6s~QeC~0){jXij
z`Y4BfPVD<%mp9;II50RFZKhUJVZm+fR+U!-=arg=mKPiOr|o9VdA=xWb_<)wYxcQJ
zHeNs4I&qYlUH<bXh$Z&jNQf~;b)5I=((g9kn@)=zS6XxWRhIL7HQDAfGVTNlC06pP
z*(z?{4s56~Ch0)tywftf-B&wD@5LogE26lzlDr<yX1xKzsXfNq2G<}-4+%gzGl}Zd
zUUX@WF+DibcQD~w#axB^`QOFyS8eWE7LM~$mgWz6Z}k{(6oD&-Wz(PT&8GSq0^Uyq
z9aHdbj?DT>?A7hC+(6G#g27FX@DHsFo<w%2;k}UF07dJ0@lWfQ4jakbueeY6jXaxe
zU+xWaS>2T|jDS?;ANrv|0XgkMM=&?aTXfQcswtdU|G87yBd2vJWy&v3-;eIj+TCU$
zAEa>I#`|uqG<S82E0nE7KTi^pKm8BF>2tN!pP!uNh*`To^E&{vGg26|q5DNlHHmVm
zid}gFmu**q*appO|A04NEZr+B>yd2bT@?YxFyy$qgdeo503zxzC(5socbkZK@85Tc
z_^I@X-sscFv>Wnkz;hg^k&+z8m+Y0Cb{KOa>B`UVm1vgzvt@rGRd!jSt(-XBWc4(l
zbpYMe-5{M@0p==J3>jMaLi*DuPu6ccHio%=aP{Z4a~v&eaIG9aVRjPX!t9kxgfRB1
zo|V-tjyC>-?h6;nEdgmIw}n0(R+(Qbyr36@TNzn#SrIgc*6p1wuhAED>?94=ASyO9
z9$H^<SAQ<W<7T;bB`%!^H)^;<@$u+Zp>PI2l@sYjvHn8ZKT60Ujp5N7;#95Vkf>XK
zad#1E&AaNlgDYC2im6*;{LX(F&ypodwesoLr`2nB=okX1uk)|M)U&SqTm-Ws8U2M~
zP1k=^_7kb(ipoCWe>t@V2_gL&Xn*G(wS>;<g-yHPtt}n@e=OYqhfyEVUiGBfV|8o{
z0o>vF7oM#@u`;<PW>xC;Aoc$>`3hg2@ITm?JQOBB?hzxr!nmg0d(Nyd&a#jDABE?*
zw>Ea)r>O0#YFzjT0i8A7?_Yr>bSo&NU17~ai6+s8uB`v#6X~Iw4b&-<d(3|gu63;P
zlLvdbctm9pl@MT($dm_HZT)ED-Bq3347&@z;O{wwNE6pP#QkX1vG~!#&j~1{)sbmq
zz7i{!qo4Z+j5d<|^K*`!<Xq_IPx!SrmJVRfmB&5fS4vQT?-K&6OON^=OE<%D^+yUx
z$LmJ9u5!-s^GAE$-B}UaFM^4c@##$#CjWwy2FEUqaa>cm(hZdZObh>sH6YbPs+fpX
zm3ylEubZP|uBu#NO!JL8j@6ZE;UHtx+HbL9tn^ADUssm?eYBD5e=Pm=iW9uy@!Vm}
zl50@`(%!K5o=A)w={RMpDkqRii3Jj~%H%^f82S~`em#}hv8HmIKs&|w_Bwksm!Cc6
z@yg<Uc@|Kw-*Bx99rp=iV$}-mJztvE-5e`;irP~(b1P-6I?Nm$D_>|mTK})qoA6mq
zuSf-?zGlgX&}AAYwcuss6EVMu`?%2;BJ>v~|Cgz2tkCX9r@ijsPa}=3H5gz<KG!Wy
zbb592w{`zxa)OCfzJSoz>LA=uH&?%X0eXm$g%bpIE7mySUO@eik9@*-EqK+*5v)78
z2n3d?_t_V!e{NCPEuyjn3@1ru<t3u>bcg0%7e&pPXF|z2Be@pkwt%CLWaEw#ctxKw
zC$19yt53DiPR_wAYX>5oC8^NN$!g?wt8Y+Hn@Kl!imS*td-D-q^}Nrli2^H1P*re$
zI%1C3+_M^Db$<#HlOVVD578@I5r@)`pI7ZHOuH-dCx2$~W<OaGwp+-3-m9xO@t7B+
zf7<IL_qmEA#~|ukXzSU=t*R4doLp96tJBEpxLQ*Y#`q&Iy87_JxPDZ>Q(X4u<l+&&
z73{CVySUm7K~1pz9d@!?S?SWwT$`E#e>+%JAWr+6*A=FXMo0#4M9W)JI(Iab4d+h!
zb~W4^TKP638O-m!Z{d-JD)3A+&*sz(iS%J)Yim&j#{rynL5gQLs_+4YR1`R)!y);4
z{W=#(r-{JY#BlJ5^g}x+Au12C9)DL-{gY{hGi!tFVfgXQYy0-Lz-^dlVwkROPnrjo
z1}8oOGOwSzonwgP_N3v>$+#mrOlun!(r_&oSkkGYhY14!v3vqJk)KJ5_3?dCy6q#E
zxwT~tKEq|R-Y$cD-m8LD5)1eZf>>0DpDX0Am)!GE9Qaa<6C_b${pWPx+8U99yd~s|
zq@3Gtuu{Lql}X*lO064%HH%i~&H%Z~3;PFXx8tx9T2mWSC{H5QSzDuY>V5N=XVmS{
zBLP~K<wt*DgeN_BZo~S*9@-xPhT8<Kfa{`cKWyr5Y;YLOUIepW&keNm$OpwQKV%c)
zx%44s^_F~95N`1>e)SP5Ryt8FL0mykj}pFH*wulcApwS`OT=PpkQDfALn8S@>acpa
z7T(*<7rCYjH3Zhwm-VCRvA7d)6}U6uK6`pn8Pp(Es4DVzVxtn)7i59JORO=%Q2une
zaJ(<oJ#z_5gwJIQ`4ui*ws1##Y`L>_h-b^&Xd$fCyLkcmrj)G4Cq%VKcLMpHQ3fO%
z5%G4Mu46#Hn)oGwgMd!U71o2v_xn`hu*u&2w10TF_sV-92>3npKWy%Ch)iP(CmRW8
zdgiL_$WA*w@unRPHAEJRSQSAbm*LmO{V`A{cx(S3!87Ylc})nZCw>IPu&PXYJV5j_
z0R10-M(xF>KNiuy5-w}WBled{8sK5m&j9ovmbm-|wQnq!v<cxAy1At?4hq?N<;^dv
zod;n%!T?iY;aghz=>dOV5B<#eib5oct3(rD&m>N+x{DPZ!-SLFy^@yNJ|%v?@E6E0
zFqgBQ4of~hS3rO@Dy_AsTBO|1E`)vYgZ|3Cv#y{Xu!65EaNk;dp#M+1?j&F>Q2Gh_
z_>e?iAAkgKb#-Fe=eL<|!TW!W&lhPJ`fd-by5}IUu)tjvTRw>_u)wkdX7ebvFLw*V
zo8zM?Zo%5;3KKy_d+l%sgVA9I1{kJ`TEwDYj~uLkV{$0R7&ab?8$=W7MBI~D?w;c-
zPDM1m&AmzYZJ2P7VK!;$6jtY#!gol)>t56+KoyKX|31MgA=)kUSp$8p8Rl1kX^-yr
z%9L#Be?ZO-@*A}r=AVO2lS}6pMtBY{mK;mNIkI7PajV;@&?j^hX%KbSJLvqnVgGIs
zz(3@nFChi)>ca%}+?U0!)~`hddZt7@fndMn$sabS1H)f${(8TkFi=;^+f!So^C8H;
zdP+1n;*$n;fuSVpj{v5$s5uA(c(zxD)$(-CVL@bzUV@)9a@TVkBJ$}vF~WvVBc?Wp
z1DAIN$S=N?n)+%RMurr~cL@u<j7`2P(r+l)RC5ChaC?n6r@fD68v86)*nt}$$A185
zh9cqK>Xn+?4FkaZCck@yyRA#)ciurzUl86!k-Eu^*BzkqJRtnE=-4~pKhA>xIPI1!
zjK$q5ADBM<X1xIFLBSOU-gEcUFdl6B;}QK9zf!?~Wr%@(Cc3vru;~v&^dIInz5^{r
zx#+qv?Gw*sJT~3?b{umc;2XCoDg~Z{0a`qLlCgfj-w*g}nEteb9(rZz>j?@~og64^
zpu#pVMPy*Mt?;VQ3*>iitXSR?OMp<`+$%4oZ+uk>QZ-yU!<hUDykH@$B?2}DsINZD
z<fCnvpZM0)M?7-!us^0iAL|^R49!3SaK*D7+3|Tx=Pv_rkulF+o&u`<G=YT+Ja@Lp
zCmH0!D)nWk3ss16*;4m)*&i}mIbg5x@e`8KAaSA-Y~<=NK&EHN6t;ay!wNptWsIbw
zUG$MqRjY7a9ahhP!sEzA{i?3##liJC2fvqPXvKDcUQED2Sqa4g=>MSK!}mvY`u_Un
zhzKY|`zLFAg%AX{`@#Pl!NX?1axTcPs5Ps%Z^e7Oz?<u%d4C-{%;*F2rff^9hF-SC
z=V5khH!nqO{=EkNiJPmXGPxuEh2h!0-f@W)YYOQ|nBn(I*9O`^P~dw$hFBHVT64g?
z%N>4K<vh^_(z?Up^*%d?)!=Jo5ZW~Od0l8fY`-nwzOS@84`6rMVT8qn*U2|VLD*aF
z=8QKNb>wvm`1*AVDdZyn@*}Hj6rrG62FPddaN)r!&EsH0`mOse)tnPqfFX6CE5H5s
zryIYAe>jocz0xg-IH?8f!y7kWODL>N5}^v(dCznC-LwX%F9`4C;<ZNX<`FHVcdyh(
z#Ag%i_uGa12S@1~7I#HeVfvPr3L^|CdT@JyH^*?;qXjEMDa=IdZ+uY;1}rWH^hdcp
zvc$Ge4@AFdU1}DJ+}*tKY-`Vp^JAG0K0K|s!rl0p8@SmgQ8;cOa=jEwe4<+Ez~5$S
z$2TnC{ryKLM1$Y&P&L6eu<OXc28>N`@uh=6eqX8Ngs=tp8>xR_nh+}#rB=FS;my_1
z*iT_CvG5n{U#|m_>fq4&gboatZ+L1w8HSj86MgJvTIbr11i(h!mq)a}5u3u0XFwtS
zrBYb6mLsrmyft%6v31!2*q~XCW_YFm0p)!V!j0c}KZK=$dytG0B?^}a_8+o>0X9-l
ztKWfi4eXBqPt*$cfd0f&q9*g#<V4mL2o%7J1q|$^gc1OTEANKiOTE2KgEhQ#4A6&8
z`MVJZy9#%-?b!0{$@L)6XDu)dNi$LKKNg3v#3!mX3i2a2=?w})?u$#Dx-ZNr(QvhF
zcGZf*m|^yNCE}3eIz)u1-vlXNF>eE;Am2=@W6)CxE)VLyTl*N&u%>{+NJVsn`S;1!
z+c5V)fooM_unLhezd^gX9dv{*3z5SHVoWq~>&)Qft4;mK)WOeXU1Ojp^(_+c|HgFj
z1z3xJ2kQMwR!%((75Q@=1M)2m42iIq*L57o4*&prV7?fTAI8Vjg7wa$FcZX2>HDIl
zq6k?)luZbG&mU77Yz1DpT*&{54kZE}Li}}P>+=e~JD{DlLQ7ulcEEfPUhu?u^<^ye
zHGt?w;Z1fwy{Hx?@*kfV2G*;zkpUZndoYwz%nL&q1FqfFeTy_6yzSOHD{sXL0IE5;
z^-fOQ^6n?ro`6`&-3ndWQ;t|R9aYtF=YG<3AZIfc=2+iIpn>F1L}*Z+5Vn_JLWb(J
zXA*Q6!YnGHiCLnT!>W#d9@5-|2Z#oi0C99P@Ji%-Ptu$N;QCaeXPJZWoUW1@;Ep^M
zTDdn1@CzA`Qwy)IA_t$0oEo3UYaZ;mm|!k(-Q%fYsD771foMXPMuWK12Hs%5&I42+
zb<3Ze%%<>im06%K<e^J`9K68PX|Vma>2^8n(Vj%JgB5Hp7abd>XKrD?u=<3Yikg$1
z32JLl+_=M`8Ly=b;8_QMd0o}8{J_bUR4q<L=3oy{duNLCs_t#dQwjxqzrPK=fQ;4w
z&?l9yXkILRVu=G|LTwwuQO<F6N>p>Ea2|j#Ugq^HBwWv|di0&L<RgOdKjJV^-F%5T
z?)1^Tn!TID4}mopHY$PnOXnkJMWZ8x^%y2Z2$nU}9Ah(sY()pco#hYKfOM=&WY_d5
z8Tv>#ES)2<E;|o+w}T|XGZ)a4{Ht0c_cjei6nVp2LiKPOm_Vn2e9lIl&Dn8lKrol|
zqa)Tb5=h7uVIF6g?8;QzPdQ--@=tiU!kFyyjZTY=gMgR6j5{sgxT&si#0oH6Jz}9y
zdD9HRqW%q)mp{yXJ}JyO|AEMn!wJKfsSEEo%dedI{q8ovgZn*b??1L-h?pgCa+4-K
z)WGW~I7P<+-0bEqxec=fUVO4Qe8PHVAN9qXTfiFzfVbz1PE*6U(k`9$3p4XicUt<k
z?(c4zJgOFYSo^ti2#!nWm!nL_%@`iY$nSSPy#4WgYg7DVF0>ML_cJaWec-bFfY@G6
zO3|HScaF+9i#+uE5;N!&Q_EE$-FR4tOQ=?8aj5sAWkgJN_MBY(kmBf_X+Ym6$+_=o
z1~})t{Q1oz<P$X7B=o;UAG^N|2cLH3|DG$~=U33CyM+F06gShexb`T~cHI4<+t&Zj
znY}Q8DW52-ot!qYWKe|d-lFIdG*1Y(9BP@94vq-2ANyKdYX@CE&IRwj+p;z)8d`a$
zpPb5N$+IxhApTvNv@kIh>~TeDWnt!bcpYO>_Y0|0M6#+TcgzEVPGnuI($(PK<8_)#
zW4@=18d#!b!k$!~;B&zG@dZCQ&fP#GW&$XV=q+i-nwIv5$@U)krQv0J5jz~aa_XLQ
zjYJkWb`xwL>S}6aRZ<OJ+#6_M_4|AkYdm|#!hllWwu*~%_eFQ@%=FaT?-hf^0?|S%
zobqV<#_s$xH(<zN7}68Qo?xqvKL2FTi{=!4sc)~j!%}8EHrRUqK0B_T+-&F@{)@Q#
z?_Rq<A0_l1;qWEjX*<(g;8b7_{ZOVXD@<>tIo7?qNRkEM8SkzHp5ec<_o>7N>qTir
zV32MJ5i@ez?KpPQ2Z8z9jwVWsgi0A#g<u2$&Y&jG&Ptu<7S{TIrEN2h$sN|XvTr+X
zGHs#5Y5rdbb>)<4=LMAl$2@Oe-;)_$Qc`53q>{nH4mWGwU95>wVD?Rasdsc!D}L++
z9=7+LyzURXoKOVS!EDEE#7$vV&wS5V|Iza2()Xx?A8BxoVR7jTGyKM;J}_Vp4*oCE
z`|}3&_sBjY-1D!1IY&|uf@jVwA4U7=DG_f)TxMp|Lw!*%;!QRCk_TjepWe=T_X?;(
ztq-BP$QS&<>YU@3;xr5LS=2JG*-rGmh+>$<Ug~4C)IRYmINNc}7+|TwOI2^r>;A3@
zA0@OA%#aBaFXp%Gulw4ie{LLms(kp_;2ZPLg9Q7VbgrAf9l$$+5wi8YDTlY0^0;|#
z#jBPyqmwy=-B^qJ^KP3t(^X3<ckyl84-(v^ac`p_5}pZhJH&a+5+hhD(}iFvUjD9)
zl&X(W81F0RBg0a!05YvB-S64cdKJ;qf?;X<31DeW7vq!?f>@LjMQtbhb@h2mvCQ%Z
zBQP_lpN-OW8x9-k58;>*Z@|$zN6CAK1#pbR^F@OqmH=Vm#R(}NNnbYgv57tWtXMOi
zVg{p4399RX4vco?OFzH{q>;i#zuR|+LzLS3$9Ddok4R*j10Ar|co#H>Ny&=H{rtA=
zL}0P?RyEL$f|-O~x=O#tCWeya)04ek6kK|eX@>P<3%B`&uqLb_(Yo>~h{i5Q6&bTr
zCK&UpLh3xJnCMqb3I`@d(j_q;Kc0&QQKsIi6`z~<ts~eL6%s<5yuV|U0yS1J!}_-K
zUHe4~ci?=Nm<+7Tcd!HSn5EFliHn9Xz3ymdks0jlGr!#UFJnzVNXQYHY*I|!PK=xs
zGuTPU39yrMF)29T-thWnnfI2$Yu{gkn5ZM`eod5C2aNiZ_(d^%+8=V;hr{MP*qNAK
zn&i)l2mel~1JvNNdKCeT^|4CR&Hm=uSZa5i9M;<8e4w2SO(p=7(bBI>HjYEaB+o(3
z?Eo*q&p2^-Y}(WFbWpief@m;pd$vM{x-;AI7sjNb)pOFhKb0_^47N`mod_ai&#fxi
z9dKsdHN(#pK?@W~WXB{BSJ}ZuK70*aC5&z+QBJ=3Y!Ah$<92wzBsXbeB;bg=^vpdF
zc#3(xZMu5DSl^BT*`vG5o??5`6P~zOyjx%Zu6yPU2b<Hwa*Pn~^tX$RoHLxlWm1Y~
zX-zlw5_hN69SJ#1u)(`RSPxU{%|W!n3!p0QKREIe*>t%Ht}?Q+z7lgKznF2S?W5PM
z8@5A){a`j3FoJ7vr~K5vJ&vatX9?+3m;4qly1cZgJ3H3hEW{uIy0=DK;t8OO{NQ6`
zxvtx$WeA_48LGphP9CWY5>hp6cpqIxzjyuH@Fh)rj>s`F-ubEH1cRJ8L4@i~dlT=6
zS{>!i!BL2je*vCH3N$y_PRrxT23gNjK``&48JyuS2H^Rxt0?;>zy5sb)~0`*Ru+@Q
z_Q?sbj&6(TxuswDgPc#d9n4<3uC>aqor>{lj6!x(<x3d;V;fQwl?6R}sU5l>{<2>V
zc7j(=EEr=ihI)T^h*uWCZ~jg(?)o#Am!2Cm-51pWF_y*-|GF893v-;!tO^r<GW%*L
zDU$p_KVFldJv73C^LRbZo07kXB=7R<9Q;L88njuFvLXCE{nvMO37VSZY-$WbwiA*O
ze7c*)w*Lzu^4Hbw0T&<@YA1aN1do4XZ6WkmgIMGuuYCJISje+L@3Rkgef^$gjzR)U
z%P^LWm2bldA_dR)^-fsz#ogd)&zZg*kk~`WV0%zDg`MM8&2a>up6sOMfw;dGj(8(^
z_%S?!zkCM2XBd7%z0^>fH(VpnHd*Tf&hPqI-!^qdEN0!0ggw$GCdc6YWwZ<qSjLz2
zW#A8f&F7c6u&z>rpdiVsVMrkpSfSy^Z14Kx_1@ce?D_(D)wBpsEk+@cW}O(&?s%~E
z?*8Gornk@naLix<2i6uKs&#eh+oqhgB<}OSI>Nf*#1)YMNRR-2zmt!VzgV~2bZqBt
zLc(yz#krb@%T%yWc#hr+d+^}Fmvi=;8h;n|K&D13;=R)6!QM$cVWV(S7Ol>C{@MYS
zmTXc%ys?W3fKCk>$dH5jC+>#e>NT>BqeV8Do^Nl;Ko;H5^T2FL>+tps``B=IY^2C-
zkpiFkZmx;@|AbGEya(~+qpqFREB8B^(Qa#Gx$le2rZ&|epU88REK?bIDs%AvVcu){
zN}G6czUMIUIw6U4VAdYwx@ak}VAin&1G3=al9>~4emy{O>ZYV6QEBz&z_JHm-~&_w
zHW$#qVK6YA&;RAX*TOH6At04-2nd>mIIvv|+HiO%1gs^|;3<ge|I5MIs|uW)RPya@
zrijWnrGtALtTDMr><07_Y|^nj1mV-K<P}#mkApjPvm=0RW2?6W?QO0OMMk4V2Kvs8
z^!?nWshxFlcKc|ASs<J!Tl%o6X7@#FkUNP84Z+(9zCpkpl*Cr$|3*np@ZvBl#d`+{
z+4c1NI83%K7g1Uia$Mw;`4R>+`W-0&2mL`qT~x;DqFk1l?&8H4{TDH#T_t6J<kj;v
zz1CoT3n?T5f)=XE0_bq?`53{hVg~9h7zg~HcUoX<#(Td0`S~f&kusHLeWEW$$nUh}
z<%{q`Y;v^laeM)JZsL8}1P!6z?Rhx>smgB#2sjsEyAO|wl5{%vZi89iB<|C}Y=>`z
z%i`|;^8(bzNZ$@P)?z3-CZjd>>+30t{yKX>?uw1A>I)T`t@c%bGaNCgNgflbbDhDt
zct=B&e|ceC_p>u`O_&4`BLhQK@V4kpZ;DU*oD6=aa-)Y9{#gRCNL=7Rj|HQV(Z#R&
z+ck)TEAz^-*5YN+d~hqJzCphP;aEQ9$E4Pm{qtu<VNk^7H>HEkDem{F9bzwvt@|Lf
zGv9`2r8J$d<oq^_JqgHsNCwC5J^j01Kw;4LOmq1S&U%195V-D*QcXg7p5Q&7#<!AR
z+qLg7gQwsaenLx))j@2ce@VlMs-lmd0|M}8h&K}1#=x6%vJ1yVp%1R@>+yA0$PN^Y
znT=FWT|gwCQ|p_1+ZnNwbkaMhI!t1K>Jk%{3&j5=55r!W<>!N8F=*J6*|%WWGqMZE
zVc1~XH|FcBOV|y=8dg>h-$eV?E3l_eMXZDjxRZVJPd8QtocMkW?2c9FtN;Xhv52`s
zy*|L)GRaZ677=nBFs(SarY*TBIC0Grf8&LME$>bUWGQ%ALfoi$uS8r8jyIvjUO(-R
z-5|91zFm}<Z8+5Ah_iyc@1;~!#aw8>SVnac%q!Mp{DT|5sWHIWVuV5~Iajcv+h%$9
z$5Mh9rFrgrXV|%O=iIKBE*tw2+6wl=R^W0cgc#z*8{ps@D#|9az0<_r%_zVS1yh2(
z?rUGy0H4(z-X0ofiA=Nz*l_-&FCoJ#W+k8L4kyjmJjO95hqn?=i3tC_oALn=j@-ad
zrdm7@e%>V*5pF(WKD<si(o2osj4l2y=wwC-u4U$z*>>HTKR-{Iv<o_XIvuU6tD9_C
z&!hsyn}h9nsvgurF^poe`Gm-OqXPusnfpu(G+s5Vk9jo`Sw6dQ?)x7r?1mL`znf8k
z6^cv?_pTpsHHcw_uYy7Yqw&GP#URbR8c%-B#l_`jZ$%(7Wrt6pvNR48GX5Ml9z;zg
zAKg#Nz`{aGM$2Q-FVD!vwvT?CWZfZ%7H)?%?F*4TfrCGD2?C_F9P>0bU1I`9$AcjB
z5xV>1>+u&m3h~adR~O48&UPctGAI!;oPWCSRgvq#0V*<|xy9)YSEhMrX!$sNj2I({
zk-I9Spy4p7z{tq>to#SnhGL5rgXh(Cd+7o?7>RW7NN_CZpwwWve6*}s9dl3fn1t5C
z%%~ZhV7n^`OzGFXgD5p<dFiJs-QG~CIORvG*7uf^c=RZcF_Js(k!sw^6Uv&+4oS~2
zbhPsoqr@S`ai`>d2^|ega$hr_d(6HuQV$}b;^k`~AsCUc7_>ryc4SoKWvh{Pb-<oG
z+?DoYFBK2z<Z(lub-#do$(ML^h_y9O9YleNEjq+%u#sOwC)YPT#x=YQEEbdf^=1HC
z`Xu}hF=hq_uqZ3Lqa%Y{G@aEp2Eup)1>v8IuK-D+l5l2sJ~nam;DGGghbL$50=!q5
zGcnye2~TR-YuBEKoNitx9O?NFh@6=sA4iM$A}-8<3%TLvy3;#c%C|JFQ6tvmnzG?Q
zL!2)B?*^m&1=^%SKnFi*+>&}x?fHStanjND*BJ^Pi&l#m^#14zHDs{F{xzw$<ot!~
z(uu1`{6Fz-5R6tH4l7sdH@_W;bl@=@iq5mP5;8z&|MDRo5xxD*(YeV1u|-`{Djdt5
zbpx$EmBA)fyrp#Ltg;eh9tT5ywaxi93`&dq8S}OD#_;#7&e9ZocSBH8g`hF+*mUQn
zs=yC#o(CVu<IOJ~k33s9nv?@3rIYaRo*yaD?C?vhgV^pXfLBxbe3Za_Nn%j*?%7~d
z5?$6u?{gczU-CMPQX@}_+;A!qI7_pJZ*uyNr)!>f`FY_<?4{m)1w2-=z;pka2*ys&
z0LFe(*KWHmoXKi(ZOk2X%IhkH=cO-6Np(ZA)m_y?8uuPLgFa7AJ{7H;6w3DI^u}-)
zQaHDS<Cok4CfrBc^cm9RJg~A4%d+_W&rY1jK6VZx^A5@nIpL2C@N_qL-T2#VA@_dQ
zh&9BiMw&<}_ro#%ChWIJwuM23KR-Q%G}YXTyg-I^CkTffC0MN*^(lz-Fptn`FEp$9
zCQ~muF?JQZdqE6bQcAyh90!)Dp>5&Bk+WQf(J}reOzW8d(?Y9eu9^%JO^5U4bzLQ%
zQYa4qCq~n<7tcfmF$o1r*Iis)yU4KemW2_(w>Mg($Zdo|iclPbo3CfCUTS;q1!`!M
zY}zG`wtQW$&4VCl2ODzRp%o#F6lm4Q(e!b*SwrgLRF$TNT2DeEqlWjhv;$=Bx|7S;
zTveY)<8~6DY~?;`!9xl7Hk(?_2yW`qc6W8O*>k{FapO==$?Z5P6;Bpf2&HVG|7ZBE
z+KBAiHJfoRxBy*<Y}6W#o<6w}fCYd>nZAnK=?#)quVYR>VBD5%;iaVMz%9c$M(oVU
zp;fp|CH=Nax&5ilQDjjX<~5Zaa_MJb;<L|a9QBOU`=8P+{kgTeMdZ>PIAG7j%<QEQ
zt4NXH-;rmRaahRq^atvjy<wt{cayy{?J8c0zW}HP9ZhGgUk8w^d(9u0&^|RtPWcM1
z)a;}s;Sk|mT8wk4$c0Rd)!N&~r=RLZ#BfWR)a%o?j&*l;ugM4;m}KOYz)ox#(e8wy
z7(<c>VeI|%ki6UTX~EAO0!JCKwQVFcXzoq<R9~29=_3LXH>Ia1<bqBu?cdx+GzdmK
z*?L*l&P22&z&^9@+Yh(PAks8|G`=z*EkPsiLY_8rHIL<>vOl+xVyN~ny8Ut?>N%Fl
zAOpYIo2v&;<Q?9zLDtMdU<#S1H#mj?zAweu-jh+s*7uj>V>S89OLk#{(5jF8Ma4nf
zk2?VxRKAPg*NGD+etxxw<lvgUeMtl-wdbL1U@Y<%IS(PeU`t85pRD(dGHa&Ob#gC<
zoaYe%dc!e0kV<CvmkYE(h2xbX?7o_gHo3=sZEVT-{|{&H9glV2{*PasQAR~()0R!4
zvYpv`7qTi831x+>v!!Iu$liNK*_u|79TJ7iknH(8UPkKk`8>XV{QkJ_Kkoaw-rnc?
zb<F4Uc^t1qD1g|d1sqV9i2C~WIRDL$1e+<3n|BvwadR;Ie1FtB7@wF*D)ATot|)AR
zjL!h<d_0=-qxqZv9}72;18;mtm-apc#p|BeP~OL=@WiId{{;An>~~J{-7Iu-?;zYZ
zLg+Tgf+@m>I)%Oag*EkTw(U;p!F@o*YZ=>V1`m%)aIK5W-6L>l2c-jDyCz55G6rF(
zX50C5M=%$0XdV7^tEQ^?Z@KNd6OUBR2wr39VQZfuoYIGiZxC8EQmECSnB#`1&_XN1
zIXhm^7z+3H33ER1Blt|N=J)A>U-Q+!x?~w>wE41Ip*UMZDdVR_3WRT23f&UBG<~>}
zq$KmvbBI#)@E^wJg}Yb=94xz4?Rst=lzQkd^myAxML-GR*Lfbi?bO%+MghEhphZ<Y
zM96-0n8YgC>T6jzp|w6eFZqV2J4}A%y8a<==XeBAH+xw7I#j^xh~$_n3>wpDPk8(*
zA`z02g1UMDw{UQg+p05aHza!(J+S$h(OJviLyf9M0)GL_HxM*|lU8rsXc!MepdS}1
z_Tb(0@$j{(%V`}(?&X{ppR;a$m8YkSA9^1bZvUzhykm6#8!;0o<~9|LsHq8H99&zT
z>@by1oQ1+K8tyoI#9u+@?~M&Qp=?s^d%Tt+2~hhjPw-F%&ph*rbpm$C1KR;Lx1Syh
zwc1z5{{17L^Agsou8syU$|eHpV>2_{v|+rVSRg=5dn;Tu!1(^b%W6E3BhYE!Lv42v
z)ppBKC<!Q=Y~?V63#>mOYX$)~vk2EROPaf2>2z^Kl<ac4y5U-xI;T1fSUtBhH=RGg
z@aar(tUu(@t6BBKs~M$Gsa-6*&a78-qluQEbN1C<+TeqXGWVhC*>wj4;I%(o;obf3
zciL+E(eFJ7iHT9bn@0tOXQt*PMQT(xgHu7NV$SO{J%B#cCD}gcVyEFY+nNUM;O%Y)
zj9CJVnM{u2S0-xEI-m@$&ebGl9W`hkjcR^e6w0Mnx0lw%AK9H5e!5l~!=1VDfD-$$
z>%Zl9JfbENe$xkC4G6Le2p%l#jk*Y5NPZyvT692#>+)#6gn6|eZP>&8N9)#Bm(drh
z%B}));ClD&ox|+}kv|8A#lXQ4deqE4N3S!FUMG*7m=z#|AY_bC56<pAt+@`@|H(s@
z-XBU&j&sDE{c`gj6smNF2mc~-c$pyTNcwBt`wvsG(T+H?0GcIxu%F~^t3?^d`Sis4
z14-o&$ZX((`v(=jckL?0z~lMarGHMyI_=aPZy_d=9aVo_ieio{f)w7WaK`RCdPa%H
zZQ=6jt~VO)zn&P?fWsZ%D4VF&CON^1O8G6j_!pyCAMYG|K76F$k<SQ5I*G#0-n-<;
z(H?z)dywJ{^Fqv^(0o<~7fwC)3h=`!;Si>mZ(4ulM-cRaTbAu#0uO2Z`MG6v<8|J`
z>tw>~-1u6id+~);mUwLV65NfhXl|Ys91`-5OZEgN53jYqpw0Ms=T3^v2ajhLZa9X&
z3vXH*9hY^KPMDV%N`a>H6Fu*RwCoHQ<{`bvgKs64m5GRiHxhg%`<#3Eo*i&e?P3%=
z4lAbu_G6<GJe`aVtP&1v@6BXtH6goUx}#^UT4^~sXJLCS@7hIn&c_f7$rj+|IQDDN
zPRUuWtUJ*9_Rm5Ml^UwE`srHW+b}+w8EsSCs2q%|eM58$t}?UBz=c8hTRnnTxaywm
zFXMrAhLdXkZESs;7}|+B5(q%CSDkka-mYAZ+ILJq&HW_b?82}9=2UgY&eGe~Uu;Ym
zNEKr(I<oTt^wsH1+^Mbw=pZ8Q(n<#dHv*!p(BXD63eTkQ_kHw%_AJR%>&i2(PE1Uc
z0(|S3LjP_k(=TM|(7P_p<dq&B`~z{~1p;^+!bI^-gbhV#ONh|Fp4H+9Lu9T=iE(2G
zOuvctgOWk9P14LnkFeEY-h+==V(&U@-6Yv5-LZ41#cGUTr1!wOA)s{5;sm@-;gesU
zrZFzl@*h%|@#mfXN8USaaRRX(5QIK|4LiPb^ns|liX=}e+jt>sc%fa9wSbz`d<Rx+
zn1mt}-C<T!&r^A9)0hSNz4-><Dk^2Sy$U-RmWdP`y3C<8Jm7*jp$_j_42Ou=&duw(
zWOw+iRU}I2+bNL2hlk+vL7g(kynV-oTK|bXxrnnqap;{WDAD~o4N$fCEDfJxu++Hv
zfcck~d286KqZ2))NnnQYhSl#LV%agO;;YZ0@#hIK=-j(qs9@<3)ob3D#{srLN(%Y(
z_?~cxKkc0F$&=cGlG@0V`z@cHwGKUF+4Yt~f(CO7AF@2l^^bf_njiWvNSxR)Aqf<H
z9<u^C0<pOP#pdryT(O94PfM9ftqhC+1wzHCL<WCsEFiJdu!mO>oQ6D@>FCbOw164%
zaw)h%e|Z*FJ2Z{0fIwUg*m7q(-Gk=aiE#7VsHvXRR^`3+S9x#&%7cQ6)@VI=4JCId
z1F;`kF?B$$(UsqsZ>N2$B<e8!6iU^)I7!d%p^W4>9SkuQr#}#W^cT*IY$w70FbAN`
zw;>HnKZvcOMqbqunFjd4M+WP%Uk~lZKdXmPDhS&^dNBKh1&HOyh9To!Z$FUn!)ALI
zKqlMr+|QsRr>}qR^vU0O^8ZnKJbR2!L7ACAbN1OV8~bQZ;-39S60Hf+2$)C(;O<7#
z$*<)aIyz5dnf_|QRiI?Bkq8lj_ftjhS042k-Y<(G>tbObmtN5@6st5YQ|At!x*{cY
z#E3MP@#N*#c*Stt#m{kx9VNz}wlEs@Wmb=$fbZq_@hzqf`nv(h#~rkEoaa|VTAx{0
z9Nvo1PnfRf-QVHe=+2_LyT3vGAePR3qafr(k@=!D^xYdM2-tI$X?^H4BxY2g;!OU|
zgeU%?*_%;kzK5o?#1QbNf+qm)7@zLJ4~C-bu-`Lfe%yI?X8Wa)l(D1d*fE57>DoC~
z-UTT4v&x_OV}>4@@IH}&NrnIrBihhHHKml>G)p}Z{z03mzH{mpgujZ=b<@qj3Oh>I
zNYZ6{I2)8XTZ;S@H4rLqS{<H%AX%E0D5&25Ycar%bic?!&FE6y8x7C>M^6w_bMI69
z#)zpU_CCxfbG0EdeC0s!Uy1!Cq#N%b4SGoIN+?sjX)yrDVvTQ)McL2bf>al*|0i~W
zU$jxE=5Ms_Kf7Njwj?VD^*pfIu^4l(MFn#I9s@4q?4rooBPK9)P)Z~!hH)>|eQeFC
zSW+16$l(F7W(x;Ea3e}X_x9|!>rC7!Sxw|2{)EN~BAG@O`am7Y8z0Ez_m7XwCH5Am
zB7;^W3@S1<4i4g^U%Pe*V`|s(5~76IK#j2VQlzMkccHZ@-*_W;cq3e714aR!fzXL1
zuQwXB&cpPF#atY9h-eSW8D1Lx6qoP#>-)g%Sf=e~{2yR`N&%deaw0Tbs7Xhk2v&hJ
zqg%W;ZFmYc?w@ZE`K=^es)oofAd%p|=upLKj>#|0K#)+WC!Ci`i!QbpuJ17&s0l!C
z(`F<O4fJVLJu|VyW_Oc_uc&E}9{%ghtq`x<9+SZ6;W3Yx;N_E1;7^temCHSoCVf>n
zl(>NZKrB)|fOBd`Xu+pG6<TQPmO5lb8L;kr8-Ijn9>yjTbs%icVLv2O@8l8>4Sjvi
zy$6o@gCUM^oY6FLU$qwm$nZO=!*no^{sJgDF1-C46QO$oww(XyV;uZM9`sJ7r-53r
zP-Xy@BLJc~aEod2P>E&f7i&1rPaI!r*o!w#lp0Gz`Q^=@#*fb50A8na_J>&~!q8`s
zAgWA@S}o93v>7y*%&MZZ!E1deEca!-_!@|y4p!+Mf5M(Yz91QH3<#kZG0LD;>EPcH
zQty3_bMRgL@Clj<1DjgrX%L0UZu)nFK>t3c`JUTBTq=UwNAw?VAA(KkPMj}B0-pj!
z<lKC1_`$coI>~kwS!U$4Dv8<+`v(UHL*=iF3XE9t;V;DnnMRd{2t?g>2J8WNpTYm4
zBpDD-jtY!x@IFna*}9JlJIFx~nNYw`0r64~K@g+%`Zov?8K&Us4n@E-G&)0EchY$?
zsxMB^m6l#aFLVc9sGmOsKNt#{THtPGC?6($wQUpy;ndq)3NZ%%vZ{#F76ad(@TmNg
z&IQ;Z533up1J*=~IzR^-b`<`kz&Ar%H67pk9A)9V*3gPtVnTeazRz$MBNLpnl1}KK
zpY$IloEl!mN%OiL$bv<$8r_6K(3G8+6?{7cuEoA!6c?VpHZ8C0FhAZ&{^D3T!7v^a
ztXz<m+f4)hI%q~XXu-QfxxiVSMJd4`ybiJ_RRFc2<2kU8ta_p(_(%vZloq`lu^*>}
zN5!c_2OWSUducw&pZCG;K}Y^y0J}a13OA?0xldicy$kXCHhopQ*x1>dwat>(zLB{5
z1B>{cOA-~zp*%+1tIP9#A<8>Q;{QIYlpe6({a5hDQnha|JA2hZiEi>K_FM_OoFrXb
z)nk@u);sePA+KM*zHiEgAeqgUw9gyxyleVtZO2jJz6z8ft-k}cnv5R}SHZQ$c5SLq
zX8^kEpfGu(vG53Mg8{%P7@5PlbLVV?)rfWi1qX7d*B{P-L-`?x67HCQFSI8BprBf~
zbt1doOKuE$+_J?V&kmZqoP$;rx<|beN>BZ7+^XA&q1#2S9n5fvUe$*4s!oi;r!al#
zy6j-yo*7=j%XhxqO|kr;V!7{={m>jiTnqw`w7bV=dfSkp!T>16=Frno$efcK-u=0u
zB-q|M0P<ZeG=7OzJtZaMYF*Gq4uAoZxDG+7yp#1h|EVbsBLe=iTDmO^Mu}d%jov6J
z3O_j8ptKI^{aMNfNoz|Tm7##TdLlkx0B^K4%~+)U<;#}>#m8A6{Q1=Wh#1pN_<;o*
z+N4D+t{K#oypV&oY!EuqCUAH7x=NraITY}o8ZVIYkg!AMho`3)JHH42RfPfC6e5|c
zHAb}2sDl8_Pga<iwNZS%Odl>OTk`tQ3K%tTG64|kH(D(r9NZ!zc|tiNv7X_t>S7z!
zquGTg!h@>i2M*AQ-g%O#_d=z_ZX_N|+7)gh)~IxI6?C34WAz~+VB$~%l(n<aA`EIX
zx_|J4K7TMI&r50Tlp7!zG9<rR?%34Y4Ci+j&Yi<mF!0TS5&A>9`4*YA5H9_E#u2~M
zVxX4$Cy5Rq*{rXy#~a8NLmOvzvJ+siXa*!Vv+OP!LDd%|K*Po3^slLfF<S^?cjtq@
z>c9h_60ji%y$%Zrz|Ho9oBet^j;RY~(;PTkx#&|7nA-|uxGeCIhc-!tL#M9?e+cJe
z`G-`fL0xTTkS!Ld5k6EiA5eCH6HBQ<GLTR`D|BLoRy`Msta^@YoY+CNj&Zow(+n>F
zA*ws(*nkS&_OLEn<n<$s@NFK|a>k~n%#8Pxb~xV1s|a0B0{C+o^r9K4aKdt^rD-sC
z7TcykHT^H7u`T9=^lZJ6GmLm4D3(hBP(N6O+pUptfy=;w_2+DdQ5f77&NCB50&UJP
zsA4j8@?QVb=74rqFy!7#tteY$NZiPf-kDITWxxDR<OCOx%uwD>`fAsrJPrDeEyK<(
z{&B=<rb#x4Ly2uC`$R&uTt_IpSNcDk2pK{RKbRmpdd#%!oCufPC`3DtvKde&()+9Z
z$%34+f8xe}IHVn_^W~GRMIN6l3@QD;69pdAT^2dD_|*EbIaNxauuA<U6$6DByOkjH
zqEu=f$<9m;PpG)!|5gU~F@S?{kM4^Asy7-boPHK<zDXH;iRL0pJyd^zj8#KJg8`V_
zIDRoY|4c04HKwaev$a$y$F_r_2bKzj>0RNH!24Ix1cBg9d=EKSEWbs&Qjv8pcizg~
z!gQ9|RF=71`b*le*BYbqZ3Y!eOVz=Zw(F*POx_PX!0C0Z2tASWmLX!O%la|;G>@KJ
zbf`ermk2SjbSV0c$@jO9w_@UHi}sp#)JBp2#6**G;1X2+f>I3#@*V+-%=lyzrVbQt
zCatOJOHpq$pklG>%r{lss0da1e9nIL_XQAK`>Do`5elLno6q?Gz#C2s=w$*nUxw45
zL@T>|?jrg+2#9!j%NTUms+<1)3Ch+^AjI<DvHgV=GNVX#_s2=W4K`^AET__um{nR@
z8nppw=QBdxuJN76zZwbGxZkSxf9#=x#fKC8t8nUbgC|T(GF}9I+Qf#OAjT2QxS65Q
zq9b(o-UE*=7I?i_g)~Syd;iiIJdJ}o6)wdylfB~qG}u+=;3)^bYuGnz6sI0=OjP|`
zu^`@*3gy&|?>)t7U*)lD`YpLC!Ym;&1L?wy<a*gBlXi*~8z?i_Q&x^01K0_~)v1N#
z{eW01KJlgzy?C`TItv}rOtYdXpl}<9bQp*5LioIipt24B9*ku8n@_8Whj_tx#Wpn)
zK|O)M99J8b?bKntO`CNXG{^wcY&6JbI4Z7-HGroxw-+8Zh@RXztzybYvv>5IoSdwV
zzHdUvjo(HIq*Z30Hr!li4c%{Zrf$J-=A+5cS-sKI`WyAwXG+JZS9=HHVy~iq@E%aB
zN3s*w9AmhKg7XZjIu2RvO4?*<POPe2P6Hr6N9pNF`pqhS3P}E1bd!|+Wrdy8JfII|
z*ih=R2WZafjTC>f9a4OvTt~Kn>CdhWdcbIXsRQ7anJo}JO32;>b=dm)kxIAfcK9F@
zU3gUJD?$AW=;VsQqWihqF?GPL=p7bu%ynboJHI-pi6C?U{x+zt=N`^=Ee%n^Ng7EC
zO}Z}IvIQ6cNQ1-^=CJQ0kim@`-{^cMR6_CrLox1#+?6XfVAolhnMw)@3R0*vgOcI)
zms_4n5&X3iR>3>ZsR=}fch2@{Bobcaqr;1_+BJBeT*V&@QttBQIP_NK8T#b~k?0||
zKRyH?+1o#Lhaf6^C0z~xRqI5Le9Wf=-i0jyK!-=0c@Rj8IWWi?k6pVZHc1O$3qxF&
z=N@0AHvKEhr@*heJx82X2h-%D$&3&cu_IJ!R^_u7fNBeM^UQ6f5p}Ln<SC~>35<+5
z6`!n7EuKp0^NwMcAqGX*Ray~nj3`vJuEy(y<%g1m^O}0C{5V^Q(pMCuLFfnHZ=RRO
z(r*4)PqQ04W;lTa_FpUFOh=GO9|4u-Ud<Vdm(9{bzn_oxjgk2Uqjcw9JK!dyHtI29
z7=2a^P18*2`X?WABZbdhl~_VRpUhDvop=KDd?;x@o9Ol9>A>%!Y7FRp9)MVgP&HOd
zl&bwML;HP(c6Gsy1-=B{JLJyt!_XuiNTlX9kVuu9h*_x(>;lT*;@zd!m($2uRDvPw
zuxuDI6)QVCZ*)88ZKK{kVB3MK``96HWRD&=0w_GqMy7Z^4Ch1`z>)n@BKppRAZm}{
zb|@z3WpASPSWc!7*t-DeF!;mc|MvJeFq&iEZOy8Y9Y02qnG3L8h~U8>s~sVJh9pDp
zjdEQ;@v&*fRX}%`el2rqs{PA`r+~^~C$S?_28cr&WpxMD8St$8HPE7*pw_>M<K_r<
zm>E{@e=KGMf}4%d%ov+ppQZOg0kU8qo66zs#Pc8S80rS;#XKGKQSN#QL9zWcwVI09
zq6sS8fcQ~2`mE$SUCWZ>zP^USGxj+a3N(;ZNFF(J<51MQ+kc(~!wR4Z!m>0;fu9v3
zL}UJbj-sxuH1umHfWQpXad_q08sywce&;i3F2^TG8@d1*_w=yt$cX)S<8*%Cot~aP
z_WYqFgTF4q4pjjQt0FSO7Q<zM8x%6S4X_sOoQp<;jub)arQE^#Fx2yIhgNJA(&d)5
zXg{X_k=sTt<@p8~Lq9;!_D4;sdIG{P0lO}_(V5ST@JlAfDd2A9fPo8QQ~3MqAd>ag
z-kXlH=){~*<l%hjfy4u7cnIA!y>$2LC~!-3fb@AIC4wCWfwT~#$MWTN9Q6M<vYQ5V
z5sVsIT1<OMXo3+GkuPb^ezO0IZR7qellGuF?e@2W(u0E@U6ta{BnSAw4sczP*glLx
zzN%(1bV;UbD~n5Ya?*4UtC+*|88WnUd;=7D|I{cR)KH^1#f6!HIP|41%3HVZ!Us7P
zKri{QCXiVXeUROI_W83!J2Gs3kX(?ijerV~SDFp%e#^)E5VOTSj2F@O6avi&E`+Gy
zQ>!7R-f<zjPt?4cel()&`;YMZgCgfGVs4r+N6*~whmk62mft%flPXC#ACXpFevrND
zZ=4U;D{v{_8n~U15z6*9(+AB5jtem*B_-K_sUeI<P9_%S8S?GNE<Cn3IH9@sFA04<
z2kF@D7hp8wXYMD^_j#aKV~1C>OT!Na=)E3V#|$3Q(2|?qBo3dsp|er;aNY!MYb|Ix
zqF&N5??l1>Ir@?nMZuAqvX%PA4ydzjr#|+O90DdPN4ODuU^F8~<sUIsHj?c`WXICO
zHBX<7<B?cB>>^e@M^3}<tjhC37`!=S{WWsac%jgifH53$)Dir7jY()orVV47u7D$j
zAxCn5OEkXt)7qT__(RyH6l$}waLXJio4%F{bA|wvMy00e)CC;?AW>|$E^kZ`ZXVf7
z#L`=o+<hnjBd>ms>mUdk0OD1MJR6kov-lrcVLDM$$YK1@i+$k)QpAxqb!_<{o};li
zw*So^=dQYk4(ziz|6w>VW|XUrmsCZQs|cJsa7@7eW|qQgTvt+<;q|Ow3Oq6b=#6Ay
zRwX2s;xC-s3M$IMBAPhSa3p+h9gtA9)&WbAgSI-<ka_&4Fb;;Om}dD|dtcpry`57$
z_5vE{l-ptkru=e4lPGM569wIS#)pX0^%n)$lK^O)2Caj1(lp!viDAEa1F_#ALR-ME
z-64o?MnTBzuqWJK7mJMJ1hw;M!YIU6RNRBpE0&YZ%|v0Lu=lCHtevgjd3&gV@DG)?
zh@>C1cS39>Kr)QDtLw6R=U5Z`#MeGMz6&j!Az?#U3H;v2&c~@i?1r~BwNY3=yrHh&
zJlFpvs#0}z`S|pd<a<Bb7-#(P5k(w~^VNZ#z{P1Il7HLb{Qcm&Pv>b1f*!3V8B8I=
z!S0eO4D}2Qh8Y!6c`Jkq+=_>oA*!C{B28C;1);;Q=g~mTz2%YP-hPOO2M2t`;(R?b
zt%s;5pfCQUc~l+RjrT_KP>FFA5`C@@!@xJ=rBVPef1>{PP~tZIhjT$iieGu1ATW1l
zP#ss8u=|do%P(m(0#bNMM*;K!h{b^diYBxU6&ElRqYvW`VgEW7qRB#fBocU$SedZ)
zL39_C?N@#5CclanMs?P|_#+W`G1U0@|MBfT_s&JatZV#QZswJDUhQ`;v#E+LCZcM8
zD~YqR`Xk5*)M@0@$$0<oZvYGrRQVxY4da7~XfO_18NW!oX88&cca<aC*`!7rg5F|R
zK_oTk4;3MHo);3y9Rfm_Y}PAFs@rKOiUq`P&KrUz`jcC7FjNO@g5s#1LnBmrtI`%f
zk*vh|R$TW{f%F-;Bb5b;*1*Zi^{;CU<KOgfEh@swdq`-;yMGV8-8ld#szP!^>z&{L
ziV(4rbPp$rNCiOo>H#D%<D%Fgb>!+pxwc@ptrW!aPe+#x2RT@06{mif4E*o@fJEb`
zNAX8?f)4OBpy*^bau8;Lg7h#dbl&aTGncAu!DUqO<suzB>J&m3e-FuSx>}gN&ErhN
zav?W*o9%an*Z@<?b+ZOiyIT;A^|oI^)tF*~()<G=E*L^Un)LkdKLECnL33O>nR^Z3
z;^sKXy%xCK-~?Ge1)xfRE?~Tzwjo%B1}YJXvd6?4Fkia$QLEa(FZ6pv`MEcyjj-G^
z{IX=Ff6NBLhX<MuTP#@?bOO)uKz}HnXdDeTlmYeD{vkE^0Wo!5?q!fiFM45Ja0LxQ
zieEv)u<)Q}Cu~gJhYzA^NA9Df4%HF4FR|#mgKv>=^G`zbz+ycFDc_)`PEHj}qfHeb
zz8KT~!@O@J4u_x_Bh(LTu^SzH#FKmF9TDJusw)x$J1n##NLNBwN*>E3vm(E!eh~eg
z;L1NbTPz*)V_uih-%MnI$U8a@a1CN?lX7!EMH~*DhGS8HAhvP#S8$ptg3?8n-BiDC
zKYe>z?XNGU3}5U^>(S@TEvT)F*B=h2x!AhXro;2yknl3)>koZs0SKY)PRls36`wZ)
z)K+{p`hK%{`y2d;&o<5IsapEO@_vVjc77iER8-(TRAgl6<n*Aiu)Otvh*r!+2U@Iw
z&w0ihz=~WfRQ36H=B?87i*4fL2X+q95m@>VSs|~h$tZ5rzJCziym0v+aW1wB+J$#m
z<ZR6_#SszF3oM`sD8|!%elV<{d<twg{15M63|SAo?a0z&v1rc>fF7*|-0#}8j^KJK
zn7{`6OERi-Lx@`N*k#!*E1L>~<9DBZ(c2IT#6)%Rr`0nvf=1gWh|atQ>7bOZYd~aJ
zP>17RHRR#p2?Ox71_0mj$qzIYl&Ju4jCa3KMS)fCo{7o|nyuHjZ#$oV`OQNUFX0P6
z>Ns+O_tH3G0X84{d@A+OmXS{*sk7S*<Fz;>gSA5nJyMMwPlL~iGl8OxIY9~)taL>%
zX8QR|o=vN!RLSj08jfdnAZ<qd4P3jd_)PZuI-c(_K?Yzo5iDAKHxZEtAb_oh#{H+)
z+$VyqpLO$pwvx_Q_fQo6ZSfxB$gv7~1ufwBM}lK5hn73f20~-n*&870!l9<4!;17(
zK`KJEFZR?%zH}~+a{d_AeTK5;1yo>wHQ&-}G23Q2UV{IG5DVc>2*4&2Kl%Cx6Q2(y
zK;82vc9Krv;1p=>cvArV@kQykdKt7FwTbdU1JDD$c)^(D%5}5LJf*hp)jd~B|KR~Y
zT5)dsvGx$adYRcHf>wa{uW_C4Vd*fqZJ^@Ycy8+(R+S@%=sx?L6HWCgNpjE3>?J?_
z&o6aHzx2hVpe_7TZ}`3+1T;V<o5*q#fGe0xJ?9)^a(NO#`|;9G65%*?1&(-bzxe+e
zk<dIYG5(Lc5j;AXa)ko@!4crklpFoZfq>w{piT*fB~J5&2<oFjSraG^ZTfxAzF#qD
z{dPi4bStC1M9BUVp?k?D8=YG)v1q^X6QaNEqe>Qjpi66KBuo(rAPn!|1$KyS&oG8W
z<Al){q0Kd^4P*ucum2L6`1VKfME}m`5iucCBp&%8G2fIrr1kex>H6WH;8jt00L?jq
zxy8nAs!;C=BXT^IE2~aLO<fPScC#TpKFD@nYK3iYGw=w4e=V{#X+nQP*z_N@bG9c#
z^$wr{pKW^>$^ok0dtg^*XE2uVKj-Y3!@DzqlsGWcfQapT@+;fqehlh3<z<&=_+ivo
zIGVU8Mf<;$9pyWG+QU$bha8k)2i%Podga+$L$;#?FXgJnaH3xMxw&e|O5qH~@4W-O
zC^<^x!jiX+|Nr;f;#>&E3FoeqI{Y*NO(u+MkSf{CHo}CDJY?sj=g(W)ekR`944yB~
z!vX5udFWs&0m(%(fLN6L&10dz3?EyS1K(_=>v!Z?C_JF}HA`kn1Hn-9!J3q+XuD%J
z#n!NLzy#odX0Si~Gxv-)!)Bi#1l;H`-Xk|8QPz4w-19y)j`dF=gmM=t6eR3K_j6F5
zRE9$Yl_`uA`9izWPmu9bp{?bPq0=Dc$fvk_KV3VwW>+f#DVVlsT<$M}Z@nB&+(WxN
zx?uYtHXQaEgd@Rzg)FcPT6pN%h(!a%uCs97{n201$^ir#0Q1n(!g>)vcl5WPpt|&>
zt=Yvcls)e251acjvu}IQ#xrTrgxtoV(fJ>QhaSG1NCvr96Grs-ckz`izkWbGxHOj2
zcr#0D)44!i20eSD+hFDz3q%w_vX`HVng0H&|KVEUuO?YlN250Y4x&x~{^D8m+aDS(
z#O>EMun!Rdv5z>@)b{k^6?mkVyrc{b>gOC7sGVuNud3{f^g<@EIRPK^+RQNs-qTWW
zQqguyDLBAA3C3?}?}bo%zCGJuEL)~&tKQN_WOdSU$pT(95m;OEAUfce6f=Cc*eV#B
zLEteWco!#1Cy$9VlbomiH$~C%E}*&W4O^9hR?!Rd4XMU1A=Rq#NiU$nV3X&Z!0QN`
zB&f<OR(|=pD>=D4Y@QW}l)$K3yVr!Z!NbLdB<6ejhP)3>c~!w=lB&92PfzY;NtI_V
zgCW{=Xc!ybrQTdv0EP}X{@zPXuv+2e2N&o)JZ?iZ>YS2e#)etsgr8CB99Bhhdv<Tu
z2F88!1%&O}^CVe{s$hjJe#D~XxVNw%rq8<ysg?o5L|x{!1D6Wgff%iaaWNhkswn&@
z%C<e+wur_6dha~_4LC{)bRI%nN~=&bB8an=)ih@xPPDV&$L7+7ufW?6u8IxpyMm%i
z*Uh+U8m0jdN=)EEd&L9lh#G_4SB@1!pO(Lwf*z~iHQ7osQ_tW<D`y3UKsXW(>_UXO
z`Y23<YC)$YFa5FD6^^hWG-4wJPOCynUYljRHggYNA{n20iR9j0RrpN@aA+U~215pc
zF`6(_)8e?^no1ASix~aeJ@)E?F55MwIC7%R`A2_wFz4{rL~bx=m|Tq38jS4ptD*sV
zePb4w4yD<_Yn3(^))C%wj8K1ApA7!g8646OLxudMom25niNjBJgy}(zd~sTC+jG`Y
zSX4E#O-7#^BI)Kc$5HdqSU&=L9|Sf&C4|RCBruL~4zhiW)(TO4lklTC0#GAPpV-ly
zJ_%qcq=qd1_5LVx`zn7&^s(*f;Yc7Hgt1Oe#RYqlAzowYRKXU0$jjOsA#l{EJ36M0
zrl0s41SA{8mTw^6Pfd}iF4^lG$@lX|7)VQC8l*Yf6YUOyYG>*mHnW#S=h?L6<j}ex
z8R%Y5Q9G}ZI5d#%${{o{re|(B0n`AXRH3L)<1;uty9M%k)T8)OcF&9k&;diR{;|?W
zdl>qNcd>LNAelF#jJH21=E7OTm#^^O3f#0{6EqIxP%e0JV=A<~_)~%KlmO^jn5b^r
z&@=yP(&SMGfe^@HQoN4|YP1?%+8#6TKwE$o6Ji7vAwj9YU$ZT01$wNOSfG%m4G2Rd
zAX_xf9VorK^_P%G3Wk73-cbM-C@6jTN%XXpIVYxq^yYPXW<5uM%>2=u9$QLeV3wrK
zu`PqF*@lUs>L3OiLAxRFFwxXEuZWHYDsm%r5+WFeTm`l&Mbeu&V8Zir`?(a2kgxd@
zi>|(s+7G^>yNiWNgSY$j+rl*bXKb3uHGnT9LficKcI<X%?1*%l*GhdKLfBA*Lo7*T
zG~0s_qpq28ZUvnIoBP`D7?}UA8Oo0o+lm1KNVji(d9*PJd4U+UGq)#wHK?hhu*uXK
zcGWc(F7D(%|18O5`^cRH$e-o>889St@Zt#N(zLeC1}nV}R?4E42GZ$qA`=@?h1FLb
zfD0vp9K_+-Iq2XvEqi<2NgtYe`q1!QV*XQVQX0X66kQN8n;3olg1G~X#sT)g!gcX^
zjbo42=0Xc>1^lJD7c_$|V5r^Df)sEy{369-S^R5fQ;k+fwQYJ`=SL1OjtYQ^<xKRC
z-+;)Q)%zbZ>Prk#2}J8)7L0|~1h^_Fi#JSY4e=9xZgXS0$W}$DON@ASH5}bF4{uKY
zuKe1_T4-9zJfav-Pdk4$lxnjd@|T3XEckZl6C4auvi^=gDo6-w3jeg(R3tU|T7Czu
z*PseQTof8E?{)d@13O}tF^Kem=^zc~xp8y%Yjc|)gU%1kaZvr54LcR8U&*-b>Oa##
zQnKOnPkRu2EJnugSOw~;K!*&2z70=)e`DJGj5krsNuJ&cs(v>}f1?{6-_0Csa(lwk
z+)s{1k>B2fQUEcrL!%I4UAfvGC@V3F87;g91FeAomn0by-r4*`eYfBj(M)0^Xtlxa
zf;@>uCj{u>3~lH(UV&efhol(dZSmUD*zf5t0&JV-$O}L=JGs}QK{Y}^Qlg#HQGc|t
zWC51^yd|U+b=GPXmvCG7=J|eIY@u&={}7D7nn3jsE#PT)G;J+z3g%}6VIxpIZuj0q
z2<p5p3pWMlzu5L)Sna$P^I$yq*-|g`z$O6bfs%6U<?_<0pygEBJ5jbWRj!!ui+&pO
zOTZ55si>6;jlz^{L1K`91s15s!3Wfaa+M6%x6Ca8Ql&`OEq-A(^o37w&m_JlegHP~
z1fC&RdnM@ilG=8);N<EXxg`l0f#|Cs+eo{djvaxfpmu{A5}onCrs52<EHZaU0N92>
z5|Y9@0KPQ?hy7{ZW%Z={noBXrL<UmCSJS@Se6xQVlE0UNrfT5|>>f-^O=bO&W(2bR
zY1pl2pyi=@Q&aWT3W(qm!E#<PxkZ3I9VP-t+JjTdlug4LQj#7mhgE{IGXB;;3)@lk
z^3lbvbuUf2uRmLVf3+%*eZv+&XAoKUY*i*BXoFMuE}?2wg=AX!EdGxl1dkV<S@%m^
zYm!;*-@h%;9EONHVj#B~Q{}$CY=kgq7uf2UZ#$%truQOy$PA5mkH82>(**c@ZROf5
z54808Lk*0bdot*fPZ__jL92E^P=`kF`e})lXUAAzb~aO88AeG50n`lTYZTE$a|mGc
zZ5PC0OdT<h0UfX&LbvdGNKs@jd9mM;E8fjDf|rtX<YVe2B2ylzEZx%9Kr04<mRN@6
zx9B}g>Fvo-NQzf;Wzsk2e*aKHtraM0!D!d&t2Z?8#WJxcs5Bxi<0`L<!)(}~pZiKe
zrAkD>X2!012$A)uju@wq?GGh172r)PZa7gsZ>S2Cqk3A3K>AOZK@I9%Uz?8>@iRH<
z+9*0-^Y$^h&!wmhd|1M-OS@cbut9h-7=e4BH~yw(?<cmDs&D)8bll`OS{dJCY@@VJ
zZOPH?iP2qeuWsxfp$8@_pT~c~Ihbf#E5{P3`>Rs@>%U{x<sk2S)Ac^uS7D;nhs43O
z>kHL~vx4e2J$uRv-Zp23;-``{_@kx|v1rJ*H)8M`6R{=4Y$`<`{p`s<_g5PZMQ)Tc
zNK2ESFzB1lu44>qpW>sFzYioE5;nnb$s&xYNv~2mfrUniL>k+ZYyv@7TPw|)7>XxK
z#*PVJC|WwvPME5l^!$38LdEKwUO2yb;Fnv~$3x>sX%8qWDkf!SeyGt@<u93fP0Tf@
zB>}CQbX(e)Xr@ai&DQ;vQ$rsBQCO3YEg4fsvjnLf)-u+kl~L?|9|duNfn<nbv)vne
zbu(z_Lorx4cwFjtGeR#109dp*md6K;EuH^_gx7B6Y5NeoxrI%*50DWih>gs+fp`Xp
z>``{ZpUxrhV{~4H<t*i)_~<EKm}@GAg_Thv!N&GsVPQveNHkw@US9@zpHIpGfm>BK
z=9Dw%jBdT7e4kO+YF~bP(vZ<ZR(lR8j=*YKcYUGT6PyTO?s3K9Pp4LK7e+acrcHnI
zq#aFX;G&EDt$FhL%0}7HNhWt&bQIRN1%ZM`Ct|rJW=%_r=9w*KbUq6<uzoDwtp+~=
zRo2|7gq=PR?_`e3wCP)MDW`o=raVR}cbonl?!;n=0_KiO&uZkxoEer3LQ^2tM*Se=
zC0hVCYKNN=Oy|ZseG8TFU#G?%n7!y5C45@};Dckrp+S`Ez#RpFW3>p}U#eDMS-;Gy
z=Sp5b21?6N82an3nP^Q5y74)e{z6k?b6duXha|LC!K-aKZp=`^8fh#q=8?{;i_ZxS
zd3!A5HS79mtgkx${&<#)Wwu?<`tYuIq5H^qPsF-acfGKVxfbL7^b`U6hSN=<O?F1N
z-M_|&fEbe6Hg0HDX9(}$R3ui2m@QgZmxTXm70nit4W<CvjP%IqZlMKG%$b5#>Al!5
z!x)o{=0gu;!Ap>ze{o-5jTgIPuX#@GRmy(lNP#8-FPrcB`JHD==4i#;^wFXQ_4WTU
z&a_3tWi8U${3Fjn^q_Si$Ivp^{bymvz+GZ)IcK9v$^o&_7c@gI2XCfeg~rskrRAaI
z#O_f^%t80P3AQH*0!mV0KnS_l|5~aNbeJ2+eV~LA3;Y8=IxG$9fCVO{2ySe&#A)GA
zo2tx0yzO>8@4sY#$O_n{(DJE8G2XIu>{hFV7i6dtNI(wl;R7$AG+3q~_RF`DtAb*h
zOu=t$3UJ<fBvYn!U&(Vp;z|vTD`oAekPAvu2TQwDMdf8e%7^2hc&B_{fr&KUQBC)}
z9vIN?KXzBR0?Lo~W4XZ0nlYuVwvyYUdKk8;oHE*-NR-m7V%n0a%hykNVRWVS*6SD0
zQg~etB4Sc+DESGX8a3sO_##Lwx?ZUr7t*vv)0WQDM%uUQ%7g=^_SkvaTSx|+*(U{W
zc<zPPAvv{Kwhovv7Ju)SzTO}@ukQqCmG}2L)$;e8Uw_P1JhmzwHJWwK1K~(*T?mI{
z=O~*Z5W5%eoNjOZEbg+9tMAZh8t3|2d<Ak_2raAyj(1CEAxw3JwRgu1-%fW<=x7?R
z0_GQjHg~?CZ2_%68z*NlXyF<Txg#2X@Q{^qP1tg(o~t<^u}m!fqt3xgletUQ@-yy}
zZSE|slV=gjm%S@Ep9`{JiA4u@wITWS(sLV&^Po@kMWqx$^qZ?FZU%)Di*oe)+yis~
zq}kLOH;%nts6IOD`x_+rM}yV@PJFw%_qeeAke+TK0aPJ5F2Y<5n}b|9Z$d{`l_ZQL
zNFr%qD?s=`Ii!$d%<TuYrA+Q4B5ay&LGnS)ctfs4%#qef3wDBYmfzf&U8|4ZKD3LU
zeOL-VBsIV38IiMkPSr(NEvPOF{kKmQ&G~E)AtPzOVx|Y)>=D*k1;$w~CQAQ=yGC06
zzDwB{mDXErw<)FK+|rft*>a33zb<4hi_c1hC~_CS4Gd8c29*J)pXl6EPQx;thX;-y
zNneKfxM0NoKjEn`)5-84p$v|@qa5b!6Grr0{n=9WYfvE_QVl|bN<NopX8m+VT_8zG
z*bPP9t^?Q}ZTCF5?vZM^e|ouh-4M3(G9_eAtQXzrWPqffV+wP~GMCWoa8m^gYaGq-
z)2LQI$IgBzUE%>z3A?wpol~6k>!Y-55qVQHQ;kqd^!jM0vk{+;7GIf?d~+^D-8B*8
zVfQQ_81B-5HW+nu`u2d@?wvP(sj)J?OWGVz%+)XV0Gj0UAH!?`l+E#X$68zBWxNj^
z{QcCbern0J)iU}yoQejyoV}<5zNbozSdWcbNb=fp-H*sB`wjyZ=$maO{zEMs1Yi$Y
zFaf8;1hcExF}TW;Cd34IGtbv|qhqZJ!Crc$f5*tnzA`_EW??LCx=LIuWbax4lXgn=
zt&L0yHDM#(l~wNVb1nPu_6Dnl^CpRqre0o|1Zawf1n>9#h~tiy{=bBE`%Wv#T%@at
z6PY5>8pQSaJnC)3oI5S!M)!%1VHFC!18tNmqy%<sBrB$c>Jy+W;3mOKIiy=3#zXS+
zD8WMlD!+wA*u)UJ_pVdjZdG#j^^ewWnlz5--oJ!<fQY2-(w>#3lViD3iFR48z^E4|
z310I3y8VYvR>^6O;8O6VgSCyC&4)m;a>+3osQy8m$5XW)+@*wRs82o$cut)nC!in<
zA#`&DazN5@wN}-QOo_X<#6F{G0$ORo;G9v7pO#JW(oS^P)H2l=u|XV`EMfnv{uKxS
zBy9&*zd;r}qZE=g+KW~5J9R+o_^Z{nMwpkDojkHM5mG^T`tu5@W1z$$<U~Chf*Twp
zcm<vf^RJ0{_?xv#rU^tort2H*tz&UH)&p|lrt=RgNMOgJR_|SWVu_I^T8+aN=?3Yl
zt%XMk4zaJpU|jv~S4M+4y6jJ<cB^W8i3Mj=dioVkHD+7!!}?TQ8M&OPs6x+^dzFp%
z@8R9`s|PX4p%Na?-tL9W^?8C)?mDrl#HyPPVq3_m9!_Wd_$(IzOSCp7Q2ScY``Vjv
zDY$CXp+3U06SL(_n0_09wup;lWj3ABG8wj?ZL|tgIoO5Cqu*kBILRElR2C_<^)Ig&
zYxbR19%Nr&cDVz<faE&q>6?93Q=E*Cu=pLRwpl<6{c`>BAzbFBT;Hr-+-i3DT-PTz
z_D91Y6+boID*hTT`YL&yCaso-^rwky;>)^H(MzaL1;bnVz)J4Ope*IqS1TUpdj2@X
z;wLrQMy=x6s|(fQ2SSKfq$`rV735w-r5~WwN$85(ckfxcHa;hQ^da*Vu_Gedrg2LQ
z&d)@gY*Mq>AF((SmtEz6b0sR{P1naKd$9iIN?)zmLDLGDD}Qud>lsFM;Z$Af+CGss
z7KZg4_f?CxYwQaOFPze>?S7IZrxi=&8X3ae$EEziYkjlkaW**Z+q%>D2yrQk_v`Qt
z!zq;xtSVj0Ok2BP`4bX*s@Uvh*c&R-7dfih^U{lTJfG*>@^B2d#fguQOhu^Sz}PQf
zr-6x+SU>)#;>9mfKAfnwp_6H}pPb6(N^IN!{*_g?0GIv2Qb6JWaP0c#!Nk{d3iYH@
z8C$3H%5FBSdwq4%eN~~(X^TztxQBVwdp6!($g$>m#syU)!Xj-^OJ?UWAY53*g@%V<
z>WXMEjpY>@Eqos&snzGX<f<r|CM#-sVk#m*t+3$gN1~qz(M9CtX_q1|-HG3=`}&?Y
zCTGd(YoSP)X!-#Ix_PwSRc#fg?V~$Na3Zlq<-F{az_rL*4-Ya?WNPC(q&^4E{^&0L
zSGJ<?H||L@xa~1Crn+w3cmDE0r%}b1kJB??LUVBRG4|UtXEL=ZfzcEjqTXbcs3UdS
z=dewBeY0wDE|jcf29@lzUk|*PSDsmX;I4g6mAytmFwOS7>Tr|gD4Y6&2M=TfX9v=N
zNgwqJ$*9VUEq4wKv{JOsOk+{PAG@$itVuO;6neB@CagWDdLpSrQifx!toSU!VOEgC
zjFPx7+@q4nl9cXkS%?nzh>lGllGxt-fm<iUTVFo3N=}P6eGal^0d)e;RNuVIut@`e
z4`porbrK3wUz3*8)?P_*z21>yst^pjt*e<^^?-(%V#zZZLbmy{w6%bAz<36stSDTe
zNx}%+9m>2*-kJd=!_Oi$(a}OI<ZWVC?On{HY1r$uvt69CGm%r4VDN*WednKs<ZFdj
ziT8UQAnyByu8D@mNu$0%<9GKy?1BQPVE9IVDbWd{A}3fd*u^{YoKNTS)U4^R=mkWm
zJ?!GO`tq_bJ1*v%Y8@J;+LMxzLXY1l+nL^xC~Y0%I;WfJ&Atr1+kL2m_|yKQ3dxT4
zPmHD^FBe@NFNj+ewi_mmGMFf=$cx14+;zUF{;2?zVwVL2Xkj<>pjt6BXyUwl<K3`*
zP;#(UGa&wjUG+ui;3Q^^fd4(%mhnr}>{IL`ejXRXCE4{EiEeXH=MDII^vtr(re_WJ
zS_HobltOY>Leu=|ifbocrof#4BHxq%`4)~s2eLQseOk(t1Oxh2<r$f}$=@Uzc?DpC
z0QNjNP-|T2sn#l4deQ}*sKB>VJ9Au#?C>uJXpPz0RPt1=<d*@=E2HM&To|(C$x`}^
z2v2<|6C>H0;i2VmZ-Tb)?rf2H%L$76xlpHG%KxQVo-9$`pxBmj*BN^Kwnr-fTY1j;
zG!RqmtN5<icFJa;=90}@Rm#kR<1Cbsaox@4De04dvGDoE8MaYa6T3^d0L8&2-KOb(
z*RmX>@?4AT1SwG6#PMn-E9C{N6GDt82vL6}{^}CPtQPYuNx&;(I7*q)^&Gd^47G6J
zv+Fh9E4mt+yJ$%)9BRhb!+(+y8y@@2c~w7CdDYlBH!|C=<+ufVAGc2B(qXYZt+f1(
zzx<B8Ze&@p<uq<a37Z9PJX^L0HlcN0#WS)$>?YoCdlln+*LQTDdwa#s!a2Hy-kRYU
z@!&pk%vZ_<_w*6mtDnnUxlc1rd&t&Hz+z#Ob97U~&VsY8ntV2iRpN4pWlqUbJkB!J
z@qBtnd|emg*V_?Ff2yKNo+XM2hX3NwlmI3N^T_Dw*6z1yttv%Z_Sw=#2tL<m?k`z8
z(&@avkb@VEVT8g@)1B3LTsuK#|6<=gl0?|lpw|d+7Sj~PQAX9(?A&XOG=zA`j}){k
z6Vm$k-c=b0t6WLKi%w?*73Y=I=&FoFrl}*0{dee2%0l-^QAfN^N{|>(MQ-Mu2mw`^
zyJR?i>c-;A??z@)0^{e!unc~oJO|4DgQm^q*K*^rRY7>sMaC;uXuxMr5Tn~3Ov7NX
zrpl@2;2xNv^Z89eMYM7+9xqynVuc+x>4a60yGiu6c>=1Z4U?o6iD|zD9j9XPq8%yJ
zWn7&A&Krc~?7YqHOD!=9eQjp%hpNbZO=|q<-w#1$t=Knm>iy8v`f}y^vN+&M^24_W
z(83_hE7mv$!?9;-Z;V~cU{hilOg_n2^|f6ib9D78O5`QPdvn-H?}b%J6~`yeK}^n2
zV3lTOB1rh{Tq7UnIShZISGSbsy7Xpy7r=!M<D0}~RLWq4)GMRKF@`06&IiKwz>NdV
z7>D+R)=4s)_Ki9~f%+dURWANG+CLv(#lgwgycj5ms|qUi=`H5fW<0$~K=io+{`qxM
zvVqRmgJ}FR-vwjFwPp_NRhQ~L8+D^cPo+Q6*B9~b>6J<cRwO;=_>o+^-(0L|c}ldf
zGBzDNZxLFe!gi)Dh<Fl{=jML1-zh`;Jbw2wEN$wv`VPd0#Bzpnt3~;D%P<26-BcG?
z4_3GPmDC_$hi95wRcGsBLGfmm@e*{vlFMQ<pCr3@ks>WsiLNF|WXjJx)CNL;W2zG5
zUGO&gD2Mb_?M$7^$NB+4ykKN(1-th57Y>3B#hW$7kdMmch(@j-2B&t=v^5N(^Mg?W
zykQ;4=rE~47*Yyg?!DSUXuF(Q4Mb1Lzrep+Bx%}e5fguwXI9y+JE79EGCwmmHg+W=
zr^1W-ipx6Gs}d)0*fW-gLl&UUyI*Gj(J|C!#JBL$gHd}Y+=K2<d94e6_FVY$w|+PF
zJ@s1~`EjNN{IEuRm++W6$dW%{MCLvl$KA1=5%rl&%Mz3>O*<{R?Wm6^9<5E9PTetK
ztDx^w-Zk6dd}9}`@4@j^azv=mX=5CyG~y=>zV*TsOor$K5NZc!#KuhJyQsL|k}jW9
z>dfCd%(FP1O4*km>8F?i9`(9wI2o8b=F0WuB#}qQ6L}CUt)MGI`W~jc_bu=&bLUS|
zsI#iGN`)%0CFM&AGjnOkKOwuAsG-oV;l&wpR+orP>iMI1={c^)3J%|$CMRF=ee9uW
z<Zb-@rTBN$`ufj?NWu9Jf-!u29~OFSF$tfU@MVY<ptaO$hTTU&Q5A~!^f3uhPbWu%
zjc`qrjJ~s=5<DT8!e6p7UC%geCgFMD__z?d_TYN-^L9<*7H2$}XZ+`=aO$_>L#>(j
z$lT2vn+8d{PVJ8Z5bnz#Sut_Km*F7lw6^5G@%d8&`5~ufg#(NGyy-3_E41YE>oA`@
z;?tSqu3b>q(b!XZ+tWq$Y<s4zvAgzfJQ+=*+Ww>p!Zb(k{rhhgC}9{6y{icc<~OM!
z6^>DPKxu)Hw4W|bnBHdF><6@bC+k97IW6xQ#uyr8iV>{OT)&eF0tQD-y3>;hm;`(o
zON7fyPJH|}f4Zn;eog*fYtS>ynd>S3n{15-33O7&-|SURc`ODZLYB|9`om%Z-R2A!
z8gjXhPl9#kPb#m293;M|3<Tk2CWH%2*TULeQHiK!OSajL$(i7E+^yd0Jws-`zL>L%
z3BH%Yc}jc!NDMyFXVc*trdqCVI4K-kVqiT@zUn<+&JI{)bAP<xj2#6nho9S?<g9#$
z&X^;1#y+HpP7pm!CJQb!YpQY?0}YW(lXtf+t3UOHZ+iUfr+^C}8$4^F9k>V2yswz=
zW;!e-m~*pWK4~9zd`0uQnL4?a$xmB1v8CaJA|KxhW3I`H<?JZKyerZioooA)f<hA$
z4hR~68t3|C2PL?W+&kkN0=sYVtDR{7?($I1D$%1V*tVz_=r9Q{#m3^fk7s+CM%E*r
z%!TpkD9JV{{j~qd$-~hRH56$p$}#pf55#-!7&cehq=a5+y_rVaAyIGp`Cd_<!+a->
zh?;M5pbLg{?Y+V-B)hQ^AZZ{Um*JkdFA!snhaqJ4Z>EMe$0#DZGQzhA?}y?LvtkvC
zVCl>_@rP5i_C(%ht}sEr@s!65q+^W@jutV#?N@vV5^&5-$yqk-P7u1~0Z8|YnyF=p
z`RwlbsmX*tm9AJ?*5_823vBqQHB|+E(6SjqlG*eyzhwLEXQwTweu=IM8+?2H;HOB$
zZ7`36i!&WF7X$&n%%8XJDQPoQ1x3^&iQOi4TJ%|(Hucm-E|)u`Wsl^Cimxp&jzFMP
zhCymg4e;@wA{YEi3SwYZ{L<tk%#fd#su_Z;8AwFMX*M0gsmp7k6%E~S^V%9*ML{tw
z+T%EymJ-M>@Rqkegd;3GLc#s8`@+=apk5AF@}`%E2!V$v>@EeVRBExEGk2jH_3>RJ
z;EZR6)*)+2POrnntAj3z#LN6SAv~f(5qo?`TJ66**+U|@d*X`EcZ#)g$L^wr;;BZ*
z43=5wao$qNjVz~v2tww;CpO+Ulg2kn=MSVSh4V)5zvScmdfN@2VDi=)zI~@>CBkt6
zJHsN^_#oPJh;FaUf4D<%z5!%oYe^$eIQv{&f?m2_F?8R)arI@h^;D`tzO6-nbmzd8
zqL{u$q`cy6D7hZ1k~tXXNuy8)`rts{X=5(XCWNi70^Rsk!FMqYC*DAZz{qzWgsr{^
zkb5`h7>NhTy8BqR6-qrOqu1m%n^H~FNE+{>{Fu_hJB(EK%z}P=m#Y3NJ!XM-ENVJk
zC+~{GB@O4%MF6UvO~Vx)S8Z1Y*7#uhyObMU(^8v`(}V$~bIl_uG0`yAnw|Ll-IHe`
z?KK=itC~RZk6#hKJ#i!K7xeYrBkf+kYKeoo4nh)swrJx^xvJk_WZictlz4GQFlv<>
zRZWy$^6lZ#K-j@x={2p;E6{!1qUW%gEfDJW9R(_X0}|170#^R=mfA5fiIPO^vT`=i
znRBi%@6$%QYc!zzJ8CA<;)sVGR5(Hzra2ai?YAn|mMnyO6ge)ez7a2P-7^aN4Zl*$
z)h@H{=WIXt`E?ll|MfKCJS&Fpi-cc^^1y{l0>a&!Zbe5*@o08_zXz!t9uS5vKaqbE
zN#0Mul!#R@Onkef=#@Mset($9_)Qq7Uzno+@|{zh9H_fw)MZoxVTCa=+H$k9C%Xvl
zdGvc;6R0S1*SQm4okTTpTC$bJf1bXtsl{HaBq@Y^xNZMDW{T~h1ktS<`cR3_{c4rJ
zM-Sr|mu$>P>mZB#OpA$Zp&QAI$_BZsQZwDYC@aO?uAoDWPF#O!Jcv{sL2!$}Wf{#T
zI(iF8^gdD70FaP}b4w22qbSR0!)>0H;`=?`z9~Q-faGCL?$Hw4I+p#p;b$x_2Kq82
zo|6GFEn)cgQK%u$t_cZ!0PNVhu7&Q{M~gyT*`-_DLiv7s`gE3K+?LFTtO~4@k6s;Y
zI_UD8$9|Oi_{TIDg_ds+fv*pw8I$>^Cu_$IsCj05(#(j~5?(2VKJa}?VR36sAQn(2
zc|fRD>R?(>V_{eTilA_VXy5EaoX*n9FBk;tl1`b`Zy8e=OK&N=&m+X_JwZ&;PSi-<
z*nL6#zWaOIs5GE2nWC`@UuAosMpr+&28B<T`=0x~^`BLig-_J0q+DpLypy-3{=?es
z8f*M8)i_)%FI}Yv00Z7K#|q{}QF&WbD=fPqlc!_m(a5}W{&^#`Ez*Oq4S<427sIoX
z5MCStSC15@&TW@-HQ>X?e5%i^0}nvFSn8qjEulr+z-ch#gpN#iynD1eO#0Ec8Nn3X
zs_1?Oxsmu?aP!dsf3tbKT|~Ug;&ixO6n7nwVP4%T^pcFpUHNrpu)wYvNcqr2#&{j`
zi^+Rl=`eoN8HDRadK)z1f`x)Ln8wvD{n_}uwdKfux!pz@`CkA?b1=wsw)o_!LY-U>
zy%fWo_*B(yhbAHA;91etGN|Z+?7F!24~L(%Hq?Ly^@H1Ndw^R!*O7fD$}NMOiXB<T
zBPq5qzF(d@{<=~s0B2;-NuDXNJI+okTBjc7&8|Ef_z>QrR8GZJsgm~Nc}ri+R6t&{
zO}1c4urMrN3$&}_F@T+&mP;4?Aj}9UM`%;0JhSb|_!);3VamARGQa%f{q{v_vL-Da
z=0mGb>~Hsqo2&7orL@3}5xk{(G<)0cbMGY*47yr~<+HZ_@K-_eG&YReLAhE~6L2f9
z17RMoF~Yy6*458-#C;^bGWpu(v~`s=9K(TwIR`+bb(Znt{SG~M!Ysxjvp$&xq#RhD
zRF7aeT@PbWK_>WZ0)A-;J!Usmp7)zCUQ|84zk11cPo^2EkBW8T6FxJ&d-vm&8~Y-i
z6?R{=OV8F(ygr)Na&<)On2^Q*qyuBh!+~OV=5CO?TrNxjzJs1|Rbc%<<}P|6)2Fs*
zy`3eoXwiA!*)*Io2nWvB%$PS!>WvABhVgw?sBdO_mnkhPl3&LOjc_wSWmyJu*F@%h
z3~DJdgSqtDpPx@@6|xzS`(SN$@~LUOq0kZ4=Yu6<*#*Bx2(&MJy5RV=HPT|K2P6mK
zHm6goj}+EM6e*7-b!aaT=zXj9p0}(&yTbQ%OnDWm;yk=acPh_bKfCUTc;2$MrE4zF
zR}z!fMT3r9_dk^&=^0~xV-qza1*~B5cwF&kiK=H97W2a0EX9*yRXHW`N2<s}HV9li
zT#WmgF6i9?k{R*cUUA`!C}dsc5xY<oyyxHu(@Pf8g?E)VVB<JuopsnSE_hSB=&={C
zY(c5eecg?4<vZF%LQSajiU~)DCkWUI^ArioVFnt4h?CVs-uzeUNW5<_;9@NZz1;%G
zf3@cr@!<Lhnkugi_x{EtEPr0yt?49xDe=f6gaRZPqF>HbTGC$zDTw?fg$w4L`AK+~
ztC<Gs;+hv5s|aMso;E0A7)fdYlsH<LY4#X*&_U@AhLKRdFKqRoF4S2k09FdEvNO|2
zRg3dq1)3=v4<n(}1thg)SNtpio{&7S<;*9u!tm~jN>q?rNUk>&fy{*vFRcc4P>@c~
zzo<&xO?Jpsrj=oe<S7}!Bd?1}Wpm1{bu#k%R{FE>Fs2`(;ciShwEva0*p~%A`CJgz
zBM5@1_<JT&ee%;ebJCYuj}hWY$zQ@_er#|3i6>=7zoo2e635n(2=H?W8;<Qa(@EyA
zV`46!pP%c#6>m}Na>{p|yx!}!=B+R0TKjQ$6ar1lsYvp5x<jA5X&QSe3ohFAmScH!
z99K1Sjn{5i?KV*Y#a6!2^0=0c_fsdr6hx1N(F7nhgxeKJtuZXl0fG9iy>82vBVdnG
zq1*-~Xhf4dK;`Zo49@(h$yDE~2}(KnO)rI6jce?FcEO$*i;xHufg!RVxJ+J_(4E&H
zOmdb_GlS-7wi3Ey?QSQ3mGZp=oqQF#_C=*js_nCcvNBbEnA&o`=Gc!uN6#u;x?e^b
z0Xtk22kSzl?_qVnz(!I){N!NlrH3bSyUM5A6`e+w%4e<scVLU2BjcWVWP1+}`o=4`
z0mDQl?bE9u0uDTVvJfj*gxK(6o52q`;gSxy-y*P=BK9>I#qLe!yP2g#w51T?z=&u%
zC8d<<HCD~&AcQbtj~Qv{o`Qlz1MEF%yZZ@H3f7XViqJ!Ws(9zJOA95a^RiEB5Wq};
zCSCEodfr+YfzmP?&b+q8$)tit5$m3s;!bj9Yv%*N){e;vUp>pw6^<;QVOQH%v4AvD
z%wtynFHeF8y1IeoI?QjTCPk$@24g}}LWC29UmUtEd10T<JW(Rk=2?j_Rq=&*3e_i^
zFNZCwrcm`J&j}F?0#`iQ6l$VeXy3&#q;(#!YTlNp5mwQq@c<s9ip;b_vV|a_?V`v8
zwv&qJtuwR-?}yDqg|7SAHi;X}G?E>c_j~8MYqj7T99<YpY?l}bFmHhpm*BfF>v%Rt
zl0sE;_n7`L{Sf|9u#df%JoK3hVTw9D&dGkkcnF4dPa!<a6vPn?ChZ0kYAE;11DOBg
zN*cxHb|a7tJ0?%77&+x7qc$O!Q&$5Px0|KstUKz&OA+#%VOAXd0>U|`$@2L`C=sT?
zFE)f;pn}PFS+ofkb1iueDip=hcjqMGH$8ODg$L_0l8(7*f$s1_4D+|b2$+h5-H@k5
z!Y#w_Ky<|joBV9k_`Nnyf-Vmo#Q56moiSwYyYz?MDk;pwF~3+y$D|c(PQ!w_LbOSV
z;Yl)rDmG%+&5x%$YG5PK7Y_fxEzalZe~nr+V<$<@!(}4tA!t7pu#cCoXYFeH|5$tP
zu%7$)ZJ30T(ITOgmL?i#3Z-3?QE6SYGb)w#Kt^Z}ZQ7-@NP~nf?b5hfilS5+QfUeI
z`L0}C_wl=*<9?pw`R6(um#^>l^LdZgJkRq5<vkHE8CJ_fp~vh|z|IT#5kCZ~**xeO
z9+Xt`9M#KnVs1$7oT^u3(1bxAUw-PTUj#;qQLkE)d4y<LAQ|eeI8(%#g?A;Y)%tw5
zPc8hi;u&)8i#)!+mB4<OVf5L%9n0JOWap?$$$eG(+LA}=L~<L}OxNF=)Dc$if?*6L
z7q4UK_66{z*IuTsaK2i}^7wzOAYM+!VCEnm_us%Mp#54^X%}*PpmyEMWp-~{Zr7ee
z&A_b~uuY!mYuHY<ncZw3*!L=@)DM)H=>ZzcOP;aI<DUAZ89XgyPTbpJf;$Ukp^Q=E
zJbij|NT5R2xFoS*JE@5Qno~-C`OCM~Izdj{f^S&-;??Go6km90hsok+98C>O*DpFg
z!Bkyq0z@^qyvOPX^0kEz$tDZk5iwJL#i*Y8Uwc!QHUnFK!?%^e3R)7rjQqz1n$I<`
zP|WI$k|`uVJ`5q>5Xyd_Q+7hT29*J~K&abQh;>^d_Av7@{~V8oxaFZbnu`UH{vi`w
z=GXl$hxMS}SM7x*mt`}-B}lBkq4yT-Wu8`u&c2oMN)@BHAu#{g!)_nL%@Po((&<3p
zO%TzlFHEhp7w~9=veNnc)Z8VfW=J#*ZD$)0sYM?YUan;8I_Wh!bp+hXJz7Q^_>OKb
zHAbmU6H^81H8xBww`_SJ$bUTD`k9gP0*0T>iyYvt+C+;3MjE4>T?hJrSoCsapz^@>
zf4A+cC6aWZ$gQp!OQHfB%zr}Y)Tc}Ss%H{ZWRu@DSmZCh)o2kO#(m#1oPDYUh$x#o
zFP{O(H=mvMhs-A=VPO#PmSg<hZk4G7UjMcRCF}W7EDfnzJincX>oyo94T<*T8nT~X
zP^@`MVZNKfMu7ImHm-s+1CQ@t$IwHOGbZr44h))}zfYrYT|O(}y8dyWQpm~-?@a@L
z$A$wdbFT1`4_lWNSs4k-TfVFe#ZBotS^%J~kEzb1p&%4kFxijgnAMFOuwY!!>+tj-
z#zn*MTXrJCc9&?HPQT=RXAiLzFMnYOQ}qTFqY7wCZw*L<Y<dm`X<WCX@1RRLo)Iwz
z>=*P}cVAqUu;>+~1c;@T^h-l+SPxn;w{M=iOHVazP%<NvM#LwryI)WRVb6$6C<j>^
zS%rh209TOoKC)^dZk3_$3a?a%9)6ibk0Ohmklogp@1o4Od)CE@N)OV=03ZRwVySIa
zkl!OJbMq^+)3kbH0f1G1oS7`LG9yxHU`E(LC`MGM&`X)E%VlZbw-j<SVPZH`fxX&+
z{0PU6{W}Gha<D&fqcKuW4Tc-J7V;X<wPI3To4aMKguJAmuLbj?EnMRQ-KeMMI4<5l
zEaJ8aL?n0{nJmUl>c_+bQNiXNA`usyr){#V@?u2&fw;;i%TjS#j0<1}1U^&nAy8ks
z6h;`NG83HnK*%l9h7lQ6l&cAymwn(&J_nJ2ta`a_-;nxW!HXoII<jRt*S@6}V9O+<
zXgbZzWNPFwd9)c`F-<q$`mZ*?4|?ii_i?^~r4>YrGz5LrYYDmCF(vbQ!p>i9Gi8+*
zz4-sbqcve(1UKr**LT)G1dLB=TlAF)8b-dO<+qwI0DqPIY^U;kJ72cs8s2XvtI@5=
z7#23V#(N|_HzsLY7u{0gApUs@P>czLROxiCu4X>)TIq_93$?rtV}M%ZZY3{@<sJON
zzkEbK*}2cM7&A|ygMp)>RYjhw;DKdj-zkuZa<??~kT22(5z;eKdSj5p=2z+ULMY(>
z$5CR$QMwjA63@+mqjUyGDf%naGyJF$IJ5fjF2SfBUwOElmsbf1z7PV~wMu3$rM)0u
zs0?83@3W-1g!qoT3>FmFxm2=qy?tHq<&3X!%XI6|5@Y%c!`qG)Ah^JxR(vKq(ovwc
z^B97;#@{DNi=3qPFESRpaguz&60Bg>5TP6OFS$YCLeyReBEh_#Abk-skyFj#kJM5`
zz!r03J9+)#@Cffik(NA50*3|k&yB;$v&8q6z$Vk;S)Tp;QKaYC?d{w2;XWzZ6JRe;
z+U{PsCKWhLjfS+pj`eBsFdZ{8<`Q@qbv(@THgyKN(L3Z}2)(aG?}C~fwB<hW%>bv%
zgxFP*B0Fi{g=D*5r@4rn=Et8?#F1%UBTlV<rKOkWyXBLg-|axCf79r7z&-mb<}ADs
zf9rc8+NnU>bX#(?LqySXt;J%f3<LJZjAbPEBf09<LUysyGonn8#NlLWMLNeUVyOxU
z>H*yxr*Tt)i0+T-6BwaMnZjxjLdFP_`i(`WrbmM4i){aZg(JWOI#RE`>~aJM%wEv=
ztVCKBrMlK3@mLoUzaVIeCw?7t{WU235?;fGk243suwMTSS9kPKqy+u-jOrF+)|=*r
z5XUfqhVJ~9yvC3PVv~O#GHULUsvye2!x2z?2`9(3u|k{;uHmq(!WZ6?dFgX-r@V73
z+B7+`&a?@`*Q#3<AArE~{BB-{A2`Oy7)XBzWG15%w)c3d#nledDsJ@B_ClOl4Fy(t
zg<%^ZU^xTEdaELxUPlk47jfM6P<I8?NHf<|?YHTrS0xF_|L`D2h|36Mp7A61NlQw!
z&$x(s{GU(3j;C0?F^~bI${3?Pc#6^)RSsIk?c^zdW)4D6aYtMRAx1-#4M(cWR~Ebm
zV+4sKjE8s^_JET(Y3kNgQ4j!yj}hXf?S-8<uihmP;_88BqvGFt$|E$rfMFoA&YBU?
zt_%N=u+!>}R&-@mjgzubkv~o6c%SeZS+;J<Q_Cl-Bll=vzc<vsB@A&e7w5f(5x`Wo
zr~?&gZA@G+)c&5>NG_w~288_OM+r-;vgY|>9767mEZcLKOy#V#iQ??!XT<YgAFQR>
zjsL%vGH&@KrpxQQ19Eh|o#+t*hQEIP$ip9mVmM9y@1MC#e#X44K>$A^N`B^0jn{vE
zW+@3XO4JjT@<@KQO33b8lo6om@8#HZjq1R3>qW`MH>H~}elgY8Z$7ysgQ+nCqLcU5
zuUmeY-13K~EjHkmN8om;#uHUjQr7w3Ul&8GW7w866}qWs!8h1!qh)Ok+UMckmUlK1
zR(gBFRl+NWuk>-x$=`7pO$43nQ{LZ}YoQlUz6!@z<k%(q?F2}qO0Ld8&lcMGpmXs*
z^orzIeNv5}htj8p{}meP_Z@^XlcUWBMUP(rPZObp1VJ!Oj+~AXs2nEZr2la`h&!h4
zC=i7_s<1pc9Xl`3E?$BKyY}miLrJa5ahvtM3-U1fPpGe~oa|58CE=_(O^!S0496UW
z-w071eW0h~{{8ETiWuo4n#3cE%&Z%Nc$-sGJ2C+0xJBOP%PRw^7a%v8K$(=#Ly?l_
zLdK{5X3gE&Ki1Qj_#9^S>ZLh{$#R08wIRma#=SowfwK+6chYet97Z~LiTN^TlP4{g
zh@Ys}LlpcJjQmY_>X+(%j{xJKBieDipVRI?2;3&`)2g$MRcbpNR1hGqD&S<0!83GP
zJKDnI(=oWJiPEV*SWVJl$p4Acc!b1>4Bca^7`M~*Zsz>^zNbi5@?yYrJxxeEOIGN~
z=!^Nl>vg?9#0p0KjRj;c(Tu8wbg`(8<1oKOa%Z|X+ia~hGC4AXjtI+;vA44i6vD{7
zH4H|wY@Ye0Brt!NuB;kWUH|F%@lycNGYn}6O_WMxcbeA_72oskib*C-#%9O`5PBk-
zLdsAs=$Y>*?A_GSh!pMmljAQy8zD-Wq<vbBM$fsxvfG=@C@$H3fc4v>#i#R4Gy~nB
zIIX2(=iln>IV}WmW*Fv0>7c{M0nxM*O|F}Q6V5gmIwDXJv}C3KpisixDgx1eOzk->
z*ZY4Fs4XKN?RAAHrmP4b#R0oI<{W4JeSdhd$R-47klYoDtEHTj^Y{OKi99`VTtd{I
zD^o!0oMGsw&;+GaQDc1q0^aa0rZuJ6p(x-)dUTOFV<C)RQaRH4%lT^MJ&uWyaTmKz
zU72|l^5auw0freJKUX-w?SwS17Y!7Lk(?F)W%&kMGW{fM>qz`XfcWO(-6n6AU&-jV
zQ=bh~R9JTEhT&5FEQQNOcKyR}`&W-Tpc5Cj)Kr3k@#Z%QA~;%Y)$l$3$l^_XiyW}>
zo0}pHt!!6*Np0t~1nusQ@Y?6Z<9zt&35Z9g)1C>W2o)F(ajN-dNz5J5p@q<F-T=xv
zCcWoWG$!SUr&ZoD<;hN;+9weO-AT=}+r+O=NT*InX>05hBr0$FUKIg(2g)mc{dOm6
zxumkp-~qYsn!!rdo?#!{<R4#!p6ZrB*5WV76Ul&j#=vx~E70m~SAA+S{F?6ax%}=G
z;^?n(di<CZzBr=Cywgawbh}L9d0ZG<B!0Svy&|=g=_N5`tS;3Pe>r}AXb8;y7m@(e
zZa<ph&z#W*Y_G#TQJ;l(DA@|j3nr-r7lk{mqebN=oyTWm?am$BUwi+{yQVhgP6rU(
z1-Crw9bEUwPQu~)m*eQWRgarLY)5dR=KXt^;8IFoT$sPZ{mY3Ta0#Q44$v5HdXxbe
z8kzHOXU9J&M-D>Wl-@pJOLWX2I(8vDil<-5$T}KnAWQ#S{F7gFefAg-8WAP;&?VZd
z$#EhK1yIluvM-uvy(W-idOgygDMxW~>#m%e%_@CH>$cV>!>z>iZrFafL?|p|n(lQ;
z4195TeFVTPOq#n`sM8?42}2#~Z@0UHuD`SB@#;;Y<<T#;FG}>qQjMzi2Kgq1Q%%cu
z^-eWcPV9Ib8G4-GP`2=)z>YPw0ukMUeKHZ<r8aI$SVA77_7KN$Xx~gv5PFi&uFql>
z_8&k1`>zwpgUQ>yoSzyRIvQD81+N)Khtv-SiBc(Cwn-;}x-EDtCnM=%G0lfDy7?0d
z*1NvFSmpwn*<Y7T`HrWXU!t_@3DLgk8%6G^4(7Z_O?O?6^E%iZ7rz1>ql?W+;m!j9
zuLL9VYfx`I;{4WmP|!uvRX(}1fzJ5_6U#@P%i-GZHDs5$OhnbotTF^>lMj4$`*{Q3
zuW~ec_5rd@q2NO(tdd(ht$;A^=Y*J`JX`(ptHsZb%jmjFWtCYDT9}Nh^vQbG&4}H6
zwdHMRWn{C)NW4Xe%9GDP&uaNF?O?6Xa-<m|_7K^c+60UDc$qofdLFp3U%uN*l@=uy
z21o#h5;K_4cc3ok!p;Y4=Ww5uQJWf;Af-bxaM?w&9jKot+axr7g}S|@Yv&N0?Rv@k
z6De}7ajpT%aSV3x99jJ&Y9;h$6GX$NtFnlsOcY1nm7lgM%jihyzJ5uZwhB1i1kFWP
z1TW1jCcTY=I$Y?kmo$mw#aF&q_ZEu7A~m{>5^n~WUWWxjB1F|=fAE3@^IXgU8}%sZ
zwYx&rHCH~Azvm@={KEa<TBbX3_n(NS+4~GGmFZ9rG?6)W*Xa2r_I<Mr53vsQ1Knmd
z5nQm-GW%p2*8ZxsVlm1!VMTL-V=tSLp`H7UgeF$YG)if@I+mysqCQGu>XRm&P;aR&
zXWf(^q-W$u-(M~%;iEB)Q3G-oh?E^|n}&gfiCn_^{b6V@W#1FvT;k$q(P7+HjX>~4
zP@C^EFJz)B#jZ2kEOVT{TBf1|#>vGcKg7ZAMgfUsk^Q#tv%0gZ-Y<5Drw?P>w>n6u
z9b((R++Ypmr0Yfw?7D5D=1k9Q36+eRW$d=LoM>aN<iVhC=7rPrBUKVU>FRa8dhdC&
z`T)A3XK~Ygrv|NxA+F{{>H#cbjxqw9^uWISa1z)Y27yIH{qgj^tLk1~Q9ZqO^r*rI
zCn)m=t{EY1Ue#Ec?^nE}iBxE~WtG#CXw5;~?u*=9MKX?~Mgw?dMOL@)I4{-;L0EBy
zXkRPX3Wq${b)OYCv^kcyNowD)@<I+=7ZqfqujF?ee8Bl+#OdMW(S;#GUg_&`1g*Y{
zV+cV$&(1TSBxc?*Jxfd}S;aLLu_QLhj7jC9=entS#(;0!LB;p$dAp)jq?S*DsrTI7
zoja?nO(lI1wT-)C9D?hu!&PS^0zs3QyTVM)rcvj<c-%s#>xR0^`eHqdy&w&ZRM?<t
zp<L<HuU0bvwL>CbF+c9zXWNfr2F});pQw?CffmNaDOj*(4W(|Dkyo_5!bWlHF3+CI
zkSg_?kK#B=RAZ!c?0k6K_S>Q0tC)T2?^7l`(rcGLTnAsy5BAx`EzPyxj_HEh1Gq`q
zw>9^cFWhHiFY=6@wd~OFIY?>n?VUt;;Z74A;pLSZ>Nr{+Z{NWbt)-WfY(u$qBtDco
zmd_ff6EZmFz|QwvS!OS;P}_e0+VaL8ry+^N*o$|WjRqlj<pI~V(cQ$tLo-%nPH(()
zsa;Q$X-$ND#(m#&jZa-J66Y4~%x{^*zzJzdqQ1uzXaz`tZKK*A?}WeGU*Gjb2v-i&
z=CHt~ySbQy=n@%Hx?!V;=`FkAcCqZu;qn<)09O(gnvR{_+JzY^n7@L`jQdtdd(rL{
z+9^fbI=-K8nTtS?aB@$8uVAH8RUsFr%Vbb`52KsyRyDey^TGp+^(Z5+u{bY9IoTil
zRV_j^0JYHe|LfN>7)6=-hHpEaRYeewUx)KxhN(n4l!QtvH;6q&EHefVeRW7h&B(9$
zJ}Lq1?^=#{PW-+0=03_5t2eTNJ9K&9rpStnA!rkZKM)F9SJl0+kct~g5|YGkK}Au_
zv~kzjr%(2IoGjd;yuL<|B{D4NcweahmbTo)6oHX=-cw_0+$>_wU@)-{*L*68LHn!b
z$sUkCCT0=NCq%aX4n)eg5gS8@V`K{Wep8_dRq=OHHvV|h8!Ni@dN?s=yqCCChwwrG
zb|MT&&O=r70nM<4;gT26DLPwHxIi>q5NIZw@~Iv9T@*6_CLCz^;}>Mva>)reW8({X
z8)mMZ!q@}@iDfS7g(2H*rAqqV%X(_mZmC!~4PmS%`k&(nF~=~*uY%+W2yjfDzY-OW
z2@j|7k77&CEuWkNu@}l?E5`S+P|pI)OH%Q9F*)=4INn#V(KkU-JcJ=4tY4xhR!rs}
zE4%I!@H@=x;wPc$Kb^%^0w}f!p@!I55PRnGSEj>2pfb8;TgN5BI0LGV#tly;-@uw7
z>e!ch>$Fh;8K>{RvXW=nA6}bzJ2IRV?=_)zX`EJ);@4aiTz61`Q|bJ2pY^p|EJA!d
zVY|2U7M!BQ%c*7B&Utu1{zLdo>(2Tq$xW{ricv081BiJWrW@t!fq`Pdnyqm8DuBH_
zXrD2y`FJC*f2JAqJCZ2PbnBx25@#Y}vDJP09`T$GL#=dTexdQt4|G?71iM*Gf4o4y
zZ1i;QZ~dsQeR$ismOgjOU3lBKC@nQw#OCCEs`5dT%#cXSNXWO`2!*<B#pd%**iobM
z4HtXrC5^`n!vENn;3|FbE7xsxx>pOO&iRyA_Bb_ioMU5Kn4nF(VWYI$Wz`=7!u>C9
ze~8lD9`^8AO1fV2-Qb=|W{FGdAcA$}!{b|!;f<V9g<BQ>i_5+#27#Jfk7WH(JwkFP
z!rtgT@k{XeUmyu2$;HBrGY8FIOMFs)NSAt^1oML#kB6S5>eQ!}0<gn87IG`)botUh
zPT6P%tXJ`{v!Q4dshV84_wBu_h8D*&u)4?J6KP?dxnlL2#OKo4p*R!*dijS5NA<`Y
z8hh2&HFT2yD$kJOL=cJ-m7L~lQAA=E+R07ZD#Cl2!X>?uiT93H`yLv~X;A?#CtH&}
z74YWKrQe8GC2U{zHG?RgliAkjX>t(WHflWw4cuo;NXn)t(|5?$OL`MkLgVAXA`YvH
z*nQ8W&}p>At0}|(8l$!|h5(JsSj`Za>@x||H$60+&2WidBJL{!F%zCaMS$g4avs~u
zFS|6M-A!;XWEPo&?V*3_WtQ;&4kn8?AUSF2j2*qrgNDZ6di~TM50Z<#wz^FOObj$C
zxC6#vT-DHhBQyWKPl{8+=GI(Ho8i?jw3BNi%zlh%+_~+$95og%a%O#t;Ny;MfE$k8
z$jSXBo&%HQbq5*1-wA&ytf1~MPm{<`^Ud63uRmqMD($}WU3M4K#LM)^&sQHr3OxTT
zufHzTWVrFYpGAZc)q|Dm%p1fmhnkI<A3K#EY5Q_=`HBuI&aFG%$oTT^bgcehGH9b}
z*~RuVW91u@N@;^bSw2Zmetu%e+i|$$%jp%3YR|pCmh@#Cb<cLTx<B7dId*MQWqR_9
ztL~G~`@Mr64?FzrbML)d<8GKMf3oG_sN1VLn=OJTMV4=1sV?2Zo8|UIr)+9qc5>gk
zN<$alEp>OE|CBC}RWHc1V79p6Zc!1&Gka9=V@G9|^I_Ll)cr-IM$oqMHC>qf1Zy+}
zziT)eoDa`+N^G-!RT+keYsISD8+W~^w3JC+tkWt+oa>iNK)QY?f#*bVNC5;)*u7*T
z9>jkq5Vzx@cVfejmEBoywDMmwB71;l^c>&4qL!)K94NM|vcC82hi=wVQcpqB<kL_4
zclXV;I&%=o^jXnBtY3FgjRV@_a)zaHC$V|)CYZgn^L@+7<S1Jopj0a6X8GD}PA3kG
zR%bq)j39{Bk0<oA#{z!IQsJPK+@OrAJL;_*%}#$ucru|RtXTVD>zq~5k7ct6OZNTn
z8<=$N+FH*sDSnArwOCfX?B}i@j;}B2tefN6c$&7gPj4UV#=AdXY@Po3;rYtv_LAjn
zt|zP3t=M4R#%;)^MzeGO3NMXZ<__Pyb73s4%sC?hE!i4Xib^{#Tb#bDsk_{)bLeY{
z?{zm{qZL6r{15K4R}#9jU&JEsLs^sU<p<8Ux@5Er%4fg0Djk_st!m7fkSsK3PCC<9
zIMVWy=3`@eUaXR&L9&P6C)N#=^e65f{;?%z=)(z<@v*7i{(#rKcP1i?z9g0OR0Jgg
zrpO)wt<HQR>%eYNdmGUsY1;Lm3}DY4`BmF%54w9{04MD>o_zojx=Mu8+bUmQ0j^D<
zGHl0rnP}$Osn1)*y2`?NGQ6b2brj6pbbgEejrw_xcLwMFBc&xSrahGg7y3Nr!-1(!
zyhG=nLLue6`z|qL+ZIr`t9~Jt(sy<9Ie#ehD)dwUsOuC!rYl}|J4ZPgb)i<=^!mrq
z7h<vs9y7rWCa}MzdyT&*-pNzgr}HZ-LQJK2zr>^ND38)jgP&^40-MLCS#gZI6udY<
z=2ffWu1NZvR9^q`+nS|2?mczlo<kOdIm&C=vQJ;@^d#~fsH*Z*1aQpus~&#^_P&oV
z6-zOcx$BMYB8d{_{03PXK7xo;x^PYp3nPqN(k#%togiicHxvgnm&8=3pLUFHgI3aF
zfH!)r#N%_)fj3nm$5K|F{Ij>*`QR&p@*M;=+uzaG{gUb0R+XQ${SazUjoQ1*HLTzD
z0}6pR6fbgQDR>MdKAT=;Xcv5s<UyaJK6X1j8X?LwPD%;m1%8|N2uacWf~XQ-o}gSZ
zqxw%=3sdT>Q22Zn;Y%~;VxD<s&LQxO!ET)&(ogC*$LQN5&B{n^S3&|Hk&BAm9MgNJ
zP7Xq5Mfh`4`q@DdG1%jy!5wyE80Z=L{PN0^@{t;Awns?VTPo^KUybyVW<FLnD)Fa3
zUAMHah}Z0<Q<AM^*2pnZzL75T9{ESBwbS{em-OE^nZgik34XHEtL6BCY8&bOletDB
zIz%f0Qdw)28gG*sC`g1h1p>C1Sq1Y-9~j#43VIDasABp>^rBmRd(#9%j7~B=A>OLx
zgeH$a&i%Jk0P!W(rF|mny`%>MdWz<^|4Bh^RoYBuRnmDLpA}PknG6m{iap7Gpk*|0
z^u)@wb9*w{5p|wBG-&g~iTS=523+rYx9;6m{wy;n2E~2OA?Ov4#1okY#Q(Iu>E=g?
zC5|Xnuvd^7ppd8L<P>z}m_Psj`+9(;(DE_7a_jfl{tVJSe~HG8ds=e3+i@emkIjY(
z_QUO8uNXdg@5A2{?<bQp-dh)IX!mv$saU&3vog$B4-U!W-8IR@#`R%>sX7z#RrtF&
zclo}*JOE%yIDGE_-8cJbv+wdy>Ue5084b!ew<oWzK1Z?Yp9Pc(s=cwP5$_%!j(St!
zdRreh4*JPsJ(b*NH-`jRMTP{tn7JVTsEf<)XlTB=D@tBH0Sc_tsUHZB9$=MKN-F2#
zQSymvOpy#jqa{M&oyCBS{i-ZHo#DhaW0Z_9x=S0o0laDpwG!_HeYE!IK4%;drrh1I
zsWraW_&CeU!Rp=rT6JixoqBrn{r|KwwIwkqF;~K;a{&>iI$0%=_z&cF_DO1xKKRG&
zj~#ZOX+4ah*Qv7+KvB3<`m-NA(4hYC3%bsV0GandD&<Hn{=VTBMS_;cdLcb+h%@pd
zzY}S^Ik5d;b6^Ae-HbWc1@O~+=u4|Y?#o>9a<7t@Gub(TzMfVTVD!;XdoU<&$bKM6
zp^{g!w$m(Qkjd_3qx}uNeK9F6_OUT<*assvoQ@qSH=W#CmoQ^@l19Qks>GSC@r!Lu
zyT>40`@3Xs%BtP>d6e=>xEx%?vd$IXm)~mG@}#VanzibTg3dbkx`*|;>ll-Iq54df
zbjC{BQD?BY)!i$)?B~yC5|mXm;_7#4wG=v;q>|k0CR)B6jC3|`TxA~9C1WLRs^fe{
zP;#1uKh}LmkKRuA2=V?j{^wscir=z}gv~uLtjLmI=KLW4?nAwnhv|$$)^peP>$aHF
z?6?=7T6$6U;nco$5Ur8|8k6@yzAx=l<eUC;70V~h53#T-DuguGUy2h%NV<zd+7C5j
z!s-p+f;9YuXZVq_>sSBG8E8n(P+wDgmJ<g40dj_6G1>-eGb6u|*M_@mq9spMh6-*D
ztZRSG`&e{TL7_i8lt|%|aRO)CV`s+_Y<nx?Uha})ayycXoIZtPZtu8vCJe=}>Ct<m
z6T#=E`s)u8E^4qT&(?oW5QO<w;Y@$B-+1q*jF6Jo&u9<#gA8!Hr_TMLdpGx{;Iy>s
z$?@gOHlz!`Tt-v9&u!%@kpbQNUC!43%zb!zTz1@o`D1?VSMzU|yq0rky7xI|&l)nw
z(rNGNO%S;A+^Kq_{p+zyWlfjGFEd9yU#YvbZ&!RCkM^G2y5RedWDKsAeEgF0Xn$dh
z>dwo%59g-df1qu0C9<`wUh24vLM)3$qnqyQFQ-2xpNq_i#zFXbvHHV|D)aF+uevC4
zij_wD5k$X@d8+ivEl2#IZz#XD(Bp!;!4J<LKkmc0n&n<uq3Mftz)j{Wx}F~l)*{dY
zn43KbD3acybqfn>;1cwb&aMQ<$eZ*`NTQ<qOk2g2a<Wt+FVg;&{pxuZ?1B%XKZ)8=
z>4@R8$G#;oUCbrkBU38fj^AD`a}hn3{IKEPEo2$zp51AlAm|1}&^_ouH>u@z04hKz
zAUctg$;H(L!+CFpc(aCBo`;~`a@{v32&ltR{augk%}rPW6cpYo4cmfaQHe~Ky};JH
zSWfk<c+c-ZK(>S+WdggP9rr|Nr+clb)A$Fq2fI1Ue1m$%t|lhDl<!S2zt>;)zC%Hu
zt1jmSrXFrbZQ;h=A1L|@kn~y3Ih}G3vT4);QE=G6GEi0!1r0~dbnRXBTa(4Z(y4|j
zu1iR~DMdkQ(HHzrSRoJN$oS=x+68E_aS#n@e3vr%vnJ4B&N=^1#_BQ2Pu{wY*Ocrq
z`uUA3b*)wUj>C4d8Sgwa+&6ySB5tb)zDWWYcVvQGQAf{lf_fiwM|-!zgr+T7%X<xr
zMAMTJ>bB><XTC%Knbd1p99<#&`ChrCrX(h+O(U>*jk5B!kG-TbQjv#ZV2)Hw8<FLz
zB7oi_hAec2$m!_@vgsQbT62w_Q=o457}MFBgbdzE4y<dkQ+9NFmKZ-}H`3AeHj_V`
zxiu)jezgT7fSbS4`C>XO`X2q^s;dOE;2>mY81s|vb{>VxSWg89ylv>$!4`!cYK^@R
zD=<N2*A(1)w-y7Jj3;QKcUM%LOYY(F$@RokdxFU5%`RhQVELP;Id1&^U{)1DAf^i>
z{I24$`J2KHG^my_p+r_o$z9zkaM~>Kd}+)UFV7g2T7KSW-Qb}2JUm-ppB9V#mcrFN
ztPyRiqTsgHkW9GU(`l#WXj?t}ka7~BXKe_n{WTs_DefIv9J@)>+#fB}HW`EC%O{Nw
zu?!!|`JHb6PX&vukDl|Ncn}PTQ~TT{B1l8z>&6GUzk3I%KmCdmRSvg7P#L64o2p-w
zM-<KgGh==9q*_%9!*TnwYG-M(5PlMw_XPMoTL@3x`8hCg5?fMJ&1Zv-`mZ9<R%H9V
z@wG{qA)a4Ss`0pdP-IV1YDily7~!!eUr<g23q<f7GeL!8{fQ1!cgQ>~bk*Pe0&OVz
zh4{CZu)3KgYPpeCghRmcRgAQ2w)^b9boe;sBs}NYCCC5$IZ9V9^Z_#q-(0-Sg?OS@
zufJGw%3|lghaS;iKFTg3H}4QQt`6T$8w0j1vqPUKT<ngIgxaf=e-*4!8PL*X!Y+i)
z??N4Lw@N4_acr_7kZ9fx4u?|3>l<1{&cBtyj8&b|J$CJmYAvZ<8S)D8&4r?8+nbbE
z%~TJ~2fK;~+kPzsbY5?hD#L)K;qG8m86Km`FgBN5|NA^57z@NFg!XG<Cbn@wNrJ3V
z-5o~Af3N5dxLM09q^}ziCj5899%Uf{ICDP1cek++Ngh{RI$E}sfID(DoaU0F8JYpc
ztT*nakk51_g2){R$4)8Ea`;JY!-(ahiV+t$bd59!!WMOn+B0;6IOSnvS|sOpd^!xQ
zWeF0li41OZKa=;n#6AaLNyxt0w#@lBZ5*)Gr<CU4wf=jX0Gq@oHv0qviTH(W=S*&f
zu=DIy<NLOnY<H7&qGe?@jIPAkay4LOS*|avEN+ex9ztFrh1$?=b_%6WZytqCU#zWJ
z?G_8z9YqM|Oc5Y`FIraAN<@YScsy>-ngTk(Up@5$W_j1Y;#wFxuyF9H{SN9pcMx2i
z{GM-l0R)NyarDP6OV<2hbC)b-!Y6E(25dvoMu;-WwfNW}w}<SO`l5G}D|G&@rjlN-
z^6_Q_PsWW7g+@Lmk9U%T>A6I6-!G*V<rQue+58m!LTLrF%n*LlHM(L>h#qP;44QIz
zB^}ln%(F&`<{eH-P~Z`7;8I9wov7T59&CLRuCMT|1X%R>iWvSgjRppvq5b2}n^e%j
zt15Wyc5$T9==Uvx;_eZAf@v2oQKvlG^%O^=e7l$QZL8RAd-U}7^M%;%xm#u_Fm*)b
z#kW`@-V4VkWqHtU(|rub+4D)p?9{`O7($0Ir&d<@r3C;>`k|T|i_N%Z<0``j;ZuIi
z{H_rhqqgt?<|jYrfAAempL8&7sh8RLXMKqHsuP{VMX!k2v=y>@7yV82*VKmji>ODy
zdh)717#Jk5-qGXvZssrD&P8jPnglE<oYR6U&L^9cu9~X9b+{%%_%TL=5;A@1gtqQx
zmYTi~e)~dUQT|Q`1|Dtk>{DdOxxW++Xc`pdIXH>-Qk<5#_YN}lM(W4(hwci~lg}E_
zNw_`>u$>xd#|0_)Ndv*{gy~eJPUn7g1T@dZ%DDAF(>J>jrO;Ep8SyKmA(|T4X8(}?
z#Uf%A$JsLz+@SOJyQO$}D+($kFdUTtp3my_$4Ggk{yUh6>|E&$+*X?gx}(b=N-sFN
zABMz{Ei3ssvwY915-REB5Pts=VJ%HSeuB3<@zfEhCa?ptO{-3%9DaW5;RU1-hlbt(
zlR_q(7DK$?x?6n`8Iv4FO|hyjibsPLaiF^5>pIB1s@|=Ik{>5Re43v=XxFIVt<F$5
z5o|xiM8%%V&y4?MYI1=2tU_bvVDZUfS0|mP&X4+c*^{|(#>hNOIoYF~1#$;BYKj9Z
z^(zH;<-n~yU)UbDhS<wt*h{IO7XNw8e8RX;@n=LmQ%7j@dwXY~<cZzlPhFp@X3zl5
z)LT1xc_2lMP?Z%*sP`EGQIm)*i{t2Ox_?3gxrh;k4bjyFwRLsfx$}F>Uzbph%^`Os
zXcP;;;{nfBxBHYa%7x}@T;nB+lG~tGax5PPBUI{SxdW5}5@o$8#y~$(LLW};yI75`
zom~tnYY4r}kkAl}Wex%W!n#$PMUO!)d8C0$WYBzVxyK*x$%D7}T-b494L9Di6YnYU
z8CukT+Ao54uDZ*JMnf?fq~fSw3`|f#PoN4xqo-QxLEJ6QH|2wXoI(qL0T+9aX@~52
z>v)jNnd}=eZb3nG6$u38_MH|4=9fOhV>;yk5eC5+_gAwK(QgYbm^pnLemp!}ogVLY
z`2VZZD#mD{h&!D>Fu5e0H#KmxcUjvQ0Udcp6H*mC(uWR{X8ZR081AM|^w9x^-f5wv
zq=YF6G$=qC$hn&x0H~Er(GDO@%v<GJ1Zh7|@FI+qumJDdD119iq>g}v;W2_mc#ZWH
zM(TO<qd309$R6Y?{T;jgzcR8VEl}xYc>L|>^oMm$U0y4zsdjJbqPMEAcZ2`pK9+Vg
z0QK~CFKOp<=3VLl+>?Sv7{r4V81HF$NMj^{X+%0)hWAj*91)Y@ErP?CEXL*VQUBhn
zXJ?rSd0j=E!>9_n!A)BdW+q2Etv)yMK4Eq0p(E3}%xRBI<F>EfC-X-!Xc1<X`R=3w
zxCE=a4orPZl>PSG@4+`nqgIVMYoAF*ybW*gQ=k=&h?sMt(_83{S(7M#KQzy}t)AMu
zx{!>GsJMV=Um0`}wYwV1S4i~+`C7&txbmi@QD9kUVAu^0u=9SiU&$OM99mU2-;a`@
zxx|0G8R`iBef{UmlaFgDDgH^-Qm|W<$sic6Q^xR*_xIli_6Bb5vG)^Pf;WhTGJoOC
z>SXKxJMtqIDuEE&f|UaZLiRcVv=-7Vaz_?jkE}95Qr@6<dcsGbBYpX=pE|G7I{&BR
zxxNx%>-&35DQI0elF^A_E`Ei7gteZs#|dcdAL)`WBATQ>EyUz}D8@AScN+;$azhBu
z8aK6bn1K1nkb(La!M~?ok`LhW9ose~be0sc*8(xP>$h_xf3hH~nPCT4WJt7!fz^y!
z`FkPoC)sua)huFpcgurR)!a(GS%32Ii~TvC!)uzf<lVnrjlk%JJ!Eu4d~6EWp9Y{b
z8TR~3+m5D67V;&&di^6&hZj-$i;nsS@ocf*)$c=PLK=@TqN7Rw=T}T?5LnCCf>>Ro
zr&*6NGRdyJ@%W$1X^nJ~@3A~Dwq%YQLut-StGoPp+Dg>0+#O8`>5fG7pp`?f|BHjA
z=GH&IVw6OQ2A>28HUG<Y>?*5^x5R>o3)ytIuoEQoQPwz8_mIe29EDnN)xgV(zl|aO
z<VAP_Hr!-w2Z*(k*;o9Z9s5DLK<`Lhj&y+*>T-=Te?CUJBA!%OGb{&du?{!;`|GNQ
zK<xeg6=8zPxD}Y#;W)2jLW5Vame5@J$0uk<)5xm~i+{pzf5D6%Q3K~o&VTN_Cll$x
zpUOPa7fgVC!I0=4;63=C=a6O7#P7`0$S~UkL>0B$8=U5slzzVv$QXp<OteVJg(Wmy
zs+u^=2Ecpkz_A_M@ccrnD#59iS9{0s_HACd$X1+x5id$|X7)>lVvLV$5~dW@h+3|l
z=kNM^Qve$h1Z~eq&lrqfeIqslO%}qwo!q?`D<vGqV>S5%yi=J6r(q6_O0uB+uSpnW
z6+l_S*S~eXGK)QHAxAQ!EM*7TL0QY?uwMM|cR_%oD7c1{#854a5Tn%O#Z&V=FCQVI
z?3X(K_8v(c53T&8TIVSng`8`>8Os5CTG^5d*cs^-lcOX+Q#{^<;$NV&%cSx6v`rt{
za@z<~hBR|U@096M4P|pq&tDnE``8wXkPj{f^E;=a*Ke>_KNae1zqh88eB?{JND~ce
z7G$~-zqGtGR^5w%#{95G2;VUha@ax}J$9N#rh1pdkT#Kb9|~n?J*`F1&8R~kBKa^?
zS&I|_?PNt_JcLcLrqZ9s<2oT)ynlE)1=wlFL$Ez&g+4yyRe2ciKoy+~7a0QvTNz~;
z^*itUnBi!-A%v6IX-DC5X4Qg&Z6ICtP2JNGJcQiqz`#_<Q+t9B;{U&2#)c0ciE>!|
z4=6NbD)v{^6Gx8VIoNc(&@)v6T*?G&70O9Y4pdwPEI^NpzFfi@d1?LpCW-=c8**rG
z@ki?U4Jq{=xiKE1{ND{Bu-s&pgO9KwB4k4vj~&HmUa}#isq^@q`<tG>N@H4&k~!I6
zc(pNOFL<@*kD_8c*^4V5tYYT(LJfiFUc*$*KOUc|G!-AQY2S4edm#ntSd)mpygc1#
zbP4i!V`NXJlo9k?%8CH{CZK>=d_fU%5)fK|o>g`(MmT86rB|a02~dwk^8XLp)5Y%v
zJdO&PbB-&*KHU5e1*#(i4ILQs&t~B7f<W3T$2*gna}e}v4&qj_gVYcjd9G0DQ%mOg
z6GxV**cGO0&jT~S_qK&U1g;+QNNhPdtq{B?Ai3fmuG{yMOs`fzeQ#@^g2HD4Q5|}S
z@|>PJ5?z;*Cp5Lp`j~<O1ABl734!2DL>)%SSB9K0;3ZKRRN@YxnXO>a{xDSvqG6_%
zJ8SkqynvvA$q4F?=xa~f!nqS+cZ+!UWYoTF+f2fg5QO(ui`3b%3ZK#+e+HrHF_ihM
z5W~n_E&)JM2!~(T0fg(&xtVbVR}BWLp&OM%FAvvRg1Fn2tcJH;$XU^co0{t5TmK(7
zwHJn$&sXwV8r&2Qu?caJDV!9=^x8mWik|Dx25n=*qfSzWf<Y7iRGhBZ#FhYQS{;IF
z2LxlonkQ}0A=eK<<WzNeoy>NUD3OeT$o4aI#Cb_}tAH2R<10CW&cM|>U9r$emP2ey
zAD96o2e+dl1F9WXnb(Y0y}2V{pGNJN9jJBE?0Za}#$`uLdF4WfFWV=O1ioV*Yxog>
z)314-9SSHx1psE}$4J<syt~o+k$L?A(R{P#mzFyeK8o9aa?_H8vowaH8x=1{9P(p5
zTf6SNi0U}YUTZc^?KG5)gn?21z+$k%OS*g8g7yL!nPzDdtH!=N&hy13;XYg=H2*Uj
z^7=($AD&)ILEXEw2#J>CXZBCmJbzNwN(>5+VqwtPeBUJ>qSSfYs+seN5}0Z5rZwgB
z;BIg-MHL$BtwdYPn1kavTFqJYQGI{*4bx=9&kpV*?L*hm=PR3&&U6+Y1FqLB@9Q{P
zD65%oC?p587hTL0_q9J$lkO$AX^V*<5lIN<J63rn=}|4fPa#Akhp*vB6WwKK$+Pat
z@sDOd(~B<J?)dL#I`Zp>>H{30G_;<+-2VA;rgGd5x=A$>jB7wimMiE1D+DwXhvPY@
zEDi@hA{=<`CFSjTGSvSXXloS`MR4-+uwuj(_2eohXQ`N&reE{A=3Ow1vPj4YXCLQn
zpk82Z;O&-}dj?d~T3zUoh5`*2mGuSZTg+=0PxT!yX6Ux?n#H>1g=>ByO5+$lePzoq
za5W;&#hUx2B?A4_8xuL&v`N45)S>6uc{kVdn;v?}tCydp)PVSXEejvP56r8yLxqLk
zsyta;DN)NqrwYR$YOe2X&ifi3BVpx&TfKDIigHY9H(hTbQx(j6I;p9lBpw5eRY)5#
z>7RxGdsdG*o7}pY<t%?ICJ#68=0XNt0Py=T)V6P|ABuNEr%@0o-=k+qjlKgmcv9mA
z42=~au>7ZnClFYhB^_+E7i${lb!E@u(YrD)D&#dmthB-Y;;nV_-Wao5Vl=cqUH#Jy
zd?7*VlPTJ-2vQxvIN9`BTJAgTXqcyZn;p=KF<JEhxdcGgBzmXeAyeSuMHS_iX%evE
z*TWJKHRyq#`b@J`-V}txxcGs&=?{JmF({Bm4BWSqNV_Ris0c(jH-W<RPiSU`1_NDo
zqz^?{U7W-x{%Ha_Ida7KHK3-kZU$E=5%|tHk5}fBP35}k7)UP?qC0(E<x_$|v#>=2
zCmBMKu7;w)`CB0&I~Uj$@sBKIu)FRA`#&tYCJhjT*ryr%l#^Giv*!81U6<e}gxECY
zI_6mtn}jCvr4>>){^?5`WpT1Fli--#60dwj>YG|R5BAcIMtiFBCvs$Lk#rDf%Ogm1
zQ_efaA|_qrv@4+{UQI{%^-0%#9QLwgA4)FH*|5fe;K=br_MBVR_=wx~TyJasHiqIv
zFGdFoYn_osRlOluBi8@(mGdBRmSso`DoUUdRH*NbR;}16IaU`lqh4gA{-}t&QNltM
z_AAPF-&DuGXX)Hj|J)m;tD^9!-l7O|A5Q1=Z{G*Io0Ah+NA$c@51Mj~v}V6_Eo*Pq
zpm2%Wv;e35RidQJoIeCtX`~G_j(l3sWmr532Ixgp(ma)8L6trR44hXNJ|EsgIme4}
zm&<et?JRO`8|GWzW6=&e^J;aw_^DE=0SLgRskGEvl79U4&(hx~8VRA`P=X<W<-)3o
zjH#8XD;&W1ijK4k(<#d5pM1I6ffkBy`;`Y*TZS=&Vz-vH_ec5}{Hglr$>W792KuR_
zYI~8ti^YHM@0UjxJ?A197gxAvG?Lv)NP!Pcz$=Ciyd1hxvdT+Zea!+Cx`RCW8u`lW
zCx~;Ae(BAcnY{h$0xfv4+_Qra54;WHd_j4wLm={@;9|;ViXV)qvw4TxR&09Ct+!vn
zrT&Ovj;)RQj2$U>TO@00#u87q#!}pQB89l1;<+)EqB~F$yzZ;4g$|RXV6(F0Xf>|-
zJY=S=Gi@Af$AgC|#g7>IsaGA#&=o5Z%0`L(VG<s`edgcm%5s*32&NC&q_?6_L>u7!
zJIocvIZRtEn^>MdFL<<Guixf^x+dM9(c$R-s3qVjdssI4c7ta)B8qly;AQqS3Ji#1
zu5NEmWA3S5AN~jafT@*lnT2LT2r4GqJO9zfAntG<OWfNefwP$s{i6Cjx|;N@AC5<B
z0U+h(i0P_rYS-7d=As2wZ^Uh-B$P$iGh=WSH`MJNz(wOdmKIDvm3BQbuy5M(R9@2H
zW6VdGzfhibp_t9bFhn6VA+V1r6{l@lRrVT9^Sx`hO$=nrKyTigbK~JlY8EmCNE9Jd
zNdJ)_hDdn^TP5-?iUX#+l9Eh{BK;fKKqc^rdjbCmKzNrhRK*;1pG*R1JTL%`=m}=1
zX%_mWa5VRq@0Q>Sd2|$<wBm0glJx=ZfhQMUFINJbB`1v4;@!05Sju5`Czv_atv$8T
zC&^r%iyb!C)Z~r))Sk~*M`c$K6Y7=E7~M;5Pmw2$bXA=^p}ot{zV5zd24)&qkb9O&
z&knNa+x-~B)ooo#K}cLA)hQ*Jl+Cx$P-Idc5o6bEG4n+laD&U5Oj-o!RyCgJJP<OO
z_LP=c)z|2w>8=Pa0wJf{65g3^b=Rl3QTd#bl9As@``XHq7`$l(0+`j_E2K#eo$qLh
zQ6nbD{7NovK=bO%;~#>24EJB$qqWv2TtHr?!oj``%SmWXM?v8Km<F+T8^YvYxfShb
z6bvM0hw>453Z&<vwGvl8AvxorWOW_g?}VsuxKKSpR0$?bf{;hxGPy~lhhCT*<&^eP
zUZjFyq<fD+THApFOV+>j!zcrt)JcbF<pm3XCLIU}`l#HV-s6V!l&&#Y$M-n*p?`h*
zzj<}8E*{!wo%KDluc_r<=Cf>xtp#h#h&&-Q)J_a>o7oxp`7$Pq-}{B{k$pwLwKjRd
zc>W)>7;fh0o)m(<(t{FAzN?<lI1V@FoJL$zAVe|vdEe(EV>O*kA^jiYSl>x-01wVU
z&&32`o2lFT(i;ugrxhd(fg~ifUw`xj+bn8BMoU;(W|EN_)e5{n9-YfK&mDc7ETy%s
zFWBVxLA1_Mofbgyn(Gj$!P1)yB9ueK3O$Q=Squ52|8bJ}@s&+IVI!Ge7^+Uzg_;;U
zt;py_#+nhh(H<TSbl_8xp~p-h&jrwZ9X|bUV{-&YvG)N_B4Xs9h>>SA>nwJm++`@c
z%)+ucop<&}neL`N=ZKJ40BUQxN|R<}ycVX#`{VtUNkLHZJ6cKsa9S1cEPP;iMfImh
zG&dBxxr9qO<vim375`1<T_BV>0?_qRh-k@>xVCaqPBJw29Kv9_=-pMi6W1;xEXdwH
zTn8AKGS+Q)*m~Vp1I+4bizVb;pyzt3!s==>`3as9J&c#?Z?gWnOqi>0n?uAed{l&1
z%+N!Gz)HX~4V_&q3o8;xa_V}HXJI65o!r1w7^Z%bRD2AWrisoM@ahgC!opb!hCThX
z)%EV|CMP`S^HGvkKE}Tt;7(YAo+H#Lj>N_;5e4|RLBN_!FB$EAd9XM${^i8*n-fq?
zFm@mPiu@>?w+ND<Y^Ylcd+|0?36p{`dian!jL?HTjxb)=pBLrib}sBt7|wiq+3S*!
zk<OwwZ<>A7KPNUi>Qo_e6vU|zizLXI@E=wfLLT^5!paMjh!8SfLn7FGh73D<%2J#|
z2=thk7A=J?z&v{5td$9o*8v0LWaJp=f|ck5&9@6=A-iP>$-l`rrj{c7gl!#w0O1(S
z;2Ew8bOi*VgtB|TIs%{KFW=#Ldtl<T#`~Mh!VH{KpVlmPc2bG`d3360Dv-b|-(q%R
zgN2GdF4Z7K_p`Z*l#&=cc{tOCS^jMlE>GN>4|~j;EyRJS(70Tmh1q(70m`?hSmWH`
zJToaaIcmYL%^>yDI{G&rL;qYs?*_?C-wwnzb&BX($SXW2W#OlFy7g>#K-SQS3`9$(
zJ~dvn{O^J-QjUjLF$tYn`KUj5LgJDgFrM<u3?QJyE)jCfc`&_f_0}(kk{?BY=9VkL
zKG@ypGf*xM-7T7m_|2r0%>1;x-3IKI3J@ML)EJ`T8*k{SKzl=#HYCpmO~_Y&x0|_r
zVzij;>p&5px=FwA2f!*7XrP;!1OPiGW*ofLW#z9*6<Aqmj0Xa0g60kTZJN%Yz8xVz
zY5J||S6S=U`Z)B~s-Xf%jmmk5CD3?DhwhGDD4rM-x<B&=WfYaW7DpcI;)ssW5Qw~l
zT|XPLeU-dX$&vh*tes}4qnNg5kUaczzngZlZHuiBBk!rOc#aQGFlxipnOAaGR@_o4
z36n7O8v9qwX01gpl1FkJL@gRhDL)X35@%v!?W{b@dd17mq8pWx*Ip3XB-->qU(%vA
z>z;YZZJ|wa6F7nukb~t%Ss^}Jf9e@TJ~floSwtdgAx~ISE-@RUuF#sjVU1%0+ZMrb
zl<anwL2W6of88g@J(ee|=x%hyd{I@IpK}*@4W8r+V2&oGvbe(jhWe*sp8eQa#x<0S
zBs;c8q%Uy%)|>%GvhlQ@M3GG5Y$`v+kM*w&AN3!8EJ2}mgPyB8(=6)3X&3oJTj#!p
zsWk`Ec}YVuAyCF`R{I(zxrHv0oxE-VjKX(Xj~5u2Ia0#e=ccBD5cD)9NeAldP3)AD
zN98YT+dYiJ?SK0=QOm+YcRJ2XrncURG<j<2EMJUUmDD0o8}30~rvy=lp#ZCiA8#R6
z#IP};(NjFEsafaQn<q{e2ns#-&UO1}ThdP9{tgPHXyYw;V_VL2!)FF)QR|YlVoPek
zeF?R&yuZ-gHf@Rrzoj)?UA&i(d9U?d<^V;Hri*BCL|nWRv47bmAcEAy(H*b0J4uFh
zeSd1sFCo2YQN8m2aPvl!OF$>6IsABbkhw2z2h+rfgUy-NQQvZG^)FwXkoWmPq3|>L
zLO2wx>!r$xUT561q3lqAD`)xRh<ralMP_X<q|6Uac>JRU7=*}_Q(tW~7dXntcYV~l
zreBPtT&|D*qL74h9ZEx|c?L4CMLqkWRFysSVxhE#CaB0mchMCFl?qzyP&1&t&e+v?
zG=ZAPV94rH+A(fqu{orW%B!7qH#|vynF_c<VWICE=AtuBKCVQfnF#lBBzekq!<HTP
z4$T8{gWsk;hvE3#p}PVvWYIn_m#Z4P%<c;7cbhoh7cUojGqQ(h_Ws;r%X;&o76SLD
z|LwcXv1g&@NV=jV)3OAyNQ{UY74G1t7t2pV&>wej)u3j9tx0{i-B)j)y#1dGYM&{3
zubd>bm+I7v@vxIKh)nrcuHBO5F%aMTX-BZAWBH~R+Kr;`kbT;TwaNF*HA^J*X;-j%
zMlW3AX^4&R#oF&BfoBUco<enHuZ6!^@YY=%@Yb$PZ8B&0U^&%+U+iwv2!h&;=UT0A
zfQ;~*{D3H1g%tB@f|w=k-+pP%I2mck*;^CYviey1%Qb^>FULO>!-DTl-~-DR7@pYh
zqLU&p+v+5f7*Bm8u_dSW+Bw0;oeq)aOZ*?Xf_#s*zrncGJ*38s%DXSbSXl=@>bEX|
zNKoL!mMXrR^F!vaW3-2cx;aTRyMec)DAkT)%$}I-`H&O;!2muWZbp7wIgbH+&xJl7
z(3rjujS0y*USQ>(g!Qp~F*^k}2#~(V3s>gwZs52G<h{O&RsSt|y#57MVrthmoP&KK
zBRKZdyk9+3U++%PaS|b6LIi0SU079sMt9xXwR|Xsq#T)!YrJTopKd~8Cj=8#jjo0E
z>HY^6VL&Cxuo*MZRt3~IpR|9(lvn*HLEB0Ko0sZSM}nwPtlrI3=vV6;;p16Gi8Up&
zVn0z^buQ@EbWo3*7+)nf0r!yg0u_siV)9!b2M?meM$pOG_wlggBFiZ>W`{27WM!z0
zJ3(!xiVQtVhBnoXQ!x|4l#}`P0!7;v;V50Fz05s6@?X0!tG0FgwH=nT>k85j4Mn*X
ze=e;oTN2iGmqly965=!@18Ag|)|y6m%3fO4|4>tcuLh;oz_Z;!8Zx|O_(G18ux%ea
zk?N8LSct#YGMv4#WpG*|<Iaj7+BZeA6cj!Yb8GHYNK6o<=#85*eoCPoYHBHHbuxLn
z;q%_Gh3D+EStL-DWRj(P)Mo(@Qa+UAngo?9|Lol?-A9(P%wS3*5=%KL##HE`?yT#B
zh7tgCYyeHD<$a=I0wWzI%}5@Oi@qCo9e^0exvi&A>$fHD=S}VUH7SIL8Qp_zFU0{q
z&aDw609np3Qlp=lVET#45g~u(<VvWv51-X1S<1%?AGCGXF4{nMgb|gW&6W&9)o236
zIJHc+cmIJR?L!Z>x9_dSTprQ4It}5DeQg%(6b2jSGxL~FIG9HwZBTwIhxuo<H#FrU
z=W|}Z*SbEbEXN^cdsbPcncKN{#riPVr`z*Ea)iLXs)muDzM?gkVA-l2mlx%==Li4}
z(|~tl<aH#rwzDbW55OL7O(b^KebC0^Q5f<M58%`WOOqc_IUbpp>OoC_U{r=bf#TSy
zs+8DhjX=Ljz0M~D-Iq%EpPG9#>-ji=n)TJWjUpE*pOUd3%+k&ex9qi+ML^GmNPvN*
zzzlLi4l^LIB=l)J$S<1Yk}|Tb`QsLZIZgUX>kqB^&$h;rZEf_OQpzPaM3kveqt-Fv
zhG*IG#k`m?`eT~om-hscs0jc2KJFL>rw7A*-6`ysKDNvT@_;Ol&hb)%wi1H_SqX@l
z{4#IcEZ%|kNB4OzKmf)T;|N<#lJ{isJT1%l7aJr)I-+IK8jT1Ve=DE<?ubpJ+0Ga1
z@I)g*4M|NrFkH5;8dX8W-?@}Utq)1n<Eoz+@JmQ{NGf{eyrj{P2_z->Op=;^tc~U;
zdgnK-MY76npKe=!dXWabl^l%^waEhrvV0KDpV-K|IFfVWKDwfN6`~>emZydwTk0V|
zh!HE?KoH(3Az3KR27#=qScLJ&v0uG)cmNR|Q1lM7x=kDjBDh}mcPBAukqBtd*xT@<
z4tLu7sPCHpe07W|PKp&yihg_o+n+{Y^d3%%%#htyv2dIeGn|zE^kf8s{_GU!?-bqI
zBvvU@gbJZB@O&<B3g}r|8zHLi^!^5q>IO4$iB+$0r#{Z5>8x8f)b_+Fmk2d-zC&9_
zK+Li!tMRSdls*AdpoDpjg+Jl&B5G-rbh~7R9V5^)et@2={q4X@f7ty0lX#t{>{Ti<
ze+Wpi4VsfspGt%#54%Mn9c%0O>x1hmz=5?I`)5I6(WDs-q8&udhlr4)8}KW2YY$$t
zc{fqv#U-a+H&lN&9tGa3JvM{d?A|$x5{w$8V{K8^cm|+7HS{d4f*BVipBF2PGXJxb
z9i^~b7jAi+moU`p@~SeS16my=E@IrXrE}(&aE0sekGi+vCHi8+KYxCUZ;(3D^ekA@
zM$CdyUf)*9=ce+q^}Gf8aS0DZ@2ToP(~AgZ)YKEA<6G0fskGgKKH{iP!`*?I*w3@k
zPe)|>#J@>>V4ie(JNPvX5ua6N_tVL@$&Y4-K2Ie@_JFZ1+83s0dm~*t`^85f-_<bk
z_e`CkQ}6nM55AH;$U1hP#l5$M6=LqLefIPF`TC2qz_}3Ea31O#ZpL7*e$?}pi+G02
zR?nhp_W?=@TF)hvZ2s4%vyv)b2GzZ0U*e_5Ug9EpVy7lGCEJpvRI*DJKcYZB63-B;
zaP@W;&)#~eezdF1?$CP5s#24lOapL2{=1b!1$d#N5e&qtt}m-JA7Ay6K`9Fhi#$ET
zhlR!s8U4;tNs8fT4G4%GV$G1WW+xCVQs3BJcLVqCAH9pCUHd;2s=OO$KMs45zNG(_
zNUc8!VVobXBTQzTM?Kqk!Q`!5*cYDBDSU}l+7qd7w^%4>%hymmQMhn<oBgk6U;Ic2
z`G~~0)9DT>mrLW-mt*#vb38=7l&UnQgygkX_F~}4*qL-PpdKw*yBV<U8SDp+AyxVC
zfM4HB_L9#9B9w&60(FH9B5NXtYg;eR)Mh8R2Q_7v5jmk}bGBg7Kd|XO21+7UI5m=E
zVHMdj>5bGvPMF2m99%+CAn!S4el8$^_21PcPsp}|e1yqN`tvR3zcj2LIK3XPRjX6P
zj<hlyuD?}S*A7($#nZD7o?=>yF(hFLksv5}gVVz^Ablmm{c#Tyv@r7jYkD4tI4VKU
z;sdZiwApdf?laqw@r(76Kok|2U6<G{UPgyLkR&d;Gxd|WSKBc+;u7Ny1E~{lY#*y^
zS=@Fevh6o>YwzGWIyiAM_E=~#;l#91rSzIU+W3@&#%R_;OGgpYOzyPoZ2j`$WgHvD
z{K;{;jwP28ZF90t!IB5nUXEG$+5PHjEO~GVoH!9Y2}L2Q0_0k`j0sZ4CY~r_4}hYn
zhwojYYu~$V^Tlh#iJ;UGfKXcrd(w-6$va2#qJ%$Hst8UHqGzeHU(jb6;${i7@zeGb
z527TiN-yMQ*o6cw9^rBhN^Py)fE%Xd#8y*{vfgIa<-59sa%{=c?_KvXtIHt)LH%0s
z$IMQg6n;{|^{W}?7vKO@R?@2H&y{@WWLp`c4zo@$7UeRp{m;#yJwcJ4`raq<726Uo
zMzTW7c|zFtqgKiBniy7W#2FWO8jtdVnsN?0$DW0Uz^$V_;3;;DYOT8FttBsFx~`1f
zO>T{@vHtVYq1}Ts0T90EiltNuY)jw9^X^SuO-(_2eX|$JgJd`pCq$jDF>XIBOK@+W
z3Z+^13u-~%df<yWU*lQMK!qeprTq)XOjd{i=c`IKRS4c-h+2K+I%mz!FuN5C8@;$r
zqs&X2kIF>uETmaV#mRynRNBL3M@2z9%It+wS|tR#$V@KL2X8*qyRH9vG*wd`10$RO
zzp$5zsJ7dDqbp#b_6Y65SJE!S!kk(X!WVRvyx`H%yF4}0N(%pZ!KGXAg1s-5;;XPP
znsi;#5l0*Mn~{$g`;-s?1EP;epbI#x!uD&tfBoa@Yw(=e-O3pi`<LLdusXgD`fxRY
zY2KPGzKW;kKAhdLz9;ooxLi)ybG4X)yFBD0!UttL-T%onGz81;W)_b~$@XOnmhSp4
zeC@6ks$Z$OJhOPN$F7vVH^qJ%SN4xDmf<OhsO)1<oWhN`!hk0p<-NO%{C$pG7up3u
zsNRNKWwi!z*)igPkq0ilf@^TcD6xT)9pAc_PO4wYR-s|de{O$C7!K%+ZF|2cE|(mo
zqaSy@@$2<?@GB*tv6{fWy%JV~d=C)`)<pNoU)X85hZIlIE-7ET`7b~D{lAR)PAoh9
z)tIXvWU(}~!t0#3jO>meX8za0ETPm4#S;Zb>#o4L0%ZDg{a&vmX8f26B?Qi-2kPUm
z4Yfbw?~zd3j0llYP*;>w`Unq?dXzsIq=!Bd4<=EDLp!pL%D>~gl8Q<YYNmID5xR?(
zp{7%XyKRDqcMQrQL{BCGogUjx`<4pG+VL^z+XWP@B5;T~%t8$#buUi+j46_@B|+az
zH|=<aWt|xD6Hb_pZj8Wx>~AJRLVxK2mo*q+E`hNml1Uc5U**1m3%|7rdWbb&K+X$<
zx2ptTmS{dDJtIj0IzJm98>uf3@4jckw8P-cURTWgBf>mHo+kA@QlVV~C^{FUWZg&M
zpaH557A+5=^F1;lHzMThR~@_0kQ6m34YA~bfMO6W-KY`kD~lWJ?fGp_W#`==G4yO*
z7b#%lM9MFZOq~q>@!unJgR>-pFJQuAGX<^UayTcgzcw&0qtv`vf{{3h;Ii1FI(U6q
z5x`i+n07-q{yqfVfPIx{-kD%@6`=yXNnhniz(BQtaSQTY58wIMC5XQM9eQMsE~$#A
z<STd!bnll};aies_^i05A>YSVcoibKybkM^Yx3^+7-Ac^U8Y7ti1Z3-COo)LIWwfS
zJMsCQA#gILl)H3m2gZ>voGHNesoXo@oq$Tzp@|R1O=-te0Ac3X_kBX^suWQ%2nPm*
zjwaQ=f|`Jl(Q=a}YGze{`3!Mh+j2}+)Us*2xcgi_Q7junLGNmhBoE&R(Qb@#Xtz+O
zS6H-2)wrNO;-@w#vMs$Ly?<J&pZW=n#il<m8||nxZe(%nffYEsoOf_iDK(bVV08zm
z(6&}6hG>}GzzF4I&sWi`e)+48k}`m0G`Cd#w_`cRN~C@J;$}JV?xli6iKn8m5M+cf
zTJ}c57J;2ied>U)V+?S+Dx95L%Qy2Jlo#i_WP1S90!SBcS2F3JHQSR(U2?lDXAWJ~
z?L3n0*W62~uvZJM=eB1a&w|eQSqhp%+USt>X@<*mkCr4fUlo3DLr^5RE==wP*{$&h
zxW5zG(9;=;-ddK4<t$Cx8bs=(PN#3ND9lR@rG1ehk!qcW%ymTN#^TXAr3~+&`bE>}
z5;d{bOV&AG`Yac5?LXJpRn1c@Bs1Bs7XL;`7C{BQDuq67y1>%)H3a$+<04P`x_o*{
zM-QJkLIftsq*CqcHnZIuCR4Z~QGTif3F%bFS#gtzlu+BI#E?~o0ez4nIGsc!3Gc7F
zJ*>v*xOJ2Fw&Y)^b-%hDdVw_6H&^YeEx14=EQm<>tIihG4HHqVR}U3`VDPk^>e7}$
zY0C-)!T-hHn}&0_w(Y~Qv_gX^Ny;omRLT@088c;`H5kg6F{DIMNXE=VWtNf*85&4r
zN-`8e6f&j~Dn;*ps-EY5-tB$Y{d~B$?f>Qf;kK={?(+Lx*L7a!xgYzn@B2ZSE1|r6
z7d)U5kcV>(C1U9%x!i}^`bpt4M95s8*ik}zg_YedGaK`~mc}XoZ|@&BV1M<le{oKH
ztHI59D9jX`Ft>}u%L!#y8VU@x-WlyR%rj^nvORMug{ea^p7uKRuMOJYivj25_O{|^
z48ZU?*Wvw~$>n?N=6~J#I3onG(`GiRi>$$hr%iU+5b5{w%P1qMbDC}d#VR$@s!kwv
z)Ma9ARJ<RP$EFu(-1Zcjb`A30;M>6bUSglnw;aKpb=Gmwi=^9Z^2DVYR{snH96rmH
zv@z&P_h`Y+SS$tBGh$CNlcQRXdo?;Fw!OAvWU?lyad4Z2;Y>n5%;BwF<}pB8yyr|V
zKf|mpTnq{82#jOW`;kG(hsi*?m*k0tA~v++!AtZDLwkfrOh&vK_~PnAm<6BahFsL+
zRQkQQ*Ie@ZUi!Q*i&lq*{-YG$s7I_r`@Vd?rT+6bNdA3~6bQ~#nJ-=qOhJhwK6l5Z
zPzuUioQWE_I|lJuX|E7Pd(V=F-|o7E)k7p-wpDo8_vr8*X(6p0^ziRmBQ2O0D%jEy
zO@+SjTs9H@oC(tBmdt(^&w8W<{XHVP%_8B<+rfOV&V4MIAmMT}=mih+7%_xe<&|6m
zc#%*XhU87<m`A-$1$V-mPiWkjdLB77rmH2xhsEahV!A(PZr}H`H|UcvZ!*nr#{7*9
zt>N#+J|vp3>u_QA$I>Oy(-8!=FF}k^5hq@RG~KB*r+u#i|HrGM<Bt<hFJAEp*+-(V
zk0`WdJ7NJaQk2fdsC-Mj3m_|K@kxLwd<KHOMi@C9lR|hWX;=o;lf;_K1`1i0^S~~%
z*28l+0NbVsyVn54NXN#<6Cpz(_Pu2;WCOft7MH+O!dr|1Lb|(qUE4@r37{pbK{PA-
zD<#lSj^?caGJvqrNFVLEa(VNl_VrF=7CWY_pku#+bL*8G_wQ$9L~x&4MMui<rBkKq
zeCnua<lO&wjuf)MbHgLu=^p@{MCbuTRu#!{;E)&hha^^oY+yJSr>xiv?Zxe8H?zu@
zCEZe~X7@1Ge{jm9$9(&da1%k~OvVebsD6xMJrW+`Pc(qzWOC-^d)rKHuaNN?HwQ*Z
zlNrmlr!9ebjx0|n=qa5h!gg!&x4Uf38o?DoYZL{>ZBJYSe##rYTJ`5D*3;uE{>sry
zKs@LiRra|RGPYBydw$ghzb=%5`fJ28djSs34dwt#Q6YFR8HsE50FF`7<TUXyXF9yT
z-z_2u0Z%5Bf@v>}p;<lL&cte5d^~dZiPya+xuOb`>zpA1=n=7~CjG)dql92yuZGZq
zY?R1ok#1{GrL{m3+u@#P4!rlOQ=fjU5A|ypJ9aWq#%R)o+I8tD_xB>-l&-NjatpcN
zL)bSwR<;QPim_0Sc1n_-H7PmoI`s+652NHWw-UR68Zm~2euvWUv#gPz>uKGm%K!`4
z0~T(utkR<cj7_Qn>pBE6Xe3^o10<vv?+NB;bUSpiTw1w{$11qAARC|#HK)L?UJ{)Z
zK*Llh?H(Xwtn_60(p~yfo?l&YAq@Dj0k!vl)3G2tc6n;jrXy1k$x+u<au%`}RXC%D
z5B;2vJD&Up@cEr3Nccl<s~c9lBtgPq-q!$@fslFA3$jsIO+&K1;TjSAB2851{_9z%
zaZ<K#tY#NyefxV&j#AzCJvLh!vy7sWiBv=T*xf7p8)rBGYLqC+7O%|iT@!EZi&V!9
zz%!z=BC~^3Jd5nZBr+H^|KLxREjZF?YPAM&0kmY@QbyO)qj(BnP*FM(ULi;#QiP&O
zXg7-JN$-q$2$V8#HWoVei1AGA>tYIKwEg{+>amq8zLU4JqbFZTnSG|C>^kdmJT_@6
zbf^W$tA?EHhW&Ihyr3nDQ+xmyhdF+Jy(n|y_uS?=Y3)c<ej)S0{D0_z2oG#0h0j6-
z$rmo8`N;0k&bfvZ==M~MLX%|H4>BRSGTQBGP0XDouVWn)B0uUgMRN2yUo6*hVg}#)
zeR5ZHU;%$J{mAcyegePres{#+S+c;~026;Ek(?XK2E~Dvvx`NehDsaOZMWElMZ0?J
zFbaBW0E*3yI2PJact!m1x|*{jydY<!8mjogKI!);GRZGLtpybDAv=NszYTfgKD!?=
z$$iGdy-h<}z9z-TkJRlT=3NXOlj*Dn!XtF2m?JPJ;3o@c3twbNl9Sd=fasEQW;b~n
zRr)?4qYlJukpjuZ&EEW61bfQbJ~!8MzEodG3N@5k?|HiiAzUCf{An`b&%X0bb*jqy
zE?Tq5ryr>~M*ejdFHA;*Q?(4U)vckg`PuAV+ad6!%KtCN#{cb>k+IAl3fx+UgA!~k
z$h(OIhS+XpHQw%yqDUu0g@Hu>t5|Sry6Y-7b1BbP3otBP*ZB2r`t_rVpY925J%oqY
zO_u*%*6XskLszK@b(Oa6ddf$j9u@RwlXcyGplZng2_)u-evvEG8L!;<ry=}P#TQS}
zWfgOXcM(0oM9JV0qmO~PNgdf*97G5&$+6?GU&C3rJ!=<^B<YsH!!Ls9($5UZzE*R1
zIXA?a6UGt*iy6swF;RL9mFh5_rl&CTEz{avtbtEIb`>292chxOKolZIGM)T=xl~%(
zR_2wq2QQE+CHUgc%rsulO(bMb#JVwnW$-smL(hcOb@W`!f#?8IYAc}BZrG$Mj+g2R
z=SMJ=jfju8uOo%Tk|5u^P?5B-i?&{8<EQqCay&Uow*-;C24XL+e7bf4n_^T9_F^l9
zWi7`Z?Oz3q&wuvk_kKr1lzztLNLzScu7p849%L?zW1>kN>mK(U8BcBqXQKXIy+I<I
z$lXZ5>twkSVFHP%1@%P23<{9PMt;JGpuP{8z6f+vW8FpBqje{Uh6Pw5(CmW;qtlC1
zutU1yX7;?blT^ONV7*r4u_cmz2~0AL2}#zQT+2T5SZ33C-?HrMAcPC?H$NVIXJ#Ir
zyAohT?G8Zep6oyD;7&f&Us+zc(vQeF#P7F4=8-2isUo&2%4x7dyYU|5_DFHmBHf<K
zZ#wj+jCbOiRK43xmI|=M01NK%4-wrNFheQ^sN$kMY(@zc_LNF*NBT=ku(8|h{GZSu
zr9w=20I#V1T!z7I3&()H_-D5e6+(9+#ljWbC!1P3wzepATbC3yVVaN7qq#&Is7IAP
zEBRr9l+aHTcBLGh^Fpz5iPn`IdTkwwXHc|R!Rc^UJwpY_#y5LTl~8ARz=ENJQK;*6
z%$Dy*Bt`UCc-W&&B9HgWJ~XXxUvg0t+B9$QDW5Z3u8Q3y1{53Xr>GF`4#x+kH`7^i
zx^nScF7z26OR*krcezoLia``zdtz>dWb5_n=9@_-X;M-8^6xnQ{F-}9AV-}$cIQj9
zw0-MZen_mDUqx}z?=#!`WGUrY&V7D<{iIfT#z|Mvf5Lj6&j0?tfq_To3y;>)T1FXa
z43Bs2cj_N#kbiUa=Z|cXMs-)puG&b0hTqZ)?T4pI7uv-GWY*Ax{odSf0?WsUT3jJ4
z3cV((8wrMI1OZv8Un6Y-?&Obx9c@oB-jEZ7sFRmLJyFSRFYR@`lznMyc7ad$1h5r}
zux-KtqJH;bo&F?=!^-Gcq1zf6as(--@Aa(11)gc|(*X19@ane-rF0~cd%U8_(O;#>
z>O$S2#c3ApOM6?ZQ7eUxD%UEtRucoq{3A0QF6qqC(9!VlDbQ=LP_;?rzP7Gf<W$%X
zN(G^?2Y4-le13==nb{?P<4*fi#F!h|`I)fuf0jy^+RsTx?*%xk(A<sNGIh-s^LE4z
zv6j2`KGXh_1<*dY=@9xreLtRGUM^ZPK{dNzUT}Edh+aNS&7T9KHK-M4^7tNTHgC#n
zaqIh7_OW_pCd6PigXuVZr{0deov)1%&LV@70799W_vO-k0FG-~E#7<hVXuHrM$iMz
z%EN8T<L9M3SM-JEU<j{<9T_u+;^lOv6)M?<SK}z&?>4+T`=#@U<^`w1vNN%Hdw!st
zjwL3$^h7r&3~GZON2Oy`aabmr3QtZ{^nAJZGLA{pZjUz(uxZqhY%cM0`^7D;++lwa
zS3#TNbRQzBi9c^<!e+dg?KxYh$@$8SVuayN#L_>H(d3xkKx1b;g|?D`ZO4rgljS-q
z<w1k(%762M_8_SDXy0UaFpXcg`|B~zjpopQ`Qhml{4EJP*x4422jQBe6riI!HNzHY
zhTnaMszcg|gp;AZVUS8Yv{q4Mf$ffSD5L((&ln1Bbz|=5LjTXzpM3JuGEvb6J2{PV
zb02zmI_;2*<UWma7`wB?k~FWE@)j^Chm0kZ&UrZfgB$Z>)TcE;hRLq;nbE$?!}0rG
ziU@EvZXpq(N)nmpt~<AenP|0bzn*;U-1@#`nlHN+e;+kZzv|VHXm>IJ!!2Rw_nUjW
zX7hwgRqJs&C)`d??$1efZrk42!GDjZ3<X$`>i2snEP7aGGxPVpp!xHTFgp9*uD<Q|
z^KRgX3jrz1D1Y)*`OH*q3;K-?a0Xayj<Bm{x4vA)pYTCqKGqofn21`4t1mBMBkiR}
zeea(c+eJ0K#($5;cs&R?W{)ajdV_*iNr<UQT~un4`VdJmx_y5-#L)>9OIkHEf6iIQ
zy1*`+|CeJnl#Hgwp23hg$B^9F4b0zjv|RXCt=~~~#pT&-SLumEgCk~xARp*OnQua|
z&QRwlVzgXPz^wJ!0*N=#SkO#L%%|jL)|T9N#xL)~Ct^P%iYI#u<Ij%+WfHUG(+O=6
z<S}D*o@Y|~=adse^UvWB2FYB$6pYW6Qk~0De_`mYiGNQ+)ycYY|M=(LH@DBB<24^I
z?YI`)sMarTv`__{$~u2-a%#%0$7(wwC-L89qUr%o=0KT{4K7+yubtfm+kq{uH!v?h
zqe2+wHm~PryuP$qQx7pv!by||dPU%ID-1GSL9!y-)?cJ-k^USyGtt>{H^b>p$00AT
zufztTCd~YpD2qEvVx)46k1}^ly!7da3ODK2w%@g(BgFtG+T~HDmaC`6b0bcQ-a;Ah
z<U!mMwJTe=8rwH@FoswBaIHp@L>RIVPtE;dmQO+>k#W_TrN}sU{2j>n%N`rEuklQ_
ze#(j2=;58Y__A+b-6!Ce4Ghmg#yOasHHRHD&ES-4qb-D-FRkM%<~QJ+zGus`#jY(Q
z_)S0oKh^3bN}MxdIhIA?vP)@XWpM+RZ=j%6{k6)-yGcWnaic(4n`cV3nk`J&IuX^h
zdV1P5x+QXI16>8(2aAeT{?hV?p*ccE;)_+vJph*=3hgcozt!d}s%9_Qfr3pM5i)^b
zRb`KMR<o!&<w-4(uZo-a@6WxmC&9UUvfq8J`ipM8p^0W%NI<V|QxW&%HvnXwai=C4
zi)waej@-QM{1VmdBW8Vv%FVrO=RqkEHaIKDWURZ)B|WE0>Fib4#xeN9cE7w(JPpz|
z&`4Kg_1}v+;?@GZ@?C@UZ^#Y<uQ5Rdc{N*5rO9z#0opmpp8PK_4Ill(V<|Tmo;p?1
zf3b?&*vUt(S)paw`?@G#EKYOw8Kc)W9UxJL0{kj2>8S0(ArfNJjEOBo>Jx0T*&-o<
zVUAdc5-}8}!W|kPwm6f>`D}S6kNv|`%=ZWpl6=GGCXif^2~`^z3`u6oYV>}aAhxur
zl4Q3Mi}X$jF2U+l!n?n^>o{Syp_Mt#=7x1D*y?Uq;BC_TM!uKB;wI1DBR6Nes0?|<
zC7R=RJ~r8}6L*ZvO=tL2mKcdMeZ9RrYM`~uS`5Dnv43~{5s?MifCj4VvzXCT|13qS
zsF$N=sVW86D5_7m#voQg+!jlZ?z~r*vmUM2<&>mD%@C=OX9pjI!4{@lX42ClJd;a<
z7NK-Cp3(%2wubV1t^#I>2EZKBP+H}3HYPUK<|d2^qv?<vU+tPkM{)<{I({yO>QrYz
zb(Cna*p?;DX(iwNo$b7G-+WRhqC2$3h|AAWt0@I}t2zXZEB)z${MqG}mNTix6ISN_
z_j2y0R$!VVWV9<FS;p8jVt06urmTH;{(Kp~X;;}6f4u#C^m{exlF7q9%03g{Z@(Mh
zW5}-x^Rhyla<TI$-^>JRg!3yS$U9i&X=p*+d+rI0Y!Y|M7HRK<pKrZ(h+#K)LDmWJ
zMoXZ;u@JuV1tTC-sPAlBtI-wBmfZ157acBZ{Etg|8v2GtUo^>)wG$H|{ik03w~xs%
zK&FwyMXT_2;EW(6utsD~M63V-y2qkc$ml!SKB4gM;JYTApcillGc%;>p)Bt!OVANF
zLugmQSa6t?QWwTWvJrZIY&=Ih-3l5Y$?-X$&|y4@wH|#Q7`$ZZ{SF2T#**G$C{0XD
z`vYXkYsUzy0pFkjsB;@*iAdmZLO!g{#sx?={5j%{x$&B*m2E-WSB{gk6vrYW{ty!=
zGos{fQ2We0*18=Y+&94RrOcLI%LY49l=sO|Q%$nFFb?f&8M{?8i!EE48WK!(y=&KF
z4GGadJN_aXu<S|t*tU{L^vogRRp>iz3qSou_4plJYK~<q^WsO~uZaeUuSR@1$b;ME
zF@DYxD)_cM!|kTMC#q9>hcX^sc|rLc47L!Rr^2oIoR(n}#`mIw_hUvHY8k6-E`0ZS
zBq#kscD_h7LV@FI#1yxmQxcrYfoI;=4N*v*MNv81V%DnrtKdXmS<GLLVGa(3$CK$_
zCOWr2rcvyT=HjO1rN9GH-9h8dWka4&>-4p{N%r)kxHEK&_28`jP%(Q1wvg<c+6nVX
zrJQwMy~9)R5Plnx{b?JbheRcmj{IchKdAYD3wUegPsL5bLMb_9F05d_bOsO|s*39$
zw>{&E*g@GUrpEX(*Fo&*W)-nVE(zPI%g$cy+w`*EiE^pvvr-4NeiFKcVrf$(uSk@?
zT=9b&BwbiDRXSl+-|jPI?-K1v+G|J`2lE_}16F_e%(HF{Z{h7>#S1f5TzJC9Hs^e6
z&XJczA3I~O1)Q(cL;93&R{H-AEZpx*T1~0G^R`K17>7anY%<|%we3Y<To9Yen0c^3
zhAR=j@})oAc#CM%s_InYL?x)P=QeWk_rZGV(83SRA9~J3BU36)&TT&rn!_#Dam%#R
zyySRg@gkf*0U20`J!S7{Rfni3X?W=!i+KI}8c()6oU>xv#IjkEJcYgc-04utog$^2
zZ^$U;e{~6VP^G=s|EArQexA0!{Y#?VT*>|Vh}2bNM|iyL`9&k51>kKu&z@%*dJ1gc
zeHG5r4D_m9pUJ_y>*|`TXd*1NJrR+lsZ(Kz{JqQfZoy>nbZB<6yS?J$9+=`4%rycp
zIv;5P6U<8_X0<by0+2@aJT))r2K=*KZk`I24$rr5O#VaNxY>n6ge2zFmK?AYaD<M?
zh+ZN0WW#DraAV9RKZ>vyxl;<JBtibPAu_!E5mXh8co$N1!{}8c{EsK`(R?ryld4);
z1;#pruIy;vKZ3)61E>A?dDXN_|GhuTalW#MITR?f=gyn&^A^#~4`W$a+nCU9(oZ|)
zH)p9zz_l>lsck2A!LLg-y}=(9Sp)c-_XS@jyLreIS((ltT02OtS&zmRqs5uV9S=2^
zRL0Fz690)ArQG2%bMa&Z)6m2$k*0hi`{kPh0QSx0pF2q~7lOdT-2_m(kdn3gGj0}`
zQF$umyN5Y3CRe9?varb^DlRj{d1os?J<4w3xeDhaTE${+=c7#qh#7YXt0@lqSdXtn
zB<icg@-8UaRNF*MVgZV|$f7Wwk5BO}po!XKXBPdGJmAlh&&8<`o`Sy6{E^tv-|`(0
zZ7SwR!k-DfWuJ2XFW#MXX(7~{Pi?&=zZa{=|0$CfNSn*c>dKFTen*>LbfkkFk3|T_
zXvxS1ZTmfV7h=J{;ezysh?!LQo`uhQQKGYDn3h=QDO^6pNCpT#@P!xciGE!4M;wmz
zING^T_Pa${{gHQlT9S8lKRk1fc$VIM!3+SRFnzWHJa?}z8Kv~jtZ9#<eKjiWW8ctB
z)<|8<6u#Z@pbN$8@%G)*^*rJAKq?N2w8x*x{ygoI|2}7Zqu|h$Xr9UP(X>knw>AY{
zx?0=^Em0FP&a5h~QQkoITcX6-eVnY+1POL?PpMvz3f()j4zE(VV$>;I{`~{kNi0&l
zp8ymAUtwSDr&j36n=YzlKgLe~MS@pi{xTt$n%6Hh(-PodB0Xwc?zjb76S=A+zYtS!
zm0XzYRv1{lTwL=BnP~c(6p~cc4o5tq2<6)BgGixLFxQv74!4!kTD^Y}k7>rr+-TZ<
zW+ZQ=fJYz6%p_XVn_rcGMn8btFU=!o$8j`j(u6CMuXUb4^IS3pT`c1C3SD_ey#I7N
z^+k7e0UOtICW1P7`w%DpAol|aseqR4RfE{n{f=94hV@(dmAlesJtw=g$rx9pMU^(E
zNVv!o`7$w|nS0$(D@J3h>ohmL%5L(`1y7ke6^EDci*M0U7D1CC=691u7La88SM%mq
zpK6GYvUTo;&c3d6w_J__)SI%8ZLJRp?pIhF#DiPx5}={$!cRmg$tz^0zOb)G$8xKZ
z+;S1M*li3YH`6IMdTu%N@cx~1x|P0Xc>G%3_+g|pB7PqLDd#9|cAwetm}eE?A;)0p
zbgYq#6xQ(%tB7&E_o&TPpx7Ax>73<kt4(a6KiE9AI6-2*&<9K0D>a7gYggJv<{gT(
zdtyl}4K8nO)UB0!JT7_=-7)&LB;20sn*e*oSXNT#L6zkP-f;0J_g1stGOW9`F5pTZ
zN!T5ybwI3SRs)*8YM5j7IB8@fB@G{5K?SL-b0t<Z0$cQ24;%cI#A(c@^v#OsC0LII
z?gbq8<*xGS&?-_q^r!h%-$b_O%#Z@*5f&FMi-%1Mk|$s4EhdYfO2M#PpL5Ax&v$t~
zK1fT}AB(@rR2p}T5NRWmSPOfc3WS){ra0hrF!GW~^B8sxuUJWNQC}`IUqwf==LbkY
zU82Obv>n0j*#`IJV71yxa=IAZB(vOw&Vv0fM-%*3Ajg8n!Rut+^WomJ-RB<FU)y&+
zbYfd+netCvUkLQx30@%G6cT@MAWq2CVCw#7&GtBAJEPF>N_kdJo4yoNw|gkwTr;d7
z8)~X)6nh)uj&L!r(1hTgm(cMDA{3Us<VMY#>~~d6ZOI9Q$g-l^RW4e03tY4r2}!09
zz0VN)_#sBDsV3X>_N9G%fQQm9(DVzdEM`CwwM%!=XYDUzc8m-k+bv!I?^;fMQ~BCq
z_Z`D&c&xj3*T}3}&cJ~cI!sQ2?#*PWg{_e3`9jZvrAA*bN{Nk3H2)?jO(B70IxA%b
zY7jNy+#yEvF}*4PE*iw<Ni4iBrNd?!lp9N)r*pH_L7O*1w0V7H&)X)NVgyH?9bhs1
z6js5e8)q>dM&|#CPTecMW-{T0CZ5Kj#re+43r0Uw4YkD>3ma9<y>oqTVZ#h5oJWW1
z={ZLrRUIfh+O^)Fg>a*&l}*@RZj$D!Z4objH9KE$8UNfDCZU=zlxkgfGHrmXPLK|h
z_`3vcaut4Zr)iR<y>(_Dldjwbm-C;mzXqKpjPW(0CypH#vZnrOds`s8nmq;gW2@Ul
z)(<&o1%9j@8KdBa2t$_cRI2Zq_0oTkf~q>FzL*-p?HCQw60!Sj>M90pPE`U7+k&r~
zVga5Yz_3}bReNbe{K(+nBz#XDt*ONPI%=ObE-tQ>R65qT1$M3T*|;spXCvdrXd+~_
zO5HRct4kI$mG^VEQ-Urse?E~O!Z%sRScRgl^dvr+Q!^20R9I8fx@cVp+7b~!4zM>F
z&-t*OB5o{v-mYSF+tT0s8c>Cve5gk+7H}h|{9`_s3+mE1qgGS;zRGJ{eMx2wcMV})
z8;Izn@Y$+`#?|E0*W%L$Po_aALGB&)sfh1XKZ=c`A`6rkVr5~f4DxaPDR%t{FDKqt
zn$7;g_;U}@f$HdNR~}h<UpX6S9J&hj&75cB=bUs=aelgnLiWA>a8`k`G1pYXvrUh2
z?&I%TWZ3c}M_`UB_wzOWw{z*g8#m+=carNxiIUEv)EoWjzHaqdT;;bD&qjCR|1=}=
zjn320*_OVciV!7ndjY)n?Bklevn9ytyx-Ni8?L5Z>1C+l2eaNpm61L#%hTOaoaFij
z(C2kufnmeHD`NV3^n}FJL<A>Wxp<};4!>=f+l0OMVUoxmR4=@&C|*>>xZGDP-3*<O
zD)?K4K}i-_%6D06(>MH0uB}$SE4`k4+LJd?@c>!OZZ*3j<jALe#HV?&v_atayZi}~
zuhL55GiI0@jbQkqkl<B#TA6h6vO?DoJ&zI&(<lw<K7VpP_+R06CYgq4n`~mLX1{b!
zMQmxQCKhBv2PSg3%FluRYo!`Fln$XObD{I9%M;dV3|LUb`6jYWjj$A0Sjo=^Us;h;
zL}I`6XoGKFo9v%QJ1wwupY?J4gp~2$e%+lb?(|YGS3Elu>d&C<xbjcK8uoaB-q8@B
zEK!>RwJ|Sf4WL3I36CO&DEPP{sFu8u#GUt^=6`%mezyHBb9}!<V8t*R_jovSQHhXe
z&m%wdJ%MzsCO=*K{!J_ka)w_gVur}lSHgkocCm~<=EK1KU1BfwypQFxZbd_Y4Y1wv
zaf}H7crpWF(n$bSA-+>Y$Zkhix^;?=w?@n;Bxhu1HUJB3L%kk{*kyUrWF?6JPzN?4
z&-%!-vq+~UiP}Vdj6G&A5_#1t@+&4<=SlF7?Qe#gAXZfH(Wt!GCe|UJS<7>QH1;ox
zzm6c`0@$O5&~C`$02#x$kp8pWqQZU5y)%%1)B~Z!aZnHR{^&+TCBbV!xkQauG4*r{
zDDwRT0AlWP%n=gjT@Dg|^?8Ee*F=s>cu!}L;kU2P{yf7oq_Y_<*xj9DE;T_67sl|p
z@l(^uW_dBzr7croIbO#Qxl(A=!1#FtxRM}dl985qsM6PFqo(&#Tcw+8Ap5}$D~7o(
zBVb1Bp(ZLkpKQ+=v6b+_fx+^7iRM+pMS67agY#oze^%*MEwV_vnde4%0r|d}N)E56
zPZSu=j;Fwex|6h<pZMjn<A4zH2LxZ5|3N7B+Yg{-3=QHtYO|3M)T4ckIMZhJ9r7Ep
zRhg@EJg$Q|Fh*$DTizZ^*xQM>q4@2(+o#~=axfG30KC#0cpCsBya8lO&BAK~1EK>4
z3z~G4Bq9;whBPxrL?lrp49ouf7ETQDVnWhs@*5pb0DEWoGSB4X98lb^S%xm-WxpI8
zu9`aa+O<Xk%i3IW14F@8utrAejv|*k4tAp^fZ+^MuND3PfkPUPP52<2Huq#e?r74e
z!yj@i4`$+TBvLLErqTAZ^OJMNw>_#+XAL1TZil?E2CR{vLnG`!OuS=%`OUS3r<(oZ
z#=1#*9J$QZsgqB4bbgJ2NUK}s?=Ro`Q#3bHKIEn^i$qABzv`v0D<uA`d~Bn^eDsLN
zdt2E)(|^MMa$C(w;hVEnHo?aZ*>~hH80=Xf8mVFL2jBbi{F}Rrt5~j#334t+r91FY
zx$hIz$>1!q{eBJc``$Me_T7%&eH|*8t&|ii#aZ*UfbpUucCM(|U`wHWpk{IT&Ato_
zagjZm`q@_qM>=@M9`*jeC=-IL|N80j!C0$`GHjEl9Iw@P;&#Yf9<W1f`(Y1$A<oe|
zvn-f_uJN5^D;3p&yAS-l@E>VKYAMqaiv1#<yu5Dtwdv4e)H-^z%)HH<FLw@83TL;a
zo|)?|_T#Mo0i%SIiQMzGI&~OaDxd!WlDh`Fpr|&=YVM>jz(HA(-mgx|by<E+4vM~-
zV(->=EPKH(Fryf7Drr%05Pgal4F|qsAW}Q!x>)%q`GeHdJ`Swh^!3=u-s_KjulD}J
zH=ol>O&`QZaoa~a))6^nZx!X38v~Y6u)90`^=Iefj7VNO)O1nbEqj|fbf0ca{|U|*
z>!(4C<bk9?mA>Kf8gJ}<_i3oM6?;u3-0L?)As6P^sqRA&r)EpRVr)zC#mM#zB`@>r
zs4^!-oUSc;B?l5|ZSZB|oTL+Pa}7Lf3ObG!NlLzYRi>dyRf(<%uxPv9wsgx%RH{!s
z<~!lZj5oON<r2;AtY%XyaaWErjWP|qX5Gt81)J%)7h4MjGFs^*ihVb=U>)}A$M31G
zJwP$Bm&^Jg&}gxrD7&sQ@27~cA_%wca!{!4wLNrZHYRs}?e=H)|LzS7?8xI6pm(j=
z3D!8-EGc@gR7%G~<N&>xJ2WCKokKqXEkdca`%P+}>a8xR%mf+8cSXk}WZvwjk*KtO
zqg)w&X>q?Cq2yk#ctcm__RR5Llw1);_zT+_Rv!s(!#UmM6ZW!5xypEdAV-lQn;~J<
z>wT!I`>#Ipe$DeJ2z#~_X94<!^b7%?HECsUWX397L<m9F!~{IO@)aHk7v!mT`EsaM
z*6v#!4F!wd*tqe-)8)}AcL73+VIB-HY9yJD58h#<bg}hg9)gsGvd>`rouGV3P9S^s
z35%D;e>oL8unOhHp6h*siSZ)egiMN^`7hQy?(DRqQF*+A<MiQ)DeP7CqR)co)OuF4
z%nc2_N}lN+>%`XR&_99LcoA7mR+$?Sbw_S02go?>sVO?Z`NP<?-?YKTe8UkfgJW^i
zzMBVp*Y&x`h6g{~Uq$5oD%A_l_XgK&`Q61{s)U2F^EOBIGR|$Wt&s^X?>{d3ovL?Y
zi>LH=QCB{4y5Co<T9|UL`@C+s5;0yZab_Fx#HYS!XOvZFA8;qwcs~`}n{zDLXF-UR
z2p1*a3B{PY22F$%eUKj~1~3PUjC!=}+B)XLbUWdt<6+)q3(vXAO3!bZv3|Qpk#d2l
zHJgRn4hdXyx%mt<qL9W&Jq-~--x@@pR_i;u4&000=w3KTIu@x$Z_WyV-m@<&&pHiq
zkCxn5?BN`;sV83=AIoH#v-<KXdCpiW%ZvoZea}Dhi5B(moniM$YM=Wmk*Tb|mrnv~
zruq$fXJ4(y;khDvpG20@{0Lto>+Cl2jsEKjFNMDcf`bAT{_9Po7pEi|lVS8%FLhZA
z2yT(|p0=vFufU5lvSb=M7A1!Yc!LW8dV{<zg29~?gH!Am?8#c8ezyq;Bzz>7%T|Gt
zo~W(fY9!C$-s<t-jCx^n<&#|1PpaPpi(&`K88Sj+W<!bk&0T-x-{l<;XcA+h|EL1)
zeX?ANu*0mfzV|g~jW$LBG$(%cJE%-4oZ|5Mx2SVYTAJ7hXsDAMrf~r9a=yMZ@+Q(x
zk%Q)R)sg7gAFyWZ@HlS1fzrU65T_!aRL+k5G#vf96f8H1epLfvF*ZfTCPV|6AE<ht
zc+;$02c@;u_SXIk^jeyTd5ciI8lA2brAN`*sEsI|G)rpVYrmp;!sy<~TR+CM&K90=
zQ>L8i;pH?bRHZl)4io9~R3?8pgC%S@5TXg?O=8TO;T#9KYg<Tr%1Bv`o{vJ%l;N{V
zfSePdBiPpF@)$3vHdfJ{Dtw(bv+@_x$C}h62Y<QHWNW!{Z*=F?Wj{l&ipHFM7m|vD
zzd;q~)b-iQ2j}E0Hq#e4bE6M+#V<29Xg&Pe!k}tG;#pd1K-tmB{-Ebpa`~bdctF6&
zRUS+M-+)t7^0>!5N1JB7N*7bXIW9Dtch=5s%I8S&deBtz9Hv;Jz;U|~=f}34rfgO8
zZ#|60DMJ0!MZZAs1YJF9^JZ(}Xr<cr$?8T3VGk+77x_;EC@2Q*oQi<jzI9%2nJ%Ia
zJ!3B)`?Xx3!{EUxb@Jx>CNjyu`Azj)@1}iSgIW8evvCF(#<btDnVkpevk5Qvb?v>K
z)Lsi~EbDnX>2<X!$xW&r*{3yNuJWtECMcehmSEHxVE*IuUlhvs)gRbEnin<#qts9g
z=>nsk+2i8QM!8j!#2zbkEp=dY7PL%Dw%vGX_e;#m+z?W5jO6SfVxcGh#%uxz@x;$J
zX?;2bCLm*SFmF}hGZ}?dB~mcItU8vi>_yaNHt_>k>O@w$b4*~GrVM=$5jHfxFEh~C
zej{8g&QuYfTx^NYMc?B*cSvMvK=R<7khRX{>skf0Pv$pY-Ybr|%ovhoRe@AM7a|}{
z7S5L9o>aXU5U`Oz7X39B5^kw6%vr5uV;Lr$rE;n$yC~BA94LmuY<NL6o0KxVd=#S1
zH?w?S>@G~CdjMUYWt-N*QT56_^B}~Z#+gF-e?<M&bscJ6DRQ&S+?qvGZQ9YVNV?Um
zf6VuH%~Ikv?p-4k$G<`3s5*+!PzqfXy%EzW@Unia-6co;HI|g@p&u1)m1EwB5)M}W
ziifJP5zI%z=ZtT4Ky=yVUI~@L$&?uVBP|Hh-#YHwL?bYrYj7*kEpckSMWsoV_9H!H
z&RM}fly$$NxUqi0xFX~)n$;V2N1zv(aC6dRq4kpfP)4kKmey6LaoV41x8}DQ704Zk
z6W){>Gi>jc-XhS3n8?2uHpw76aWqjv2c3XmTH9_uf~+UQcHZmcnrsl3%x_waRp1H>
zDz6~X;U~P=r}fhGiZ3|_Rn}12X?{bmS6rTrc}60O&8>%{)VFVTJnwemNs_%G<<+;l
z^>zd`ef=eZOBgk}m|Du)FRP1=VH9$cmzbKVcQQ&jY2UwgoW=sI!yS6Jo{Jfn(ff*=
zL7)u6PKpg|iDSYW{M@Q?=QkQk`DMbGIMdMh`f^CHp572u-$?5MyGPP+*<XU7#u&bN
zl}Eeg>qmx6aY<4pVYxISX4Jk=|8)Ao^t`#t`1ACs6N&<*DBE~Gwpw}}JkkjXya0eZ
zZHvfA!UHk!`J3zFNcc-nIHXvbHw<q(we&II#N1HhcA!ejM_+Q2y}@neu4w(ES6VbE
zBYoS8eMf$}xfThHpOfpQp>nd4_mf+8x<<BYso15U-r^6|dH3573UXc1_I^-V#2ch3
z1P8sR5<BI3Ym>k^5X)5CEEP`+1082$n!V0y&<md?^ExZVEo1Tt=HmzElT1BxO^6&W
zzJ!x#DN^nS@K9h}FY9}{evbpHxzs+I(`7O1oH=Zo;0}6x_ol^&FUyJm8L{m;8IY25
z4PS*YO{Lc16`<LAorg$Q*nAsp2Cgw;m3{C;FzvfGD#6lj5(1SAaR#eci=YT>AH~ol
zWFsiA@dxI8cvPfe7GPyNWo)4QxPem9v6j2mC>?%n<!}s7M?73!qjC4p8`7e<aOKQQ
ztMcPZ&TIVDnaC@cwIW7sxyS8i3&I|C-Fc0C^?HM<Rc9*82UA%hC@44>4hC-ElB*(+
z<G=4+Rh)_~(g@X89x5hGOV2kDhV7L3rETEX8K!#z3Y5p_rv>~!jyzj9AsQePSv%Gk
z)EsqxzJ}MScj{RCCM~n@<E%Ia{3@x-`ctOLDwXz~Amv2ORk^YPVpd5_>ALG^TJNh?
zWmhPaq<<uwc7N}O3BQ#driK~ysGHX5_#$j8i>QP4xBWGBzfCh)f+3eSdJdwL475H3
z-V+VB&@Pv^H4%GY_)A<@aM1EIk77RXTQRo0#BJ}@swlM*ecjiwuh}ROG1pnSD4MZ|
zbN1EI?gwJoXI`%9#h6E~g^KYTc@CA;Jb<*=&@ZsO*+`kNnY@eYbOLd-XpZqk=tT=r
zPW4-RN}I;GFi=JP^4gYhIC)Apyt_1o#Q|J@j)qEo{({5$p}Pfjtf{`t?-pgC6Nz{e
z&qA0*s!KB7pVnA|y87}VR%&X<?U!X>l&bfGrl<70{?IRG5w`|r#ngixQI|i5>D!As
zajBIR@DIFgDq?T$f>yz#$~yLz)`U6QJ0qedSKCiwE4t}*47L!lz6L{kxyTwBu!vf^
zpF(24!(ruU3SFwBN!5ae3bBl+^<V~#&|ugqHnFEy&R~j;x@*<KeMyXT?d+>{_yHE)
zR$!g$d?)g4lH&U8-CiAg@0XO@UWx&xJrB#t4T2T2tOz}AwWX6T(|=h02%7lIA_WRH
zs=f(P%i7$N7Y>kPTrxU`eq7Njsg+iqCisi?qoTqYd`x`jOewbk#3--5`7<^J$f)l4
z=FWA5Y*Ob6_>Yv*tmdbnIJTPESN&Xm>vHnmJ>;ZcTy%eSrqz27y1{as{#4b0G7zDJ
zcCmS|5@Y?nei4uHjtEyMuM55B*E|dTIC%XcJ8gLuj<VEPSJJ9z!k*Z~<=F4qBqY}2
z{WC9SBQtGomGVz$+n>)>0!VRnP7o}Wk{?FPEReDFDohMnaGl*!K$oWVilD`PuU7zy
zR^u);lJK5tzrsRkdvS|~w2x=$SVd}L8yZR^$BIr4AV9$u{if0dos%R72{=3D{Z%F+
zK4Q|D%1Ux_EtHVwY+AZ|!;gkPQU=IKQq>Pjau58taDl^3Z!61wPF3n3;d49$-hPz*
z`B|2M)sLTh9V2;nhKwnZZ**_1puy8HS<cBw$lhOl?hW}JogipxF4x5aN~&~3fiohG
zddV6K^e6l_;;*FTM596!CPX8}H=whfK_lCKS3v_Wb{y3>O-ibxia0QD*6VD?foTZB
zxmj>mvjn!fULdnq+(-Z@eu06~2MKde=y{FypnJW1ac0~UYLah#zW{6+=0BXb^U=VO
z%r{`un5rBy*<X1G<J7q+R<?-U!;Z&Dzq;G)vNVYYhSwx|UG8g8pB9w)_jw2Q!X_+3
zW3;@Du9%PW=}vDV^!C%|rZIX>vz5)cm5(Cw64l-VD^<VjNkC@1`-j(kWOR=yN$X^y
zqdsc|Ax9LjnvboV5idE3L2?D<KmMzlH5V`@>~^#2)s$y}AFkV;0)FD-{3!C7wV;EC
zC=p4LiL-k?(n+s&xA!2@k&4R16?Kx7@kXu)F2Vs=OT?QA(-*LBrr@%!G3m|=2@0$+
z;~iKu_cUz!`Q@H1{g+>-^WeTO6GG~hOIuu=pVQw|r^9-bEcqsB`qe?~Mlv;=|GmcP
zUSC_)4QH0MHa$70B7}hNiUKcRA}A<;`sBuq6nR1<|A(}ut>EEcbl%DL^T&v+#vCln
z+sG*}4DaKv5k&AY^<C8Fs)v<21sNI3_fs?pE8W*6l%@2DLahus{OKg;PL<Iy9hz+$
zx-m4*qalQ78X8kJqwqY;n46ELdT_`&XxGI}wY-K9kg=n)6b>2brsNg92UV*s$K@CS
zHiCZuh4^xv%SxZ!I}ewmTBzVXY^_3R;Ja!N?aQQT96GVsSl>04n5dro?jS`UG`iRe
z)wWD6mC0QcT1xc84>boV;;cnn-YN*{(YBy*7r9hl==`)yxKO|!>z*scLnuUUg@;#{
zDhYPFimN9q=hyh(-_JSl#guZh7vJH%WTLA5!z+MeCa)<XsCZs%&z@=8&R=t2X=-2V
z_OfKZf;z58JGV1_Y`1BE-CJR~-zVA%<7UHypVBm>SV^CYfoqyPBn5s{bbB~l$v=BB
zIGet06%i<;DU;ADHGm8cipMxIA~ubZmER1RtZ#!1l8ZkMLZ`GDH_+azP=>B7fU!j3
zdRmG~buwI(jU=3qB%qC+J>?VVC$)n}Q|?HtwNO$#KnC^gQB!y`>~0R{xuNzQO|OId
zTFYDpGhjF2GXC$b-36n&F0=+W^i)$%Pph`e@@s@kmcPv9mpXGY-y=-m*?4bx(d6p~
zJ*pzg3upAF1ZG@sIt#84>Onn{3+$=lt;rd&R5Ld|^>QtO=%qJ9)%=h0>jOV&9~9M|
zpUUoueJIa)g7IJZZu(`3S%(@G=G*u)4kv5Q2(tBY9yHty_#&FL;nfTViX3f|MRcOJ
ze^lb%FKdgO49fZ+;iX~Ol;@N~tV()i8HaX_UDyw2ds>*hxb|zW!PN<pZds#BDpmYt
zSGFIrAZ6(yf$cirv~xU;XN7pW>mJ4X_^ySFjHvL|E_E`u6f>?gaFti8W?zs$Abz$(
zq0skD1{<@TiSfj9jaWY|_v{-HmIeVb;}=va3u(XZ7EjwQDu^|#(SCtl85x^e(w_pl
zyJ&3#n5e%7EKA~h%~Wd>EY9ZoR;y~kXUb`lTOH8k(i110^X3a1$pO<!9k9~J_v@P#
zgN5IKEv^8Ku`z_3*ylpb&U{h<N>T6mH-S?;Pz0%WFh<ax>ZpwlaqXDZwRsFkcib${
z_Gg}6?$P;t1NZyBf>0a8h9Gs{RzY6ixc1)_rJ4B;s?S%{=zKB3ODOu?W!yx)=jDQ^
zhkfREH6)qZ*4H^{ZX(b5v7}9A%QI07ABsGK6Zt}A1#kP#iPuTAvAa*e9;I_#ez~@%
zCC*~cLkBieLJ?;XBQ8951AZ_jA4-eL=t+&*R2iX~R3e|UWS{BkcPPf%wz}#!|HGiD
zz`}r)1ex;KH2AKRyr)w+q}~4p?F{jxWtxuvUj}@l0y%BQA?s^K5|Tf^&{Odr2oCY-
z;JxE(&b)OC4>OG;`mt6p@^oiSV*7XegqEP;;^(VB*Spn=?cA1jX_3vT>`^?$%Wa}f
zH;;18Hs}AC{ZhRUUUeJ-+$|7OG{>KQx&4v*U+_a{K3CL)ob$5npmVE^O`yy?bNu{S
z!=h&!W&{b}lyI3%boqNK(XFT9q#}--0YOTKST8LWJYk3WU|Cd^pN?rlc_e2?(mzZ-
zHyj6N^OYfpcd}K6lPJ{ujTjE!ZcN(dqfh<<i37NEOnqfA54g7a)D<!=;#@*#m$5SC
zpf^v39cZt7mavPz#zVKGx;)t%F8m<h1}657qb=aXnn`Lp+t1$l+Q^V&z55im64mxj
z!<%CQ>e`>w%RH9!<GaeI>_zi^TW8hSO1jbyZP!Q;sV&>uE?sivGjOPWvH5CLK2Zos
zU^_dw`O$1AT0)mSu7LoFceQ(MH1Xn^&1rqJjM?(@aRtiF-^UH*#_xi(*M71m(jt5D
zrT%34AxFta4C0a_r=H#{7y;A4G@K^Cxg>U|%trT7c%t7+>DAOetNvb`-bx{+cAr12
znel|!P|a1;d|!8ED*{H95DVL_Z&tf;*M1<&C78qB<lHi<#qedgN1l}be4|=vazr<Z
zvLtD*m-Z_7#GImCZZwpa`x<HG^{`)W2Rd72EB}-0cK)@+xzb8jw>pdfj0DdhS<_Pb
z@VtI~`}O@RcKxlKoHje=p?Ms3M^wmFlA?9s5gX$F+Uq>v=c0VqWl)J2&Ey+%n6IKg
zXmmZ|ZuU`b4YzY9t8`e51*Tewt|{ZQc>X)Tq}{ehpVCWpLehTvbF=u9;ya07Q}&)Y
zwTB_n<fryZ^{`p?lA&~`ir&2sc0v7rCzAHm?wlZ%K)tDJmTikS*3%pbPg><t$#A%+
z#^*`bwof+uHGhy;30XH2;B=i!&~<ktYAL}{TodkR5;27X(a{;T-;>M=jis0(&N!L^
z!{OCe9te=zoQm7rwJR&IhTveUmh-7)NfEKYQlnI`>~dqPzHG52f=Js3*9bm7w{zRI
zbL&baGSP#b24+`-cuC#_*q>PT^K1WD_VXjo`A3t9F6f#biPD$OkmvU-17)vhZee@S
z(1PrIQaw*z2*tmXdV`?k9kBt)p8@_f-S=hf6XMPNy1DJddD=MK$04)LkpVK;%yzHy
zbnitTTB&tk<j2~p6%9I{E;B)9O)9-%%s>>4b06<$2#QReNEyV;nc<3w{QX^n%ifHt
zQNC%p<U>Edc#y%|Vzoo;(2E4EXD-U~eI2P4=0=;3wTv;7S8j+CC#>W<e+%rfi#!9b
z+go4f>5qYP`F?_P20HJLt0}O1WrzH`dq$k4ma9iVvTLtg!zQMVl<&wK>!T?oqWPjx
zllD%HO5X);F;nDwYR4IF5PmSDV+?gijVlP}+YAfGvmWMYB};J4b9~!?90DJnQ?84T
z5BB-y9_H7P@cMX-f$4pJWxq|{C>X!Q?6<jqy7AJ?JS8XkU&p#Bckm_Bo=v6~o@{j?
z6M(HC=~`%m9VPpCy|$wCczzR&oiQoDp4n!ZxhzTM|I&#&^h+||b%2s#I5WRlFCybD
z7bVpX6~AW255|^7HA$-_KOq1}VDYIf1IQeL0Cny6vaUu)T9y)IXMyxwm*w4k9_LBU
zNY5`nziSeCF%`P=rm;9D$>Nl;G*Q_|6mWCWqrvxRY*B63$NTlzT{axUAz`{N*f7B;
zJMVZF#v9#Dz^3b=I16$<u@Gi^u&DRqbr>{ZCv5ge%>SlI95#BT)^>N4zYkH>Rkr|~
z452A-GK#3mr?X`j8dC4^kd&yLpeHn_13y+IyRvSAY;3DjtG3Oyxj)2n-pl^N{}=H*
zo_Z76ablOLseUMs8_<39A8&vUCD@EhqrOIjr|rv@+Gl<n|Dov}Pd+hys=iw?c6CzY
zZMz)x&nqXChGe^v)H0OB;YMZX`uvjFe^S?L)qs-QF&eib?4E$oMvbmrNSh%E$hSU2
zfJ#Eq+!o_1e<-5MY58`Z-Oy9o;)9zIR={qOX|C?QkcR_|1Ot%i<OCaMaqMwdA;I4*
zeUYI7GGu5CLJ41)xgwPAlF8c!^)2wqF$_ysll@0f6ZNH-JnzUJnRhK$4;4EO8xV$h
zoeTBXD1utrpZq~B6XIQWZnFGjZ{;Sf^s_kK#&3U1E5)9Dl!kbM_MDuIKf?dD^f;tD
z<gh<a7RmFje_kXTDLg)99KDQ6v1jEBM*~JNiyPSgrjfNvkA#gXeWzb}B)sQncuz{)
zT^5aq^lzAMqZ*W#k2mfRY@%E6S6M{cZ|)0JMSOcwu=2bh&k-%u-=Rh-xAnE#z#DVx
zM;|e<36L>qfGX0mHyF|`A@dYim7+;DXAlEVnu*8(Pc9mI6_KT*ocbS+@-D{bGMsgp
zr;1~t+}<dZBu#Kz_YwjU`-6bsU2chA$Edsvs$v<sQ*WK*O`s|k(IO?eQvLP(w0}|i
z+dMqCJ}$JhVB{%+zdb641Be5LOc-r<uOs^KUba1U%}RkfL?=KH_m$56^ElLk;&R{N
z;o41kgQUq47F{lLq&mQTp<Z1cn8l<M3tv%a`B<l2<T;aBW4J7Xl8RM^RnE&5GLnPE
z<{u9hTo9h>I3r++kJCTD<tm#fKdqZ1Vufksj+*!hSZGg3S-b&JmL_~V`I_}?A{xHR
zgi7T%CWAq8Egn?Jcq)r9&J{cM7?AW$w_sYk;<FZ&JEu;)*j;?j43Jt==D)@N{{KV#
zKSPBtp2`gv0i<~rJlIapi85K|UwH;qNy|QH<IG$7`b#1sEAm*rIN$|szL4BS#7jO;
ziX$XQ^)U(Yk|$G}2&&#aRBweQiS2tB^6JuM;7Dn_VWWTV?KK2iI0{&@b2C)zpk`RV
zn2l$Fgnvrpk}2&GmSqJat;!rRXMYTG+zqWzB1sSLzx@u+nUmI0s8vMoW1UC#@DFxt
zNWOk*SHQZAnFLNMB_|%t#=5lLMO<LS46?1+O3k@~ljW*bEL|b*pDTQJ!pN%$9qeG|
zrNjmi=$0bOBWRbpu!PErx=h}(UDd$fY_3NUoS;9++AjZ=1mYE<x6`Yk(nKC5eGnEH
z%p%e7>U2@whbYGyd=Ywng|5Q~Y%#D>BX!rYFzA?c^ach`CV0HiH*A0%&r#Bo*Q{$7
z`hu(?RzX4}|16|pz*QbgPU9-x(gEH~sT;$bC7WjupU{2(DYZ{=xE+^DUl6m9Nry#V
zBymchIiaYyS97$d^fuXrB!{Uf`sv9Uyz!r0gY<JG_ZP*l%{817YI+jsY3unBEXu8L
ze=uVLylk**r@3cB1`vF($;Z~f3sQYsD}QG_!in&UPOOJn$YBxs(M<GRKvJnDAtr1W
z={gJ%->|JFNvTv?pTSK+>m(bLjNFN#|1)R@0+bL8&_US(8u!@EbvUv2@@i?JU1>L8
z2$NEDwsr5iur_l!N~)+UOU2;h-`gsl*pGyEY^f0@q&4*Nw8#dsp^#^*6Gxt{ThqUL
zwtrF<)*OeDiff&RgAyk+!kGD+n1O3)!mO&PXLmn2=Skb8pf@EVb{3W_uIdfUgQt-m
z!q^}WN1?|MRN*3>gqn2iaV$+h&`#auWi}8V-F*#@HQSp|6AIUh;7OI8%HI5)<W8nS
z07FAJL?Y(UfpkzYK@nkv_*12AdN@G7L3MPwDIX&QHORO^BCQh|3_y9Tx)e&$9tZm3
zWvYD*!<?CWtKySaTJTBipPtRY!8i>GZGwBrwv2TbsOdklk(%LH=s(pAztwfZ1j3EP
z{_{cH!Buejn=2e{!JD2F;nU4GT>*y`ZguP3wlD`EKB|LxX4`v8Pta|*dG_K~Ohu{d
zxr#hHWV5>oK#D2u=nGy&>=qbA@Vt6H`H{y`y_Im2kF|9y#8$j<y-2tBs~lur!<tcO
zgA*;eqR>drY!V{WYlDK?+S(k-A=C;ExpZ=}i;GM!NyY#lXwxg>KS2!E5_@SKAr%=0
z%9@rKpZucakGIb*sFCz_gfneA5}xY*FH(JLv>~{DbBy)2*(o+==U%{}^^VEk?+|r?
z@mm>PRHj8-{EJZk)LxYo0<WJ-qFU{m`O|+lrsOz1NVOnSI;roCKRHMKWa>^K-Y`<R
z`>@Z{^aT_{D*E-RUqzh>;uje1<^<Ky*ohv4^Wez=4TuXuC6sEVAqPAiWTf2W`To_5
zAYIa%BcmGB&y25`fXZPVbB+Yc_VKOl*JS@>0feG{Pl^4T`Ae_Y;N`;Ot=!r1$0bX=
zbm4!nY<`DeMG<d-*9ki<FCeY4Bi4f(^hVj88*bwrTlu&4Y2R9jZn1!QMMaNJoJE+k
z()ZV>;&{HQ8g|7zhpUKwVy|7~aS%FBth+Ee&9MOzTDB%ByA<WVa9ih<J~L*Y(UsY}
z6b}RmB^gRHUaj^RE%^Nb@|z<oIR}`Qkigo1LIS_`U=?#p+`1Mays7_C%`fuPJ8}(%
zIw9NTiEAjWaathMI?v1MZ{Uq6KI&Sx$ul76Jf?1L#=O*>X({NG2=$eq@p%RAXhS06
zY^@D6j7Ji$`&6&7-r>PD^HSG4AyhXbJ6@sQ#tiEhvmhc%dTTYz%%gJ;t1g##v+uTk
zD@eT*`NEFAm)Q5$N3HXHYqZKWK!!{>fT~tM+1=BdSx*L&_S!Jv#8U*AFGQ}`Gt?wq
zNAR$>Xn}E*;jz7vb<K$ak|Ti@X!b55O1VSUO;L;RR%Dl|@1X*~OFjF~z=ty;iK^q(
zGs3M2@IsB3lygnF&(i#4w{@CRkqDC>??)WGNqev!OvcDuiX1<s&qSXJAK8v|7qEY%
z9*ME@J_hAt;i*|W9@@6~am4;&gM0Sz)$!#W!<~?yhAL*j6nUwF#}rXN4zG$E${2KA
zI*_3$0JCqX!D??}j3Y!1!++|TE_ITY`5z|-cGiDtJO8zp$#HHV+uj)7VE~t}1h|}i
zSB!EaV;ulMuISa!KPk6x!V-wnTlkL<=l9bJQbN&lV$XkSQ}}&Mr4gzkY0dw=t^c9D
z_W$s0&0r)3*G6nCRzOk+b~Cnpe$Ah#>tZg{8<EbCRWQLNW&d2+{norI@Id4%zUdE~
zg@~;sc^&Ad4!EFpm%NK^mE+W>z7KY!EnS(e8y$)Ii2pyeyLcTaYE^6>f&g1&ra?W4
z;)UoEZh>XFNQr%)mL&`MG}Lsbh%!CLFBfQs46-3(rAXi!iPbW-;4%3^GP>H5qe=EI
ze%BD>^jEMs^{9q<{pjgm++2~`KA;bTe>)DQC58K`5cDB$1=sto<sY%w|NiC8l{*Zg
zrVZ#)5;&sfO)t8^$d`K><o!vd7ju6ZUcvCaPt5U^RTAtyg=+=$R8U@T5tw(8$1ZP$
z{<)1bv6HGNOSIO^F65oNijr`Bo`U8#Oo-J%5|9wQ@Xk@8&)>n}hEPi)4I_;_?*ac6
zRtQQHaUPH(Q{$o?!(Y8b{2%Y3)1Lx|TCVnI*ncDHarh)1B^`BZ51>y_0{tNvG%m8w
znEi7p^}pU{<OJydi{8FSCO06=&ALP6P9riTiqWi<HJ`1x0B`)wjk3Pd=C4UuGO1or
z1HT-?5XpVeg>F98an0Z?Vk4~I&UX?hM<u6O4IMFQz!Iz<7<8dSqx<UO41adKB)kN=
zdhSC32u|8S(f0HE*`F<{h04sFwj6q^^w8`f18XvY%rJBeNZf4cP^0@dhuW`;Z6!cU
z<KjQy(~naA<gkT<A?D2v#Me4(dtqg?<L~99eNDvp3_MEpu-wtNoeMc?ar%=YX0CV+
zp_q<EC$n*>oESU)nhdmQl%4I+MZdZ-!k!KOOk#wwMR%Isnt|Qv-g_}JMo+__^ibR(
z-*T9h@8-VEAsoK@Ewx+IWkSm&Sx9pX0}X7qQQv|#lYlh-OYj53W<}2Mwg0)x1)-~g
z5^>i?r8r^pP@~ODys~^el&*orOYj?KysECC44b&q6<Ai+Khi{@;n+=z1!5rd$!GA$
z7RAiIq?2$aY8#qPeZ=j!5Oe~QM0lVPaADc$QnIQ8G-1--KgN4c?pNU?&YBQE>l;)@
za8~m2FBJrV|5y<GE~Mpl6Kv3a<R9q3ADRP*<6<e<0Sua(v!4~YdD5^QYd6*G(G_}2
zk`ct*faBKN%ZNQ<)>f0TZoES3-^320EC5wBDZn)ibA}pvONf{2Zndx851G|4rX=lX
zApUZf^cdN*ixEdr0j(~dgT6Zoe$@+CTA4hKVbY~E&$zY=9r9sSoR;t=-9O{a|1~zr
zp!EMu<TH3|xx})an7I*>H1X7VLjHF~iMLpu(_Ncq$y-sxWSX6seR0>!#3YfSQ~FP|
z$;>9Hlk4LSiE*v}hJF&CU2YxCpQLUC2n+wtY;2z~ue5F7j4_`f8sAYA3_&>%y+jWE
z_@qp<oXtFt-~WdnnvOq7H&N&_*o1xLt@FQVc@>zSl--)LLL%;m9@vvY135M|VIarj
zqRV`#APF+n`jO~Pn#CQo5q~q!=9JZKLS4tlar&dbl<MDWHf|vWi3Z!=(`BnzMHJXi
zjdHO{x@KtblWW+%wD2zchZderAn`g5p`a!nuO6<kJ+O=CvgKnoa@UN4Vxiz9@y$yE
z#S=vX8G?N$-VGaIu*nqb&Z7aF!s2S#g<$1cL8_^(c}``;H!FtQKEk4;g0mz8ai32z
z6Dk(ku5okKiG<(H{Faz4T#LvDo5qhM4X}4<ci{Lpy90f26`at`h6ln_M?j9K!pq9T
ztdUR*XIc|T&&y{|iqf2U6Ud^p(Nv&NwRWdsvNU6nB7vh0{{Tnj1Q;a1Dl_C_exkv@
zaHgA^>5IT!?`%N=5C26vj@Y-gmzpfYWi}HBDM^HX63($5laC4qFKwRuM&@-U+UMV~
zu0y0ipIa`c9o}G@+hkm8nvbcbbMHwZh+-F^YiYthBMd~~S}2KH`Df*W?|!9e5*&N?
zo>d6ha`ZYY)waVy3byB{kM3K`e1|21n^ykgi!FyQ8o_L(qIYgf*Fk!=+P^nM1nDei
z*3(lsaIlK<93}1dXGN*3$G)VxeJKAj<L3MQTYQoA$z$)N?j_x8D%xc%Ty^h8*XkPl
z5OYv{fU!fl;+ro%rNJEMJ0tQfy{anD$fUsC{tyfQ{bc`kNAbqiu!PGKAS=>Rz2C9^
z<ZD6DvIE0#?K7N^_Hlg)r;8iv_LMt5IenD!{<h%vbxl#^H&@~}kBReR0B-$uimCK(
zFH$@H#zZ^*gRiQHk}l`X<BQ*yUtxEJQaB&gmHkSy#Hv(~m-CPPuu2;H51mVe=IR;_
zty^oPKgyAL+kC^{F|1%gnn-qJQ&zQ#0?NF=OqW?~b9wx!Uh=aTL>ueRijQ=nc&%US
z#DO1okEo4fc+r6eJ0-0&F?%>jhUL^dO?g)(4Ae`bA!ApdeLHh9`o6nM-1`d~Q_0Mx
z8|Z|;Ds!=nG+E2i6&)`2%_y7ua$|_d%8!M-Vw<+B9qlRLTfj)K%>Lw!404>6nAZEw
z&_{<2Q);iu*;4v|^-9x}dcWILVoatSEzgt2HV2rxc*ZWSVoh5*IcKvmu3M#Hh59u6
z1Y*t%NbF?gA}+%a%KzyRxV*Drvz#0e+V6D>D^Wk+?Aw>WK2~A~)hy-xypTQx{BV36
z+kq)sF{dnD>yt76mBB}N=4nd65T+y}o9QbWnfhS0UqFof<I}5=rwQrMqcIBEYTk|(
zPRjf~4oE6M^KPfEi@!|y@ghu%;G^@nlVx^O_hv=G-f!edIyU~BCy56J;nNMs%)SNY
z>F33A9Apx<l_*d?H5f@(bW{Urvzbw;XQf*{N6$?4nO^3QwNU*a4PC7sJU6#1_iofH
z2Jf(y^_|*<g{bS@3G{NDo>X3j%=d2LN6!VJTr#G1_Q_C$Bu2k;&zX%TNYON~Vgh0i
zAJ+fhqG(9Hz~|rGa8Jg-o5@(#=IMG5WP|qX$3x=@c5P5PcPiui7{+_3KqA@ja;J2(
zng?xM%5VoQ9^>8Z|K>5`Iw(B>0GLXa!7N~acHgH`eT-|tm`qKy!JPbyNwcy7Lbb^X
z5GOe#Pi*-3E|KJUL~{L=gz1YPDp&3l99{@o^^Nb5^c2Z%dy%HN(v$f!9^>%RW0d<h
zkCF4*kNo&dHJe>AYr8DlhO0lI2Xz|;80lHi!#7q6mvOR$sIwNS`^E9yVGF<y$F3+}
zGKTtuW*9!?5T=`b&XdW76+J^s?$lFBOp|cD0Mv5}?d*G#g;H*?DhZ4+zfvb67`%i@
z#(KmOR528#)&afh`#m)GB?tWe`f>O=EC~A)?;$07DEbCe+$T9$?d*Y|$FH4VQ#fH|
z0~&T5=SAtn8rdJtB_P-CJzYgfJ#=B&hZ{#rneYQI%q(T=qO6Il=!5ahhg3u85nu*%
zaF^@#>?z*C`lZ>Nd@G4SbNq^~<{g^Pd<DuEbnGFI_$j*4U*@iN!q&`Xcx2~4;n}S=
zZTo*nyYg_VyRI!Mq>x0)Jj)b9)Ilj6b0kBiG8D-?9wHSgWgZhz2N^PF7L8`gP?RAu
z&z+7b)3>&Ur{}xg_xtDj=e@e#r_TAE{o8x(weEG__Y$p<(EHI`4};hVYS&gS`99-&
zb{xhKQmQz#LL_+9oCo}^38ac`oViOJl7Qr{!WN)ZeZ|0?r-7V$Runqr=5el(D%V|x
zSl2+YvaIL<KM<xfwyFj$L>v}R1aglp!1yDy3m(wA^QDbgCVvYcnzjivrrnQ9!1{uM
z8qT8T@0WOTW#dwP=jw+2gqi|11%5Nw2s7d)G{r;EnPk^4P*~zQ^PFq}#MO%RA05PZ
zcj*$m6z+_n!#2Zt*~^d)cKtZo^=l(Tw@65rVFKjg>Q9lMARbSk^nIFUz!#GBOh7-6
zg`Ec%khL%OA%_s8>=TP5K62?gA0)6$EAWY1@fNrOE@wh!=iiLZ+I_f)Wr^<;w@&tM
zlfE<t+6jG#vCh(;m-lEMIgXrGGoekro@Wf`UrF!V|M2P{*7e0Pi}S%~^VdmK7#BVc
z87|<?kP<pWT^WTe{69iE%H~%AH}N;NYf*-Rhpzr`<uQwj7hogx2IMmhu~l_>NWHVk
zBvSW&vBKJknZ~?^(EJ01=4u%k&T9(3F!ruFcL6U2D~t1o?U#kbNnLk=;+YP#IXV?E
z9y@~$IBGw4&jUd*v>Xm%6dBnSn8wKB^m_ODiZu2e_!~>QPZ{vMvfe8VyhQJbCyzsP
zl-aHp;b7%1@LmOIdsV#NahR~1eh+NXq|GhL^m~i`=ili=3X=A3wS;wu)uc!3J}RJb
zR9_|^D_lP;R7Z6il20jXO%O+FU-?v&|2g&ZD=~|=T{E|v10k!)1h@L-VALvx_#3ew
za<=;bt47o?%4inZfGN=gcpaW5Tu{26G_(YPW}&(!q8}L62Pws6ri<D;Ajuf1PNrm2
zHcjiuHf+y{VU(Zzm|N0ltz^0#tF2dL17*D}v|%P-3+sUvWo}X*6cc*jVS}bkFIg1p
z!xZ69IlgOyyQT!{m0H4~(W2a(j*ljgs+yX{X9&6d0O?}%u^u_#CXXCi&n5P?14>Y#
z`TV+sMRV*Kz6xDSrMKm<oklysJ*oq&Xi`ur0H<BPv%pGiqZ&};%)#fLkuE38hhx60
zJui(kTjtbi;mMD*FH4;P$onjec+K2+`Lr~b@N(Efrff4ysQ<t2mV$E%fh|Urp7{}`
zenY+Qkq47ssj#lCmjnAO)ewueO&@s=BmwI|=Lo)>AGoWwVQJq3T}t;to_4lH9_tn|
zaY2;itp7fl|M`~Wp=tt)vyD#|VZ!q1JrBKS%6zXWcHFZu5cnF`$f!d{zdH%9oC3q6
z1SuyDMUpkCur~t(t`qX`vD}!IIk5JdA|FFVi7!bDAUn+>zO;b{mM%!y^_0G6Q#QHo
zXZd5&Y-57l0t|t**5E$ojgPQU)vwi|rmUnSpH|UQnNf>HW;yj@(Dt(k<&Z$EY#_S2
zm?w6u6f}x_{~jh4Q4SY}@fx&j7A8fyk3Z-3h8Cb6w7y7J8c!U<Ao59DtK#*@JB7%8
z5cvKk;l~CI#Q=J78;;ihamDb1EVeiGirRh4DI6tHK*PIHY#4V0RQ7hCRTR#Gf`XJ6
zj*}kP>=f97WmfV&f@Bg|G<Yzm)|W0=M-03iS`GV>bjhp}ThdE7p*8U~%6AJKU8zwx
z!v~7-W-w%(m=MVYCn=U>@%^f;mhBA`>#jlc64dUgIGEuEkMzBSz|)~F7k5t*1<whP
z9N&EwKTp9e1VQ}eO=k9lZLUKykqqyy(9g>|!q`iHJKA>^e#%wKWs4%cD=JFCKcMOD
z1*QO0QM%-@^yhecd-F6l059{FrL7eN*CrE2tu<oYs4tGCXJFi^?TW9gd7Z@Pp(C=M
zukG7S4thhKY|PHa(q&PSz|<RjH@(7|8sF}Zu`XQe^gkq}0sY8INYlEnl=JPY&}euM
zEyjxGM8G$U*b`arH1--FJix*|#HL1G;R!}JMAufCBx>hxt~4tC8}87(R7jHLZ)i8m
zUS8(a;bR#Z9{mn4n>++r*^?7KzQCs1&jL<_lReZRpR<0UrTvBe2OIEmN}?Xi_X4(T
zcce>G;zEy$Rz)wSdZzSiNX>GVw-MoWtHxvRL}!{eKKr*375$Qhd)&j_4-N_VB$Z1I
zzP+g+yRGMYYogXgraCIfiR!GjEcT7Jj5;UsAN{%h5Jp2u-&AEeZ-iLLEji>?W;N7i
zaa2suPu_3L(b3sY*1*gDD*BMk77<s?t?&V7w@|v@b_;d>2$~M=uX<rMyGzt@2I&_X
zx&2uqo&$wIDv``qr6!0~O3*R;lD`^mNi<;uR2>Diha_;NyjqKSgXvCNy)9S3O+u?8
zH;Ti^JQN@4(<>XW)f+J}auE}%8MD~KRk4j%ce_joS<J}<)g4peR|lWHuOeL@e%~vv
z+}L229wbs|72`b4YICeNGQ3*Y(Cx=iq!y{y7{}*3QG9)}^&#c8oLV#=N7!^51V^!+
zZ6=qzfxc3(JVCLhFo%kh$_*Hd=6fyrbB=hfx7;@b@n|g$ePuiA`|5?Q{hYxOhZtQx
z?Vd(DQt#_qF3or~H%4?MMl#EkU-Mos`~F_HkO3A{fnDfJ-_hP>N7)#dGJ*aN_#g4j
zJ-s-4WQ206ek$I`v#hp!WQW%E{iBZ|0}nqte4@3sJwcWujExj!zq@p9x#dR8w;k1x
zQn#lKCO+r#XLz3m;9MtIjI14_tPGP4Oh<Lr1dDLqD2x~=v=ITW#2m-!Fi56<*u3<|
z;P&xX<SxD0F@-oa>?SBH)$MVAZ*q@80NFl}+CVVY;nsjr;MVC8c6bj~fk?6sClll#
z)l)bWfp=?6%$9w#r^}*2Unh~<1iEwCUy*f{yD-5U(xK0;Iwck%GHELN$9LCXDS^^_
z`MB}G&GO4#3gvK71V!dO*zP}Q8bjNqXs)$i^LW86-I}W0^6ySFn&Pk`$zQhhc&8sV
zn>$4EXJoJ;FFN+WmZ(5idNz0uY`B#>x-YKNvgh%&H5rd>b9%0eC3DvKf<%zw+}cIn
zAB0PKBKFQ7kg!%~r@a8#<mtwF;C}oV2M~In+Y0DaNo#L2ba7#7_5}m3=~bZR8A0+0
zigD>f^s)GG5|+Z;)wi%lrH<z}*C=rQ6I`5j_hH84L1Qv4+&8E1Mu578A5Nu&TW|!@
z=tUT%{D9M|?hf3!I7sUplaC=0B3|QcB9v;nI*Is2W=c|t5DDuhil}arF7OyHfy}wP
zs34ShVMr5*Qs>QA(B*7j{2gE{YJ6l;NL-uf&s{|^h23DnhUAkqVPsbi2x2BTchGgx
zft7Sa>a+`#zjhPsv37Jq#r<A=JFs(NH;t3U3HE5x{G5t2BlY>aRbo+9XCp#%O>`cD
zsA{$Q5`diSrR<+DU=?6!CF9z*2pmbXRfYY!Zvl0}fhtBYLMw=xWmUF*&g6aC9Ie57
zQg4{Se84fhhGx8M3`{|&nRVLg4{6RsW!JFEi~yM%Gk$yoyl@-~=iFy{#urK30i05h
zvEqL+cLwS^PrH}3(umRFn~Wx_%yWnF9pibdiIy=#Q@rE8O#09wApSIrGI@4^UJ@%e
zy_YCsr!&%%c1$@i;GZ^$4M7SvE19sQ@mRr0RE=VaBaiQ1I;UYY`<Fh^-eo#>3kejh
z6i(Jud6a)A;edN%^YWmuj8$-&&FrcE8b`Bzk+v(+WYkG6hWCSHY^YnRW&q!upN`el
z2*y?XhS-tJEyqq98r&u!wPAQy75IcJ<j9HfzKf4WfFs5z*!xwb<4U}N7l;b7kO{i=
z8qnUJz;FCCjLeK+PM3D)({uXcNd{5tL0%QjO=GNv9Zt0@7a5|dSG8sL)rmCTrG9Wa
zh{Zk52#B@jYmmui;ufYd@XX|_p62EWhVQnVnt;k(I#_kF>oFQJ7~6c8wOAEoSWQ9)
zS!nTS$Z+iAX2i|5h0NbK0a!whROncQ9>HzRVjvG4c!f2h_UQ>20kaD)M)j>Lu4pEA
z=8DohI5XsA8a>Nw!_Z~ipq8!636|82kwyG8RC6zcl?U;pZViSy4z{bN>^S!ZbJ*3P
zjY5#^4Q!OxI*z&}hfIK#uSsgqHyyxMiyO=Jxi**&3vypAvF~kqwAKqyM;k`G+p2dZ
z0t^_#Ossh^r258A0sF-+oa|^CTIU#5OwoiwG1yP^?peRJf|KI|LyxXpU?COQ4Mhr1
zHZ%PevK=Z=er=JI(mVAtNculu;GZ$k)6^Gw6EJF`cgJ7Gr^!#^OaF?D4B<H7FDilr
z0ymrH#R<sxsVbp5@y%<14uXx~!HwiBZ{E|meIk0G_Nb6+mng4gmN$64YT*rR^G;x4
zU&(i!6;LJRhC7m4e(9~fqcxz!OcZt~GSq|Vy@eY%UuS^=@u>HP+_~pIUY!8hO9GiJ
zHOP90oz^MyRyZXc0QWpa_Jpb4BFJYQ_#C(lV#sLh3wrkVu;H($`w|4VsJ2m;FLGDl
zpa9$);n4Wrnl3{ZpdHoBc8(zCN6-(l4lcE7WO(-3zM^}0t_h5b9*4+`?5i-Mkx%1}
z19d4&y{rVp?QkFW{;M95P4W(uQ~As`2?8@#l>Qi)4?lg9`!OAqKL^8ak)b1y57m0`
z@Oen3e#X);NmA(^#~T~xFMgosSKfhmPaFMY&iGP#33B*n4KI`)*EvxEY?m^4B6Q^e
z2M0_?*EU~SnD%84CTi<K6;RC=zaNxGc_;G=QFh-6v>&@YNNpgROBMr^a%Cu2uuur9
zzB%ZJ^=DnE!C6BG))v@W6F{$X)Bo%>-l>eM>4vIsy4;d{LFG^@jQGev<VqgQ0Ls8K
zwI0f$(s81HW4&bJWA4M|qrLu^3$U?VcrZNiSW=#KkbmBP^Q8Lr2Z4Bh3@X~%UO99-
ztA-)7=_zMVrR$CG?;P~iS~cvoq0o=v<+BD{pjVyPwdcQ9!@q+BtQrm#kh#)jsrm@p
zlV-%>w2nC|lK8JOdII}c50>M~DRAygW%XNi;(6s<AEs<l?s^&|*oB{8NkAAATX_x<
z%J*96Y9Jqz<|GXL&doF~E2u_KyLJ0cZQyt;h6=vm70-^@DPf(aPQFk#r;nVt-h2yU
zPLRBOHopo5#QL36Xo@rF!^58{Iw?kDIVtlpQRhMQ)%|)oYtFn{m7%wy{+RsFw`-FA
zJ3RW&$oDou$c=GT^}<!8WD6%N%igtn4~8EgrJxV20fBEOI3P~w7lTc>YM`4e40l^2
zIY21{R%=((Z+Vh>Ti(h77JL@8y2wJ1sqzoSC}Feb$MV);D&Kq%%+0adH5G25=5W&b
zzPQyLS*SCc9^s%P2|4i7se&d;@w`z~l>#}sWH_9oo<YQT-&pfo`@2otS?HIBE98N?
zqJ_b&EAPB&plh%s)<N6eF?wgcl=F~E!h;%CweUR>WDBqDc}czhJ5>57siIZMMOg7X
zZwYY+Z0BzyC$wg7E87ELvTsumfG{>YD{tah3X%r$C|Y0@G@W!I;{hwn?^^=^07-`<
zwpcR&Q=9qRz+p>(^0ipIp1w&c^8EJu2Dh6wbc_xsbO3`e0~}q=W1E}c%Kt*cVI2cN
ziI*GX0&%B!ws655K0@r}&dgyNDr;#aYe>K7TsdEbmqii@Uq{T0Ll1xJs>@HE0toP|
z_-cH-DNK&<1YvsNzNIPznE~981=mA4zC(byi%T)p#PP*8|6|Q3ux=44&8QI@kp(Tu
z?qsoww)l7$=hZI}Zn)RpjMV$Cy8S8xfR>VDP)FLI;og=ilrvfg$JH1Ed7MC9v?3?1
z2~=8RBH1X)4(|OTw?CJD{GZ)|YNkg3Nbi+Audz-6i)f=JLR^)}$e@Iv5LyEQt4`FI
zeg>*=g%mtvJjhbFeF$K*Zi-8R`H@E|WSQeQ4&cd?z!R-=w*FLL!VWD+`1KI*N%xu~
z5-(Qu9BSor@bBNwaK5@(?|6PdfH{iGAB#p<)(<TqbTrm#jE_?M)0L9mDff6v@*__T
z@=ri477bIn9;JEg+sO*eKyjBz2kEhRPrKKZK%<y|R%BiUHA3Kl^6cK7h^9{<GlL(1
zZg#*prRXSBJ8+YrC}Usj{v0oR6)sVPITC`p#E&+EN{zPC;zQ#sSp8%o>5HEp#II;F
zU2H3_H60_9RI~bkyX53Ba~~vXO^?dqvRUg(p4wsE#;L7iu4S|%{6;*a7B}w2d_6T1
zf3WsI*dTyfZ?~0(T31r?{}W{3PeXaIMP+(C>8mWaVD3^vt*Uc+V=MV)oglVljfUXA
zl0_Xc=jtHK!+M5%nrxvR(@fue5&Ay9`!PeYCBVAo{A?9K{UN`$3`~CQ?!(G_o+oT<
z0V+skFRuhl#w5DZDzo3iuhhz~?^a0<>fM3&n#f2e{cKeRlhFJRmrFAtU%n*Pw#HUv
z_jignwza9OS!jIgdH^h0|F0}=9jE(ptJn8Wh9QTgHKHuZTeY)^Lk2S&)f4Pn6RKI2
z8r!R~yM{d{@ZNMgn6+qzN5=bZO^Nj|^$JF5ti0Z%RhqE64BWM@X&ogNiZk5zvX`LS
zvUMDklQB=m4dRWq$awU)++H5@D+x5|xh(Qb`1SkUOD60-)c37?Hn-y7o<obxa}GY<
zkvslmxF{*$xAUJXj4~9l+iUw!l$6-@IQLTgtuHIR{u^jFM+mcHYU=W^KQ@l6DK@?{
z-T(tz)h27ev{Z59huXK&97|pY;djER4Sn((OPGba>pR~_0K)eSFzio<5wK<pBT?k1
z^a0FRO-Z1Z8KSK)!CgakXC^V{vovLf=~K5lb-*m>=$)e-;70rmY*-SkA!5u2F|L4X
zOb4)L*TjK3{mDL-;|1^lA3^s#8+ljqDB1?_)$6bODF;%W1IH+N(fP=fmrdMCGl#6L
zt5#zy_snppGzh(N4hY&1uZ`-v9D-Recm|C7c}Oxc&|n9+8rm${lDua!iXHlAE~0^V
zof*1({Ue}8T3EB>mxx!*jKnWPdwMnH&Cmn!NMaIq)pTh)ls4&j56*Z?Pl@qqfn(TK
z4Y;=Qv)tarmM@PaOc%tz-=eTCV;N)RSpryN+$rC7ZiP^c)s3{;q=K@1{ce<3RIqq+
zLO#t|b2aMw&)n?yTj8&O-E382(K^M5N-`zmv2MQtof7~8nD|F(%GPh!M@((i#k@U*
z;>gGQzP#Ua)E+Pbs-s2;+?Rq(FSEek%s20zog<X|h9OzrXHJ)wFg$Y{IKiVRUmtc>
zycfh`{xwX7(trf}EeY;%<(}*rzfL8h4Yw-4cuyo5Wj(Dq-4_o3LlF;WWZw36wy3cO
z@7yv%-OqD?r%NgDApoc4IH*zA&Ode)m2;QR$YnO0;PzJ#e`E0-3b-toyDYBFDsvXi
zxXf?9OV+271XWqeuIH#G>Hre?p3KP4&vh0l_wdFEP3J}_2eL(`<P2VM=r+9W?W(10
z3;L!711M_rpzFA>H4y5gEJTO#OCYn^>{aDO$K$KG>#)~f!)<GOcE!f>h?LBSE#%C}
zmSdunQ{`Y8w`vxtO<K!Mp?1Ay$oTI3;tg!8gERCs1QNLXF;*@Xxdl|O%lY2^*yt`8
zeVv4|Jl`FFa<{stp8b{OZ5BLXtf*=8Rb3>0H@uI*mVl)AJz>VX6_O8;5f5qkehDCR
zDc!>K-nJk7ixk*T9l1wLw!qnm`D4J<Ew)XOx7*ae=PHN`(}`h5sx%tIImO>m>WI2L
zq${ee;vn14DhE${=syLYDYR~0B}&?MXv0{{vu%y)pv#yRBZiqzK%#IKGVQ`iu+d_3
z16*s_hUePIEv^vl&E}Vy!Y!hRE7<dAfs>YD70}h(t1n56u$}yxA5IuQpYw<%?X??S
zUJ<ks<xM!<oF&>6@fk1!qj|xYd&5t#N)VfS0uqe*8)@9vcs~pkmteUtB-onkWYY*R
z%`UEnI<lIr)CFM6OrYw%$%%~zOxtAqd@JeXORBa5?O<v*NzNQlQq~wz(%~;K4~bD!
zb6m~;cf-TmN7np5UU8xrip{!3e9l>wgVaJ;ElIk$?{HENG8sV~W{VU#sszxx>*RZ>
z&V`UM*S|Ap+Y>y+rD#*nt!RaT4$FFei|nM3UtG>5^2k+2QB&mhJ)OP=%5Y{+G0_(?
zZHl2}<YkyoCy$;j&Q`Ndw2TvFYEw6cN-#0%V1w5@fXYwce6@PSUQ2>W0|s#8Dn*N$
z$q9HlhD`x}l5RH*Etu|}uB<7z?xTA{FbG=M{Y9?9>rd&tSw2Lee4*fT{>24UBG_<=
z5U1n~1)xUQ{IB$B*NEA`cp7_|?L+!3hPBf2hzu`T8xQ21<@kEI)(Cn~)jcm3nm_SZ
z<nVW;03@B-iclAYkO?mlqI%->q2RwS%4K%=2)QCoH%Lg-(N@RHd2#*qWC}T`Wj!gZ
zvK+hzw)&%20j51RFQtB>YrpcDWKTE^cNWhJ!ozAu0ID2uvn%B9FQItN0Ocoi0iDfI
zL3zMAWpuq>eb?XV5W5R3k5x<318C37qxQV=Q5~wkzoqUjgur+wzK=wdE;&KY*-EJ_
z@~=br5I$n~Vug+%YU{&e5Dg=tWAq2;q8NGzAF=!~tz!&$Qr?^l#$$dD=$rn5FwozG
zW><V>(FMFREaP=3@NTdQ{ja|Om2~D|BYd+Y`V8T>0ZEtBMTURX(wp(x-;NG`ZGzE_
z0tg=>XieXoecbfdb08ZP__1ng+97Aqk2yVjLhH@=8?k?2OF-(h$>2taib;bY6m}5!
z7#R`6rGI~hHU)gf-V=h1_AHcBw~K%zF<?s8!TkHiK*>5C>P$K*hQQAX5U#mel=u1J
zr+>ZM3-A#<Bei|q!mvm2S4-a&uAEGD{rg+OdeM`5tZ2)^UE>Rn(cP~UMf>-c6zIdg
zow{;BHKr8pF+~yLYb{>%mH&K904j|JiZPPPz!If!WYY-YQa=9oHhoDCYkKH&mA0)0
zJV$ETJF4NRbQFF6d<qEKk}`Bs8eol*!(e+b=v>y-|9g9=6vMl%XBhVE=0?9o5`N2g
z5nkwDpK%#}%dk=KX;b(uQs}pQ;EdYL2!B6|GZub}v8`}}kr8^i7mUYJF>WfMf1Uz8
z&nicaL?;`==C-kdb8N&Qx$yUo*+S+E%YH|K<KZ?ai;F}UadHnwU})N3533^#>ya5C
zBL`_}1?*EFhw~Fbe}72^tjEFVa^|D$sFpOkMulll=Uo2Rs}n)Tvhk^MH=LIPuprio
zg4)~v`VQJUJy<d4XT#vRiasQ&^xan(6P4e+o}cS-5^a!js>`tP{5>Iyek*ti-t#|o
zo<i~ySk+S+29fgUGd@L7Vf-|0ZLk0RtpB<F{m<X{Zw3E93*x_u&daa1`44!Ev`v=I
z@0S9gs~?<&hKI#?-cx8z;~Y~MF){}K{wc4~9;*ZsJNRdhs2thu&Ac@F*H2+bSN7m(
zPK_K?v)agr5REYb0@8nNQzS@kJ=6VeN7R-rWX^^VPen>ttxzENy(IqXXw}CV*f-9g
z4|xygcG%`^{@>@8y&pk6iwtSLaBgkUxn&qx`qwUOMZ3^tQb!tg*eD63EaTR{pMhQ5
z9iFqnz7}o&Yj9YGkNx#N;8_|(;D>fkcS)>l?yud9xP^z=|K&&ql8}~h03|{C(`_A;
zS^j<w1@|5BI})c)WJGWwf<QPGMuw06Kc2}%BdD+lu)%m+Be?#KgbD+cbLx7YS4>qT
zc_AhJWIPt&`su%%Nw^j+K`f<u`X)p4&qKiDaO=W(v$S>2wnKMI8z-c-?zTFtf7TmT
zzK#cN<BVtgaJg7MhQnfKhy1Mn`jj$GSdg3|`eC796o4c}hy@SirvAM<R$<ZhUL=Ko
z6&44d@p`rEz&~#lAklO1953xRd~z|+6k-@T(YmOc_~T!{_8@|-QblT&Nc>OG!l|ct
zs?PGS_pS%8HB~j{tMv_ditw{vPP7KLhl~Fe!mcUgpjFQjRxX1xdk&phxB;kt|JY<l
zGGO!L--fVcfy_P^j5AU)&tyEJ#t^kT3omNTBO?~Xo9G4hH;f%I%mk781emaEgY(9n
z9Yo*+WWhB0lVE4AtpUma$QSu-WzTZL-2s7<r_rhKCzk7&a`Dx;3e)c$jaGqOA8pyr
zn8!V^Ji_pv3R8Lb$$zgfU^+lEKMUZW4zRI>bOwO?nt@nkfT$v>ME5lG*FDX>fVwsj
ze=+IXRva=|ZC``QZKrEly|YcMxBq|BkN}=TfC<ls`mHJ=YX3bZ2_?XR8)mJPaYIh&
zz>CWSprAEpKa3+c>HEnB)a<BXC-oL$lCVLc^j?=`v^$FEL&jq^@{4u<y1)o3@bZt@
zs%Ju2V}gKwf495%pX+>wra34BCV(J%7ApQTK$=aI?}#}LJ>R1p08?FJ=oCnQ@Td<r
zDf|qa$mA(5feU0N6k6^p2OKI>Y&)K$aFeK9_i8$69vc2dIuY#=-Gs_Ebl;wQK!xc|
zXI=i+Hj{%x6|MCYYXYxj54;MA6wN>PAe=a6aJ7)4xcKY!Z@CFdj1|Sp^$oDl>TV$X
zgW^a7d4bRjQgVj~spQH7`y;;>zC6{O_a6H4-3i%kBJkxm;mfB-S(*M76<E=iE8oNF
z!IyKPFMp|-_9tW<r}Tl^E)#Umwm^dMt%?;j%b`B<<=Qp>QLhOA(scDQ&zuGkp{Mz|
z>;PDS45$Tq?RoQRbq+-;!5(b8rE%5<Rj((YT3T=k-TAN66N)1I<G1*OumyKR)MPr(
zr~Bt*B$J`B>!?@oSvD8gmC#gc6p3{kf&T0FplR_H=$3LXX{1B&9`E4Ibw&B}?$Hxk
ziy6ugg#QV9VOl8i>R#4igs^fG!hrgdNbni^`8Tu;NVULGeF7%4KP3VIszYvN$|AN=
zmP^qs5iC*N!DL8i+y8qvshWd6+!m%y7Pr7*Is?_!vqr1Lc)6?P6Yo-P^d<rl`$By`
z3N>6(hN&iyEK|rZ;y$%v#83Wx4IM)lX6XLNOh_=5P$0RPsQza;&Cozv)DO@DyPU<1
zbKt)D2}*lCIKiM{t|EP<A5ahC;uLVE*^LOf&tQbcuvd~TNG@cj@?NbXGE?UHSAiLb
z=>pw<mkH>7?_zr?QgZ422?QvLH0Fcx>th7oR%lvt?^ABr+e!op>JPVwBEac&K}Yqh
zq8D^2fTydvFUMJgoYuI1IKT04N>%kkvN;u|SGqP^y#XrPShj+0xxW$2<U)uA9Gh?l
zcff^_bzVSWUSsOF;-g}^wD9`&o|Is;<AH?tFeJP$Ot+r-SHc^CiWxdu88U#~i{J%c
zxA#5xJ9}fI5Bm}UZNyAK_5_v_)JbvAmq!jfySn0jE)#Lrpk-=04Liy`u}LLVTAVDJ
zGG-arK$%eKJq0*69eJ)LEC8)V56bX`_UAC|TJ^f_%Sf`_yI7AIAR@DU#I;zUc-GMP
zE+f|Syv|#wySr()3n+u*!_7f}<dq&r?`QV|B4Y;lzV3Ji&I&wMzTj-y0Vwj7>)<^i
z@lGPNTN~QAFp_2II97b80Sw#G12$tc+1YrTSQGctFa+Wm(6-K<0Y*`i?M6vapMRoZ
zoi`jX!t>iy<<}uiyZ?}?Dmi!SpI;tw&Bu`2jH!O>2I{0igMQ4;O6uIU7*VqsKg|5w
z0}f9BxbE>YjCyL-T<iRw?idQ6Z!w{6)E{qO%*>-`gcGAM4!bU<CAj1F`BD%k<Mtc|
z3Y0Dy6&>*<5tsr-a8A`w$7&bz5$$_`)>*r>VfiwkDc?-9E+)uBF$!&G9ca4}`zci%
zL)7W6an4?}Oi<^|QyxLhL4>=_gYLKSkvx#nG@BpDv<opTM<@9Lxo{9f@ADs{qj>tT
zfo~#6IO1WB%GAgAQ!x6?Ko10cg-^igJ`-%2XEyk%=5MB--0F>3C2n9#uso}dFG)s`
z(@>>1mHZI`c7<|b`ZkYEPliz6%zD#k351g6-sb_PN8!%6NWzFqG279>wwW>b7f{yS
z$UkNfd$|514=~WuH_Sjr&~aLsB^9t^-8Orh&mv~*8lbwP*%@9#Cb72Czi}3IEDeXG
z8_9+yMaD-AN%OgrmCG;2j02Rf4Fd<R_M}F~^0n=C!AAOs@tEn<Q?Yr~^O)sH0+9@Q
z-6eRF@4gk46fPAtzJT&=ZS<l^7Oa!d<0uk$I|*+8P+@9pW+uN<nmapufc&S5>*@v}
zwLDqWPD-9xxuzmOIo)cukS`sVw@xI^AkLtl6F^D63ZX(Oc@4n4oywKo;CgWaHA@OY
z4|H*T=BY`8v&@^Zk)s{hbVt$z{17Jq89j}RWM7=fZ(}G~2cD2F1eOA<<*i)UUA-a+
zOiU5#eI^aw#|5(D=_sedukiw3yzgRt-o>^(;YY86Z3?<gi>nq@wB*`&+myt42B$|}
z`K+VdsQQXpGqFN{jj*v;vhghp<4-~&8G=1|{LHqT7XhYU_((c`X5-0sNfGl7v1B92
zTL+(K-gaM7q``FfeUXQTrN#OU&!RP*d`(gbh#z`G@5mPDIqEa_ijVw&o^)_jMv*_J
zZCios)+*pV^SrQ4O&rQ90JZ27;MS%TRg7rhyaaefYnu+kjH#Le!-uWazY=zU6@nnD
zxbjE-$&TiNu2t50!W8e-y@Y`d$OQ<sZ|fcq?{d8mvW}qnI%JZh-sTdr^EOX=0Hf)?
zG0IxAjEzxf>gfT+(l)$?wYv`(=JI38BRsdP;HXb%`Td9#+Ri&Jqbx@s#gYGfX^_ed
z6q(zLW34&*YFJ&GkZQsN3^A<VmyE%cVIn2*ZRZ=F6)XZbTl{FO7To--_T_<MuuD(e
zC!Xw7ebkc@48MdIDv$c4g!gV%l$u*bTyw~O-LEQV6y9EaE<?0x0f*c{zi(V$Pd2_g
zix^p*5cjqkCc-X9ncZn<lwCHzdXQl#c;8|pS7Z>Gn+lB(9WAGo*dBk366aCBVKtHQ
z<D-Cx?W`p7Uoj6oceHyE!i%f^^4I%n?4N*FGEIIB%bqYa9wI4vv`!nPQzqv9Ck%gU
zAJ%eL@j*E<E+b23DXEXw^K&L$ho(nfNJMA7I0hHBHV!g_RA~#EKbb)`sW?7{atLXA
z0WHma#HxB7YpEhk`6T>&F`k^M{Z%vglw8S`fE=0N{VIl?)gvPL0R-EXS#@reH;FH+
zHo$4!U>A#QhsI!QHL*=?BD=6f2YQdhAX34IWO_kWI0sv_#O>nZ8-Sox`6?o%(JCS|
zC8vDipi5m%F4?y_M-%G^*3c&9E4Oh87KDUqlw&x2){!9fo|35vuosB?1rGR&@2Jz4
zY{TRpt$$T<K;=8c+$S)N>pjh!Q#;lX4SMa-!#<4lTa6XPfmkf1W@}kOI2jv@OOu%U
zUTCv(W?h7gI}LS9&b>WJ>Wyx%jetl@TvGV=v!>c*?ciXsmirzb%jR+LHOCY6HEFx9
zWYFrReJXLtg&Xz!*>s|G(xhzaHb?3fiRc?IgwTp^hZVI_uz-x{XO#&jYla+0$k7Zu
z5-M5$h3|de3P{RYH|lDq`21SywGB&p-OrhKIb$yd9+tKl_5k>K{m>{u?syt7H8O#x
ze{k8nE)nlTcz(VOkDVxL*L0DN#yy7<7m+*UkXEpL;V!d`&f2ttpG45@l@h>cCm?5>
zYv-F_@krxtGDO8*@i@OHGjyHy90#7nS33@9iEid-^n!D!jR1;kvbLMh#CUw{ZiDB0
zfVwI1_vX!+RibKM{*qf%KCv)Jc;yQT<AkW+a<JhUvIT=X*-7Cps<?JBT~6o{^K$}M
zm1{wXYAU>P=&}^ks4)&N9$9%aLTqBO%YvrxPCxyQ*DfaMhgND*ywY|!=Akq)$_Oxl
zjQTEs2YxOWe_;(wV{(u$m~2lE1?I+?jL>H=q^ml8X6~q*S8f~BZEkQ)#U8PX!IuHH
zZ`aor=??VT=>j4>YHZ1CUxK?G=nKy7=Rk3Epi2VOs(xcak^0$-u{^upNP-C44ch!G
zLAwu(!+aHN-!s*`DE0ce5I(K8p&YoS%09bCRa&Cb26{p~4~6i*zykk>Z9@Z}cHO!K
zOdDn7Z7T~VeOt++#LBFCrq8GY??4yc>*9^uFr3Qpvxegl(e|f#hFVai)XkVN^)>Ip
zW4ZH`Z4a8LR=kkSC-vr@4&zW<6@72nr%s4YSdol1gucnP!K$3dgv(Qpzo{en56TOU
zLD}&2G$#mHex}u7%v+YV$6?^)0uAqRxmL`gaXht9ZI}lRJXk~u{NUH?Ayve{GtiOb
zLc{e9T!&AoWyjcg*o^9js4+Z^AB|n^l>>X>%%fS{kb!u#R36_dbz!zY2HF5;fh4ui
z$nLL>Br@`5pEj@^6-CS8awVvmc)IYYC>7xfgKq*%%l8ER&Q)kl6asrAa+RCt1wr|=
z*bS}hXgg_LxYX8yyb2#Q^|NAdPar%u$XkNcYQR2#atdoLh@UZNb8`VT_Lc3opHGFe
zb5JsfBPkZ>3<TPC5y8%^6R0*e?xWJDNWYnW6li!}cit=Q;E9(zV6P;ffn`y9bLztH
zW5&+m4^c>vDa1GpmC+FKqLw|u0!gop>t3ZxuoR1X*ZB!V%S|q^lJ14zn_=p^Y*k?k
zqDf>3qbfHBbkh)^e@DUz($fiCYdK@C1+v7b4z;v`g)qx&bn{ZR1JWC49jc{LmHZIQ
zod4y-##g*K^{?IAAVTY49y-KMGlnFq&Fbr{%aCkEGjqkagIcQt>B?ygjm<-Sc+P-0
zAt45_L^>r1kI=09;=AKJrzkz12?dlw>D>tz+Lgvv>V$2oE+%~Pi(|?|{}onbciJ+}
zGzx;2MH+6lBiOK3kk_4^@*iFcCOS@02hGgP!8NhGMUHSFX$bQL%|@yTGURQB667fa
zG6x1dC}BEaKjl~&h%g0ag3l5ZE>7;=oD%UnreuIv(}~AMu0B#rZbz-N>}tQ>AK2Eh
z{I=-M&ltJy%RID`#!lW{t-1Sk6Hd?}(COd6l@1@uzEP)gPBN^LehJDNr08H66Xr2C
zVj2YdFD3s@n1ykoL&R^iIbpP#%yEyJC>ufig*tH32^Fi+!v>IDu;h`E)zhlAZF2w)
zd?zp?hH1umdbcr3gAK6>1~g0SrNiPKv6Up<Kr3F&O&JP)0;-I+J{+92(d44zBOXI_
z;i_-vW6Z`m4I2*ou6iP|`!g5;&SPejh?~e*Zw!Har>K<k%`=S<nG3HekXh!z@%k}H
zy9txR{#q9wI0bH8IQ4d!S}WsVZsoCvluyV?Z=)KD+>?;;YDwC(odjd3I=1%SD4s-!
zh^w8*=l<D-VAYb!Ic{<6iX3Rykq&m9)_D+VE31<TM*)W6c_B>DJC}`Zk(6;5ZGAYq
z!8cvHvA&W8X01!2_4ur*Iqv035Gbim@Kl2o|1^PsZCOdMFRz#XXt@tFE^ZW9A`wY^
z&LjF{X!QXS)?DLT-rt<qV<8I271&lZRx^Ofm|C?+fzTp}*9)j21gk%Cm3`<Lhjybr
zkR4BD<V;!`lv~}BY?aUxV)}q!@oykSy4QO9%wOe5kE!FZCsO?V0K<;6Qz@XJ(Bzr@
zGK9IIPI!Qf-8OpfR(u6U%b4F>5jcK%Jlink`&2=J<(@4VGKo?reER@b+UnwNx}Ds!
zRA4GH>7YjR6W`m##dfV12zJFw1WowNmO@|};$yie?IO;%3^@#2;*=Xh-O7;KZKi}n
zl3z)opA2RL^4|HucI3~~*^oiEn{g@O#D;}mNAJ9&NfPTF!LYOBrAsz$UsTMiRy^)d
zUz#H-UJ19DFlW>9GokO&s-~q7s;G4uH3OQ1_c4|w+Ad{sY0q-+bF)xjQ)WxAfI_UZ
zUG&h@`T!Hl&A~M^D{BRSqW1PJN8AlV9tW|1cMVhY)Src;G*68C+~-fux=$rHi2i$H
z2b9G%INdaqz_*JU++h~&+uLT1TXktIO7&$Yx?T%pT<9><@sjO&%PJK;-1&mF59;76
zlipBHc*;KE)>hK>TkZtSpRrZ^ZpB4U9DmM}pz!Yo%soXQv?LPs?N7JYRuF(WLv(!?
zi^*YK?2B|U4pnyeh)n`n?WHEg2AOy0I;<*6d2kUtohYV6tv>pTuZw-iHJcqz3ao_A
z>X9%dKaSDyFYH<&10Qr7xaU8<aO?xu{@@26W%o752!F^Q#<lyaC9x_AJK9Bt&e(r$
zH>3IVweBu;L-73%Qx?9>Zizb^6}H?lJmeMR7u+47vH3be6qeR)ViLKV{z1aHv0q?}
z?5<;GlcCot4Q9V_t9eOt&GOh+ep5O}>a`Gi5w|85XL8Y+IA@r&&My*H3(PeuDaPfi
zV<|zT@B+iDv*p(XBBa9i%9iqmX%!Vz)P7{^-quzzd$h+DC~#DGs_N<}k0T?q=axI!
zc5Ls<DytR$PLLx&=l)Qzsd&}&@%VbdEsM-9qu(<l&ZL=d<1P=!Qk6269HpWDeX(lq
z-?C@th4rLc^CuLpZ&Tm@J<FT@TVaIxmR?<K`1BjhYGVkH;|f;BavhHK-pft*ru{}_
zRCcc#O05q2_OVfFS&P)8iJp3@4A?1EiqtZ%`^^@W=qcL7hUErs?rECw?*|3UCxso9
zcivZ#cTdx~a`-ch>?6>fyw2Anz`SJHruy#AOvc-pEbqhlTSRyPc)Zf17;U=_F4sl1
zF*Q3dfn~jBi*GNKU`+upDgLaZXWH8szDs*IU+=EMX5H(zxoc`WKCh+;!gWrkPWV{$
zsf!TpmIew)-$ZU&aF@1PI7ElPP>eN&>|@BN3$CLz3zfMLnObT4p0Sd6r-*(;e4%+q
zNxm!oaJdJww@_irs%niUq!htRjqcwcsnL{N%-ve}a5>tgFRDG8YGcB_2D4^$Uy^^?
z?Mh{Xt&K#81+TGtA#-gG_hyd%j&$TX)B3i5I)d!Wq%gZkgs}K=8>2C;Wa%g*C-XyV
zv8#y)2kL%VS1Vfp>cphN$sVR<ZB<f(rhboda$64<b5l~gp{=r!A66TVzO>XV|Cpk{
zO0a)iiY0YbHyrv{IZH9iHQDP*Sbv%5+S;B}dCs-HPT4MBj=BPnoz)^Fqr|BH*$1o6
zzxM$&zYbS<J5_w&=54L*^g9WSQ~+uvu8YM17*e~A)v1bd42W7HA+*ZZ?cpZ{d%z=j
zjR-FY%&*BWF3u(xh#s&^cQg;3;;SmD(^sUU=ZKtP=mI9G(U&;;xnR6unRDt@CXSVa
znjPKNyp(63fKnV2LcBG&ePb-YxC&K<r%1){>I6=O>#DA?Z`TKgKJHowKWwjXtG*C#
zw){gC_dv`PfJ3ctlOBcrNV9&hAuC9Xs*(f;RCo>d1%1<+o8IwQH}g=>MJUNt6rx!R
z8iG43LWLbECFrs(S08VqynWl|#wEB)XVP{JXB9g-CO+^OFuxCUx1}njdaqB_etrqn
z0zs^@E$y$8moYNqGkI&%xF=%Oyv{8RzGy<0xjgIvoLk!~&*hKzqH39SeN&m8TT90t
z1s}6HN2zk>16QaWjn+5gkGG8izYE!PKTSAl*5+%si7+ggY|0y!{y~rNN?&?hT6Y~w
zZ|pa3YXp?vWp$B~pm`jZEdz73)?9I2Qwqn`lW+UHOAFdcV<$m{%q6@|dbgf$tIzc@
z9$IJNuW||MXDlCY6c@!%mw81I*VX%iO2^G$V8@g0hT0qUvfZM`##35IZMds^&hTLZ
zMGXsH`=xVN;gDyd7N5W7%+Xp)@1Vw7=NYQf(d45KS0o>Arx*+MS3r+j+Uxa`)CZ-B
ze;Wl9NJFQtd6A%kQsL&@M`=2KT2^~yLKMgKg+6bGl^Sp$zZYlH(Ga{{Vo2{u7e^8a
ziEb_I3Nm(VDA#~M4`vSNa11JK0K%P{xtvqRo7Zd-mJ-@iU4;+sougkDSGj&i{oT!k
z1rP`BWL699oVE|SY<m5058>_s>6|)WV2x4WDXUAd)?U4`^=jPX&7gDkKW1MR$pR%c
zA21`o-!QeTD|Ix#OFYO`#l&L~;Zol*{E|xaE>7h;C!bG_+69kGhp>!SBF<N-vl0ZW
z)Z%SsddmE-6!ccfeSCUuw4T;wpm-lRi<(zo8ojiXbwp+Xh?E*1)@MO(J%xQ2c22G=
z_h#-Z&z#AV!dtfmTf7i1Jj(XZX`iWTPbB~Av`dl_1|pV^)UB<kjSQOv9d#}C>EWq5
zSP|(Z&oR!k@06U9GvD}lE!npE!ZpaVChSGTML+kZ)QE!Mg42f1kYiSSs5jx9J`>@2
zY5i4co)rv?Z9E-G!zOpkvBz(1((F{*m99LFxtw-k<)US1*$6q_we%*ZZaHB}aq?)W
zCCeUCVB^Fy+nf?1`W#8j_|+GVXQUush^rP;7iF{VUQL22=fxl6FQvwcdZ`;N-(#p_
zU#S99R$p1KdRpqkxA{nFhFGYl!by@YMfzaA#;64{NJLiBJDHuRi`B^IWpX%nj7(=K
zI0-VNLt*+$DSgb|U@q?OYBiqe3B`Pp&n~Np1<2>)Uwt+ke^YB0Yeq?QOt*~Ix3i2B
zi_~|*2nS3iP1UgK2F~S@`66e?KKQ6FblitU=MOK}xk-Sl5oc{AqeX<r63(%jGYjT%
zhp#p5p!5{0VwSx0u4a3G{X(nUO4Eqwa4B%Go9y4LtIBAU4NDuDYaL;4*~RZnBl7V)
z(#zILD9*>bq62Dh-$(WP<cI!5(Pl9fL|>vWxaK|UI!e`jv-}S$B7-3;$@;>sT>Ew2
z$S-VOiwsFU3VX1En8mun=fvw=t;oc!@@g^6q*{4Zp={zVuk>T-Yv|)NfoWB*en?my
zmS>NsCpP60b%d;9Jj~H>*Ps~Fr~RH^@6}WYbSul9hzcDaTDz^@c5V*RMgzQuGnTJv
z<>rl)KCe+0EyCgD+8@)Ps1Od0#|D(hNL4R$&`$a^gnl|5d;;~ooVNpS;YX^%+Rsjn
zN{GUdyo;bOO4=FT3DXs{aI)b8ID9Q5rEGYgkLvv)4x_*Valfp((w>2ofT4ZPcG&M5
z>-`&o&UK!oBu$zFHLT-1q`^AiO1yY|?g0{dYNd#zpgRdL*lMCC|0QPiW+U6A8Yfe6
zqi{R4=gt`yi<LA#l<$ueuWO|8h{(OqZSdHn!rkA8+ss4ttk*EJ(<GM(UG4p$ZTAwS
zx}hK_5=TVPI(}(3uY4ZGFdmm3i*a1XF93mj5lvrKkc!Z3QS?CU)~7crDFp2!7NXdT
z=C~Dy%iLUVe0XWoYb7Jyv-W7)Lf!mOTKJ^!2k_(SS1=NDOrP^&5S|m<ttOr8IoexZ
z_Wj$87NVUbzxydp925D4wL_t*@~j3LGc)z?7H0a*t)Ig!mUQffasmppMI*jL)(g@7
z;N!9Ef9U-}XmQI<8=l#R5t644kTwcGj*kuPZE}6RTlwVtA)5nIc4K^!v)s);E>(}j
zjp)lo9376*eZ*?kcSGsG9n%GkfD(Z{=#5RUM!Km}4Be>N!153&P39uk<$IoNpF3%F
zp66HU^_sTJD%K>5dSBi-H`<U0x7+HdHx9kghnSnxcAMaM<B#C|soregzIVQ0+eDy%
zFkXJZ46Jv#Dm2MSo&ZeMFH#{P>v>{4NptET%dKQYhOe-*hI}`d_|Dw=z34+~14678
zXv9))>|ga-K5!YDt_MH#8%3y0w1W|%eAQ|w>sz()f!k9BzGaVyjieRKIn~`TVcI&?
zOd;-6s`rP}7h=bV0PFL!B&WJ+9akm4Kri!vdg?UvQi>-d_4=R#BFNHX5t`+FL8tt{
zrHZK*yU)cM<9%1)fw~^&RIlSIj6X3s(Q8Q{G5h*AH<Iu+QTM3z3r%)}R$-Q%dkgKP
z&OeZFUSkHZLbb)pqwSSn2Key(n3sjP*Q?=f+A;}L-lay+%bmtih*b#p#m`A5Z}mo9
zNYx&bMfFr9Wk}G>@uy~y?XrpwCyZTVsn8^3o-4^Qalh#L<)JK;pwl2QFcQFU$C-qT
zmTSIlvp#bt-r%E5>zPyqZh`;M=QCtamS@+2yltUBwZ0+<`qcuxR9a1oMn7&GLlrdm
z>zqps`f#?K@qnLMuiCvf#riC8tjKyCB71PIB@J)eP*^$h?4w^g`^w07e9(x__mO5@
z+>+BqqAC1I5A1#}JhxB{@Q_<!xX=p}z1L?&lV#P+Y4&`LbD{o~FDN*Zy~He>oHR)<
zlEF%s&Up6AI@Y$+EH50Az4mJ3<uGc1O!mZk;ua2#f1S=(4m54^6aTzn>3LA1p7H2R
z@!UIe-o)iZ85e8zs-59Onr&9O-jxQJg)xOF1=$kynQw!M&~<v|@n)}n*S8L&e_8v=
zwlN1xc)B{i9t*94hRF~wQ{YHk2u_?}^>#H6Z8j_Scdw0$Ee`LqijYAQH%$SrAZ;08
z<_9%CpGA}MB&BNgp~96h_J`to1$=hB{%jGtQ|4j=fAu#&VrK&Cg!5djrdhRG<UQ^z
zmWyPiCg@jwxA*E@_FUv+fs>&<MvCrJQCG%^-yPSV<?d^)FLKIo#;4TwlNFd;d!EiI
zJY?E4<5MinK8(m(C|#MNQ#oZG+M&I8f6tP#FVl)q8hBlMqt0LnC|X*~7}p<qZY8m-
zcGOnPPAR<upsiR@Z*{t*@wdS=$D|*@PZS82+ZxEIjb2`#F?`cr^>t}JEcre(K|JY>
zakYb5VC%e^o_XlZ?)s6()l&J(%7Qpq(5Y+y^xPHGT=eX2dQubC*oxw)%j#V2OlNIm
z%0v(Lx|w=8fOF*ugGn?GLjhe1CrxnCp5Q*NHgo3wWVuj`z3I1>11ob5l`D}5Q{#8_
zpId7eVl1bc|C=)g>Yv?*TD1rIJr{U(k#-~U1wP;<W2s@w!gjpRFGwKFgfwHv6!Q7T
zVDPcKjKZ9%t@gt(?&=(fZb_^m0tc0BNJ(BE-(^KP_1<LUlXFA=nGF!?J&I|}j$*yq
z)Ne}ghCcFrjdh3et=_0RaV`k}8P5Szy|%qw7No^@@6EKxd>ItMAI(!H$bG#lksTRz
znzS4r``A*eoXRd2EG7Ia@7|uF``m6~@>yv9ot{+UT`Dsjhtma*yUMnl7`V(H>5s9`
zJq0t4LY}5KMU9k89gTV0Hl=&ttX*WFzm54u5&RYf4yv2853S3>G@kr1HP|Fz(B@wA
z3f!^rwnBcKw$2*i)lc_4uW5Fcl4iB~vkUg9z)R{V0vuH8(((1P@v?529?<YynartL
z_!PG<m6C*lTr!F$ZxmDs6j3f9W)o}VJwRL<_9qUUPi$lAlJFu7DH!$v+kw+cOW#}z
zlm?_$;FqowUJ$o;pe6~VpUrj$+h+mQW3Btla;K45`}+BwnmsanJp3rY@?ztCt>R=b
z#klaRW}XAvv~Vuse(4GQ&~)!in+Bh7^9rKSnKoB&r|`t>^hgl$*@YK}1-&DA+X6Bc
zyQ`oXV^-PU>XvxRioNg%qk6(kzpp)smS|;O{9KtY=Slv}=NqKm+#N^Uc_>K@MbYlK
zy}3JnFyOEK-W__R89O&!>W(o*)UPHMK>oDUSfGg`-j)p5N&*mmw*fX%oIaK;I!?U7
z0TTNQBYX#tj9g^{3~;u$rAy@1vJlh*-R?+%J`3XMUX_?^Lb=uMT4;H4goRen{saPg
zn^TiQ`SO&h4_1i<v@Qb?Bt}-=gWP5V;3{wVHDYANr-%IjI(vLDytu7}ar}I}IvOx&
zo<J-m0<09@e9D*Shz<182?#8y5C&Wk<a%Yz(#N=}*vdY>2S&*r6}msm(?j?_jR`h?
zy%R-g>**OE=86b1|Ljfi%gLEH(=8XTQt2rCDuC<uZ=qy&+z1I<_6#IFsId+7+s53X
zfTuEe-*P0W0nGM78V$K5M6^~wxUPhjxlbE8Kp%@L$_45;cW+On28$k`d6;n1Dqd%6
za$mj*jiXmVbFn<MuQf(v1SY6YpG6(Hg?&JAy8X!)x$<jpEeCj;V9$D)FzV(4B7aI?
ze)bhM{xIR3F^%SF8d{_A1$wSp<bJ0pux=2QyieMJlIR8el64jB2AT2Osv%Gxx1Q?N
z#nH%)CEu`1f-L8`2A`rvnN_&g`0;~Aj;){#n;=)i?x(yjp}Vx7e+DPEQkP36Q|1-4
zwS)Vfv6ek5hU+GtL^4@xwao%%aEdf>uq<a!{#~OdbZ5)mTt?Np?lo^N;_gkWk*)ut
zO+v^nxVFMP+!9;KCxE!!a3-&8M&<=+4xV3hBY(TypdVo)-MA#`r>l~Cp#Fj7CH_T+
zz43ma*$Hkwfs1llEQ9Lf$`7az-?9xhW-o!Veya_16pj}P%vAu5NTK|keHl;Q8faUn
z>>NH9KZTK<#R~fi9l`vny9iAJGrZ!*v-CoE14MzgC1nzHK7(mgvGp1UE<Xo;l)uF>
z@U->E7cf)fkOY&D6$UD#@hP(LWL3fxhDDf|w2xm4oFJ-F)nd0da8o%`m%1SqKrsHc
zM*vw%X>G(ZGL;yap<{0cJLnoYK$O4BtuK9zTcIy`LH1ZBOfS2p6cHI$?aRm0Y6e=M
zNmbF@p@>@!{~}H~7rp_a%K;Xk|M<$<xOnVXETwk{15SlCA+jG0ZNOy9v%%?sA@ITY
z@V(Bg@r3&PN81YDKqthO>6zo3Y}d)z7RExVgw2*A^X8j4ijDijF#Op`yv4D3ONvDl
z?)QJN*_SXz+kEuQWgUax+x$rl*g>!^1zlZE7z8$f3NAhbn3oA!aFIV+P#Uy8=)d3o
zkSeeUFqErfApY`WW-@9}Z)?A^Qt}QuI}TFYd;zd$Q{nSxm?1(3#yg1FS%qC=ai+JI
zdwmsZgv;%&z$vJ6H{?(ZDzn-_?i`o+(yAN)lcuy%KS1VK>eK@9u;6igTQreDk9mSX
zB$3qxC>@)3^;Yo+!#>BnTSEfJB|?u8Zpp8H0$b8!He4RJq_Dc6?ni?m`w+=suXK;@
z;Wz|ysbDPTlPnfuYaRmAOd{`L)OTQ29>NZ;OTOUO2?bmlNB3n&lTO9*x+&g+ol?mq
z9sEMQokJ29ocvPxz>2UjJ5pSYJ@hD6>!@|b>RVaS-m8|TExSFc!R&$k0bV5(gL?n?
z(Zso-?IbKPjBVX(ujU>IK~JoXv_{1A{;2rQGhOSxnTGUl)*qoXzcJRI)Io)M(ixYi
zqh^F8yaM}A(sek(;JW*7{-Irq3$RzcQ*jI+9tIize2x56V?Vw4%@ATSM^$#VS+$(N
zsYKxTHS_hE+8{T;iqMr3O}Ly^lU#vEcFqAhSf{Q#anqi-t^!ee5~~nEQ~)x9s2BJo
z?_|~@4dUK+9;)=>5V{_q1JshWft6xMWX%I$WHd&ao>4ah&14j(;x4S{5hSr*C&X5F
zxki{?zXEeBzL00D%nGKdD>FNtWbXyF^F)1~Ar+&@-TnK{Q%yAafObsm#tH6O@J9A(
zyBsP|vJ64;>Kd!9&gnjg>AjI843YXp{Xhj#?)8Pdk&)8JDf9BhW-lSPNRm`XR>(35
ztxhr@0g<`5=yLJp7LAd8ZH@21U%+D)IB!mowGTXQoF3ZO=F$p^&MIKvEx&RB6bCu7
zb$nI7e>g)WIb0xgIr<A*Nr~Y2mQjeAi(qb{jbFdIvM|s6MLE!$N(-RPM7A3G=j))g
za?lEFk8o*Fx338i2wJMc3<8yc=)HGyGWxo6l}fTsb_RY@mMJEm5K`RNA+UNtSa@>t
znBCd2nVnox{qPU4xA~`=+nOZKEJ;5Iu_ALmjU-B00iuqD4}$Nh#{aC=71*zmBI@0m
zvgh|j#cb3g=1w(lJ9=k`qpAQVxAEp!*!NY+y!(2WIOhy5wm6`75*I_Id`_3L(iw30
zZy%VSDIjYvjU>Bkj2llCyCJjyzh4VC9Zr%s*8x7x7rsaMyn&V_15cX7{LazcSG>T@
z))Z{LNNan*)O?kMkBM^}E{TO38tPRrL2?k>>+003ym+2)8}rtieL78gocqh`C}q-z
zFl=V@Dn#w~nT&3|yQF)}7o}@&&%*%6*eA7N(KED$18b9=S&T079ychJx&h?7Lp*62
zB2XKV96!iAp27cN`4iAaI5ESeTNHU80?JbzknX5p=*h;U3G(b3R;K5|M(<*>Bc*!4
zLcd#b!U6lytw89pHLI0X>5R3Es?OO>M)S$to7&<ieD>dpG{}=FC~rPYZF1&k3Z@vs
zLz$g1k40s+a`?Z?Y)fqg^7Bu?r9*d|)gg*?BI^v#dw&I0^06n-7jMmUpZ=8kNjm(s
zEW6S;M^Fbe<kDSfj;fLq`7u7xZCe;r$@zJl$D*dPZg@n^T&)wm49wK4qKmVqmJc`q
zdsUVEGk@=6b=3x1xiQETI>2vm=)go@Skz3aGVEy8agu#JpIx<7RB4*ih;w?GV6-&Q
z>tb!Vc<r$(jXvVutA0AZ==qrFcutijK9x9Us`}|GGA_jJ%PO5H0%oVA_k!KVK8Qle
zz}AoNx3>3w>z2&T4J?tUwyKRTE;KdQ_w#top=zIUMviKe)Vq+u2WjZHsb^2hcE|8U
zbHxa>p5dv_XMr_aGP$<Gzo}pN8IzFy;g3+Hdoydi2a2Z+%ha2{a>iuj2fCKL55pYM
zR#o~0D(o8Ix65$S3g@NU_m<1({D=sym(Z(M+GLE}*#nY>2h^W`UPiw;B!fUiKG+xZ
zv}{uzwpRMiW#wcV@#h!N->HFNIC>A|uCDgkLwGgtdqsbTeEuYREU0PgQ$IY)hQFuq
zW7~x_6Xn*6;|^5L|6(xr>;VSz;-|ZGwdmnr-=V^+M+QdydUjCPGMQr2kfwAa`n((1
zFb>kt%)e8l$v~{l$z?X+Z}0L)+<;6#7mM@2MNu32?1hbGa<4!NQo=qALNj#`0X4f)
zg~|UN&4&sQXQl@iRmF|+ofE*BNj&%y@`p0ibw>}l!`UK=HV|Yop(dIDUpRee_g{}t
zxCVwkDkw9w1=gCz#c*WeMn}I7{Ikmk?+8wv2EMpNCp0lD9!yFA1q?@};|q_8Sjg%n
zB>zRbrG{>L&es8^Z5b~&uON8UDzP;X8gpUOWPo2w38$gzECy&UXJM?hbNH2PR4p^0
zLAK>UK)%9pvE@MqFi9@Px?Bwe>6*jw2B^N{89>DVDkZ;PnZxG$1iv}pR+Y{3dt;H7
z>t^XYHCeqsGR<+iE;!AwoX>xNQRG&`7(so_dn`xKy3O`s@{+c}FR|akq~w5v2jJ~s
zi^hKCA^%L2zx7k;L((z*!#Hsu{-FN35NA*f6@zlAIVL8V`PVDm+_quCh#Yqz>-2~x
z(k`hurPO*ig#G5Dv~kQKlUyXiM;15r{Q{t8!Q+7acm@dV?co?N*iaE<Lq;?3gKsN;
zKbHIYaCZ`DH|!iZ8bChREdolX&*yFt{w1i8LW-^Cr;6Qld(m3g1tg7Syk7sk(d@P?
z2d(p{#XOO9n|K0eoM&Z@nFdU++4GQ75SVVzxkQ+jJ6|1)Klt%n7y*7*8Zw(G-&XgN
z4GixKEiQ<~VDarS4W<!Dg2W+0jL_U${0Eud{I+~JOke@c0RIA{Xfuz$@rQ%G_Fo6?
z77}*3GJY5dLINW9p%bm4d+r<keNib`BR<dFqh(;ex{ekYN#g#sz`wnO#}V)8LUTaU
z^T?0~e|?7n`yWaW2rW$EF~G$|`tPRz?{A=zg9ovOHuC2oXT;6DHvh-}3;{O$_uvqv
zzULH1LFf!n;!f}RU-#)ZAoK18zms2gsZ+}6;JDiFn{*TN0sRyFl%E?UEDP9CE}&HW
z&tg0dJ+xmf%iig_-ElJh`&s$P;8}O1L}!2n6J7@%<e;03)W429vQ1d-dl_IN52t;)
zHNdosk@VT$=#2kscn;O3&?<Dz&pdvx_3!1prtsUA!MYl~6RE)yt?QA|bpKEVU`H+>
z;&Rm$#d>(J(!h!zQ962B_3x*w`=9v%UAbv<tVSs4pn*<{qJ`?uTOMg3TNE$@w6!5^
zcW^c^|IY&!8BZdMRFKWkumSmW{aNmn$ADoGHJf|y1i$K6UC7(uPJ*^r-&_PrzFfY*
zuvoa&;wy3|qcH{KP#@pEskxx=j8eJQIxm;ktr9tPFf_&peX!A=-sK$$43EYh&JTH7
z$zM{yhatl%o=TkqyI0!e-C<fK1q_m7&z^cR&ixu9iky@$sU0{yc}w7ALE!$@9SrZ6
z742cv1nzZ3B;}N)K+#Q`A~ixZK?Om>$4^V9r1PDGHnpIGf)*?ll|t-7o*)P60%u@A
zXI_F!G6(^joZmCsYoDgO4=8Q~i{s`k`Ts;2czQiT&53%(xOq?NrGuS;!P*D3a?;}y
zfk^T|mEjHL!Ry#Q@6-YfygaR_oV0qm!Bfyl^Dt$NIbsi<dYWzL0}fMxHg?|)G26<P
zioD^su|^Omx^GKL;`L@=%klT;PfIrFvt*%+g02(<8j=-bsRay?`9Oz!TJO6O7&=HH
z5(jk1%zS3A37}onozg`ta2Fulf}J%9G^BXR(^3K1JcP+uSe?KfQj>Qmx+6l0088N|
zFv+OQox+Tq@{toCl20J^je27=HbxT_Mz(-7gGRGB_JDy^F$mewk_t0S5TS$2qECVQ
Zr~cIPe%b7g7e6xqfv2mV%Q~loCICKJ&$s{p

literal 539484
zcmeEvcRbba`@f7bO3KVUG^}iqeWZ{bT6Sd5?5#slGET`Xt1=o$86~SkDzYOoitN3&
z-~Bp=a`3J9`Skw%_wMm1r*oWpT=%uF=XJlXtEtHCBBmwA!NJ*eQvSF)4i1?;4$h7o
z!tLN4o2mUVI5@-|<}xyBCuL+{YWB7!=2pfyIP%xSBX?+OoTt4!;?Wn3Lv&N6R~e~-
z!=AW9&7CqFiN9BYkVyJ<$j*?CR2Yk|F^=2`3VMcZ`$!C52O5@0!u))nky2Tu4&p`d
z3V1lr46c0gni2C^UWoJ<uAF1SxlNXS{MI&3JS6+4o3i+>0-fK<qh7RcBTCq|?cBB_
z+^UBz1YEs}b3D1H=)=M*oR}!v^UAaHGwmynQ^yJT8FAED-n{qR$3y<j`C|U}lMh62
z5=i&EmmCziMN}Bfs~dRW+Gq!Uipp3dkxJaftnmlRnpbwek;UnFuahoFiSumNJ=5-S
z^*vHuyjQH>dfue@)<EbndPPU_jl2%8n1Fz#Z@?ZQnZbiNYQ2QdZ>@9*Oz|v-Q7*sy
zs!5>QaD?4e|EAKd*Od!Wx!pdBCjOqH{zUxc$IK1z`2~<muOj0ppBP*kU#vE#8e<(|
zJ<ql~D9$9Vg7reooKCwzX>-sVbKPB*@gBCtF1vfl=V22!?wXjNxx#WLus%SdQm0Qz
z$IU(K7!QRV#qz|}Q<-&lIRm4mDxXc3ckH8Mx~j(!s@BCrw!^jR?bO-&IQ9jN0+tX%
z>kwWJwNqj)hkOuG5}g`UDdoXDqf>V>ub*jBw9lu^Jh1Nqv4^zm<zqY_gEgO2N|W|f
zH1n1+CSUS<le}*hSBlJ6T#0PgAcEm30XK~PGX3k?yVngppB->hKlO+=;JvKP-OKOy
zOgu3a7_u5fh6!KtNeOw|a_zkb5uZwB@3U!`+?<5Hse-BF`yV)26x$M(4P;HTNN_Be
zaVi<aUkbV?1jM;sz}cg=jlm2@*MP7B=|d>if&cX~^{q=}gaoHxjE8Z>VffK-;)u&X
z40f^M9yAbWB7RhCqqbw(Po{&=Y#VFwKD%u?+k}f5M#(09w~g)%x;$gB(-%g69ASps
z{$R&GnFu7qTVHXQ6dQiv<?o3H*eLhP*qe!DQE?zXBygl|e-z;DA0lI#pr|J7$kgxW
zpCDXqN<$EKLG%gUoni+AstSL<Z)~p2cZp3)=IogB_oQF3{N`sT;JLjJw;kc}Epn8N
zI6<Cmy8@H(Rl`J2wZv3>O=5*Wt?L}JrxPqr1*ZyWvi4IwqCOHdf2Hi6x!Iv+;vs+1
z<8t@idaWiECbv6Mx=ZEl^CI;kX$@eM_9;>0ZKhES;`H~sUh=>&!XS{BGO8fTC~6`~
zENWYY$6>0E#W9AlnTRG=D_ZA$&UntOIYccXLuJ!Oa+i+Y-TrBBPY6lrrAs0&SiU-c
zmHuk}RrxE<@cy_fjU_GLmd1!&`3iCL5mJ|>t`=SQ31QmLMjk|zenky89x2y!yh+Av
zzYgn8va)L&$2IPW@dZZ7>c~7fAhfrNoM}({E%^g@w@BFU9$-m3bov%)827Dt`S|ij
z7fGJ&7P)zt)%L*1{@k#Uuy%De3i3eegNR!)U*sRA+)WZs^whk^exCEv?OBD#w;zSa
zgxhcw+}oR4rYYpK=SaMkw1h%|3d5O0Y<Jie*c`&oa-8L4Iy-!d@pR(DrbkyEU3_Gv
zCX!%((pZf$#o}In%7?RDnrRvmkDFCxYId4S3Nl)AS??Yax4s;bbf;EiQqY}fiLBs;
z<h|ra-3e6pb2QVmh8}3~;Fa=Tu%p=*q#U$A=v7eFb4Qa*QzDbJ_=ckc-A{J+YxV2(
z<4ze)9hwsBk3SMc>rLB7n?%b+8+xQBhAQS;jG~~6d0Zp?=X;hf&Bw9=vqQ`zs{OOM
zvi4_1nm3y3H!wETG!Rwa?6qnXc`WecWA#+6avfKrM3UG8xw?sJ^1kqa*v}rXHI1G{
z@2gPg7U{lLnB$#Ls8b|cr07#oTn}qbC{B3ak;z<xG>$SnXVY}ysNn-s!78(tr~GR0
zjC}b%o-@m@d{H;+cm3hXC+|hWsKTT*oHRnr3(fud_6r0?r<~6@pZ3Mt;fcsc@m9~2
zxm_epZ)%R2owB~vx;Mx18O8bGtiib8=plY>@mJ2yp_E?&t-~!F3>r);pOR*BKefs9
zKJv~?<;~q%kywYqFGrg%HVtOFnLW0kyQpo@VzN;7=}m~~S!<V;IKvU!q7Q`)%h`TO
ze#I9eOC;oCI>p~Pg)oHZ$KH*7pc1Si^GWv8$xmW;J<d6vp|QjpxJOb}+F~63ktk9%
zT_s&4T~24k?zz39i&%%y(MHE~gW<TBOT*iTb!KR1^mZr9-Fg?wSP(ZKr&Fa>m7?db
z=b2HUSM{ZPP;gp$n$wof_TdcQ$ZpAWNlwpZuTzEF3QG%hy&rkkc%SxGUEH;(u|&5h
zhf9waL6Eh>5Z4b+5zphOC_Q6JoWR~L`Ye`S9bS>6{4Y3W&99xfCiuXbQJlxDEy69d
zo34tw%9e;A@oqI&VrYp{$+HsgUGMw@uAU8$y74sl9AnMClQ(qe-C*Nhw_WO!Qv+&D
zYQ}p*Xe^WrmAonf&b2o^YifHTa*6YM`diDg8{f6b?TGk@Jji9KXStg43}7{UnQUXj
z$Lg*sQtN3wc*$l|^S-pc^lPbH81Ko<lRIt|+R41k{#tWUWjv&7Puy->x{-tJH}pR-
z?R`P=0l!M)tHuZQ?Ha8bxerK*B(74vxo|TyRbf(;i`UwvIbgDS(yOm?U*<mL=-G+~
z6`h8sFXheZMA&?;ePVMx>4ALhouQb5BOb?T54UpqT_iD16LEZSDvv`r%!hU4<5YRt
zr==GMOy1WdaVVB5W^|wLN~`jS8;M`3<DcYjdlvFz{$u0e5$)G{r=IWCiq3y#6#nhe
zo6~rY@y?9s%jkp^c55M&rO&^9m7QtXLhM9*mU`NUdQ?~DruxmSRMXg&yw>*nMbuvH
z*=^oS(^Kt%bET1rRtg*ng;pKbFBbiVc85y^*(!Z$%6@NNmtNUa`MuIbQ9^Mh^5N|t
zkyEzOxk)eYp4Zb^wi)=I7Rf=+;cUC#Cd_sqZRbO)R+}uZiD{nGZyt#?v^6M;Ww|_Q
zAG^Tn(^ILd_lnkKV37OE(wtmP=!^xfMU17g1!w<Lr<YDu({Bg+7OzifEa_Di9P}s}
zCLbASWy)O9DRA&w+BSrDhqhz8TS@inw|>H;bBq?&Sr$*s4F$vLw+H+P&|kbeInlcR
zwVY_kvk;+W!^MvwMp;Ism1+)p6Y@8nCgmO}%wBM<9t#o5%`qy{TFD!^vV6Ti%z{-o
zmL-lu{D`;c%B0z=2VS#E{n5)Ir3|~P%3=$=xEFOA=Q<{+d1m#Gw!9w}_o`X?eqAv=
zGDP1**XHGS$E3MOoef5f^cf5T_Wc%x!=43G{vU&LhF`X(*toZeJA9n^e6{6n%h}<I
zqB<|(`Qd`Nw{dMAiynS60>ee#$OVJJZ~byna;2f+OlQSwyww-(%-Jdzs1&d+crK(%
z27Aswjc;5Tclk1+znwV#RNQItMh`tHGhtW5a-ZI^8w5Pt^%V3%c2O_M{9qE(i!PY|
zu{%R5J>+fh&isf%MxW>3(#l%AM(Ep|g8YLe-QwrG2W7_Of<qUiG9{h7LzWgjt=?0A
zjr@?&t>02e;Nu|}CnCxHP2pu&8V>P3I|X7K+iB*dOX4_LbU0+xN>hztxXNDSN_%Hj
z#T)!o1IclmByf&%;la9SaatZDBdEqyPZW~8Nxjy#t9W`kaE#|>Qd5ia9E%4_+WXG2
zv0M|J4ux&ATB4GYXM`n}vq?`*Gvf@oX?QJNVPnEiC~G?Omc6U6m|#&i*?n26VRmMp
z(!~y(WyuxtUCneh9B>;?;PM6<pF3%ytc=3}UK8TrZ==N_0I#-zzmnVNe!iC7#)gB3
zdLI`DC%_yBfAx1%z)$E868wd}gZ_zk%O7V4_}^ae*EIom?YGJ76Y$nv@5lk4;Ye%9
zoIDABY8crY8{0UT**ZoY%oYM~5ZTF}cfi4+ItcyUc2fPoC>X!VT=SgcIb|hLBU@|k
z3m0t-jk#T|?VxdR#9c+fOKW4t3out}D;oz<R|yu>Z-|1|(AzvLFx0O&T1v2-Q&xk?
z*xDPzj&KWd^Rh@1!(cFR`->)`>c?eQzYhK<!D8m<XeY|U<Kp7N?IOT!Yj4WKCn6%k
z!^_XZ&(8&ZgUi9q#_@tHmyN@I^dM{F95;3_vNyMLG`F>ZLE~O9v~_ZnU}1qK`uPXF
zPGeW|Uo+V_tZoZzkO%sNhmV_==jYhqtK!hRqH5-@##ZN!n_B~#fjJ}v`T2OoQD4~f
z(XT1dUp?($Y%gPL4Zi6p`HSkSUta&=#xJ7gJpXG>J`rKT^>eQKX!Wb&JkZ|P(E?3q
z)VqMvlEmUXKTl1PxF_JV8#qXM^W&<T;3r@+=nq~C_=gSs6THT|d^ko!dl3f*j&t(3
zw5IE}!B*R#*we8)M(fmB>)_dr35)De$3$Ob^1U{rl_ok+o8WJT7?(@(ZBpAGC~HN)
z?8`@-rCH05d)4N_^-~7>4;{FZaOkk?*?M``ObcAWi{>y{PsF#<JocpL6g~VcZ8-un
zZ9S4zBX@gB?uboPczv2%lI$6M6IRkF5@x#?wkUHMXB#d)AsGz*cQ1IK7Sl_U;?dFm
z-7r5_F`9fCwu27w?<`L3=4DtwAZ*vav{|w)gWZ2kF!WA2_L6m}?RjwF?+*Yqo-h62
zzjB)J%kUGpzVsZvf;<16@!&{5iJxctcP~=q;YgViNPp(P8)l8+3GJWm_?M_pNknL$
zN?6PLFNhPgH)+!8e~10y2`peRN@-gEe`h>c`D?~C^8KY8XdZx-C$h+w{N*rfU&j||
z!uxlqj|jxa7tzEwr};OOBm6Q@W*+~NX8cc?t)2Ei7Tf=n*>7}Zbwcr75eM44t@T?-
z&IjtKyLX>tpOhxew!agEt@>CXA+&F5FR|TX;73%z@MWit{o^xSotx|&(Ec5zr)W3X
zCHNj-u>#M@Dr3tY2F~D%COuu+TP$;)lB}yb<L(T$%S0vuk7Fmb^83wWo(EQ<+~OsL
zB9{69@fv8Po!vdwTWBN!FlN-r@^&ID_pXKn%k-tv;M}}SW&(VXD>_ss_F$>$4UPh8
z`gdArZm~=TV1^io;x|wB`*+L`#Y)noPWt?|+g?&_-XbyiHRm#+o?w}R_iuu-vGwr~
zR`~aH`Kz|8ByO>YWc+)vnMC$ak8Gax_Z(530MAnF-lZNDvsq@~`nWZpn(##?vxD?l
zej}J$eFQ&(Tf-fFwl4hC7W3D$k_`nZBW1iSZ4Yfh{Vn$QYY<1G<F78Zole-X#c%nN
z92YYp?;`UoVGG^7o!`BKct!w-%cHIBvU%8p#Hdl#zH`@+ZxZI;4Z02Y3n5r++^)OF
zwm23gBGj-3mV2>e+yrTaS!4y>1KuN>SA!@-GeWQBzJlY}%B=;&U>V7r_qPzvR8?5>
zzB{mh`Oj@in@RNhcY60;bAF%y7{2@0DgK-RpBl)-p>pO~|5J>cxAWU~j7b1-^>T8C
zHxH{off_aOt!^?amJIf64_K@?-O;OC97`0~2w3dv&j+?*`QNYt+z*llVfhOdTdc+l
z%?Q<v{O;F(Gsc!5H9!oer+ij+iy%WS9-bv2BuzS%RuH|JM8AKh`XJfREw{4>#-9iE
z>jfb<T6#x!-51|tJHO+F4^$8*SXsAjR@S%K(4$^|rKI}1A+}ie#r@D?h3tqAZax-X
zuo0kY{L)ThD-GKrjbMv>vz50ok7k6#4CBeYzva;_cJc^fu>6?xpktd?lpu*PGqv)i
z@7MMa*i52dN3fN~NKk||ADl?XmDu7#G99$^cF@VXZLy>+X!>)m+guQFzclCV-$Y!t
zQ|M9mx<tqQZit^_{d!R@3oX{FgF<055eyVi!|u=y+mhvfv;JMgkZpineYl0XokTN&
zjQ+>ogIIDcp(cvKj_B^Yx_L!#FSL=Y&z&~G+DJx1##fiXeG_S6aM%m9^u87j8po2j
zY}ycUpJhAAZ6dDPJ@lvthxsM2WH4kfwAe4j?6)@)q4+j>*qlSyN&_v>)bI19i8aVK
zvu!WnXvpRiM${<&=G3?RC>6zE7J2-{n^%OGM;pmh+o4LVjU+c@d|weyZXzv6McJXH
z*VmTj2bRQD%Yle%Y&j6JiMR;>Yyjy5g|Ei$STY!l4qEJUga+FdL|{h`Yo&~>GzbDs
zeap>{BKgo}wyg#n4cQzouKST#axF{^#b6DN#uQtuXbNp4)!LiyU~MFIA>->p*}H8E
zDs-Tw7r&kN!)67ynI+e~lVAlA_v7c!44a09Uq_F6`^CpeEEx<g1TEGSFGXz&B50t8
zZHn2F<=3rkBXgRmK^kE@yOQ$8Phju_9yBA&YsvCp$+hqVT8IXZ`#!L3v7&w`+&NU9
zSePx0t>28F2f3mn9MyPRP~kNJD%K$^yl{K7g4@iJ>)wHXfruNKmL|4oSj1lRsBD)a
z`><rN{{l0V&iXGf`!6s<tDye^vu(Kl1!h3S{uh`*KG1)G8Sa098OlfeFEAq{`!6s9
zD&fDt4CR&o7nuDQn4z)+{{?0sd-Pw$48)QDRm`B&#D5jD|L+yEA(|f}J}VycOCvt<
z&ck1$_1u1hPc~a-WZxh2a}?p{zeUBPcUMJXs@}0X+BCVQrpC|n<Nl3lqhdRBCYk1w
zC|&rb%<bk}8^$}&p|YswnG?CQiXPL0CwEKE8Z;MrdCWItc!y69x3(At`{fq$y#2PH
zX-<Cr4Rs;%OuCNSD39Bwl1VDa?m>BSV(Lb<E%I=`&BOMdhBAYC1GaTDKGV<RlX$$x
zZkYFcinX@zNYHCvS)NJJwk<dk>oNQiFSyfq)BeAq_kZ-O#T?zHH0U1);(r+Ak>`*t
zmG-j*9y7_i=!70r-VruE?{noIPtoFdmhs1kEVG(>gg;U@qHEcd&>Mau6y^Ne-EB^Z
zmZPd<`IG+oLxp4ks$TsXeTe~#w96N-R@p6%y-NWbWlht{el)6*y)n614=Q00YN1uE
z*b`ea&-y@_$CaYm#aaL2>_)vi!+8VTRFuSnBxOx`PL{NEbdT*cWj3Z?twFgI%Ai)h
zIyHjjjDny@d&{j8h9I8pQ}v1A&L4dPrwrHlaOoXG<Z$7FP2rEj4^=Q}2fvR_FAL@|
zD5+p@>L2HZx1frp3+|5dEAOP1Z<f7#SEtBRQ18~|K!_Wu=WZSo6(#?U-l2j_R`|dM
zKB=Zc=bxLOO-y20q=m~v>1P?1BWm@Z)3oa%ixz)uH?e8IeS?}u1>)Vh{)Wdbjrk!P
z%7F-l)%wrH^Sibv!v1#ah-Fk}I^)K8n^vKx+x$fQijJNULM(6g`;GRUjz_1PjPTAr
z6x<*vu+_5L9+&Gem_+}Bj1ZEUp{gQj2q~reRT48gqC;7Aq!YV*Xd{aG-qBB(M3o$j
z)Zu@^9<&jqMX_2>N8C}uws_WsDtAiSmpq<C>|m6SUm7$G5JucIrFhYM??!E_OM5KP
zjpCgyUK`~ShOV_sV0nx<h}Bv<OGUl#S38-!p`4aC^>v?*m6es%>bkUsOU}Jl%?XFM
zZ{Wk?!|0rS)Q>lFn_+f;psYVLq<whZE37M=20kY(48P*ZeRd<wrQy-(zBrHJlkb>(
zJToOMHbMz$R#ngI&vh*umh*=R3CU>WAzh1zVek6EnUT*w)RcK2Y4gF%sOH`Y^MTK4
zeRc86j{A1?`ED2}RTHi46e1f9u${738Je=~5zj=6Fa7uQ0U(@Xjw1y#$>b_WnQd-z
zY^-@%py4=)(_99}Ht_Io#R8IPB?ylrse9hTXKxpo$55!`d@o9en^eX$U468%u$c;0
z1{tm9h_>Tc?+OJYdG%&mVow+HT}-^s@|<}~j&<K#a@C6a2B$Y7cVy6os0(I_)&|&a
zssd%CDkvjYc1e@YJsEp*m834#r7e0oE!t;;BqOHLK(Xq!f>Z}A=fQ<=JdBqgjP1o2
z5iG=)pdQT(^*Rhh&CC4#t%DLp0|Xleg!!Pg=6Jh?=w>VT+jn3_kk%APe)7aonl${>
zXVGf0wy}`Of)OtsU8jcA*`F-=H%yk`hMp{@ot)#hLvQAhgi^@x<+`%BS?D55v?<On
z+qJp%BrS{-NyU1Ozd!#u!))VZF6hbJ^<`{-0Y?7Wv<dvs);KK>cOvjbF6O&xBo!?U
z*{a?VwpFogOs_23$FOlk0rZ*icHWx9-C&s36%IE@<mzJl0gUn1+^A@!=|zY)8kEA*
zbzbsYw--tjOvH}d?R&hu(TrXM+yuE0Wp_^YVafF1_61<jm+$$xz{)S!mB{ZueC{dC
zx;cwIO*@xc{k+9S02s$A1ny1g5ejTK=1vKLG8*R<7Yktld(R$8+U+x2V!kjvWZqYs
zWGs}*iFstzhUgR1%HNm0amO3!y<ykz35Y#}d1^fgeCcEK{Vtrgt3M@|6ig?aT;;Po
zVV>+W<A2)k(nbT`Phc|NJ{J}i)}ZPhwTD22e*3%l4bu3>J@caK_W`Z8q@5)8<rl)Y
zKOVd5@ad#!yw|jWpX)1o#Lb&Gv+TaT$+G(LEX%MoP}nEkb1xo-Zh_xKdrTBBD+Nz%
z=KbHj1LuY|-t%3kLX<#cr$%EFz6mR@%kWpn&*x2(vMn2vKXSi14BTN`XP$W~jJ;&3
zqZQ{F8yV#-Kr_D+6tWYLHDKdb{>)sSc+$&(yqD@+j@5a{IqQja(VJnC*uX`$glLSa
zHpaHMmIe&^ua!ax1+rlQyo!lrxC=x&F0D@Hon^u1!(U%6NX(F6?1~JHtc?3#k(GN)
zETl;<L>yxV&;jm($=b>LOiS`nGLb<TIAOC~G(2UfEi{kykMfPVqYIE~q3b5(D}V)%
zeM}(44WO8<jV!#KW7FzhUbKLRv7M|{xa{<hGy}GWgz^&Lg2CY|<%xXm<m6G=#rs3W
zZ*jlq@ICL-PK_Z85C{bV2YqO>-2$NZJ6=Em%oMv{zqDXiMapnh9Af507T-FkSnE(F
z5!qc8FI9b14+G$X;75DSY)1N<T$JBVbQuSs1ce7xw=-*>5r}AhFz3aUA#ZwSte7|5
zY#AU_h``tqj0wG^@h@T=TL3zLcWMc6=!{4X48J@mO)3%QhJf$K9r2!d1sto%6xnVe
zc8oK6iw1Ere_mGGh4H~3bQWNLF5pqT&o^G&VSN@^B7I)`0GWEn7t?s<&dyF|OASnY
z`!9gq4M>;>?ajX3vlP)rFfDfMFmEq<zKkKN2xm0(y?S^l)dN#I|BhfN4MmIcrY?k#
z)sff}X6_7x@p2*tGKwTy-3QX1&!=O8g5WP#p-rzE9c6!~;5I7>_;yI;nZku8_X3p{
z+k}tLLH=UC|8ZT8bqfbw6&%AfAXCti>NdRpG?oISb`8SPU42Sg?+*(=%;qa4oI!A=
zzhIFVEym0KWRB_FTw85L$d|asf+gLw4nP+9M(^$~k6-~ZwX=8#JJ3<`hU>cTdc{!<
zRs$j8g01kvu+^d+@eswdBX5ItK?3~oJ!D<TDy@eY_{$QsnZFSA*xCxq^%GWrc<t8Q
z&L277t9<FTGAm*Z9|t{4%p?zYtsiM8I*=^6Ow#YP{{25o2|nRY7H9@o!R_&MSb+K&
zbtY>v(5B>n`t$zv0%w9wb6n^?GJ!GkKbip*;3OW2sVVlQ=hr%%l%ogsFpvw{+Sl<@
zK07)x4gQPB6ddj0tH%f{*ns~f%c8$t<-s}8s?n@8r*30<<4@vu3Q~=s{hFq_0QOxk
zM7?GA)NMdV$Zup9Q)F7Vt$#wGi(u=w+#;?~Qgf9j8mg7-p}7ZIshGdhVl@2hL&GbP
zHQlE;)*sDJPJScaL$-Doz|S`~z4!q&=jg@RjW7_|_qJBh;m3fs@#@?8ZtFEX*($8}
z+=9#zEH{${qVIzf5o1hQy3(XOjK`30dC;lF9C_l@`hik)S3}F(%S7uZ{?k!FWAH+7
z_c=%7p{0{Z8IMN2o;j!s&bCrhX`Qk|TtNFpyg{dhutq$aKnF5+_#^kcC1?ZG)2pOB
z1X_!q5;TymC;a@X`(AZ*66@5)Bj8NE&ku~VXOcn640EO~3xPeD)maQ+q^JYXQunRe
z@(0g)>aAz&%}N~o4ox1UVr6WP`VoOR>#4lL0p5e4+4>B|cM+?|_@+o~hhM9TzER>T
zXRc6ZyHXqkbv_;#$B2Tnj;%VvR$|InYF08-);m?j>GQ#I<|L5Z&oX%JE9kSlxNn7t
zYe0w2Vb4)DgQ@<8z7VnDqobLa?TQJ;XcL%ag=Z1@e^Cg75<d?XaSl@G?`qAHV{!;Y
z?mE_<H2?7-ye~-LlYC0Nx4Y<6y+&PA(>Z}zlTDt_B*<4mDuC%mFXejG_D^)cPL9!0
zpZqAj7J=yWqbvHg1Fu$kkR{40DvaZ{%e?dmA$$>fugRLeK+bgeNPTZN_x6Q0VS4%d
zo2Y65-2y?LWxh%9|8T-&=R5RQmWt~9LW^$eEwrDlOHW}`conyt>$hDLTt(nRsYq&n
zbofPQZLn}#RC-?jBgd|ra-TyjWjF4$m<;MI-PVBo!-Shey<9ngysP@*f*OL<mtOSP
zlve~z*Jw^|0au!Sk)&5AT{~HaTKgIPjX-6!FrX}B5|puS)-1=}v0cQ`LV~wi8Yv@}
zTv%1>OCQ0bALqT;$&@ma-y<=_9J2{$tacyFMw_XOWh{w<NW~2xw(qTPA#$9}V9iO{
zFW2M5xbQ?e;Q=h*h~?)8)-5?na}zy{FjL}<1ceI#$%?ro%=wc?|KW>cL=_ZW)$G%$
zb8-(FOnFW|{v;}$AhW}`JU$6u<g5KhuD<R&wv5jrzHQRLm7v+$%zKySrZd`FW`3>*
zQAS`w2d30akdZT-jX#JlV)gX}A2>hn*>fhFzKah$4Kjm0x3QijnZSJ)pe1K@;(osT
zs;_?ZHZQ*UPqvPDySTz-F(M*pW%%S^8Y}7BSTo2W{CQ{^4lvEBCDTvRq}2MJm)vK+
z1w5AHWBIw~)fcd*fH8`1+}~JNSZ9+qeg*we{Xt$cq)dtmf|TKUzof%1dm1JtRaJ>8
zO|7Q}(oQt8n;J?k!eHghekas7E#Xg7`E#kAUYqH)l*3N@C@~kC`Kb(_*iT10gI;e7
z+H?@*O2AVyAxYNR$diB4sXJv7`e6Yo!MF^fdzXB=X}atxk2F5Ge*OCLo#{>H7-j@K
zgnc}5wVU|x4=O}0y!a_l6;nixz4DXG(b~uoDEIp*Rt(b6BbAJ!WpNT7wM@&S0h2B5
z%R&X90Mr+LR_|m+(enJZ_Hj-SFU!ix*48E|)&){>)&@vUCJKs*YRMYqt>;xZ320x>
zva88n?_X?vmYA9zfk)ua&{Z9+3Q5Cs{7V86&jIarh4#g6Vbk^w>B713k31<Lft11^
z=KP>&c`9SaoWC6Y2dZ=6;Pbj}Db~(U4b;7An@WFnrvambg_r{9_>OxPo3zuW16;ok
zaApO>^^`!xDnbkDuDa_>A0j?po(1x*P!%_tPVBAlp+kpI8eY$9szJuH#0Yn256!-P
zC+UQ&TqoME12a4=n69y&8!mDKZk!;{l>8?YS^qna$Wq116E&X0147;=N;weQuLYW<
ztrZuRpLxs-XMwWB<Y>umrf%)!;|a02BVHpObw;5Q5ump9RRu?Peb5OL06Y}U6=jbt
z5Ap6qBm*8QAOt+*=+jNnH9A?Rno_4KRq*QizPQ0J&slXHYfkjl$T7u!yR)p6P=K2f
z>#XGt3ej~S71Mkp0i%v^kp)mA?5g@E9{;0kvR69-*%XGfKwp)PFTLa_V)p~qcu)Jh
zQAY2)<@tfSq1^Vi61qYKKnW<rP)DvRezUvsPUA*OkVBL>e48I@N+==Nq)B5L2}B(a
zsDC(Ee&JmRL!UtWT!rJe=Zoh0vn7QqN4@Or?IUbwX+bjMt7DyV9}p|&S9=^)*KH5c
z0nC*z4-(s!NSImxJPYWIDqMzfLi=-BW~bcyk|Jw+d(BMyOqN~(l!b87T;-FZksJ8U
z+`It&5$Dz&o#oh9o1@Vizg{0*fNq=6rM<ebRR0HM0tcPFso+itgwPIA9e^}32is=|
zM8@kc^`ET+boNLe0@Z6{(Vn{w?%q`-U=gF?8J-7Q-FjlJO{=BWv-TTeLVLnjg2Wq3
zcYpH9phi_cdeObsI~}hvc0C=DQgQ{zmfG1_SJP;dCATlekx2oJ?q;9cXHMYaFSFYX
zX1+@X1$CajYx_gsm}T;9fC%_H6bG>;f(-&NpD4FAF|-7`;g>zh`itMGW0Rt%dfJz#
zbAdn1n$-HrrD~o*{{Y-TtHY&eL02o&uyhhn3S%xf4PZl5w)aoFUVnDrjLcMExv+o`
zO*&kLdpkrHyI7@8QtP>1Fg<CqG#5MK$=V9|?)h>d&tBNmqT;u2kLPvASWf%&3a?j6
zWQ5E>4&U;3>A(Fjuq?20$nj5_aRYO}b8rvt5IqiD-m$mjbtF9bNJ!G^+P=JUu@(Js
zTcb8z_m#VQRU{Ev6x9m=6scu!`O3FcjM)}XCJU@*R8sjXGHFt`jQD=+1oAeEnqh)W
z7cAh8ZJ|gj@YW=3<K>vV1iikK$M?=j@IZm^Q|F;($CafSRjE--xA+Q3M3I@Wk8>YK
z%CrrnxML7!gHhP9b#NR>M7Jq=$&QqN1AHaVlRtV)>)FMVRV_UrVQ=ORm=4ANpqTsw
z)U{eYMhfb7FJkskUH~D$mGgo5B%t^^GygDd!x8^}0q}>o2k}LZ@3IHkbdp+-SFfG;
z?6y4DQwQXr8PLCK#2tZyFh^wgEDiMKwAkG7?!df3f^a`L{0j9h-D|Lb{6^KO%^UpP
zI||h0BGt?m$P%c!m1Y0qLtO4B|6O#o$;x-o`yS+Z>a{RE`LfE(W+lKIL&gzgK=&Ox
zEVFz9WF&lcIbp|j3A=Dbwu>I$X%AI=+^4^o)+w!=dV}z=m>F*E>k3m<fiPDHpK-Bm
zxDv#RW1&7uA0v52#<=t`>LdtJd=pWDXmHii{h-|LsEM*GQ=eu(*L%}|%wY-4y=_z2
zis`hh*JF_6CfFeE;xe#o7qJ3`Os^qJ{Q<Wx-m@i4$slT(T#Tb8Bx};|xUHdbvPx=Y
zae4O=X2$vN7(tuBp3w0cRTr<QXpElcSgD|p(C3L9k%<a{x-eDAFB2YmFt~vaqQ$mI
z!n$eHXRy^H|Fm(e)aVF3Tsr@%7BLG;8`(%Ffo8sD+nAmSN3?Pq&5;021+Hwz<8sM0
z*Y_&3jVQ&qzOSd1r$6a1?k#sfrK93@o!~<^Y4IYJPjd*mX%Ua1=T&W0S_?x=4xicv
z+&<4sj*L9(b(8Sy9^CEIhbeRiAij$}(*?=<+%UHCF8B)m0YUnZKqQDkOep=;a9%}5
zg+P@!xwduo>29|}WU$c-h=w!TTw2=NM!=a~93{Z4r^5CDh5J5>rTr@4J(m^D^)`9q
zp{_I6YabQwz-8Yq`T!8y^wV9MBNm$7XS)H<g|KTRUyL&ag?A>ZqYH!2J8KnrmPR8b
zF?GE<VEbDZa^5$9ytlVKyo;5*;~;(zs1rNNvzK4EW1VxIHO|%0?J0MmTW7GaVMj4y
zY<?iag|8L>8%4>_nJ0xCE-%jc%*{M}ekmPu5k|n|34Si(Ri*Ayi#nr4xVE!@xR{TG
z56F6v>d2DQXLTuAV;yVxkA*LZx{Y@o1t}j+IxA2zGA2+t`t1tjr+(CP!9-Tof>;(Q
z%Ve*A6&CO{w;C5qqjwGgqQ@k*#~Fq0hqbcU&H$80lKz?57NHg^awrzOdVkk`xWNhB
zAs?dNxt{pYO7OvnNH0wMBN50U%BSvg<353uiF=jejGfZQS0GR)wChdn19&-4;9pAh
z&Ne-b2gTqU@55M4y_fxVOVYd=DfC$5#r+{iS!@cXK5rHSb|41tbGkBsH#s*D9f+OT
zBRi)DOTq~V=~pMRcb)G`G5W}rJIMC<dpv}3>x|I0dFYytz$biiuH#z(QwYHMxRp|S
zVIu2JLUee<Fvlf}l=1dRvB#1QEb2S$dV3X#rAc26<!Fn~cK9(IJiRypAfY#=8ug_B
zHYPT(herN#@i?x>Q1;mhaO2$^Ng-yf?*QP%179K*DM9kmM?!-SOI|eJanx~=6|fSW
zZ9ZQJKxu{C;#hWje=zc`NY7o3F;_eyvU8cmUNf!KKqRXIdFCT9nEV#Vcp!BY$Il~S
z0iM&!4ZkyYn;A9cv_XS+!FLUhoq;lP8PrP)(<(8EHPWO5PoAqB)_K8uM$0@&NoAyH
z#oH=>tc(w&WR)g%9)fq2-~t@7llWX$7}IhsrbU+50-jLq(0kQOzVzdDoZ*|YmYY|x
z?j17|gg5~p$V0FQ@<U43u9LGm-_{tb0^BG@D%zV6_>K;oMj(Mr;avC8{?z+tJZo|>
zwv2oLr0YP-t)hGQB3&7#9Gg-^zcqo?O1N$iN2u=iqu0{X`u-zQKO|8m0JQ|X-Qv8!
zM~*`;VH}sv=Ea4ISTSMunc;%P6YUt|!k<A}(ptHYn7j*_;r$Tv%+~F9OZN!=oUluz
zIPtz*@J{<CpOvM@#70n_z$&Xoc5HcJ#N}{_DlYybvM6yk`ye{u^uXw3%*ppd+EKj-
zUPaZVmSCc_#hoj^;ZX!GAq6v-d>5Gf#l$DE&ICWwezv6V@%nQg&VC0bAq^#zhk2d?
zxPY9crRC&6qwU;{4MOn%5{lNiMGDq(W~l;o%u+wbYB9JOO&J6W2?aY?f@(0aeUpLT
zF31ej)YjIXQaB;i4kUp_+|_PzI=3jzEMFR>9`4=X4g^36m(A({Bsb;rB<9#vfO>FG
zQpZj`oS<McyeuOZPfOXQHl)u|G*(8}*<=(t#MX|Ye{Xw{PgYJ&+2w6`kD{zGsVoRd
z!F@3`WY3HvvP6}2NQ;d9<CcfL#s1N!gG?^YmRffD%I|P0rqoIO^i4c^(TG;y=Sq6b
zzqR$AxHH@IPV)09*-|N&7VBKs-oyY7MqAi6+>azdOj44Elw-0M2{v#gH}KWF?kV?C
zvzIe3F+N28pL1{J%)(7^-AL@s{DH#uKEhJ|t$crIP+wh&%Wbu>Dge_G$^mKSHQ5-F
z=9uWNyz$y^2aivUFs6W#0N>@Mna2DHVte@_IS%X~M%B@9J&+-XxI;e}TwZy^^2{KR
zNrr=(ixa>zz5X~o@6q?M?I2*xFCKfx#H5j;qPy&Ciy?6^VIZ9Lb&s-0D7%o(KAM=8
z6=Shm&>?{nQSV0|!v|JiIz?(vBkAcXKJ(RqAv9yQ2dMBpe%p5i0$6V-WfdP12eF$Z
zYvn{ds3wWZ2GNXtBiZ0gr9%huNVkuQvz)oLEp;Qj`S%P1KZyHqyNI9zVA1w|hAz8O
z%Cm3#QsUdjW&?cy%%(l~QfJv|@isUvegyT=E~JcMS<t?Z`lFcWnl1&vsOin&HkrQk
zwL`%lFetT!?}*7Vnk+~yq#sT|lI~+uWC_RPCwnsr=PjOOmeDGbq>koV=e7#P0bD7u
z9GxuC+W5;u?aN%wtxk=Uy5ksGPQotecJ7dvaQlb8^y<SQ&Dbr&kqfB+E#rsB=_*(N
z)HHJmxXp+4k5jj2t^Am%xi9C!_W|Wqnyt8)O%HyF-E}D#V<+}dGLJP_p#a<%`p`w*
z>rY&ONf}XUGe>Ct@ui1=3>n^2kDI&~hCWhjbbkPC0p!B1jvCRA3lG5)_99S*OnkIh
z!JihhNO2ufRr%6x5dKR~G)V<JRmDkVrU|Wpa>nS3?Eq!h(>fKgQV>r+8@$gszl7n>
zl+hiB{F(i-LDwL@%$vU<`LSvelPeyuRX3~@0_FbL@{7+@&yLSC*PV3rE-``bmlp-u
zn3$d4Gbs+Dn=I7@TH1&9$L<6oN_T{~n&ZQ#;(=u^J6km0@T4w?6>hK(g>edzm~JvI
z9K)=Bb)5&S5%(%mZOE5?uZ6!lcC4Yvtd;@Wv5oZ09gR>}K&WTe{!Ebn?rqG_PsVLo
z@e#2Be$Chi&WwvOro7v56$!TQW~7Hy7f`Z?muhV>_yB1MsV*CR(NYq7`4nmn3_tKU
zCKde=xKK-VhX|ovYBDEN#X~t{N&8Z3d#TiNr<4izIOuV>4mlVT;^c&6=Q!46xp0!C
z2s82-28Ok(@T&P?KonXpHG)4*yIzGMEC}2=%B3&MWHX!>%}DX1Z(cEc4KiM2cWm<d
zY6CzJ62)iq9dspJLIoiSng+#}x79Q^(=IK{@Dy|jU`*BpU8x*$CsMr+h-|;V7Ix=y
zkvi20;)+2DHsSkVC6Z&;Ro_;5Pd~R0<yr7r(3fmCyB}-_(DugfUVv`g9$ac2JL$Wi
zwIFu4JqB<B-<}nSt_zpqa|3Bo^XU_(F^+8g?lD%s=7Vrhrk;k@p-{@n)L17?n%nBE
zP47MXEu4DNXMYuN!C!#t41bjXNJH($KXcwV9P-4T6iheEW9&D<0N51b&(*SAa=`9)
zNnnRWfLr6|(RIO!6NS9a)#$qOmtO$Eu+Q9$F3^dmS9AZPcdqN$ca+!D=(BY83m-Qb
ztomg!(DjUGbuOLg@zcRb^dNLVPP*K<rIu3*DdR42YR{G%$5{S4b%r>M+prz5Y56^!
z)~2|f_D5G1CpvvexXX2#DuvtWL0Zxg9Tp<s9<Cq0zH~4u*(H1&Gb2p^6|IVsg$z&n
z($gjmM`L#+7Oco45R(oKzRJ^3B@mhHZ{Vale2xrQ*8WFVex#2SE)0)#`0eJNDc}J&
zSSWP>pDY+aBnzLtmfavMka<lvHm5pV&RkiT8i=9t|1ScpQ45}g8x9d7RWr-E6Nt^X
zrzNB}n3j6wqDj0!9ANnN+Mej1X!zw~bp)v0@aMF-4<zZ&Vwx3VfWt{t8?yRp2t4vn
zZDH8O$Mwe&9>W4kZCCiTe2z%OEzix|^J#C(c75QX$yTi)O*+%=<1<(=(?)qF)3K=?
z7}qkF!iiYaA`nWrI~ZLJz5;JKx|;^GnmGqxBOoX_sGSB2xTt-f7duNsNpzdUzL(I>
z%xf_qv#V9=gtoT!i5Y=ZXI#+!^Ys0pJ8y6B6w>G}CHdB#&Y$bP!)Mn~Jln~%a?bs|
zDK0(-MU;exa}c9M-kTbK%)??N$ZK|ZIK$YYzchz%p0b}kL<AZE>TY8${!gp~EJ-aa
zpcXi(i6NdY^7SL8s`w&GF@o%6p-k~^do+@ju7iA{2ujb+!VS90Z;sv2U&P(9cIqe=
zq0j-U?KPcF#t$O<)X1!!HD&Q9mO=AZ?><BbEWmynDT6Agz%rjIiq9ss?`_Goh0;CS
zchVr;_5g=nR~(cZ&<wN#SHPs=nhdE3{=d8opXh8}s-i}!cfnu8Bk~tgp6%Y#r52I=
zj=Ns8{M(DpkQ<=qdLeosR=`wcFmE7zZ8e|@;RgA%H4XaR-!RcHbJV^4J_`*Vp+gzV
zi^VyM&xP8F0c-15rA1*QTnSOQ4;lD*lI?R&H-M$ArgucN_xLbMI1e@fhzH%8&IP%6
zs-YFhg#e8419_0mjG$lbZlMG4<(rQhuP+d<(`FdQ`s*0k_Q($*-A=kG&t*nPlY*qO
z*JA(Kq7Q?!?JG;|nV<su4!jrZmQI9Eczh?VARE+<4kaY*ehmTK9ymGSgPeEZy6hC5
zSg;f!l`ntdPRx}l;H!&W)Kk~f@P1;9EIHDsLk`v71jO{5zZmmfcypy<LJ0=%z(br2
zJ5)M5)_DnY^01$i$NlBxr;#PJmQVKFpyCQ#URp4<x3aXnp)B;}0H}jSn#aAyC-j6p
z1(i2h;152@X|~v4`A*`iw}5Njk#*F-b5)^B!4mikf*8OF>>2D*pMj3es#CcEbz?sN
zrs_Q&cC_O%-Wnf#l9>4lb-Zpo0;07ei2r*kfE)sKNx<$ZE7#H2JKjOPu5NJ<3rr5R
z2tyVK(*7Gxf&5>a{{*rGc!B$#n^b-noVnA!c*viWfmSi%@O$w_d|Z5%vS2#uFjiGr
zX$&4xKC~tWm;X`@sH92jQdC1h%)tM=9fZ_$J}Wam@vHj>78okZ&<I)4kYbn%Gk!$<
zG<YaM*CYW2V1Y9BI~Zd6Cxr>W4z~!XQ2b!|A+&Xu{W;K*bU!=$$oynq*n8Wt@1>cq
zx6{4#S?=*ML3!s)g!E4Jr!GxY3U^`z9f(5|^46gMjj5u4Oqx{0`tCFA?CfpA2hQdB
z5Au^2n_6K3Am_zXI9F+Fl5<DW%N>O8jw@s^wb){SMxlcAwo_Oo22Km1L$b<(8DW3N
zf`cGSm7!MvrMviCM>{H<Jq9!7D&i!>P}b^N6X-wfZ1Wg7v4I6R)>z=$U$TH0Kyfp9
z+I<DL5=6Jg0bjex@CMhK{F!{ZyAR~RsW+Uz`r7%Q_}lX*vVSqt=3xX9e3gUF&E_ek
zxu9DVr2=7w0SwO9faghp8SV&X!Yw>NyOsSsKe2tedFSmd?$gDP7EJY=a2%E`(|z#A
zkst3Y{OCd5m&&2%Q*gSwR^U7UOO_=_`+bLGK#^xo=`x%EH@D5*lK#@$J(NL$S1_17
zkfXZ6TM4*9x{=)FiRoXnJ)%G(_4z!^iXn#b#`r#3ND#8b5EtI0dMN&?85RIF9DWiV
zcvA5ZszG%=2vhkWUIpm^p)VK1T)$nVzL%6lfaxy$i{$X7KLbo21k8Do{aGuzGN;{p
z!4=@31W<v{wQ1OG(6PR4!)d%+JB?$1dm6s<Wg!f7jk+$nCK&-t-bZDFX;(Gz6HqK2
z8_-`7ZT8dL@DRdEt~EIb{YyEBfq_TBey8TM?<%j*i(mNK^BvSBtI*_!x*4J3=qb$e
zM64+wg}+lkQAf%z03{O)?mG~F-T=90G=gSo`AgF+(LFKngf9eV^u4CctgVBOd_H&o
z3;PocG!*{w;WZLO{2dASERiL**)_gHE&Ntab3Y>Aoi~kLJp!dDUgkq8cdu=@7-8`Q
zEL%bd?B0uyK%^_{(g*A+FKR5vX6A`o<1yn#5wL)Uv-%PcED=hvA9|dVlhXu{K%Jl(
z^GUD@TpGXvac7j2kM|QGlt(d*OtJn?e?;T&{1Fo(d#E1V8NV_YFCYeW|7QX*Fh*6#
z_dum<rD%QvOc@}XUBP`g1Rnh}yOQ?4kq1=38h9G8tAMewLWHn)5~`ay?@J%&HT?yp
zg(v9UdlP|Tc7ZDN2oq{EbMxyx@jjA503&6R^3)r{DxZ@@pT$Q(R@EXPj2gN1o31Is
zu%H{=<b-_4E`L1REPj?_mpzEDCBFAat+<&)T_vF|^QW_+S@TOOxkn$WdT~G~5g{?Z
zDGV<hGVYLiR-nBd2}E+dmfZ<EQuOLT3Oe>|3;kKnUG`DH8q>RV-Hbj}IGvqQdV*T_
zHrhK`9FHst0-&rt8_xoXo?o&)ZqXBjyzzRN&3c&DU<*S5DjvW$9eMKE<mH)@$Py4C
z(WRxO)ugD#D;);6cIa6>eZ75*X=RKlx_cim5pSXEJZ7N`4F%@V1|n!9dARYOrzdkc
zGn1G4!HpE(V)|c30+9|y3T>wgD(wnxt@BX2XCr;)B4;a1CBx99_zJ3pnZSzP-*CDg
z(O&9>msCGGNkFN-H;e;27s>G;#Fgp&*8T&!=!^_#N@nugvHMd+W}}`*DFmQ`Y<2Y@
zZi~$;Yw_jT&dN^)OBi%eTjfe)+VJ^CkTV=X&SQ%1uMKZa9ZlfuGEaPX#66vw?n_T9
zF<JcrC`N((;5H=^KF`@vw6pLX00*vDd3Fo1VUAu*On?ehp1u)ET1kQhOnN-P@=}#>
zl%IGj=7wZ4$jd-?)BxaCe8o=v`P#_uActH5nTSg6r!}&4rQp6LiURj33uod1Ovn~L
zT0yYYrPdcwbs^mY>anap0gszB`s^8}OgSg2bqSZLG|gUBpLq=^Tm>k2soQ$v(KrZh
ze2@_udCQ)1x3-iGKp5BYMM5-4xI8uoIet!%8PBtC9KPU7Z(-V}f^lenOrX-TLx&=(
z^<aGPlP^6eJ5qGM3$4=WpDCOv00<g#$bW=CC-DSV00ir=%s19#=of8LSiQg%gN#S4
zAm<Oo|D^(?$&F$SPM+a%duh@kpPQcG8lSP7p`HgOzOT$@tcWrKUwV~6q1Fy?w}i+h
z2Ccq|Rv39^=N?58h&*WC=Y)}>`V)VWm7^Hcs)5?U@g*#vGgL}4StT~cttM1_Tt0|i
zgdg(b1z#0EdYn!R6v^qUDT1$JF?43amo?)htoC=tD}uz{b+Sgz2zburR_epv(Y6<H
z5r@cUfv%*7n##h`nlbV7;NcOdNKUh^6u%pkq7sM{)-lU()e~A;)$kz3Y&1CevZyrB
zm)aSG5wkyde4^c~i&mCKjH(l)$EI6sP2((q`J(Ctv1UcQ)C$kavfy*SY)rdhj24m9
zC${f{_d@6v3QWl9pBOZO10^Ea*L;<=P9aOO%<JwgEj(F)vVfn&Cn}(dQ}rY+bfuJ1
ztY~&01I~J}xD5Xnrw7K9hSvZNL1{pPN2Ga@wY7ET7n3`sAa5>EDG&t&BWtDS`|~PE
zI*@m;-J~p5RhHU?PIWqvXUYyq#Y}@?YPik(33q}lScVN{Y0PQbu2FCtBHRF2T5)3$
zTti0-+*nm8mv82bkl8IEXj8N_ZDP8?l_-ayUCN!6isq@XfZKVAR2Z(tA9<vK1PW{B
z!zoO=p8?5SiUy5HJL!ZDPvGtr8+;1tO|q*?5OL^!n|kVgOzi7<3_S(&LrW#UAWu@_
z|59qw1i3=yiT<<dccVXU7a7cL_Z|z2Uj&MQic5<^#((GagNhLs(QMfS^5vc2mNhdI
zTuV%~&4Ct7wG-qTQ7q-mSGhM`gZO(&z-I^`6TU?GV}%r~;wwOwU8<JeVXRjVFn&z!
zvt`1oigI!Gom2=O>CuEQ`!;y@{KV)ygAty_KJIenaFtZQ%_shwci=N9FQTA<!pIqv
zA0amms&qy~L@1E}WcQ3=Ig)v-|8WMCYylo-lf~n-?+;VrES$G*Fk$~DEf@gKi&Z6v
zi2Tq+z3Q~wHd^R%5Gc$9qVd(PWY?t%zITF~JWI+|4RP-}yAr^m9;2@Rqn*MKm4pS9
zXSW<U?h`N#3(&TCd60@*hsH9az`2_MLfv1B7BDqIdD>Jo{1@f(7so%=ymA>aF?os!
zwM<n<8?Mtr&uZ#GB0LW-4m&Vzf)GkVD<6$}MS_(~`xF9Yx$?s_VrAm{4W2TPh=?-|
z%)}>LJl_F4j37|f58mMB#=W=aS-T4$Rbh2G*_mTuU^#_5u)HBmjRd8{4VB%B<UEii
z;P`2Ql*)7`-Mv({?-D?mpu-QUb08ID^xBso^8?3YL4l15To^4Z)xjo0_TV-S*?te9
ztvxyn#runO5zR!9(s1VdI1-y)INuL-W(e?f0QLw(;O!`by8J2}lws8sKdZu(N(Mt1
zW}#XcZO$B@CkDrWw9KjTI85Xy#&2)IJqbCegBj`{cgvlI`+G0V8*_X2*FPK+_L&fd
zDkZ>wnx7eABB45b4dh53NAG(Q`)dbayLrGdf=U!NLRqKk2t))G+8i+S+8<47{TV3y
z$=20%Xtf4`Ul5dxvQ0ms&4sWb_Uu#C5A>N&@#*i@Rl?L%Ce*9Bh<5qE$VG@VaZ;ch
z2+#ue$S&&wK4BKg<Z!DPs0mQhSv}oGhG_+=xmL;jI5zms;qt^w({b3<?GhqT<o?kd
z)2F9D`-a-EIm^5<>!RhsYHLKM(YrFIM%~7tg%=wh8n(7m@Sy4cIqeH}N;V6{_^3?F
z)MstVAfYQb2xr9f%AiO8!xTe+O2)^mZ{z2!=9+ABv-pI;of<SQsu(5sKXSX8B}&fR
zy(;aBY`q4w`MeirP0l?!vNtul7{XwHrhrFZmcCxY;HVgFv~-JjG5t-7#dGjjLRuP^
zP`l@ZD`>mRqv_)Zs6k=~{hSGcH?D<n%o`Ri!y8uBZC-eAEL(YE=5ic%7t1A_K^<;G
zUd#_ksf909hQQj`MEuZVMZ^#eKKsZ$RxPu;p-azZ1C^W60UNSzM*T0G;i`As_#sEX
zUz@E7^{z(+LFdkNfSV+)0hGJzmyE>?Tgd~FE=p{(pI-_CVvBv%9+(6mwmRB1BU8Yd
zmLZvCzJnq?=e&AaD7+Fs2c&bLR=>Q(mD^g+cjRHBGXz&d5c~ONQ2qX&Q`&P)*HQHf
z0GT&s-LJy}PtfcDHF(!S0G&wEU)Ug;|Js)Ccx(|br3Rq=xBi+bMptkL6eHY85mR|=
zp?xuw<1R9&T8l?0ehWh!P!=7YyJ~CFoP}(j!2-&@cmQtxlX71-JVFbVcaN}i;6Z&2
z$%+7%GrW6p{<-0|H&+4V*Y<$+JaC1mfP)?i6(uu4DfEzb%9~#fMZE@EE~f^)cf}tp
zPxOzn!K|A9X*mf9NIq;~SLINF5W=4j^lI<F1hNb7Kvy!|_@V_<GYFe$#GI`>L0cBo
zn^4r;67?%_kU)l((DFV@(1nl^tbwJy-3w`>$lE>Wi=f!DigOF6(vE^I00@EgQnv?+
zSR-|v&))@L&YFgJPWxB#SAWH9y#S{oMUtDH?Q|)iSr35DwX5=4K{>89RRU9i{UX`c
zh94e^Itw}Kfo>GZV9W>r=;crm`EmOSN539B)Ptb|{^4`%tsZVIUsq`sS*-w)>ff8b
z^;gOfjYqyp0bQ67M8n4d8qjr8r=~{QrJ&S5jj)Z5<CJO8F|MzMKX((O2ll^`lbzkr
zAdi&MeQ7zvD@~gH+!3=e;`i|Z;Ot-uh|TwNiKA<%^@<%)Mj_V$H1DBa=;H+kaF=!H
z3xK&dYI=IX$tP#WuUx9x6|{uWj#1TjM1v&G<zA{kvD`Wd0OGP2C1JP+RzrRXj?vHB
zwxyu_L~dI+XA&xzQX{Ny9De%yEk*9@zy;;Xo1c)qZ#m`pOVC;=d;6>ML3rEhbv$_v
zT>f<_N#Gn(MzlbU2s?2cMkU$AEcwt7x-RLCn{tfTZ|;}tPOOiqKk7c&8zt&w0kxwX
zc^QA_7az%=LJJ{#;X%s!1LcXh2V{O8&_8<71&8#u2uuTg*+y?fanL@2?*Q~Q@L6J9
z$SwTznd<=AbW}xZX%Zyg-)qnH|1$a2NvoM7aD68pzQ}_w)Tg)5gumf{a!|x&o_I1v
zhmIb`57_~)<E<$$mighjVeX(X-iX?EUSINtZLgMH9_H`cHvp2{dn)XH9m+ppg6ojT
ziJv-Tj9M@vhDgu>DI;cB<h}GAYO4_VbaX%Hgt6%Em;E&wl=FjZq-^1?R@WTh;&w!1
zS6={v4HORR@%!k9%U18ltz_(~Mu=L}r$z}`Ygy%dF=5c3`*l4WEO$-6JwEU+=(oF2
z2OifDfMj`syN0ffb-t4Wb@vFp(O*wZK-^sgc9`<0$-M+^@&^{N>r_D6LR;+7)<E|p
zCqUTYk90%%6QB6NJ;9)DV0N6^>lXwZwqtdNuN9x)12-~0kuP=mMeu*Zz~iXG(IU|!
zG!p9~XbgL5m;vrU5rG~za1?aX-anh6$H~~G{fj}s;iCXdHPlIupW%x*9_$juPFg)7
zX%%O(c!VZS0|2r2ezk$374y2}(rbJ6i!$~8y21`|Wi`E7{qbfJg*55;te{^q^d~O>
zq6RV@s!P>{tM5ZCGc_0-WIDrHPz{`&Z!BU2Z79wlNO?6tB>PJ-m_1$9j*G`M7hI)D
z8$Lw;GMt!daBDKdKx{)Aosx8)7DYHU4{(y<ccuX*2nYs8yc+W6&^Jn^g{F6g18KP-
zP_LxqQlBbG7Mr{+*J_4_4XhiDR7n3Tw@x0lUSfcFPPF418Y8d<g2Lz2U*BXLsq4OY
zp7AFQ)nd^TJ{mA8F(tA$b0wkp`Pn~uk#Gp*vcJFg6%BSca2TMhzjYtbX;$D&qkuXZ
zAS?vYuZJtZCsuhWW#rypz+mD+QCyz@UEWFDSewU{3T~8w;)7D>7VERQrtV~G4EnF-
zX$iwn=ME2n(88(bi6Zg<_+s&31@r6L+iQtuGYfQl+coL*H#nDxn<x*KLIvPUziDaB
z5L8ABumW_$>6spP*w2l*q360ia=P{`)JDfxik&e@JqKAbN@^SQkmV17l-w2KPJC80
z9-i?QJfY&2+6jf9%?tld42?={z`p?!Qluw2=~n|tCnJQVZ;c-SExfDJ2%5hksGwTm
z*^~MpZRhAgft@5K7($YGR_6tJNLUKg$kqJf%7?bqysdmHOVUrtUqxk&fNR(OXKdz1
zcH}Xe?5*Jg4+zSWY6P9Rs}}tO8UdE*nQ^NSB!~N^*)~33;Wwrqd;~8=3EK-h^pJ2B
zsL3Pi5>Ys~a52$@?JA0Cz^wp2Kdm5W;QMPYAF_gPva^Os_I9bc*zaC5Z;q?Rom)Gi
z8|Qy%sd$F+C@h!C(7{YTbY-XZyV}QRF}e}fQW>oOhPMuEo!X0t&tB5t)>f6_HHJfQ
zqxr4c_+u?SP+!=$0GxsN%`~mu6{Y!+!a(!84V-hI0<DV~tQ>1F!2d6d1|o^p_Y!L@
zn*Y9o3Fu+VpqQa2P1-kg{KTq=t|__Le^to=5WU#TpYclGYN!^;rRV@SP;lWM$1jKE
zr>gkPi}O$=B`({$6RmUF@YP9C&`~3BL%|Ku>LxZnJ#<I7Ir*UrXwkR?I_U)~-MQAT
zX-4i?-Q%~k^U}cp=XUw~hUX{z$vlEk6NG7`MB(Tx3O~^UaPwkq?OuQfum#Ul;^PA$
z3e>gCpH{m5`6iNNYl;@M82&}Fa!t{ygKHaCMpml4h8->eI3Sc}A~v;>{zbR`)f}aJ
z(Y7P*03uE+orl@vx2C<;wDYg8qTfMmN0shg(=(!j5_g>h(3M}}zv~8wN!cNYCZqo2
z9lu5~qgrF1N(c`AzhZ%Z+Q@{P(xeMRx%yu{Cn}+&3)QEFb$59Su2x=xJ?ZkstfWa>
ziv=+I!v3#q{WP=3Af>B%O_m+C@dP=1(C2Meylux@5cl0vjhB>o#x4BIJq5)E6tb(l
zuZ8X3P`>c}zVxyo^bp=Vu+VFP>Qmji%{^GHx>B?Sa##S}Mi7Hu<Ues&QNObn#iYK+
z70^2B2qoyjEP=9Es*-aTcQS}(B@fsMtP{HkwDWes+}3^y1a>;vilFB7>u7&(z994T
z0hPJ2N(8NGK!6awg!N+O-(IyZ*+40y=&3dUOsa5Dl2=SfN;;V4tGO2Vu&p}h_VdsF
z6~|Z&^#5Kc@h^p%8bR~iaP1iA0uP<84gmE$sDI(GUjB=ZBHJHu@tRd9(1_Syq<8sx
ziO-}W3S=+5%rj)2R(FlaBFLM6eg*>dQUhoEuR*<WXlQajGqgZZW)vzMy~^X5_XH$p
zS)m&_qxg*#>KAw-owTBE9HFF<aCd~RWw{3Kp7=ajX!G-;rBTFM9AWY|7~8qEI0E#>
z25iII{`TI98~1G63ypVC^XSnVl?7b~PXsC&q@fPn{}i;fl|sqT=aoX7pYA<bgFYj#
zp;xlkjqh0mz_dM%8pDoZYEc{-qwomLq1oiA;NcCR%f+bn!O2*vW$@U=BLHRE)JH>o
z;~+0+wu6dM_dzUBsuk6+#+bCLc-QSkWDSyfNSSQ=ZtOf|c}2)mHt$RIN4Z$_dI^jG
zqp-5H4Aou_-LX=4-Zs%w6$%<3#UU5-nUM}#GTftB&Z9cu*Yl=8zalCd(ma)u?gD(N
zvo^aiZVFpdIICXVX8TfY1Zkp?%&U3m%VXq^L(7N<_u+M>NG+MCa7uzq?00a3=!j#`
zjZ%sxs57=L->p>V?$0a}?E&-}wigh#dOd*!3=_womG(asV%RQJkehf9eJ63X1MU=|
z{R{Bux9{LqPM%lS0BS-E>W$7QnAQRfs1({A(A_9N)F?e}o>}waQ}3ghqQ7ZTOHu(m
z$K*)sjsJ7ko9O@?;Z77p;V6+R8a+I<r?(G0H3HOpeF;eJxs0?KQ#*oGSSbL)o}JDk
zDW!e^nn4Z0GlJNckEN~oN@}cVI|ORA+1^300IeH&On*Fs^{YXYp>mLX#Bu1XRzGVs
z=nRIyYn^XMcm^G%@5=@U0$lowIzDbH?T!Gc#S5PO8u1*{ENN?gzna|Y8ZzgyJ#T|G
zSmaS-xdv1T{L<dlI%@y^>M&_ifG7+kAfb8-dJ{fycZ2)0u7MLuA9SLQkum75h=!n;
zN8TBxt<_1?R-K*t#6F2JLhull#Qr$!q_iJ}otY<bMJGj5GvKP*MNm*=V3}R=62Qfz
zeCaQ{V^rR~IA?_bP?A}2H8VBd_ML3g$Nqm4W2V(bA6$tfX-qv1@cHkG5qRJlu{Jib
zva8=optjgc;I>3+pTsk0Wd}Ni|37f&Ls$Tq^Qa!UzqEI-s0Dhy**o);*fa50-4&kU
z0wk{OAh_}2{pXamE?YRbb9KSNb-}&4xw${@ur#UDkM!0Oe3AN3wXy$HR?%Jn<N(QZ
z2WO3<%OGmI5pl%!vjIM{0Z^__O!v(hZEMOZ@d>3ABb~!J#i;AW#K(?jQk(xYa@)}@
zgo)dDq`4T`%M+;rhcSDL{=_T^W02;u&A&#BT06XocoN(~*<R>n1ZroQpS3K*xt0N(
z>Kf?XdyeAL4ghEgxgYZXX#47bF4L^-83t59QISweP!s_H6=^{f5eyLNMnR-ST3Rd=
zB&4NNkq+rpq)X|BA*D;YzH|QoN8Fj&;d^KQ*j;yq8J_2U&VAxK*SQXN6c>q(=Vq+s
zUQa$koGat|!r&;xYjd`F6rz};aG$udj?f*p3*2Rhx@Bm|RRDAF3d=a;b2*sndh6@w
zkkhc*>uY_qOd_fF|M)BlKru|UF7l>2{@SR%EjbXTonek34TV;gQy@5p`bm?d+D|u+
z;(zc(VAgk2Y%jDnlqGk*n{zySnwzpmH4z8an*|_21JH1aZ^BRE-pB~~A8#~OgGocO
zTwsV<>B97Pu;U+nns6G_&p0b7xo%GuqQMOp2xOE@q|lZHb$lDRnrdNA?2F|p-7GvM
zPe0--g7UHfjMbCa1dwcASWS=ha6&e7+;|!=_)J{QZNk%+P&f(P<#=Y@JY8_>KwBZD
zUfNMc#hMK%?<LJ3jW>f58hekNv5S*9EuAjNzk;NbL!K<-CM0rJcpa_SW3fMTNfv4i
z;<H-^Z~ohv@RMO`4FN%7%!dvhz}7z0d+J_WI@0iZTR4`twrT*J2vhE&Le)+ACdm;&
z`Aw0W;1BD?hvo$S&Cwp%+Qq0rwY_li9&-|2&f4oIF^U5ZfaI@>fh{kWT7p{M(CVf@
zQg4{>9t;=ew0=i6Qk~dxO*Wve4=k5bTtIy>*5CZUIf<=BOAR3^pXIru)&a;AQqbQh
zxR*6bf;Miy#Sw++BIc}-r1gM;#tKloM3tpxX!DG5H|C!0yhD9qKc{5a-7TwHHh=6d
zpUUnV<|qPP9oCOR4@Vp7J_sE}ApEVAnhy7qqd~=X1w}F$)0g;#B${hR9T<5|-9*73
z&jNd+n~{I*rU>pMim02c+hpfjV65VX@H+egjI$I_!`28eLCk~1nyZYSOk=>(d=C64
zFN}2(aP_((#_W$C<A3rHKBsP9yzbALSo;yFHTi%(z<Lk?nlKT3oNGS@5x8TR`R@|P
zS1G%pr&9=eK3hNe-+vO<H<+G|q&P9gq9X(N&9;c|c`*%v(up&W|1u$jfZIeZU=We`
z@euwe&oJ9*yOur>Xvxj=bweH17;`_7sD?q8-A<ApGGE~h4}qIQA$Jv^ZNYNcAFAuW
zKLE^W4Pro3C3?^GHf+l-V~&#Ffd%GpTCK=~sUH|0E+`EKg|V9$n`{D==pLYFI>fE|
z%jn`nZ&n<2GQdT!OaU%J3iZF5T#Uh{mqTZ0krD#5zv&f*dQ07V>LhM3S++yIrD(e{
zGr883wU?7ID$91c^C1?NDeb|8hsTaExCR$tksUa_Lx+k{bL2~Llj_k>zRR{Gbv_5x
zehp3JyKwr&j=D=~jyOCd2@i#d_lJ|dN}4c>vmGlK>&S8{KQlMBl&hR;+V9V*kvqWK
z<6EIKrs2XfmR_S0NxVsc5T|ubMSB;;jg9KwKwmNC%b9H|E?VH!%(|IdL`bfe+IUA3
zxGxYiDjJEP$XIp}1R1;$THvW>>=l;}F7r^3)zp=k<SOonEMdmYvkw=e)~kkL`0$D2
z@2-IKjdtjr-P`lh-|)jnk4=Ab+>qvr3FKFLOwj;B1Sn2eEwy%5DdZ!06<&*PC9Uyc
z-*C{R4@M|-5kF;fnu)&*yJi+GCn~$=>g#p1B;<{F(KgHCH2xOlGdI$_)}3zt5Plsf
zgntclaqex3;wz@I)J;)~;#O8Mz}E#r!fAu<PoSwKXuxu5W;2gDLL}UZ_Us8hIB@J3
zzTt@){O#&O5W`EjnLnKx!F*&pw*!$dlR8~}{xF(aP<DYl=15D54k;Q$<hd-Qg|7(g
z<VlcCc=P^;1lI#Ed=#y<_UJfK?+MH-ZknwlY5&zxIS5eevAia=_g$Zt<8Q|qBJ54*
zapoxkghZw9N<7RXBQY_IVekW;UakAni!V`9(KSlp>sJ{JK>Xx}d0Hh)O2Ke(l<~Li
zeUB4VcJFiUD4#RjSixy>Ao7}P!Qr*}V#Lv!3cSBC)t3l5!JnSs9+-0ea<c^sHW!F4
z|GwAu#ONPM1a%>nmSqu)krZ-VlZpUP{;JS6>*DVH!mZc(3y;2C7t$2tuxr`D1^Yyf
zyt)3G5b2?I9N+_*k^!h^42(9yV;~(_q;U<ZA<;F;k@cIe{l7R7J=X%LyhuPLDr)Ku
zDwJVFH?-_D_cfU=Bz1!jEn`wm_3swCZ^J_G*Y+~pbQNGG*89tp;ZHBI!wrmS1O>0F
z9`lc4Ym@gPr0Of!WrsT&lZNiRpKPv`dB<t)s|sCEk~)HjamE${g$s`{(9Tl}Wq_LB
zxD1N#+jtJUZbHl_iFhl&Uiw(twP150nkSm%kT!QisD;7GOrs3fWsN~RB&CHW;cAqE
zzgwrnrFE}GRO%#=2)zkn!I3|xeg7T!-L>z8D8@Sd9irAj8#dBN5!V8ETk`X33)R!2
zxpzw4G#lfTKtP1ly!h)0Jv&(PK$B@UuH54I{(H%>f7kj+97)$Q^aO<6+kt&M2k|Rb
zA)RZgcZK5(u|mxG1)E~A!RYcGcM;Vx5Be_vVP@0Km#eH2S}%Sp1>gA|$7bU3;ZGno
zBmU!KyxM$>3<Cx)D2YR#TW|9*=rNBWakGB?G5XPCp!Ejbpl6}yc>vP<j!D4s)<4L(
z#jjxoAl>RUh84T1{wa28{ho$AzZWQV4YVAip6PA#85l9oAS2$l{tS1q*l-PnSC`HK
z(A@+)ifK{Av}%INXL@!Rz6Ygg6Ad5Gzaz(N%RN&ceZQtPN5$ILm=zLL1#>{5G(<TA
z-|Wl<f~U!6w<&Frf;p4MH&2b$Tf&R`B4M9nbs%p~DP2%Zl0PKZtgvs+nW6aWNhOR#
zrYNXmZ86g88%El^=Y5jm1p>xn+oWTOwZWm5W~Z}Bt>dCmr0@s@pmswD6OytWrAXeV
zbD_!%iLW3dRkDLUxJBs(J)xk!^ka}Gdd=Sw2owG%lE^jHp(vHUI{ybx?Vsem!ReS>
z!$2o$eOHDgi<pwDK$hvK>c=OzU*I=pySD?rX#gDYevA`EWE-S>7?C6VJ<a>yJOxG%
zA5DjP9xxee;{vUj9P~*<?3~zJP8`1xNWr9#da@Enemzr%g%QH=Dk#%<)gcVqGrxV@
zuKW)U#&d+6%UZY7UhF8m^;Rx40lt9pSN5<COelmxBgcHn#3MC7!dsaPN_P{jAL>ED
z6YJ5{VdD7Jt3W;uC*7_N7X#e+W7Bwj&FkI1V|1`PtpHf3hhRAHona+7j2CG2j+FBk
zIbW<wUp7b3F1<Y55PQwYPI{B*eL~glJul+6gmZuj(2<<`+tt6gfTD@aiE3ty|3ATH
z7X|7MmSY)6^=CS60kXd5OsP$Xj_hyt(z%8hfqXr%5{0oHM}D0R2jAoQ_wRB2-}xR>
zI9$dfO{bA_U}$X3gMtX;Frz5KlF#ChaooeSiU=sVJ^Ha?H^mjjdN;h6us$pW5VZJx
zyQ(~oQ0^Ho7O1oy&<wGFO?)q-&kL!pOvjm_R0lxsB4`y&L#mUBRZL-YeK*y)^ew@i
zbR)Q12*6GK7n08s0KuS1hT3$xpiXDrxCGKDmw+@ng>j(!k6yk65$@U#72&^c5a9OE
zJ~9~ak1R`}vYPlX35;+li0fK--$t5Qk<fC@4=tgqn{j>62{K<{mhGxyNkW2jt)I9o
zd3|o~swg!y=s*^aQYq88G^+vK0eb>URrnt3{t{_`BOQ<}p$zN0F);i)#z6d-0ukYn
z?6&9d2xQ&};$6R)kc#VMgLYp;250mpHOIqs!&}$*T?03zyj`KdNNN;G89g-C+2|hr
z{Dq+WFrIfF28hzLhZSL2l5S!l$!YKr*=oifarr3UnvVy`fghvSzy2w1)l=I`$I2Ii
zfHRqhAA4n16-ccMjd$drc&=mz_k$TWNrY(v@;t~~Q?`je96Pe%D!&y$pYmUDlnL4x
zv_H&tZB;5RQOJdA5*nw{7?VUg6rwz%@3)xbW)vUX14`2Wf=S*|2esYHPTe5s`2qkU
z?r}1}Y*w0}4nX%($qqGcLHP49oDoyI;2(bS$wr0zsslEl{Tp#hdIv_P7Uk9qjdPG@
zo<Y;}>%kMGY5AA|cHeRm8&Cn!pvb>=5(UtV+mBRVJ7$<<6R2BK^~j+hEl1Tc9+Ij#
z4gCgTo5juM`7R>+jl1Ez^hVtC5e365&3}^a|M4?o9+;f!#D3y+c7ryJD>j!RU?qrj
zAtbd%NyT~p78)E*@oD}nz8^`TejD%1lk`%f#Ke2xiR_>et6xjK@r!hf?9F)C1w7=L
zb)OERDCafS&02BOcsKr$zbKDn{m`>m>B)0)V16QUuN`g4ZOHr$KH!i_nrAXAWqyKA
zoI>BhfC#$JAPIpWPWYcU-$*nBH=l-(WHW3e|2ts=1QTUZD>4!0$Oe>JCw?zbg9dsT
zlpn@8=yITu(?RCXe_T+8%?o-Nr4s-5OeO)<UXV<A4s(Mf50W;BG<;cAa{ze;D_~S4
zwwtzkc67&V&S{Bte|OlN|ANE*+=UG#$Ntfl6vskGQjYW#j&C@Jfj2%0qyi4(KXe6%
zA~sC1to6jjf5A_m<+KAhf(D3ytz8XzANK2KPhK`2ZrJmEbe@0z1~POc`-K|}j1dh!
z-R{gB9>(}Dnw#PhDc-8`=R!Puq)B#GtA-Y-av|9+JR7P5fL{eOV=+p{JV)ygK*m$X
zyMR*S6BeSjff@;`U^Z67X2DICx*4vP0Fu~#xN;;MzVKZO;Ctxezkq`l3q62!#zz-I
zzeNS1_@E3Tf||8&Dx)e>p|vMAjs<}$uRH&@y)kps+`qSwKA0+{3)83$)XCdN{M}{L
zU0E-~>ITomtvtr&idoSK#@b$O`V^B^Je76l+SPkN7GPr}Bfyn)7(6Fx$yV_>!EKCW
zu>(0|TF_*aR(Wd_=&{Rqw1o{eXuNXM_W0hR79-tAxxJ%(9;pHB+I8f_)=gdi&2{kq
ziUApa^}2(HWsW%45nnDm<j2GIVp1SMWsrxbXGAdBvAVbU7j{w-0q7c8Ey4Yr$H8y6
zk!{<t+_x5?D+$RB57=G<DBqVD<ybnVDR=e10+*d^k@6?aEdE=S8-N6xQ9$E-%p-~h
z&cp-XzsEoNLnNFNpoc7urfxeJfz0+n*#iHk7LAVN`ge`_Ql8U)oU}KgU$^q7A>sdW
zmce|Os@6KLJ%cO|XuZ$ODu0%QuG@$=kuK4>JE~|(Ztn^9-DIl#S0EPi>+|I?#-n4G
zPfzX&zhsATBGXT#&$_Vn8x&dWs8S*UUOfCw#(#PAEk%3bu~R?};9U7u8wQ&n4WRqg
zcM{1ye91H!)@(~RsIqYy#Gs&quWnwU$2$wd1?uHqg$cI{R4aT=C*Jz-NEYCdv-%z9
zxJ`zLkeupSfLe%H64|!KGrN?HT&$r?FJX}mP@ctS|G75(Z@w>4Sbnzov?|D@Iw5YB
zo430P0BlXbEi7Z|Lye&V0-lo~xRh7n{XE+8MMm2P(;=nc%wcz{g;sn1ce~J&Bau<W
z-tqSy1GLZ?G|SFv1YQ2=J_pZkgoN3~Q^Z1`3{|KuhW!^&_>bJ+j}IE$frB)MgmZt%
zo~ngb=7WI%GX$!<ego{rqMJgL#EzsaRU>Pl>gfQwH@EukpO=7=-n&9VI6Rz?P!m)v
z-UeD|w}Ao#S9}LrdI0pIlIFMTqcne<P-M==mtSGKUbr8cK?aA2zWa(l9%swf{xPSB
zG=u2)fZF-|>b&8=N~ONd0km8J^}Ilj_&~WAQ~7KsU}p%IxjUy9jzNFlLhm+$mb&D3
zzi7Y;J;t<kzK_+Qqp&->ae2?S5t~F_a225ZO!1S(!Qa~EiT!pfbDie{Z59Vdw0+NN
z5u(lEJ-#||FOrgybeGV&4Ah^eUEv$VIFdn6aXI>3yA0?18So5GJk4q*+w!lc7Fqw2
zG$Lj~Oxhp?kAG5NJG72t$0q=<`&7j<P61lsWm!obIhMr;=S46-VB9q{=(E*^ggS^5
z@9UkY^n|@xQD*5TTP#1n1rTf08qygSm93lxyhnuJelz}UdECGFQVxn;s3{Jd>3+pR
z5Vj*k8<${)s&AyxhZ%XgKxWE%4_|N(`f-ue_TpK|v0J@z!$y#mZ8f_@S<O-s;(VWA
zvxWDEIsKU_uRoE;^ZX?B@=yVL0+dl3GG4t%0h&eU1x;9(y?IKi<cTB4<YU1L#5N4{
zV&lPv#NWw905#2yuGj>NHvZkAEdwDlE9>|f<loz^L&T`<K$WHt?uCq2Du`Kdf?&A(
zJ0$@Nh-Be&Es-%35K?ka!$+96EPwf<sZ)PW_zo0-GUv8`@h^7x%<0r<$5dZ{7N|xU
z^u7`Ttn$_O;}5@y$%wRS!qL>S6$lxi-GQ3znI@=OP6Uw^q1IHrX)mQ9;8j}BNP&{-
zF(mXKWV>qKKTs2KEetYq)JFSS_fA3==EyL(=uDhwF;XOgieILReu%Z?wjNI}_UBEy
z(8j6aZFJl4tn7w$po=?mZ1migm;ZZTU}ua=t5Tu|{mhdpx#tv`c984tLuJ%hnpW*`
zzU7GmM88eaW3Rzr?6r&$Nh$QFjQT6)VLv*_UfyqF=U`z<pUqCGm}rHxdj{Pv=q^KD
zt_J)P=Ej`kk14!P$`QhHU&C_%XHB*lRTn~eNOk4DfK(7|U5Hw-b0XT@gr?`v`tR7s
zl!#iG4hO~<F#MEM`S~H%yS?XD?)vGGw8LiZ=Cq{K3+?xtZrdYDY+BO0%jX-t1e_dL
zH}m<8ca5Q0Pn8m$F@N(7ly)ufC1KyP#{9orJn9@QUdvffAX@5tmF;f@v+0)r`w4}4
z;yyglRl;Q?+GVI`T7$v)Yu|-Or5qEJmL}oov@%k(S?EJ6m*hQPJt<utGs|;y7v-k6
zZ1aWfZP3)-eee|Z7z{-tb=O%%lV(jI526y%0MgIK`H90ogP&~CiLZZPrMH6}qUzmB
zU*aoWmrR0EM#Yq@w<^Uw%=g`BJL^no$A8FxDe4P~65H`D;>@RD?*b>FojtIdN@}*e
zs8l8QFu2&%souiAUY441;F~tHlMll0xz;}@DPG^Q50UjLY?mjIR7gz2;9gi-fLJ4;
zfH~YOp|(e|c7>!W6u@K(I*RLmq8H?2`Os1n)G|-0y_Zya9L*DI7uTEQBl<hc5La0~
zkf)d;`B~nr?(s%|vv<rrE(-aRNTjd(Ht5x2*;u72A)R0-YQPn47SsiVVsb2N{4$K7
zAcLmEcyGG!&=eZv=P$u3YBk7waJa1DTO%bn2UJ7@5!<k80U$U*KtBy<rF5@<+%pSy
zJik0$&1g!jn+G++B=ELa$zZW?pFf;}6X@$na@{rIJ`d;mwJA}pioT}THJKkJ`agKt
z;J^(~hF59+Vi4?e4H2x|B$og1V}B79cI^`m%K!KjUkx@^89-8`o;weBb#F0}A<ce@
zIvMCHdJX4&tGU(#$qsYlT{<OH!JSj$j*=tCLW^M_l<^$D?j!cO))uH1&ftRj0+Jz8
zZ8iYR2N)!a(aK;<i=-Aad=0}A48<{`8iu?`Xtme>hzJ1}e%((lX=t--s(s4xO$$b3
z<?F1rzWiUl6?$`UNk~u^fE0~SfcW_`oy0YG0CtkJl);WjN@#a(veIx|7DliH4QA>3
zIT_=kriSrg<Xyd8h$TG<)b+agIwtw|&*6J~0w1embv@251+=PH=kkIU+P!oFrCQxu
zTGI3nQ*s+WNAMy*x+LI|DxXN6R7lmW58+DlgCxZLxQGMolOvr2urmHRVmvQo%0yek
z0$oQ~v?r9?c|glhHufp^PYYpU-tgzHjb|IS5JM^47ULZRV9(Qx2|7qa9!RimGI_RG
z#9tgr&5_U0h(N3hNvjM_^F!v4pS@-|IuLCF`T6V5&jcMw)a_9V9BBe0pd+>NvIVBl
zv$!Lf#Ey-|7+^LiZno~q!+dl#(Yi)8cM*1!Rm}s{s~h^(HxRCjRJpVVA~}P>2s&5B
z%+w&oCw^Np0u=(6DEDBvFct}*Y<@=93A^tDqqP_u^g*=0gX^w(&xV~Wl(k<y@4n9w
zDuGkEpdx5!4cY5+!~s@pEenWK$|^$Z3O%#9Y>($9gHCKo(}Nx-qggs9Ef|^&-@0sp
zT5Rt|Do(|`=*#((+MX>$TlA-N{Vxs;bFld>q{6wotws`yy%_I+%zUL4{Q`{EJBbEx
zdk&YU7(tM_n|muqyOSxcjSxQmY?R3B(Y|u;VzdhZao~Dcs+a2WFbx{pUp1~f=7TrD
zEN6h#;gWJSIHZdEkU>m4%VipCp^W`?rvMDlRP)5C3h6zffb=;HUXLS5f;MC)x>dmE
zIiG1f=%yEcuLpIwBt*H6LDBDJra8qk5Bj-PtwHgKWLC-9I|RBbwMTI+WZQ3?=8PF5
zqx@;UMSG*3(l)FY=LGM8$Kv4jPpoP4XQFq}+K%!r;Lm`^Qi@0!szk<|cCY=&!WomP
zE6yt`<F<!2{0%Nk&K;SGVx=1p$!+5SvWKn_R0gcJhjJkuY)-wSHqrF~2}L4`^_c;N
zMy<EX*il)jVDgdP-2PAUVMu{BPe00y6WZO~V4~4q#=~!hT`olU0_b4IXqV}#z(K{O
zcRQpHMMbLym^8d7*n9q*+lnuC`<)x9Wko<_II_^r2_`5dU#?%}bNVO(o2H!VN9tkq
z+-Xk9g7BYqun=uFf_jr0)@b3{b+#SaPTJ2#jAY`vBg34_u({5`S4e~vDKeH{r~1mn
zj0UP+Sxs^+(9#e9*TYs7W-{Vun-qU@+h<Q(L4#`|yU+|)(1S7;h}`PFZ*Bxpb*`<>
z_0--tUukv70#P%dvi2-kiJ`-w6gSFEeTndl$2)l4I#m|<&M%bSPzA<*;#lTzKe_lj
z-UVJGAZveuGFSRtys7}4m0PDCg7<V``WLl~9VF^4)*xeWLe=cJ!P@e``J-~SdVDkB
zXm%5%oM!!r8CWN^0cp)#jqB?HNUIOO>)Upm^Es@eLe4K?!hsby1g?i(J>i6OhUW`N
z*`5=!s$E7BaL^WbMQH@aU-()h@O689c@c6u932nTG6dknI#GUyd<)XXo(Gy;(n^?+
zi}Pq&CtsF+m<`5tt;Dp*)f`9J37m#NjBMf&4K3)#zpUn4I*HiI{wq>!RUjte=K9?8
z0^34S1Ms{$DNOWK((m6!;Qy;9L%oY_u$6SO6INVhWKN|bbLntzOjHx$tV<-a-%_Og
zVkZyb0}+$tm2$uB2A%Rioe;w9c$z4zjNJ4k$O(%1AkiI4xY~;J!xW7w*<R1~;@5(g
zjH*}3i9h>$;~o{81$ATasIc?HFFH%H6&{il&8=jx^UNGYImxg_eU%%#9hy2uQa0J@
z0rx-R8dUSn^^V5JseuW-cJ%{5p^q|?{V|1GD00cT(Irw{FbhuZt!Q{9E-qUyW7RUj
zaAE^Jkj?Om3VLcPT;EN!If%NRVH|^`9uY|$zDaKzf9Eg7_lJJH0VKyB3IizJ`wA{3
zBh`Ti${t5pRKZxA4Kcb1@_l6k;ec5cVA57(crz(@&CT!OXgg&z1_{OS@q?5)qKFC*
zpW*4tRf*|tpZoS%!2H_sT;3Wb3RTmyjp04u6kaNsHXoZ>;gfH@Kmfm3gEo)V$QmGT
za@f{cUGJF|EBlxNZ`74kc<Y*&eJ6r9K<2K(*y?vMZ}T?QXzSPidvr&ec>=bO3wbju
zC8<b`0cMNjoTDsXJpfa#e#Y@$yPg5na!zDyQ424k4+6``3^Eu0{p#xKo|pB2Fggh)
zUIZzb(z_c1E4$8WXN&5#jah!N$f8QT^QG+|$8wb+oid<k8P+`54Nvk3ct-ZYeW>D^
zLF)}s(Qv_rE;es<5ut7V19I^CQcvalVN-_ZMGHe_RnH>TJ%%UV7ckJWZ*_C|WCtS0
z7Idu715YENh<V00Ko%;Y8Mf_E?rxc2@7O~|1n;E0Jfd;l@lKZcbbr058T3+$VFW*^
z2T>J-b2+nzG+^B_$j~o<zOxHR&YuP;crrh|R5@CjeaI*Q(Kf07y)PA_Y%?dcwACPs
zX`ZFczKtRX<OPtzE$)|npMJ7qUeh%DoVN72nX-A=w_4>DFFGye+kAKzVhsiB=C1Yk
ziQapAVHc7ZX9w8<L^rIOLBtWXCu?9NT2IXvE=9f{u*drq#RUKtdM_X<_H*u?kK}fC
zw}`zx(EC{|KHFp)MN{DsS1X!K2x4r^X)vFQAXD{XIu@$Kz)*N?3;Fauv=jPxen?Er
zNuIm)PGB2%FCHQ|zdRx3SFT5=(m(~he6+G0c>Vj4*l>mokB;=Msz(u%Lrcls5jR=Y
zK7Hj%e6K5*ae!HwlU14G+sur@&DHX6k0{DcH_QEorJpk+4c#54@d_B`7NS)dWBCO7
z)*KFyD_cN)8N4qXCrAJmI$jaknwQT|0~$`{O8$`^q>`UR>;7~of0s$|zr#h#Efe8C
z2+3e%(nFC9&`s0<ipce4M*?*iB?#T*N^Ohh20F60hAQG!=NmulVR7*#4lAm5B<@Ql
zFAK1)QP^APv$mY5sE`C)ZNeANu<$WKc5QWOAYcYazCsj+B{b$sI<+@dE)ZPt5hawL
zhDjB58y06^*7jEgkb?wq31AN1D$<K)I*{+a!tUr@ME3#tqTZphR#T9%Ekf(;lJjz5
z&3BjU0seKR$M1F3i3GsQupYD54Z=~>G3vMf^9abFh}Q<&c?`Y?k{0d-P9;hfLctt7
zz1*y>%tK`uTc_;z3@FpgXw0dFLRHrehi^gmcJ{fc;fZJQYlk>y%E_u%Th;;zr{6lo
zorO8>V}yHfNHBnk5QV{|4mpT#Gfj|3faiOoCbENzHUy`1ypcxW*`=g6)#ul;*(0R5
zQhngf`o)gL{FD%jQG;j;&LiEgwIG=B)$HMxeG7m0iJ`JCq`eQNWFan{zbg|LuC8z{
zk>Vua=d&e8U0$N42E<uOOQ^RAur)Qe>4?ul`6SXjx(?bAtwFkA|2to=8akOa`0zJR
z5l?=2Vm?_@BKLaYnN~7s<D#NU)#}-2oz~Z@AnWM8^F@|_o%zVO(3{OF8hDP&t&{iZ
ze9Q|Bgm7tXb>--S5)bqkFM7ynIjaH=A1!d~`kt-d>DC@Ve`SA*++zTHsP;@oDQsQY
z{A0QOAFnPJgS9rJvL?msPqT1j8aOV86=2>e;@B~jmOO}C-DpKM82H`AEaO2!M8Y3|
z&-o=*A<eM(B+w1G&BiEtik!+**Fb0%Ei<Ks++_mmnisRnezpEj<zaV=7{{v}GVr})
z8?F@PObrOHmO@VEdQDJ)W_^qS0)@9(8EhE340s#*)M1}55J0Y&*8^yS69_moym=|_
zB4X((;7h6PH2TZdbVO_?d}kAWBVlzU&nEKzZ{PJN2(X4>0y-tdXbP>|Yo*hBWiR*F
zVAQ+6aMcDrBCE<BWv=E;y1n$+q8zmyj-yxM@Zl<F@5@y$Irw}jRhZ<IgI^zo@2P9W
zWeEt$<~K!kBjO<R224T~?uZyE`nG!9c!zrLgM-GDj1U-8=tD%+YAJeMP3^hucz#0f
z>cl}VjOoMfY9<ivNpqCm`r<#m%Vs^ZZW#RXI5sQ8D#b8Aq!O16B%xwhNG}l=%gl4w
zX8!gFaBiCg0F9hN#YWN+`w;UfwXyC8>UA*>7KUv*4)Ji5dBsOA-7$Ce=NJupuwr|o
zvX&gw9$1{5v2ZaAAE?T*lXF1(CJ~xRkdKK7!Hxzh*i|LJvZ@Or1;<|=!3`H))vAw@
zl3adN3^g4U)HxLy{_l5i5k3CXNJPA5Yg0HlymY8FEtl{1??JTEik>TvKw0nv=R4f3
z5&iGcFP!}-8e&#0K<m30KvS;(BS76=vs`om*V)<tm51=fgJ`3ku7VOf^j2&Op$Fu{
z`wif)mw~*N(C4WWy*2fXg%x>i|3imG<i)Zc51+EAm`?i?zF%O9YS42$`z$vet?QFA
zq>etWixO9~Vglo@pHtf#^b`pKGf)kVLMo6OR~Ky*V_nmQBtgnqgu5D~KMn#(rpiET
zx8VsAd;egvtuF{YW`K(hz7G+F{&I-2`S3Qjs@0rC><^S%^uQuH%jJ!G8LpTCMT740
zSE2I^jivyK6+s1}6c?MyLl<By4*r_=bpd;eW*9<b(Yn2qfHxnUZ#*I_x~{yB#mS$z
zK&bNNm0HqzzrwU!;>nNs1*(&e4w+YjA{oXiFh~__w&)8O4Y8+kfLU(_jJ@ygoq$El
zM^LAvk5>;cB&2cfpyteV91S67_b`|eDYyzGa!MMf{#4mFJb~E67Si&|00=Gp!jgGx
z8_CH#QsIjfh%0$Lpw#<0h~M;HWeYMaqo0IzW*Lf0P=zYdcZIY)WH6<&{`O_Qx!Q(=
zF>j<3Fxm$&;TOeuQRYZOjIMe1TxmAkh^}Yec?9=;xb#cq?FN%ZpG~&1n#E1EF~;k8
zJ~k$l-DB6NV#{j^AtRf$4R8*m%hV57qIhTrmlZ70L)k(xG;Iw7*%xXqPg$lje>hM`
z)GKzKfQ!ZBr?#uuEfltP=tb6odSj$aN9I<$_&dY5dyj|GDQ@Ghu1w>gjNpP~YqRnU
z)n-SUF)0eWUQ5}VCb$l&u>Lv`1YXozwUZE7#yOGyTe6B~2khPG0Xkh-g^YU-d5e#r
zV|jPw7o%K)dp?KaZB8+g*hr-|;8va2Zk-?8+af7vPi@a&ZcM^|3%K)WWovJM7G$-U
z^>padOLLUmhN2aXIrQ`PfgtIjwdwO~?rC{Z&Z>}(9=DIGnAuu?*j)t8xDy_CehY#c
z^|!oVPi|w{S~wsdCKJ^l^Qma?e8Ne4n;?S^EqMuZ;mcHPgspKz9BuanGE%L0%Dq{t
zD*~+K%$t+cn{T`DEoyqI+^-_g<tH~HJ?SKqOQ3g<PrIR!h9FsOFp2n`+zPi-9`Rn4
z4276Zt9Aw*U{D@&slvjg&C7G^RLkIVBIj_1WaONDineETR*@bYf!urjPCjV(xJ~=O
z(A-IjnIX?)ef1uMu0e-j3_*iJPGrOZ+$K?V%I?n=cN?O;Nax@yu+cLP_^29|Wc>0|
zQX2e!$kernWwRd`oYiSKsnCHMr$`;%G_Sje#-4%@WWCDS=~O&yU{<KdvU5o}Rj=L&
zO@)JoYd35;a{>gN0|a@}wap}nDQ@1o^;^?|RfEiHxDrBDFK_k?6i$7+!{JfeY4ld4
zyUd;YGLX1gWWob>9GyVRb`Uu^dw69!XpI7SnU2*VZj`wKrRF(^WU!2X_64EKyy1J0
zHTxUA*KI!IHVwRq6B@QAPEQF8IY-~uPOX~V0*uuKxByX^Zd0?Q?gAZxU%w}?>wmyO
z{hw5Q>wqfXnqoH`NtO>#-+FWjrqKTLm>K*P2f06>O}Ia$Fm1X$3E5t+lEjqvxO*O1
z5cb)aWdOUb_y`aViA^RWCOTb^m3SgovMmq;E>AT!8&P^sKP1$pIB#-0`<u!A0?2#{
z=_E)<3h`Lh7R<tg2BZOB=+^eMd4&uSt(G3Y2(#~92f2+kJ65B(-Gcpo+k5U5D&eA@
z35q<T5apezONR{1uqwHbMB!8T6?eng`fyZ4J3|WGOJM=+RSlc!{*j8>`Dm3DhI_dS
z{r)+?(td%6AsQ+oWR9gF$A?d^YaD&gi(=xyt=1D8TV3XNeN4U`;*=|&-8eLVk!;rW
z+h7-F+4J3wQY7XG-~wMRb?^F=`j8`YW+ap}M$pGA#snB%T<7$`H-2ZV`?$&StET!G
zyC36D=YdaojeCFIs-Zhne#SzCytqJSj7+3a4|a$|#wiYx@E|q4+xC=G%=Zx%3YghZ
zw_hH$SRslQAjqZUt_eemFx`ZiurkYAYLBv4PjhF{rTv`~S{)^+^~Q*tj8p)7DV4o}
z^1_mL4y}~&H%bL7LvSQCQk1J!htMHqrhNU081-}?sITo3G<rh4ljEn|BR`qm)<m4z
z8Vk+nKI8a4J>jSbxd_5FpY1&yNa=0$&I~zU7}~tk>gI&}1_H+sxbgm8J2xVsAi69P
z!tN{boT_qDFYg(38TeFwRHG23(OmdmB%_9N<Ee6rRwJ{o7i(cw_Dd*(RSFJf?;3Qz
zrmJc*j0PUSbq)h;%)B_J<+M&i4Chc}ST}%>aw%NnreIs>!llF4o36ruG7zzS)6UBj
zL#;)+vN%&-@2M9JlctcgQsgbx$0r#{wxM-MXbmT^pU!)g>jZ4_u0LoFqU7@y8{!s>
zDlq1*#`ODA<dT3UYz&U^+f6Zz?*{lce~>(S|I>b<Po1l?X%&&Qy0>8&Og+W^*9rEh
zVd8e|T9TSA>=P`6@Ey>ul7Vm_cn22M=-5=m1c}vU9t^@KWpZrf^QVM#rDAzmyL8G;
zcHAVd>2#*GSx<#Lnk~Nai^8P<>?r)5!pBFOlQotB;dg<RV3Ov}P`#>S5hJQ*U%C_z
zc=}VYaDd@483tJ_XnxS6zq%rCvg*)^piA0c_aLy3(NwA>9?nYQ1~2hWiu?Q#8Pe>_
zFrEGKP6g=bCnvotIu90B<alONrfRb_|7;zNlXY^;VBZToF!q^%p8)*7IjDV_&#|tY
zC;(<N`IVB3P=6T_V%DJo?~AwA+v+q2u14e-PWPP&eVb{&^r<q&@=9}n>54%*N;WPw
zWTe7B4MzHHrL)f&-pyt;y?2t4F*Z*r?Pj@2&j&_Rry{3Z&y4W;_M@w&DgqhPjjP<G
zurm#9_(YrzDH3F2Xg@Vl7WcF%wuTkaXxOwE6cR|Uc{9fv208?w(Zi3V5EY59#4hk*
z1RbH$`vUwOD}in#A_gbMhNcy=)ucz!Tr*HUdU#+KFRE6uh3{+dW3-nHlcg-dy^R)r
zL`LXhI@(eQ=&s+WLdMPO1dKYfKlQ<5u59ofj>pfE`75{<kbPof`+HP#OWhRwC?CZ=
zV8gx~tbL0#QVwYZn&0im&j%1%E8`VXvOxpq&8c8G#h1zTJld<%USabY57rz*9FY-M
z5%y-<<W49sbQS3JtriPIudhb}`vU{sQ{p`dIUFu3R2?)~MToG(RB`z+5n2?uJZHOR
z=G2f2tn5NiL2<)n^x-nP3-P9Oyd0X7WRxlfVsLBS)ZASC!11Pa|K2z2$F4_)#=Sw9
z_^2B<2Vj86i)2GsMJ$5{P-K=n#ov3&e1V|UzjS<-7CSQPhPY056R_`)mPj_7Y2y9g
zKUUC`j@$R<9YbT>5%dnFA>xx;*<%m?k9=3nuUJ>oKPp>}dM+FpjUp*kxiJnD7C`Gn
zR=KUG-p%8D1$3bZWU7*fZkwmM=Mi*5lNOOFO)CKIqD1mArg?Sjmv00fE`YG{K&!XH
z=d9k%cT6nf7WcR`yuYT`FE+1!KTas@2%w6eZ_HbjnwWt#7X1bVdpJJ9WI~Y9nC}j*
ziU1yt1Foo{;y)epq9}6**Ya#yo)d&E&6KOYoCd1LFGY6K7WRrInl5ei?ZKZny0D*R
z(gXXUPjKV2=+-vE)}Q|gX*v7A4z)dPfJ-AY=!$)cR%J?rPl(b)(sTs6S|rd%S1`@X
zbW&NJ@-~3m;!Qpaqh_C=E`<v#YgSe!{mHVYzo=YgC&zAETWPD)8dnmTBdSc&n!~l)
zY_$N5c?*4i!@_1QS1iO^!ftvAK>V%YPPEPsJpT(w>V%*oHL6mmrMB<wLP|&nY?5`G
zll+b9BPLN#AE1^ZC<A2%C*V|Qy<pNw4G3BhHU4Z`hyr^ZVN?5FDh8<<lt^X$G^H&j
zyg|w?iB@ni0+9hB-M5>=SbyioHliun`+OL)>xTY9HTKDDfEA?VQND^U{)i7e$WbJg
zIbvf%UP@{MU>H%5&7zj6=2YXb%v^ki2Au*vxe61#BZp9R=@X9Z*$*Q=d6D;3GL`Ym
z<whlDR}Hc_2lHdgD&3)}R8Fk%_5&W;35ry8BFE;GMIO36e9}92J6bxn8|L)sL4nz@
zqjydX7)d(f7d_S7Y<E`kjAzdyA+Md3#$wBm&BmO~jf0DIsniOM5ucP3Pn+=Fa0!y{
z-gxdQ>`q8hKEJhg%lF-o9?<D#Xe{uDhU~A+^jqA|PZsm@xe;R{6_yI`SN6LP5dwKO
z?z5kz8H@%xKs@9{sY4SKa7KIOBTP2RuDkNEHzs}_HIN|N>0mhOpff$U*TsoWIY3%b
zF#dz~cy_vu?#yU^Ulr=Np`*EgPn$OYJ4;Cs&Z$=DK%9~){Bh?!XtJUbWwz-eScW^_
zT$|DB1V7Z<vUG0IMc(f~T{4t9cL<hX!KnNMwf&uLl;7JxfNbPGs0Q}z0!=ZQtZ-F>
zNK={}Ti?tMgy1kuM{ebk6l))9`xI*&|810hnk87QcPpMlFyz^>``|u^hSB?h48qin
z@I$o71npLoF|UBy#|_s3;V!hr#M5te7aY?2SHDblTe3QuAHsQO?U%}=n!ljLQ0WPW
z!AKk#@&O-t$<rViI}a>nGGH{sH=xbB&sCPYNew>Y(tSc&q4$<yt1Ie=BS93y(JP-T
z{q%DWp<tOY>pMbAa{#$8MWFLxXeE2EtOg7Nf%Z*qcoAe88ZH^|je1}i8Y8J9Y91pF
z*%`ZY&2}bDvM^<+2w&U)s1rp%+=&hd0czoE*B1ckr?e!NK7_;kqPqL1YnHOXg-gkP
zsD<{0GRa)r9J~4lk&s`2L^${C5|BQM&}wy;^Q#YF8lwLB(FMbMgkN2;d4S@ceYu!u
zI#?Z2(=tmw2buFeN)_d@qb)ahG>18d?DZHi_YBU@E`N<B0Yc0J0Gql20dxko$S1@t
zLfXfW0iQ(jAWas-v0*xmacF(HjF#7O90lv$>lJ}3mE%wLus2a3==u!I2L_kuI3t)i
zef{uxcy4bHf}?GKUx-0ggkXd|6mj;^TGiC)O79p(xz6rFM>R-DTwruMRGeBcS52oT
z5SyEJ7^Vd0Sl4J<fxO{JTM=q9+>nSwT_8FU1NkYk)Y!9#$Wb8TBim0FTEB61qYX|s
z_5;${@FsU%)8HQ*DZ*?R@G%|-nqLP`t6u7>I>%J%Z3>4we?3y*t!Tj1O9rR@W944O
zYi%>Pv)S&7YMqQEh@S5(&lDVX@{{I1gJ#4T#D4qF-)86We^L?wl0Vw?1j^871EFzL
zp>pNMD%qlo?Q&20*HFh$t=ck`>}s=T2Shn(fH9bR-3*#uJ-|{0Z!_sK2{D$<i&@Ec
zJmLBhVm1H_X6>owkYhAilk|1&?cVO38~wyf;@V1U%#`fU{cJLfs(MfcUUh`+Pl#W_
ziR&4$@MNX9losf!)qL3w4-hseg6Lv*1#9(7z`F;|1xvgYJ+0f6fS9FKs}4?ZukN*h
zde-ZHsB$=9o}Qu<{}Tdh5EXR?0mVV^Z}AUg@2(-K%HDbLwUhFSBzaWdY^+Y*wt2^r
z(zaN|qw4oHMaa)kkI@WXW1%eb&%`LiPHv<LvSa>uCU}OCzo@Pc5pOlA|NNBee?{z?
z9m0-*WZ4bSRn%0;xb?W@B<z5_1lVa5g}3o)KS$mRkS{sIa}PSriBMYKopZls547PE
z2WHX!F7?)Gt^t3T$X<1AXPI$&s>0;xIR4;#kyU1Jowu)?O3&p^2O6lh$<L~A3|OEF
z@_+L^<ZnT1I}xJ7CBl9&_iB{bAW%A*()Y<}f54!c4UnELaM0A}WZKbS_ihT#Q;<26
zU>ImhdsHnuys!0ywh9%dzf<vAwwU|+{)(cBVR82Gbs*uI<JbSmlm~Uh`r*?=a9lM}
zhcrY~cw2nuG=4cbxe+Ka5cybrCt6a88ry;{RJ`4gjXHt8_$g>)dySq)1vQk-Wn%-a
zNQ}7q$pO(4Q^Zx93_HAX0(xNC`TjW9dLFe36@_a#mYI1pbwm{F9bPhRKRF(SE<wjd
zlRvQt;ddLCV9-fiNe(W-ftsIS@xRR&N|^3%pfR4K%cVDL2+<09lu5PD0yZE-*pV<*
z3+Xk?0Fp|N^{;S2eai_fljN1PrMxwA_&Zf&t&t~B8Fjd&ccs2MtzIPK$phN$a_*NW
zCp@zRlaD)1#~lY}G{7YJ6T1W@a(4eW-7UK^ENF=#G8ZaP-Xw!}K5f`rB5V`4C<h=4
zYD&CXzRq{0sQ?KKN!j<d*?_bVP(-Mja}i0X>ipXBNUl>GR;>n$GeAs6KC?5`;!De4
zE&U4PmvmM7lF+$(77qOhB;k5-J%4rm=ofzc7#PkAx6^-{&Tsv8c2ii0?Enl*s^%gg
zDsx&OuOy7DmO05K-z|It4`;}+$<P6VIS3T#5nziok3rdS;vNV~z+*3CNod%3;u`s!
zHoLNVL9ofF@*PjnJxY$aq$wzFl+_ctx2BVrR>1#JdM-FSAMsVuz|)&VNSTm8E=u(m
zE~t#RZ*ZsYwsrzD>nr3Rrsq40Jm@7tm9k8m$AW-i_c|kK4|^2Ryzycv=i2P6)0V3>
z48_FP2A(lBys(dJ8XdMAQIPWThioePq%_@sdOSO#jl_?rgO-gRq=7^{UTztr`p<u4
z%U|&zZGwmF0WhKJtSLZjXts6gx7-ee_nJ4YU`oZQ)SXUa5wX-qPJ8Z<(Ub*J90HkJ
zTaq=Z(V%Pc5_=S>&%}ba!H34%1k)tU(Kh$Xgea0^{5Y31S|t_K0nLt2z4hCl+_~@#
zO@r`}9q+WbK=hA6onp(wZ8eae|5n!;Hh^0PYTt;IpZzviOGgELr?uhbwY0d*p(tR1
zGM=$4EzgV3p1O({&7MVlI%u5xq*CVD+IEER0TF3xPko@!tq~6#gD;d}svkm=&K-3K
zTj)<Vbrv1vh)iulR-^afr(7HJ9Kst~M#9Z|#A^+K2_{*_`Oks!hxu!`x=^l8Lbvn|
z8Q?)d4^|fj*5ap!V*#XRuvx4Px-8cW<M^9AoHm~!(_Ekj7sHFP7R&-F`~Vbxqj)aa
z64-rl?=LR=t;2<dlvKl~AdTZ5*__Ja3Sd`LcLwX7<NyXBpb`1Aix{(!<}$`#f3*V5
zv^%8NGyd%7{`@oWGhk<$j*<aeSM-e+^kvRKi^C^&7y!?gsNaw7!F7d!s(>D7TfG35
zCJWD-<K&2>^pc8~8?^~AY!}Ed*tjZuD6kN%=h}{(!@vdz4C81QDIRHL79C8Ha&kbS
zijAR#`cICReq$hxK9fsZ!hzbpMf1>hR+Z0R4&D&nxXT6Lrzftw@JC7;1dl?(e8$>O
zH|YxP2Lf1jSnV7;c-hby1Vulox$u$zD=~WTc2VE{^qE%%lP=zG`Ugeyr-SjcE-h6m
z)J5havk42`^HWuUj@I?ahK~&O3uBMiWf<w?wViDtcKX;=Kng{q*JKQ=jH%BMhvsf|
zV1SlBwByb#^tvZPqIS2quLVswBEGA|+?Y}v53YR^tb<Xj72#lNy!Un%%Oz*ZeEIkI
zpt{xk6f1agzhZcUhI9w?^gSQ<{F#R*+<1C7@~~8Oc>36Ft*)?p4(7~_#x714!W^Ox
zu9^>rzVaJAw0sB@2ah1L9`R8mhdR&+<Ohn;ObSB!S7uuW$lmC&VXv-!)G#PMUld=o
zTW9>UgoPe4SV>$F4jR`W+uYf<rMY0|u`x#K(RMI}37Bkp|3qV#*roZ{rWU{-e(Mr)
z9r9`SC}Cvn-6tR`BL%uI;0CW+6O%J%^tyfR%(H#N)tn2}C?i2)9vPNP?pd~bmUr5;
zD-uliT6fiq7pHM(N%tH<G1Lj4<0m*9T{Hfx8<RX&@4ak+<sc!cs_VPJes+F;fpg^U
z=luQ20Yr~yEdBzcmKYTcs5<&uj7AYAlnBkRyBBHPR%t(nz}AwCXWG$^^Q~`)IxGeG
zD%u~dH7F4sZFK1jO1T#6a0TvN3PZ<_o0v8agMOoyqfe(F5`RQ(pBc&ZQ;Y1+ea1%b
zeIG{U1d6t#3%g&GqRNpacV1O;E@>Qna*+9RM$26UnxWx;5K7tidaWdSzhbX`YkG2U
zs?7td;{EI!&H)X|?yem6c&T6O+s)9BK)uzDKVdg$YHpMg!sGngR^<r4*ah=H&I<ZD
zDT++6DpP&AdTseeMyeEGiDlhr{G<4=?YY+FwbjlwM>G)*IqMi!8t)lqDr+t6b~@n?
zubjvPcW0=kH6`?ArbMB%uQBSu<3Gzn_}97g;azVcuYn@=P-y?bKXG{L?m2(0xFaCH
zS#r&mV(lz`5%=kX<-6pH-K=qGbdFuS;^i9gBuO%fc17ZZAk7nplY&HUyjPP1U)&zK
z)A=;9YH^NtWo$6;wdFuZ##~i^k%scX<X5xmIm_J6@rA&FjO8)jDsvU=ef+=ty7j;8
zY;l;Cvrinm>fs8?zU0EkMSN+20F@LXN*o#%)(6$A2PAM;J5ick4x=cv)YSa(YRCGp
zaB*>`N<)t(sSYoAn6b;b_^OY4&sjWs`#}0`R_VFIo%}nu{-TZFTO*sV+t>tby&jGy
z(OGmQAEdYzX&(N9wTpTi|Fwp9QvSVgpn3mmS_*L6?tpt-6&8M;st1#$kdKir0H8dR
z3Pfa2N*+_;S4RZDaZ8Ic?>ihseYom8?fDgh(|rLVO=_BN#iukqv*`n0H{UyRWAfL}
zzy5SMnEz{cmgH0`8#}Nm=lA?RG=^(<U&`5LTaUYox+@xMa&j`seEM#N$*|X1y_UM8
z_E3YLg6Y-kZi&mw%i+)uG6cAg3p6!uTY`E@-$0P97lTyP!6ACcsF;~Abod$Z+c>8C
zCpjP23*h>EgCti%;L(|-w-28Euh07-3tojE#|@Ah39)BDA~{n~17=vCV4y%m2v10l
zx!9yNH6t5Rp=K3LdHYJw&Zp@C)2G|hfsCo-^SAU0%4p7vWYjQMw&`-b&NAL-eDP;1
zL<dd12dBc+$q6>#S+S-k>pY@p`x3^oJ-gU7?~o7t`v}NCl2i-v3OpV^e%!abFzrS!
zaa<sW@#&K%*b7~c_uiSG_?-~)DzoQc?Ii$?(84gbX=}oVM1JGz*~^_V)=AcDQpHZz
zk%gALWmhXF+L$Yw{^yVWD1T<SF#<Yygoh4k`NZYr<)NkbQ-BX%ZHSTE)7nPA4C5m;
zZ*LrK&a=(C?W0=gPzp0Wv@q^ZD8%s_Ik&j*k^6MqqL-~QZkSSKdfoA3k?AcVqwZW)
zF~OafGa5WakrOZ3v<v^w&$Nkj-O5LT7$x*+N<P%RwJ+V~#eBFE8#=;_!1oH^69<EU
zUOIVAijGWa08g6o_@F05)iUHl7#OYsyX>YpP)T`n9Cved#PTFg^$)yC8jzLD<XI(2
z5s0u=XzZjzhTR@OO|a`vt|fIlzRxHb`2_;LMUeFVAasMG_VS-u)!+QJx7a|&r-t#v
z&#mlQHJ5;Y!w67n?Xc`5r5$;intI9m9K){OJ)N(O@$Pcx>YKk4J12u@anR3q^te&a
z$cY~gr%TBWYQIv*ftU;o4BQxHnBPf7d)+Dbs?2h~rN7}Tx{R>yW))p_eO#UMbT!}<
zJh?@;5Br~gjK6e-oJ8q{eGX5aJ-d4S`pbq?!@e@e2S`f&xsnTV&z=g(pYc0ARg|;O
z{^(nMUwuO=pPRW|*Uy+eU^6fddtjJlSjRy_Lz6M4EAYp$P`B^WZO@d_Zj96841^R@
z6lS5-BE!CWW7-YxM;JtNnEpU-*HJnLts~$6?q5FWkAH*T0jIyxk8bD!WjWY3J?O)v
z5Sh==*x81L_BmD0mURx?TlgF!!CtRHPU05G<nSg&BD4X0{4d{5{P3%~*zBOabp>=k
zRGU;lZd84;yYLw|-li+&SN%0vN$qF}&E%n}S0KyFXBg^pe4ED`E*lJpdD5*9Sg|Va
z#3OoPJdvL2F@YA#{nawcLuYu!MU}Dx`rUC$J+Ck;If{CkU$=~PBVzHXj7_XGb<a@P
zDWLH~)N9*EUGibixy<!<aQ(3E-*z6h79i~H3Zrw@I=a1cd?@fMamW3ypJ`&1BCdMg
zfijF>>2tm!7Q?6CXWm)%ws~f0@g>6T=_{mjEbG#;RPUTT;ZikVw|rr0fn2JMsuuY$
zYyRCo{MLt#SlC?YyM#D49(y0=2tNpQdSG+Se|@3WvU$4yFhX~U!D#G`q}v5MVwHgC
z2Z{bsaoDHhw-Tq4#|$&muTW;4Yqb!5)sumcj7&S3$+KfWr_o?4N%#Fp?e}J;J{8K`
z?93<IsVw<*45m$Eb}#C>P;i^DzNVsYts6}|Qj=&PTgKh2sh8?kp?Cci!{%}4zsE|`
z@AP8Ts1k;o(Oc$u4Ug!=t(J^4r;FhYIoU&THcX*u!=816=xQ4Ydv7gDe8S%UA8FB_
zc$`13OunEUL}z7a!ZPNV0aBq3c(Zu8@c!5_u$Lr|f0s+T^J1heL!!I8`&+7=on1|q
znf@V;d;Zwi*k5O29td@PT2MHnIYgB~NZb=Zc|4g^fd;=(zp!N3eXb<YD9vqfaQRJh
zLDW^db8`OUx86oTdXmN}67~HtYjG*gJrINJfR>@+LlU3|LLf%0OFuQF2JbrGgsYah
z{@Hwl*ORSNU;K_9lK`IjG5Pm`f4s*(`jZMF5T7$aD82lYI&qm;$VD73nCwu@v8Yd|
zln#Lg)fvQ}^ozW9%uY>1zl=O+)H9l+hVN5Iqq|^)-s6kxF5iM8HO=+PsJP63&$D0A
zWj7X+o3*1jlc*xNGuFRV7d~WwA6x8ZYHIhMbRH%>!<BO^p1n8+xHyXB0+cwN`{6Q~
zRS~nUzd*k3hBs~Qpb@RIr`|~yqWSC}8tJC}^U>`iC8cB2Y4CuCQ$cJ-yh55N3QTE5
z!O2EgJ8yKj($vmgts9jc;0{-fwz)y<O*c#=Y`WJwuHCco`1mVU?{V5z<_ss6G<xRP
zq2qZ$y^9L^tgRyb4Mu8|n-+8`8J0%h<>QH~bfVtb*!K15#zP)i4+bu|jJSpqUk((%
z|NeD1P5SRYPd@#x#NELkDI_BwhY7X^508(tBqkQ^vjaBR)k~M2+^O<!{<2TkOn`do
zfyZvYT$y_v23B!xO-==Ia~(UThFFh<n^!Iqv_|b9UCm(pjZR22UeqK(yLW&tscV#s
z(4Y5nRKtu^a{HaJ#aEvXY`)Ay9K1}&%xqD|F2jjuXO`CQ4Yf7iW-R9W|BA)bkMR}L
z?_X$$S7CYe>XkUm%D<C%>rBKXU-zaAnt>+*c+9+_W#Y+Fm)b2Iav@c&$u!Z?>d4Mu
zuS(Hryff022$*Hi!Ipa+*>Wh5MTT{8*|iJBmHGt<l?8l#P`zj9qp;1~MT#9}Cy%*E
zm->ZvXs~p?vJ1Lg(l`8ten=+PIN_yvqN2khZ6iwb=s09jzlZm`gkX5rH40dbAS+^Z
ze)$OV0r@Wzt>;Tt-;{)1AG-u{zH!=?`SI@{Zi2zwY`_(gKuPB@KR^G&-G})UJFKQf
zZVtcjj*L9901ZXtK|}9-ij{olRu?UE<I&^C2RQ1>peO4Jyp`eBx<weGiT0M8^v^%M
zGp_78%iI-X-IiyZ^PSfW9wZylzmqxrDdI7Y`R`uY+~zKdIs1Ez*EqO6JdW2-akY2c
z6hI$XGgns#V{6n$cSW9pTFUu~tgO$IC+V|X1u_AIgoM2?*={)5EjY?*0j`@?CQk9q
zISPFcReD+s55cxJ(wwX_e!<q(Hf<RSFh)92f7ki>2c<#h^V-KH;>Ku?AHOWA8MJ5=
zwwfJjou0X`w=0TbW#hiDJE(~-bF}|o3DF_3Tg&s4XEZ-kkA~1#+INDZeSJpVA*WKb
zF*^7;G|h%ai=8Q>M-}346&}TEJrZBSlRSIvy5b|DA|n%5{<Wn$mm1HAM=RcwvH!eC
z$zv&@?3W}Idasb1f==RL&`B~msIju<P6f<3S3rE9X_MVNDcrVfa|!%I3Yo_5NV;!v
z%>u@84MKsSb!vq>ol69_QVz_^NW+YO#M8*prbGc%HK!oQM~@zbLwKx-l43LmX!sal
zPNh_9!0+n=X2dJtFlgE7f<BH1usW~8II-KPMl+{juQ2pVoZ51&74Y4~2D+fMZR}^e
zmH*4l57^pVHp}Ue3-T-s%fawN4N@(cixm-z4euoAN>e`(2hJwEjK?BVwwe^=vR-^I
zEfpz7^ZK;fv&e+q>U^Nei9%AY5}SB2JUICR5JqWFo%(#ngO<ZfJNyyn-75cg0gW&_
z?T8niaFXF*JmhUP@ha!v$)`j`v<%7+w<J86RF%uK$<b{~zgZQ)qjB&&M0jza)Q_Gw
zVXiyPLw@;!>*LzDcF#7F2-l5pBy9Z1DgrJ|ygz;ai|t*SA`$?<d|)AU5&aXGf|MGn
zj~<Xzuyi7zI(H)Cns)<-EqH>OJX_nG3Wp^uax)dLn??k}ip#cpA7z}^E)x>7!#kSg
z?HLvESfZo(vP;3+dnLmXb0w8L2JES;GE#Zs&v%b6cUq3*F~!BdhCIK~<w@ZB)6d7{
z1AG|)vKG*K+n=>_KN(pRFOfbYZT>yF!%e}D7=f5g<ISRCs-5b|DC4D{w<3Mj<HYSW
zYoOhz14!{~l;O!O;F*c1>dFaNcRwOqg5p_3G4`oW<oIreeIM%upNd7^f86U;H<2(>
zEiG&p6)l(i=Dkj%sC1m7mX>cX+mRy|%e-0Q8(YbvfslI!#GjbLe=qZ7w6IiQWN=0E
z(xN+cDR@k;L#0lGPs&_Za8}pJFFo$!Q@k?<-Ithl(@v&-QdcOh2j|JedBi}WY%{rg
znDVeWE^&YFEo)ON%>e41hq>jh6L0$bf0P^9?PdtJd71`NPS3(*_U6>bDJfV^FrGP6
zlMoBQ?-|QkT>+)5_(P<+_3>r$UeKDnny6ZMQc$^_NOo;NaG`7Xv0Kq00=HxCLD`M0
zic}V-y#0jkjO30(><Ubkx!0L;xPL-4?ab7D^XgT?G{}cIqQurw1nL$1Nc;Bs-=;uk
z$>c!-faSFxcfo|y1$({9jxI1uv<$TzMqnh))kCST{p&BOSfq8gU{BI3nERnUlm-`a
z3N97hyEI9~JR2pLTXyeNL=(p-lUsn5fM`S~@$8LB9pHGR?88!^l&CF8IuBEt4B5AV
zWSV5qbNN~_ZM=<*jUnXb^srGox3WPN;*}S8N1ol$pLZ^FmO)yK;?#@iJ$*ou7L<Rl
zNyDMvaklPlcyLt+w|6O1s<R`|4?i_+CNU)I)yxOOyH+Ef@JV=XUnkwxC;9a-QPX=Z
zcgc6+g5Y<=&n*?$@4c2`*hkCFEk7Jh^-Y{+Ppo`OO&J36fZ9<8z2$F-2y^8ndMgut
zUyxT!GRMNCIZZ#PVHz6xZXhtg_%!TuTVQ@vXLb3!{@(H&3GX-8Qctd)S$ujQm&KP=
zs(-)1H{aa>RP<(Q-R>`gcuQ1rSMtaR4|aLq@#7(eimoj{Eo1}J$vyEo*V<Ap;g(&k
z1SDQ@%kY`HzpfPoP@URCR#a4^{i_a-=}1D>9CU3Q@Hn2k+*r&gsTqVh+_XbY!Ku;q
z<KQ8+wW9K(#&H-vRW?k)4jDRJFR{*JbsEbc!-Gj2V8AUrHcqtuU6;e&^UJ5MDCj?k
z3~0flKcSEsvr2iq^ivE204~Rdc-++!*`#46u@4ARUdQPz+Oy4*X9E)&=_71Kb)6)j
zZ#pWQe5X`ZCs|r!IDMC4LV4kijXW)Z@cy?2MJ4yy{)?`UIZ}AjSsqcGdtNHGdAa_f
zgS)j)jN|?6aN~`o6onb;dYEe2-5Nk2ng0w1Q!nT*LI>`-#YC~i%@6osRgJb^Qp6*!
zCK(Ng4j-qNA8$$wi(HZQ0)k9wbbPKhR07ZPlSDMsyXS=34}M@(GQ!3wS;{H9@-AT{
z@X+U~6Sh>@3<Y`5J$z()D%)KeD_DMO^-QQOpV8i$@O2(Yw^_-=-np|AKl;q-T5j0U
z<}9=LC-(L{+IL$<0cG>#vtFcqtpbI(IwUIM(b7x^)ae698R~2T4(SZYfzgh7UVwyj
ze&N`jOm6e3@Vmx6MSBGDtQMw9L4)C}+3+41>}Z_z<K*lHzNRPe;I&2Stq?Q=!htT*
z(mk{Uj$%DPVVG_Z&>3D`T}9OtE<XOXq)Kus?YF5H@d$5Rwq^%LQwV!CFsO$+UjYtz
z>?s~`jUNxI7k57bH2^dS%_&}+@7epBO~(tW7pcR=Ni`7mZhoMc^#Z_O5xnd>{nQEx
z>%ELBx!I9(&Ul&-d!I*&<-;<I5^(9?X1zRti+2NmCfVaaUJEz#BML&so137M!bUj~
zo80@G2^3-ZS5zIgX94q(`Jjic4tfPGBlW+@ff$`aQeK7c{Hb`Qtm7c8`6@k~Q_tNZ
zK?BV6g`ZYm#cS4U#>VecSD~9x1|X+i>a{p=-h*$bN_=_D;yPl$7?LbzwEgya3_8gg
zT)nm61+zz?M@3UaDlcaw;gmFakMUAVpHk2syL@<Q&U&q+m38ilZkz316PvoM>~oSw
zygP~{7}2GbC@$pO2!fCDft%R}yI`(NO4e&s!$C)+2OZ+RBq(_01>vz?$by~?0<Gpc
z{^cytYuGd6^%1UOIAk{R>5n5sb8KenWe`9=%wVnblo<$S{45I6bpvxifRuG;F|&to
zJ{*GMwTD6M;!~_9JO?qoIx|lg6O{ASiFe_h^?|<UM=@{34}Mhe*<%KzfwwT4<vwF6
zj>8$*<$y=bAj@HL5HF-9+DICO**pMqUYy~EBq>Ds5Gd`dT=hDs6`4i^ySA7CDIg*y
z_yX|c%}#PhV#d48{=HH;c|#4cj3DoKMpt5BVXo7bqfU-g_Yg!zrcd0P0K$O{pfrHq
zz41Wrt*HZ_zHJt-|I9&8-2Qc@JPVrC^)A7R8HQ5;CR~M$ao;j~rR(C;`5^RI9GH11
zhWZvs>Vl&Jqvc7!F>mgV$5Ke8vvG2As;ByF0W8ivdw0IG52~{QTDxC5ejgE@Rq~jJ
zRmR8phcbBf6T;JNTv+NOIs2aOoYI&nc~C`=MBknC(mzL~yGQV%p}L!SL3F$2^9eqU
zES~l%sCsyK4ahd2Dgtca0iL|UJ0t1fa;GWbw#B@h1J7Fndtf$S1ECt;f|C#?kM7^U
zpK#y4Tah;uVF>Cul6QEfJ#%X9)ls2O4H{&`W75fYMHT?R-fb<ZnconrAjn%6?*%UT
z60iz2HN*Nt`5Ng6P9@yyeL=lb-=YbRK7qmouRx+Wtm-N2HA6Kvy%tHBKceL_>`maR
zg8{c*sE0(8^Z;$C0T_vPCh8*2)fjo0PB^P|Wbe6&hn4Iv7)hBGPk;mktEBytCncb<
zzsIBjN1<5I^>NC+;f?!v>qdUVz|G-QT--IBRQblY4Iqm8?wDU$4`HusS^jR9f9SB+
zQ3U=SB=Zn?BH(N8u!9Dt5F}r>Ba>zRA8lV5mSvlDEeL|Nffz_gNh2zuG)hZ%ql6&c
zDJ7s-NQ%;el!SDHVo-u2-6*Iuk_rOfKH<iBd=6*keSE)ne#{}p%za<idG1(yt+f#k
z!Wft=J)hpbRpj?rdh+Hmbg>5(WnWp;N3uVJN<_l`Q<HgqmeVMh;n1RL0oAk3@xnHh
z(vYu>p=4ApceWB23<xJ|(st7QlqDE3cZX{F9xq>*rz_`+L`A8PAW3KI5(8)b@DuYS
z4eu8T8*VD}Y`ir3ERo;e057yRj-hbtJ0~mn!Ww6()ukyFPg`j4BB1C=DKXG?xgH$6
zck$2~yqv)FG!eqZ+=@q|P*NuO9Ra}3`f`R)BgJvW8bveaG_F7}u)I#-l&hbip2Y%I
z7v)QW380d&hiE?WCP|DAUdyF4SzTM+x@W!UdbXWNH@t~z9k|p{-^ivSr*5mI${$tg
z&4WBzoYRxgTn~AI*h=CKB{NIcbCsL!n#uvv&{n<YO16hxXyu8`%DGOuIQ*K9t{BkD
zcLZl^L^RSId7~I^<%h7^wtsl{?64MdF95-E1a4i&U3zH2=765dzc`UQPT{aT+5KqI
z3V?+tx~0mHAy9%<BDL1KiaP}fSZ*$}CbH;;fouc|1H=d!U3McQBltFZ0V<XWwNmBd
zgzp4AnaF_Z7`rMJmO`R8v06f96dyu_UFJv=Q(C7NjXEu>X<ibVM(ho+6t;ZK>uC29
zfwHa#9k(F)8YWy4O1_~sv%VL2#m-aO=0H08VDUc6WwMJFCR@Bs`Jg<>0~NKno_zpZ
z2?YNWUEs0qI}W74(-5rm!9DmpSOsQf7A%6!u^uD~=W(SqIpHdFR;i>V6pb8!b;1=>
zMy$yqM+b+V`HzC5w!*Yxm}UOyNP=#~*WqVfrs=2lX{4<YA3f8im?icdjG8Rc9E!=P
zsNzP}$oK4_1qZ}p{(`KJ$LgLjl1E@})&IV^^RXeI6!O$$QGaqApxh2p5yAV5U=9}i
z=*UHC_b;^Fl^(?%3X4jlPFkGfQO?M(Ly2@IEd_jI(Ny6MRcqd@6DAG&$9ju%QqtDk
zr+|}611~yMz-<mt*Am9rv1Li)vBwVCPD_S4Vs81DX-Q~Hxjk%E9Y{vMw1hLvzKF>;
zQ^zE_A`wr|ijkIh1MqkqS@2B@$%ll&2lwN>zkB*Hq+8Sp23CU2oddU7GbMT*SUY#s
z!NO6Z3()m5z&r@Ojy9Jaiy|99x68_$$_DPT_5&U(3zu>e|8l_5mU79J6Z}r03KT$#
zJjzIwoT67U(nHSxPzj4UN#6R@aZEXNX~yyq%qg!w$Xv{`>g7pSdKBmsQ8hhnH@5TX
z78bPw`;h!Lj#5nXM>vWQJw$J65EhCPom8hK0KzLz8v{q$Y#_KlnzutO6$;hf*aZNz
zN5C_#gPMc@AD?XT$}8f-r{4DRX)BQyk{;oa^XQTm`~ezBGgJj0b-Gl(6Kizk>vOt1
z`(8IZb`|gWo!|XOY6nI~*D0Aj?z1nNMV|5C0=zWcA^cYGX4v&{!?hG_itn>*|2%@U
z&83Ww`<H-W-vIJMqIp&*+tUnBnNchztOb*^61^F8pI|@4Z=amVVF$V4Y6zILGGxmu
zPq(FqAw8RPWfsBaX&<f+!FGAA4LAxECrvdLtk+xK)`$+nVp991Nv#{f!L2$O%aH^b
zx%yDrcNlJiw*m}?MOQ3w_+7`F+}$YsC@YAXr~g^Fsq?{p`t+$XG*aK;C@(x}i9VTn
zPC_D>;Bs#9Cb$jea8~*8@$oT2UfN>f;^)_To)YMNgI;Jc`{V1oQBWa|o2a)+&~ONe
z3O(?KbwwX9_Q=&(<d7a{Vx(SRVJgkT({6t}gOJ*|szUxVTTO{2A?U1?K;w6$_ehon
zwcE%ZWV?3Wt9kJypd=jV#NJh?gl?VNZO%C8V}@ceEV>`CE(gI*<NDev187;Cdo^vg
zV)=9+rzUlI`K2kCXk@6SUldFR#7Z(e{bVB=ctV6vS^6v*z?-&V-Cs5;rvWG_Gvi?l
zd@4eK8S3wk6UjdXbPqN&KexZm2LF4q9qPpf1JTH1=YyqeZFAr3?0$MFs4YWH77bUX
zz-}oLLfgdX3sxxoqd=)Q-abxo)4*fuG(y;jtDp2{>F`cHqpnQ69RZlgb%OBYnz;$L
zvsB&05UWC><t$D0blK#M;l9I<95tQYHuSHHFD2|Ke)YAm8P&d>Kkj21);Kc7#7h9?
zgrOp6dZ6;5u@re97a)(w&=2Ji>7rGc^H6)3z-*E#ej4tQF?d*#yjG*N0ggnw(8pej
zC(X}Le|9D8&6%xpY5mRbv|st9^cg>}N)K5Kk;ml+LYgY!l;Wy_loZ#3OA(k6<SQ3o
z2DZR_+d|rz`8^mdWJCOK$nR1S{m+Afzx`<q9&Vez1r){+;CdqmLoync+-LflegsAb
zg|Ai&A4dh0Rc{dlF9MK?g@VlYm2x;5DuCu~6i8?9CqNLR70qiC_ic4`!l_6-&h^7?
z)E1sQcTVaKCBNnA#Aaed0C96!!s^aj4nFe8_3@SO7x|4<TWsqeFkA}YJ1&UY%;=L1
z$1%b7aSe!XjlmKLmBI}mh+|-5lVj1yW&;lq`MI~{oK!}xt|gN|*?gp;5pUAw{~dCn
z3*Yo`ZBPp2!@vZ-33=FX+*<WWvy-8C+B(F)Sg?=SAh*cYg-yec5;Nw{jSG~WfNAFo
z2_{VIZcDRuzxnu#O#9_R{Iyz$=6}mg`P(!6*N-*4I0$V6p+mNi7J0=iOs>XFCL*BW
zkp?coD42V@7ZP}1zjZ5oXvna#d9zvqz~$+8*;g4C;GCSt6~)1!->6y7sa|IR{zqq^
z7bvkz!1FLtZ+0BZ_se}(H;+$3-F8D6coGga*JT*&>&t<Y5-C@Y=TTBjF~7tE@$*x}
zMN>=~9?BL06`3(?<qR}}T*Cro<jJ;_WJPGrnY9bjRP>;VqX(u}aj&`;2h^uwZP|Lg
z)<-6Yr+4wY_SW{#;!a$0ad{FBg5din#8ux)h>7_vo+?;`$80XjH13hKqTr=a;Q49|
zir!Jw?g`=WALKiZxzavCIBfC4lf-5d4!0_c;?m&4?+Z5P+9gV2?$gZi&n8?_lGrHk
zK-&ExnQU#Z=1UA?)prw5G}>zQ$7z`+QULHvqWbi-5LV;vgCSus{X@h)fhxKXXV)&q
zZd1KI`}RFrOPT>Ry<mL0zq}C&ZKM8cPP{nQgK&$hx|quE=sH~5-{P31enxX(OkJHD
zcg4+Agqjl7WUDl^D%;IH6E0vGbcoelpzdqV!pB~SB2Dm$?#yK+T6<VUn_4?VWI_4!
zN*BhKUrK>>YcLH_2B!9nyLXG@{6?B%b-*e!Ikpp^k3&UT1TfhufjYBQxlMn!J(sQG
zHqc7Cd+Ux<oq0My6$q`-&3xhi*-UWzTzYB@L1~=_jDA!ttjm^;Q&@v_a|X34g1BH_
zTrb4d4sAGmi8}*;5Um3KDGUA<i)R{x3G5V&NIGQUGM*RA{WvpYqi5yd1>nvcIKoZz
zR&A_%fDim_ebbjtz7Jh4j=ni`FK>C~cqiUh#g?8?`$io-8CgS-SHyOT;V68Hr$!lJ
zGG@&+Y!p4A*zV`b2dl;gAc7J(#c{+4d+R^1twn8zJh7l@)4@Z~Ox70tJdan%Twcpa
zhxog<9A}+i0MCxRt84D*fQ>sjg*t)uKSa?NlSI<#x2|(?>kauSBd}SpGJ3T&ui1Vu
z|1Ejbc;N%wD6#6KO5Umn;DQqAL?1{a^Pmnh?=LHK4grui&vsbM4Gbhh1;0QlH77OD
zt8i7a(a2C`?k>C%1rB59bMl{nS&x)Q?rUZBRVC7uE`qDM1xh&<))pL*1|DjpyR5qN
zUJlnzcU;P~)}a|!e?k;pr9l!`VN~u*ab-d{Tr$v^#f!7+qY&qGdrUbuy5Oo8F@-73
z&~9Hycue2>3?9oJx^e54Tzg9|klCZ4zdJ~h2xLdp&PW5YI|9Xhpp~eoXr5UM6+)#G
zh3uZGhlqo-5%Hk--sPtp9$WXlwEV2Pj2~tA81OzXR(c!)y(h5wdlbzwA@{J8k7FJh
z&1p>Y)yCJTe=v*&kCP^!@E@K_T{OOmg3<n%FaEeKaKRX%KySSxx0W;xpPC&=!f~i7
zl0TYdF$hB}aUBoRSwy5jEhmIc_tlUPvTg`kX*=+B8lh<&1?^b`z=u~Js)@XS1=YzA
z28tf2L}B}0;k{@)3k0r_rZ@W+WMUuR&ioW1Iwcz%PfbfAgJ!&`&01!vn2*op?~@g@
z++LxY;LGsk&HH(<rYS7|^uInAaoG;AJVl)t1k2DVh^t|0nd%Jb#o3C3)M(bM#{lh6
z)b;E5;<1%}TWLp##N~CK(<FFP?jgT(&y^XhheC6S$#=oLxA(<`uy8n;Vq)8Tl;qH@
z*-MBF)0x5`hC159fJ4;(GaP9<U<@8m9L8H5Zx6eF|KOU)+^09J)ZRbrLTPqY`i8OI
zpnkENh$tM0uw6ttuAmJznH#AOyf;5m{|Hx#{;DQiD6OQP!HVw&Z)-rrl<YDy5V*cy
znZtI}y#SGKV)!;}L?jv{XLawszVaywt3}`oS4EA3$Egh|bisdIY$C&aT3zxI*dnl>
zO;a&jxAfvcG2GnrmW4Qd>%J1&;o)KXxe@7Fkbv_>0hT)M1a;Cra-tAdB~g1AW=rX2
zMJdGa(ION_U!4^YpAfiP(g2DPEfx_BEZOpUK3h51ca)Vul!Qf-(~j>Iv`8s1S?W-c
z$;<ZRgLgv*(v%aHDuWu}_zD2b{yssDM!pIReN5yP2DI~B5d5QQy$=OWo46Ky@&k*C
zVJ2f8Sr6szbBTJS)hnA#s0m&MsOpr`6nq}5%o~)#S9Kwlnc?dyk`4nvavtctfPjUZ
zgNtVZlh#1@_G~ypNn*4~NlC^~RV7h%J=IrDs|CpaNv5WY=+KAua>Nee_p1Lk8%`)*
z+MlA<k3DDq@PqZ;uz_8uVH$fA^>MrGdsG7#-2`S!uNpj3AnZ4SOq3DkJTiH2)L+8<
z3zeg-N#_XxGS6>6x3WCL35JuWEE|C{cz{3=r1X{RkP(z2nkWrQK8b?XcC!WG<LfHZ
zHqaR4ZF5C8FIF3onwGq370pwJE|4k&?l0Ml8mfl1&@>SHL8TT6Oqwp@%OFUYDIgOi
zkG$s!jH9RcCYVb$du{8W58cRru<H`pNXV`)lY9T!L8sn>)=?M_)%dPhsu}vZht$8b
z|2}5|Yn5)K;pKrt7(dEr{Tw%z6A#~^79Pkoko^oVMY_<U{hVOT=w#AT&eh@ecW;Wp
zr&Osow%lHl(2mcAVXRhVfX7^mNmbu%T-0W!gK6pW7q7T)&(6j39cSDgP(LQ;gW(0t
zKA3i2P~67>wG@;@=OF^b(hG=sF{0SNG+X7KCJbZRv#+3@Z2-ITFv=Wv%?)TxBY_Z6
zTLkrLBb3lr(o_P#UQY_Df$-#YpxC!P{}gKZ&EPz6W+596<nlxRh0;tq%*9^7D9iyO
zX5oS9gsOQ>&Vf0_RY$lE?gP_zB0_0`po-8$pwC4Al?mXohQMf88m{!B)wnM5-RE{1
zSFPmuN+3#d<NR?9-_VC?E0W_W#I&C}=nsr*5wj>v{kYAI`Y%6zyY91cz7HbbmHfr|
z`7r3Lq~ZkJD4X>mk^}2v!5K_LLIDREGnpV?<cf4JnE)`oGc(_$kr9^ryrWugW!)&Y
zhfV6h;WrzB?;urCuSf!M3jW&QCvmaB!2Og2F=ehg1K%v~>Q)?{%C)lKb<LK(?`Wk@
zU+D5N|KmfL&c_EBf7q^j3w#KyCSbOZHwE_3$4wf-_W&V<{kSMD6|km-rB}dMP6MEI
z6>%@%2cn_q>Q%!^QQY&;jms!1hBy@#Wn)tyN}b2@^l3|#_=rtVZM^M7n+^f=7NOvK
zq!S#dvA8D2ZXsmxltMGAhpMM}X6<hIFkz%!Tjo~92e!mLrKKLc<|4Yj*ti8XM2oTW
zFK+DXsYrShwm|{|!o_Ig|9|MhEUhmkzx|i$Y@H{XCy;bNC7+a#FbqL8R_)^IEcZom
zPjzXx`3ple_ntaf+3EqW4F{`6c0s#to47C(eEF*(5f3I1ngvQZo>i5xikh$U#l~;j
z1m_gms!5;l55Avg_Yhe3DX>xPn%|lFXb;d_-OeP9DsC4Ahmb129tKyV5NhfR>?_7@
z^A3c7vq#NaTQth*3T*HQfCW2Ja3J<-OpFpA;gJHyIt(^aVDy3T)u%WoH31?fxXr4T
z5uQeaFWCs5bZj8%&+BG8Ey#8j9d_<J?oQB@;P9udG4n2ZairWN*r_B+)<Gyr(sB67
z_OVXgKGv~XoOi`T{G*<AAI1JJ2l-blLWrIWXdP=8vmfLUIF)#Q&wGOxx0kvV4kf38
z@LEBlJA@wYt3O=DW1WTp5Dte{RY#*90w9tc1fH3A4v=g?(497MZA9OJ`bGvy-oj3(
zDYbb``f+{f%&Z%RT9)bxAKsxhb};(3-^ndq?S$LPJwN90usuiMOwZqPCE+vO3Ypv)
zz$e8^hS36Ih(+Op78VO}!cdjt?!9=(Am9Z9$#qvS88d-q)BUUmD~uRAs#NF)gf6`z
z8Ux>*4;EQJU_IgW6;4{(XP=y9O`s!ztHB(3F|-qSCIdI%@WxTHrihYQ402Y?`gsCo
z((yjGoQjMr!Arvvs@&vrH}(cP|E<T{1mW>WV>N9^;V~D#?56%_kGYCYF-KqUJ2cA+
zySGV~l`(z6?scXxN_O7|Hy7sI(BpS*wtHF#8$pcM9FpC;@1dEVyUEw=8s9J_ZsQ*P
zQoFOw<!6hY7GG7Mh+?AVFuX*lb^}nSFt-oix5u--3+!an)SO{V3LV%aOEZHrvxYbK
zGanm2i%e3IK`6AT5Q&GyB~p+%N1$1B_X4V0!3;UV%_k<aP?NH{ZU;3?<&BjJw9**C
z%=%HTCUA>dD?MC*iRJp?g0@zGeA>(zs1u4x`<2JcNgGSX)dHB(X4K!kd*|*{eRa+F
z`L*5d-QDCauVEi*I22qGKw3<n$pbWb4os;hOn}RC&FF1{D590<JUkzp399*XFk7Jm
zreh@B-OJCYpp|RwF>8*|YFm8*=r_-5xrZ(fUdnbd<=#%Tk8IBRQF<=CIAQhA%Jd4F
zkZ&5wYS;Y<9CG!W;{?G}V=lO2ePsd-_`Tjkj-?k6(1o`8ac6o$AXJvuSI(>HAaz)>
zQEgzLLA4jin1O)Gk4kYFY>&HoQAQ_^jdt`G5Zje3_eEB8WM4a#c~3av&Z#xxE5QGK
zAI~Lu4b#Lm6Qs@ThLS!CjGP)E!JNk}#>K<CLTiQOTtGyW^~JtZTh=DN3iNkW@YaVi
z+}zKhtEnm#lYf&yg%5~sX?CjB0Q%R#WMf9vg<KJA->}#%<);i5)t>Dg_qhG2Mel~o
z5Hsn7UHgf2yN@3}AWCDXaP>s|!~UbPu?CAr75PW&?uio|V7n~XZrrIh^a?WfDD0U@
z6600%c8Y6XmR+={)ch135<-)hm>4SP;XLt1x`Ta6&?wy4mSsCO_7m`}JV_P0>57=6
z4|}NkC`3u9UKk($XK$l5x&|}*<e_YbaZfUUiQ4@)_A#MFNj>7c<DHJAz<wLoP-1U{
z1@0p$QTloN3p*h+HNq!``DXy&9!lU%YNW@2$;yYuk0;-$$*%1*{zh@np6QH`lk?C>
z+SRMa#Lq;A21d-E>Gn63sevZRLH)NK^XzF1IjM6%bmJlCA59$e_ky9q(<eZ=9@GIY
z7(J9tt7^LBA8dbov-9c&Wd3|%G23`l{di{%=Mi4(0#)<sq;ijCYi6yyWPP(r`^9m!
z-K3<QSG78nZMx0Cyp&C?;;~=Not;2rFqn?|(glZYcDY<3(1n@ZiFHIe;E~**I7oNq
z7qZ^G`$_KoOLj<>#C$2qL6bicKe$R8LW$p+AiA@50JMpRP@{_Gyg8guETTpL@cqRz
zg?Zi^SJQ#c+gItKk)!c3%b;3+vY%`fWYW?=Fk4)*FxgHrM{!$rTXuV#A6T=fe({3A
z)mo84f?%OqVPINa!Xom`l$cFf*Xl`=drHGBY=8M8e@SVE23vEXT<>hT$^aTXm3S(~
zp#KiJ|D_1@VaCxevSbEdsQj~000TEbbw|&})@1q+abdtKo4vLORE$zEuti4=D0Ox4
zw-#6+gm;6WUNuwwLe8a^C$Rx!r+@UBD!k-dfmfZdLqhzmNqzN<aDt^KUy~Xo6s>m&
z;8Jq`CXe{s{i*Z{go)!nS9;al#0FJz=<KX540>epIzcp5580d^Zmx8Z7R$wp7m*qT
zp(L^qz;$W_2GZoxSrkW}0`=zcvKtt68v~0NxDE-OPd~)flnHaxb=i<eA2#EYij=#`
zoncm@>d6)5ffxSF1Ep}Ht7rH;e|O2Lw>jgp7m^<X-u%y5<33Tiu+^3YrZrVYIgp#8
z0IHV;>1*PEx)Jnw$}nIDw6T*H8SWtdVr7i$C@{psz*y}_2DJ9<<SeROZA!Km3`dLR
z=e|C#4GkI6_EuPIh<s@vQLQQeoVso}o%ZA;AN2MDwQ=|)l)vfFYRrBrnOOJyiIQpP
zyf2Kar6C%P0Cgp>au+V{4FdJ!N=ere2lNxcpBV8~Yd=BRc=^&1mJJQ5k|-|4>Z@Me
zC3cV0-vezBjN@%APp5DMV9R=+WsT<$dhlF7&L)oTijU{p%$RVtnErWxvzpCf^|!}L
zK)^TfrxyT?&;S02$zlVJ^3LzTbYJNuhr7_+rKKg{nMFe2+AJub1#q7S9<H>paq>u0
zv`S9Cem96XRkL-qz?Pl{P=T<T8j*=NPs(X>E#GTJfAi`I;=?f`4S&iR5q1$buhe$G
z<5L($g68SXJ<N>nZ;`o)Z43Oo|Ghu7sSgC!yMiwKJ}CU<cK;d*XaJN6gMB08d*N!>
z!v?a-C+e`HQ5L}+iOu>(fK|+a_vS$!v@{eomZ%JdNR0%K6D%Ei`|lSoB{yj=8iWMi
ztS#whS7?^gzAk?6hgLt;O{{f5U@m;xcysIY?`{T;r#NPUeS7&rs1`7Bu3aPe?O*)c
z*Z#|oxqD&RP!%{giDsI2=cT{^Si_<{g8>vJ?wM$?3HO@bpvu!-&z==(mT>TQldws$
zz6C@ogA^EDx*uhYqqtu>SK_($p1FlI=J}p^T(eM8kEkd=T2banX$MK5_|^Bl^X$I!
z>tow`o<7}}?jyfaJSK)zltf<a9no(t&%cMFeTmrK0>8c#D3@sVg1vL62{5~qp;AUy
zS`8MF+TfeL<?YfewH@{YAfgefqs1&J=kEbER~BZvParz+`?g%=V;r>N`&f#lN6CE3
z#q>~Ts^3U?{QTrGqr83;X!F8M=zbGtcVIJO!js%LMmtj(P|ZC$cjO<w11Yvy7z`o0
zHE}!a6Nmz4WuKy14i^x>$9f7gjpo62fdR_wQxJm`XlQ9g8NIw!Bdvr!4~S5G(c`lT
z@$4#(xi)g_#j~9;JVEHh@Gwa;{QlEognV*8Or@<Yz%D<vmggUS+FT|wa&iU)PL8)R
zS4son&KMdbRL)g$ltzHxRs)>wMMhBYA%+QfB)px_w9}&z5)ysIDZrNLT|KlyLp82-
z>Qv?}uk@I{c)GA43FCu^3R>USL7Dy+mF>UYdQB6CDH=V<Uf%?AcVpYG?b`Q$dSB}a
z&@NvjBN2ZH<N8K$+`kB%Gd@57RMQki;V`G1-i?nR0z0au##&RRoDv75eM>xONiJ3M
z(@$(vd@;V;c11l&k>`7Y27i4tJJ+J~^STE<K&$^;@R#c0Ux8xi3g$!SRQj*;8vIdL
zA$M#RJP7N+&UT^u+xPFlg40O`)B#;b%v?xMwHi<sS@4t#z0%CXgmdxZ2a|<9|2(^8
z6DI9c>oD%~d5!le8l{&KeA@c=JVKwTKIZ*42j-WD=aYjOqE#)CtPFbok;IPcMh!Sh
zB_oa-6q{ZWu&5Wo=u(D;3HoL>3>+Nt;H!m;MnR|~UwWW4lPuzBSy8dtezp7X_q$<&
z$~jsv_cl>&$qPMANK2BQYDw7*MfDLg$=$yV4D*;m=aY9!R}8Hv36ECVxqls;|H-d>
z_v51$Ui+ZGQxv8F$ja1IF!3!vVK=uFwK{s&e<Q#md5&X>#Gure<4||y6XW_3n_}W4
z0-Pwj2zp<;b^UkIO=KJ+@k{rpQn7OMqg1{f>H81oxhH+$>D2*Hj5JXC(c&;!K4saZ
z2+Zw-8BWk+<ZV~?uVuz}_f;KXLJW0K@ud#P$556^M6ZrceQuO`EF#S0@wQ22^OF6G
znf$k7>ne*$CmqqtvsXc^$aTE-cSZl-{@GuC40(kOoQ`_{1>7W1Mg#oZT&~(%$P)$R
zAol>^Zv^uN<E~Hg!vMddDQ_tZKZI@(xM+eRO5c>h3CDWiMg;V|40AJSZQY^-5?Lxj
z#Y8j8+SosK+eu+PeEk3!3EVw;aOZD7*boDz9g$@kcwY_;<ol#s|EzQu>Dg`BQy`5R
zC~zg{p+}k%sN>%=Ieu<FijJ)@%v_VKCDx!%P61KninEIbTu27ciSQE5178T;L?WmC
z0!(C1tm=Vsy{vc&{j+@u9Cz9hKqob(DFZ749PDt)N?Fm_=@+D0`K_RR$O`jKdS`Z!
zPTi%!BNgQ)`fW)6<qH4x7&4OG=d(*%m>C@83PVIzKy6Uv*qUb|iv(y@Ucg-AhS725
z<TNryP|ea-J9g~YLLzL~yNVPnh{(6uhG^Klr~`@eJ&-EHq@P#9T$w$mV%<{u>pb0y
zXT%c4<>;LhlU~RAOI@)&X2T(y&BKf^MtOZwoh3v3Q|S-k{q`#VyF>))<`S>fA{3>8
zE0d%kfqNXeF8~%q|B%<FW5OSpzc9%|Nx!mrA1gG3VZdwC{P6MP;;?GSHG*(MP<n!9
zb?#N0p#6vER-kPCAf?dOz`52dbo`FBt+SnLS&B%_JANtn&$G*?eyjGT`G^UQLMv4C
zf7DUksev{y8Ia$=EEo&j1Ip+ZCx9EHpdx9cyDN~?>1Lm@UB-vNnF|9R_rdYwC5*Pi
z@#WlK^sm=OH*UnLbV}G)YdmzvO;(urFirXbpK1@v_*DGIcWn>1etC}2$Zd(HDHN)h
zT!7QfbovnCKYILF4*0Mb6A?5`9Jrr7dzJ_EJ47XTGZmLsVfQgGoK-_!K87!ml;Xev
zMm!SAkc0%9$~zI^gfW#A0|{c3&4YvvQWIO_xFceett$eWGlLM)by{va{qj+_gSDq9
zCJqI-+`i2Zb^|*Dep(i6|DE3+wV#L1Q+Ei04N?uFvTqnB$6|=xqg-llBJO3eKa$gc
z0~Wh*5eWhSd5F(8A@PpiJa#HaU!*7;bJ@<M*qLw{_OraiVH|IQlUu(KO=O@v@3P4E
z<+ss9L>qH9%opWt9jbvYwOh*Nw_okwer&npvjCVQKFKKbJ%Qbe7=MB^**9XqKsR~j
zhr7aoh-Uz7N3dz-eQ|oi$27ZKs<Y`_;+=FcW+Egw%U$TJkHG5XVb#8!;^izTz~~8*
z{N5@mM1AXAh-+8w>+1tJ${M?Md&u%{Kjp7~5Vo5VrqGBR#iS4?$DQxvv!lDpt6UXW
z9rZ<RX^IZV^~ws62ABHkhZi?sZ%ED;6Lm^LN+B=tT?)Odoi?w|SjRN5uM}`FT8)#-
zjaG71WNv1t_Kyj_2gl7Z+|uOsv(MhFfguPZP5(bjs9ez^0Dg|b?8Vtkt<XZP8|37`
z$?LWHSWgVi&0`>OXj8-F;4zW`RW`}YVdKf4s}C9T7=1NWGIGS-Xu_w6v|(SNcx5h=
z^YH)uv0zk8ks<Fcbx7sOggISfn1T>9wb#NuqexW#!P<p(#VG9npnFLTKM!Dl)}Z%1
z$k1c>?2Z*WM`ge2cbF`z;HOiz?!VAW7PLCO?@(J+@~QMlPQCAsU;CdxL-BX_Fa3_a
zCpCx(8+WS)p%VD#UJJQ&xrhJ)vH%nxuC9zR0ipYD+Y>F7br6(QN&@(34EUEas01rH
zPHUhcA|$9*7l@C<N)CokIcGbEGv_)eR!8I=E7m0RyFPqo;x4$vY_7<kL$UOVqmLYg
zy=f}aHOv91>BBy8;sm$%x+fyNL+>hyt_xI83q8w34im|1lYGMF-X!<2Fj@55V@zIp
zI70`O!as+|hP<(1>=^-rKv@8jtP#%~SStm9hFx)j;2sVaL(0<f$-rY!nWLCl8S{V@
zmBT=Lv7&(AivSLsBsa9856Z#Xx_D+oav-i$;o`-8tqXkEKW4&)V`XJy3EWy8sV4i}
zun_z6N0a&XWWDM*7l_>5U>5?h3L0M`bugGAB*ytI<MIg)@S5Gl)CD0zm=>Gb%ixc2
zCtC$%y^Soi%-hoLJ<+c`qZJ08=#4<rB(_5SPqAm5xZB*yY(e1-;BwTLFPE*pAdj#I
zr>_W9K6E@e@JTRgw7(mVo`=IqoDzHIbl8SR)swd49^Pe7-{9k2geY~=kN@>n2`wUV
zKrw#Gdu1$@9*3na5mvWIkEer!#^=}N1Czv<yw!Ra^XeYWu>V0@<x9kw3vNYYZxj9L
z>3^)&ATI{Qi*eSo9;`xO9~t=*LoGxJg2)>NLlMmkft8FUD4M)bd=r|Mr1qlEsa$%X
zZ#pXMYp`BMf9cavaBtWcZ@tJ4C~ZaWqbI5yhhh$to5foL_n6Uvmn0w?^FD@lVk7eu
zV1f5@sQ}>14+M$*NG=o~iMv<{IM>O&`}WbnK#jY-Px<{zjb;Ul!(2ZduZDs#-@HNV
zlqj_yrJU{F``G`iEabq4tqZ#9Ir>@4M7wvt6FbzFS&6jvttEDsYCnN;qYlD~N!t85
zRHL^7Ziu3fkS@?ci~({(Ge1b~-nbppB{bA;gz3L-e6mkIB&%rZeW?GWZUaS2w5fyP
z>ul)C{%Qn3Gen5Zu2U2ab_fj7<AA|3l+G|ngF{TfioSBXGJ}#4Nn{F!S}%K|ziQ?A
zcj@A9ln1x(YQgpkw_*D4FT8FR`_B3^uG*bg|7ylJf0aN;Pk%5B3!`LgE6$97`5@zN
z*W6Qzqyo6OxLU#T)$%}5g2f*R#iS%89b!j=53px+KAp>{nalb-5*8rY8bhaZz}u}_
zgP&Q2Uh+Xp!0r;pp)U`VK2($H_7YZZCrEIJy!zv-q&YBMXz8~Nf8LXUf2$2DQ)@k`
zJSI<oA#p$=hu{Iv;UCG8E_b&3TTjgV0sKAo|Mw}8u`8D3hT;8T-yJ`zpREt^7I&|3
ziPvZYIWq`AkW-9}3b;(%xPCnZChOx)7Nd|)l!3_<eabvk02DcaR(A-%Y9S0FQXOjH
zMSQX6(0m2(&7)3@w!Kk5oYfgQg&ur*9Se~0txAdyib=Lhr{W3s+3MB2JN@W|H+4xB
zlRJhywgY>EAgyH$RVxq>m_a^pEg&E|n2`>TKW(4U()^_iUc#`S1?xJ~77qLEiw<xk
zsRixij{JubmAk^{kvCV<tjLbPw4Ue%f59%I>eB~xJgm&rw|S)QzBI`U%UrnH7J<hC
zytrd<twKQ|awR8l7YPCM4v`@2XaK$d!>Lnkq7%1@-NwKMPT>OC2eKGJMWIPO{ougE
z{sP<shrX`SIXk|o>lnnV@@@lbZp>viG@yBe$KMrY^e=xvA3;pxHW1D2CMZh6C@d;R
z`X9awSuWks7iyv@$47^M{aZ9U3A#LN%K;_H^Zah_Uk4(7PJcX7M2bQN3#e6Cac*CN
zT~wWoPfD1$%--ydu3Sm>@OMpciHnIIc?8f^z&x&-;i*8WD_D}9w<g!TLOSM6H?|u_
z5sTeGDAUfnb#uFGOg8vA&Aam7r+Fd6xRh)DMuB+Pe=EvGC~}vFme8;jNZWhK$y18e
zSb>)i1q}K|;ASkWK!fldz77KdmXES#6M8=VQPO1K-LQxy&~|cgExM2_2-_{s_42tu
z>EC?vWHVn-el~#T4$Rc#ptNtDPjp}C0(A-N_}QBT(T(7t14+32R28ouWt-U(fK(e%
z!~ql5)d46*IHEpHSf;}Mohz%2pqN}5m|I?UDaw@W0r}Yl`FJ7raX!R`x+4^%J#z|d
zcwmh>GcXYzsKJ~fi#Pu#keyuW^QeVqpbG2>#UE5N=spCSr|i;q|FOA3Zvt*>$NZ(A
zuL1M$Kkbp4e%y+nJW#w*$~yxOh#SmKgQhMIX&R1)KAKy5m4D~RWBZ$a%g>LXqp5m-
zr|ileTgNxYORGs+?_x`0JeGP0k;@J&h~xk^XuB9C+esGPl3cJ)Reg2mC@W~<pQ^4-
zW)g6~YRDW!yS#YmQeiHx+Eu_dEUCf>s9j+VLg9`b?FnF8237<_f?V>qc^y8<ArF8_
zkb^%FE>EKg0w7Wbm>J11EZ}*q5n*u)gnZWt{1k6K=mna3fm7o~+N?jcj9S4fnJa9Y
z_BxU|`Nc>%W)%mUYy5et&Rm^1A259S@>>zWQ5ax*WrcC`WZ&oMky~v-$Wa$6-1o%o
zLhWB+Krd1C?E@iDa@$ge$L1eBrvP~+i-Ff-1R50&-aS(UT2k{tc91-AjCbur$PJp~
zAyXp+=BRe(rEy&c=5r(LMq`Bho~t^)rJRFjpB%lFT0%^1aF?B7zUXmvVOzrR+ZL`>
zrskdS!N=_tU4JuL_4$tTg!*dU>2i{zXX>CTw6ba)Zxh!9n%{n5Vc~MCt|3j4*q|GO
zM<b|<V;uVUB6O~e61Iv_&W$vMH+WohwISpyGadOdv(u&}YHbVN3ZcR+vKizUKe+_@
z@JT9t;5r__He{6p`<#8&-#}Cd^jF5f=ybd}3Qm1XAn;liYB=3Kd^!|(ALg{gz|eSl
zs`z_`*iKa%Js?+ETE`1|6gf`yw146;V^RC!M-V1ut%P+Hwv2|rEn(`E8&roD`8#B(
z6reD<$5xViNSwa1F6^<4_4!=%hTV_Pd5x`SThG>U1ktpXui!oPt)s<0mgYlnj7HK%
z#>W1-AO0~K{Oh)-&Iilv+0Wd1KvR8xvaIBp$-EzKpEyp0<w-lMn1;4g=<Axou-`4t
zO}O;Fv~yKD>c&HW091BH6%=YHY;>$O_sC5BNUOOG`)74jrKDwU5`5MKEh1dtX6-2;
zY0S0Ca-xy6F}p)iC@eUPhB?+L>%OjT>xQ3M*k|{_sgvMBjLoxHHFcc^Z_vH<@OJF}
zTWzrUBiV5*7lVVlEGQTJw`1Z~iZS%Wa@wf{oueMf0?>t9A4!r%3%KQF-zi%Hg;fp_
z=r|#rruM!LbiuB#cNrZtSZB;Ej{Q0<BHYHIL=$$)>rur>-P|(g;!5I?2SlMh<t?}H
z?eSY*yk$q9N2CisjSSm~ZVfEl`r`7-%)OSU`<0=OE_S-D{ft5s3j$t;-)LU|Q`Zr2
z6_kaWe8m9K{X#)k*O=J%-VcDiE27TL<AZSbA?=OXPK`Xt_R@kbt6)bkg}&>vt5Mdy
z(dh&>2SNFXyr(oeoyBoyjsii<W$N=2b{l_6qyvrjUN2p0DXI?dfGHX)IQ%_AY%9Vz
z#m}CAdxn!domTk_o5{|>faGT3E%NOPLh}||V)YixWkiH`dbUwR)#8_YX!{TUmK`>K
zq^W^_@SLkARl#LD9c?~vkvf93hfmsb=3q-maSwmSLoayROfX(`c59x_-qX%Gl3{Sp
zsW{Ex%^I@a5)#t4XiKBD>M6Jfh8Lx0Z>?G5O!pDm9XvUsp;bH|^YFp?px$F!(?e@3
z!t^?fVw68}j>b^+RZ@|Skj1zqYM_tkJG<vHOg`B(b89~C%LWTQG$#5mT|G?2dvGYY
z3ZFoccMYx$j~d+ux^NKF$}Yj|@CvBRpJq2bBqnd;vhL#x5)TaxEvSOqI-T3190FCc
zk`ZZreSPGypx2Om8etXUXGLpfThqfwj%4bkYK@n;t3IPav1r=4BeGbO1WK+*$VocA
z3y!1k{}TyOR0gz!=K}ED^OF_g^#S=j1|~MkpSZ?<mRLgjY?s#s-o}Fq)lhyqv86Z;
zKTgNPf$ih}YdwK^4F+%}2*nLzDW-lco`yLRu0GS~uQT@ec(-h~IdCI6%=`4{GIk2D
zPM$RHEy`--=^Cc?O+H)CeNo=eGPrywo}}XP=^{<_L+wvvjGGURy;M>?{m`ep(_duc
zUTOD_gP8DWz6WtaK@25cu{M)MU=4IUUF|ZGo~Q6=bE!Sc9LEY2`N^eH*}T7xQfwN{
zlsVQ{WgNJk;J3#z3djI;0Ei)rA#FPtez55jH5COTeG&35nrLVX94`kNDRL4G@=|#$
zcMlk~XTzgli6T%wG6IGpSkz=xH!S2+FFa(7_nf;S?sd8o(!<_8dxmor$oWnd=A;eG
z3&Q$>9zO6fpo6}>EdymmuthP>n%=w@Pj0W-oDWbt9t63r9zcxqxMpcr1@x#uAUxHh
z8gSaOx99>QNte1WX4DR&5)C$L^M$MY3@|uiHJ8ylBD=1&EKUuD6;=e}e?qN%ci_W2
zs@)O8t@rCYGNMB(x#}V4JATf*+ODw{nx6l%|8&wBiM3!0klr8IJN$&W!IRVWEVEW%
z^BEeQ4;vmkqh~y3D_#;X<k!!VjM86nJkvs&^q3hJE`h7ZrF7;!2TyH9MA5xnz!U+$
z#FmKxFb;BXgF2v6s2Lyc+nM7-pmX2vG_?=T?pZ-6_IVH#G(bjj9AzOIhVmKdAoNP7
zUcvA;8Rn=(=e2bPJ?C#{v!&dIw~eN$$atsB0%ayR!?M6~bQ+B$cH%%B$ph1d%qn-8
zai{oC777-@hCgnXY2AK@3&@>@VAQOf?+PosC<(Nmt)C=sKg&y7j)G=244kRnDvvCj
z5Dz5MzAO$b9;a2c`~DAl;R;POsCB(UdRXx6(rDhb9q-BTu^xG$T8+lEpVSNd6!BCa
z{KrWlHjQBcC(%yZ2FGo!+*YW%cH5^+It_AE&jr@Vu;wVvcvKbvLCS<VPpGV1qCi!Z
zHz@t#3(X&#w)0Cz4++reJ5&7t=)QDpco;9Q@vX)shBLurc)Y5#PkZ|(utZ5{PrQl2
zL7&q{6c-GDj{#IY@9qdLf>B_j<mwf(hh)}7u?9?GIFPy++P5=;g&+#LNw7E9K<N9(
zK7?d=J$32@swB|FLxFdx)2qq~Xc!o_GXa&O%zpJY5IK-rnd3{ewE)luy_iZD2U81v
zP27d_c_lB8Jb2E9q9ibB>=2`bvCx--3D^QB1)Gqj91WHxyGKEa9|<EDzJ6igC7{N(
zwh9VnY2ddUZD9V{d;VY_n)G^L74N$a=>Sj1YhM<){Up_2O#3Cw|C(HVd~W-^tKup+
zbU5>Ge@lo=v6^HRx}6Daccj7so9#00!r|$|@0NT(4s>IsUGu=k?!FQy$z5`ToOVGR
zt2z5fb+Yw2Qi-f7=*H;i1O)snOet)a-O0WPbMuxnecX~Z)*RWic7Y*FCI7ihuuRh&
z1HYBd*N2987~tZ9x>rY42Nac3UkftSG8ti@^uGNNDE!hAa5ZGL{V7%7>aYL!J_<G=
zWgM!uK?LV*i&2BEYFj(nxH91njA<DVGEkDZm}an?KX<MU{vsmFki0HovF%W|fSnfY
znM}Zax(BVXhF(4JJ=g$5v+v5eqXn9kwj6^&qxh5|FpWYs(4~%Jp(#=xs>7s9oiED>
zEPsZCt9!Y6%)n^jn|ab&Kbt7Y`hmw@|0(Cz2&0_)R@hAX*T*d4N(k%XwLi7Q=ZJqp
zr5{Y1%eY-SuNOrCotOh*oN#-459CkNfkH4ed9c81ulGX6U2Qq9lTK7k3_JB%SnrMD
z+euX>eZS+l><$9R@1Mf=Y>qYWOCr7sumvlqa`TtJ>GSSQ-s!n#e}WhZs%xPgxA!;z
z&}HP+wUx2)O0d2M^kKhV^ppfV<+d(RFk18`^R&Pkh_bwJ9%NaBivH{dcr+<sqlL~w
zlWN2=%u3q6d*t(7Y3!v50w$%!d>BCD`^(jz?FTiqeH#2gi3GUfrCy>S^gb7Gsfz=W
zMntE+$D5+-KZj5>G?*&JN3MM4EMUURj|+Y~*xx?BD6m7h_#tL&=P$VypVt>zb_4i}
zlu7R6<BSi$hpZk<5(}FkTo-=WP)Ej>fJ78?nx4Il9hJi*Gnl?QGW7L|xh3;dkgz)Q
z-39A4pX8e*oKv^o`K9A*mMn%DW5&6gH-vGYl)QW-grIsT8w9A_z|t)3$dKUf6SF1n
zcBn6$R}_(XhP{-EEF{rQ&DIQ{hr}tTVOoB#gC+=8%!onDWzw5`KrR8BKtK-xVxVtp
zQ;UZlueLOepjl2$qz<O|3|=b>WjuXW5E~e5K*_w66Spz`JtjQdSceVBonfG5pJ<4&
zf}OqR!A$!}?`hXrL!?B`zD2x!5^2b`OT(rz>J@MWCQUoF4PyW8wr~B2BeMBWhr-9O
ze|@O3SWXU#a2K6EujhELZCYwwi7vnL0?kDLY#4z<(Q^WFO&9=;Qc$gCS>=ZE_}<j8
zmW{u=Tt6(FlbS_KP1&C|zSjEay!j297tin>e7f>{GXj5ls3LI-y;egGEHmD<RL5Ol
z!gmh{psA`a->z0vejV&xSz@X-eRCLr;MpZ`_>k5Bb5fBemqoG0S;^M}<_!hxGIn~g
z%e;f0k1|h!)dh+i*>YCN?R7<>HobF2RaOQTl^{XTw)M9vO%)}fleiTEut*P8#+tYI
zn)fx_?XQhr;PXxhne)GG*+IPslk>YY=DW84@MpyROSOlei{dEJ?AZHXB|L`OSmbHB
zpvMN*>)YmGR9+(YKIO7z57?%QX6GFa>Y_xD_4AR)i&b_^Q!Noh*~kS{HeSf<iPbGi
zer1SaX8ooNi`bvn%NQzW0{AG;&JQNsYAX(3qC3*rz-e=uHTuIm^Hk+qzk8^>AR93p
z>AqyS`BoE;y#)hR)S~siPuIwa0}A`J{tk_6y=4+vLsPdm20Ds`XF!P4P)jQOgcVd0
zMQORawqk8=96{IoVk;QrppHvC-~4o*)!W{piY%MnC~XZa3Ya**CHFWy0$bQ|#@;;&
zn$~n8KI^`4S?|!1?eir<gL7-?5bVA1z3(Kv0RFnvisswGkH0*=e?3xspUnqZ!~A|*
zag1pLEqtLuyj=ZZ1q|KlK|${HImzz;l83^tJ0o;>&}@qttPpj}o!=<aqiGNPfh-VM
zdmk>LU<w9>4-viF8LX+0KnDyMMM0imYIQfFZFfwmtE{_xH*0G1pq$=LA%P67O25{z
z?!7{m_Z~MLuQ8J_jXC{l^OTZVK$lsML{*1QT@L9hb7~x3P_-2Y<j5FE9ea-+oyr}Q
z!yOp-He==GaT6GAmpjFZM%PK~(_n6YAla-pm9qn88>}ECx4+DiB~e7|-WyrGXtk;g
zULRdIbGKc&P06=X%>9Mvb0PlHOm~caOIF#cbVa%dQQ4XsPqyQihw8f@2Vrwa#SEQL
zs-kQ%#NIimC@&8aEs=xq^>J{a=_J2<@PrtlkFCq?G;0AE|HLC!A-C6xBV3>3v?OQh
z+pN-NcWt`2`&FLG)Dbyzn*GqE(c5Y)6dcn!=?-8b4*sng7*CmidQba)4@gRrLEeu>
z4CZeWe8mrn8cK|;>SHc69SM|!kc6O*9JqSiyhnYe8ye<gP+2%#X8ON>Dlq|EB^d*`
zM&jS$o^6I&mT?WQEA1I6{gaov6UWi&nmK4jb%JArs%D=P&VlNcuV2z=9Oq7KDM#hw
z$*}n-o78M9YO3~ZB_;ED*FDrlQYxkghECF*Y@xb@DRn%tG3Vf(Ar3#XSOB5xqazKd
znndL{w6IW?TS$S1ES4%%ovv4+(yFm{rN}ECd?F^#cWq-~`2<@~irk7E2nv$-z0Um)
zEey>h?yaQ&qb>#f9ou(sGiUk``hpTo8iMmEz;PXi#{-Wp+5pl}W-l9nRM?=YW`Ef6
z8l1E&5{YlzxY3aZ8K&{;$JbfC&FTPgFo8vbebLQ4ReVYzrwi`h6$%`BYaUd}?`|oO
zjD3@{Gyi&-Hoo9IT_P>qmc$cxV=&j(bkFS4D{n|(XL0E25@FPb%8Y(RWKaUtM@PhW
z3p#k+A7_i$QIbpP*lnLnVYEGhXksxeXo2lpTw^D}b1_VwvUdBKZ0ESYARNBA6kI(N
zfkwi#e;ZD4-~juFk=k?C*J&o#sP<g{!aktUB8>&IK!XA+Yn7c9{MiyAnZC?3_wodT
zB;zk{6FaOF)U)*~SwKLxl+5DPhyTO*N5(>2nL@U%g1yT7jq*Xy?Br8fDm5EN-}$HG
zd~2#xegD#M>)miw`V{l7d*}>O9X}{af~>K?Uwu3fEdq!JyVL9@jUy0Ej|LT6x#<iR
zaTdyl_*dfhV&0gc`c}W>FC8z9!++oL!@-|Z$e*3<&sr4!TIFWMT-}ZJmxlql%Jx`D
zObi<UI33tW+NP1KESO|YDms8gAF`y?bAxw)1eTjEq}#=dNvb*edd|~*>8uLoAE5$a
zo-&OO!u~3mEmdv4I@0szEEUyoyv1@qQ){Igvs;G5)eT3H%*W`+t0++WV~)Jz_K_zu
zz~x7~3G;x8u=*+1no}$99E_zO_fo)&b#-}$!eyq?5znpU9-MpYihpzNOFZH!zO?j#
z-OT^$0LbH@pbsEt#}ZgvDFHhz1*Y5b7QqKb60nb2M!`l%?JB^PdXHJH0p|JsIVFc~
z$xv>KmT1BTa3K#i=^t?U!g+{G5X>P6tgnl|fKioxR)ds(!}7z}@UN@VJ}tL?;K=4<
zUf+;vP4AJ7$}F2fb7(lOu75W}oUQD1qz2U0jmd6m56l-*(<7gKTVSrD5tLFU05{;M
z$Nym(*x3DJb#NmR#LRVo=4`t61I&NfqwOPPK;1J8OiVWktThWQBC?yor*0%YHLf)x
z{WcVHbs)xJO;w0@?lLog=A!btOYhmMu*1vjMPt?r;~A0#yQ%B=_tvc%WY&V+Jo~uc
z1cdI0KNQevY0t6sFb*1Eo(1O+n`$Ch2O??kYDUKtT>djS<0fw&r~z!m1jCo&4u?Yf
zi4Jx&u<XJcOMhDh`MdlKTl*+nl&P-mvI8ud^sj9TM7c7z%Cu+6E%Y#vKVwI<9XWnU
zgP6kYxjrgy1J|zY`1Ka`?Ng*B*;whUK6Mk~oh)?RDIjm{afXqs39zS_&m09*s{HFH
zw2mF_fNF(8rKwqLr2))4S}5p~4p18DSXZMPeEoB;&PUFM^861BC&scr<9$ywH7;8<
zR_x7F&h)evE|3pPq(|ZL)WaCe*zy$xCOi@*GuF|bIH3fL-IwcSm!X)}=8N;9ZZ2Ad
zNC_6GPF*0TltosVg;#`wV&w_orZS_lQDDC5B*NgBPM>SFFz!wJKa40POd$-i2TGl%
z9+=((n00Khx*B%9EB3~58TmQR4D@G(Wpn~OuUCmdC!9=*j_A_$SbVcydtEZm?J4g^
z-@hB~kLrjQ_eg|h`r)wmE=-cLw5&A~qIL4@Oj};^KD4;oRe$UQ8oYK1)bK-kD*J?Z
zAu!Vk2P{NNSs9@*5{OI@a2D~hHI{)EqnxeF|CG2tU-dW`2%+77%HZi+e~&IGMHAZi
zJus!}=vjx{j@a<5Go3l{sJhnZ%!4Q-dv^TY|LGySJXlo;LYZ2NMC$`9;PFsswaA(U
zK5Au7@tNdKJC%?j^|!AdJ6KqDh~Wy9XqQMhGK8(nEuL*@NOP-}v9}-Zoqc)lP^)BJ
z?$XmaMe{XcqpHgfdBjwFI{n|qXfzXS9rH&7t{xw&ix;O_&H|`3t_@>2^QAZDy-G?-
zJ(nQLFLi;9IPx74*Hh~#veF5%@AU)IYs5oNmW`4pyrYOk1|zM4tfZ`J(5-QUc09;l
zE#!CQiEkS|Ah2%^hd+4mv3B_2V)Ubx8Dcot-xRh`*P-gtr7Mj9lSD+l`VS?`4y@`;
zWV|}|IPtGVF5_NB-Y;oKG$y^r)MB3_Ee^`!l4d6@+K!{sSD3w$(gh{0Iqyy_LM!zW
ziGpXd&&D5EnT=H9;*Ad6{k+Smet)dIgd-c%sH`#JINz5(UF-^~`&X(CZVrETV4H5=
zztv=(x?$iZ&eiNBy-v{VP>ecjp&zs2sqL@sQgiOaBMArA!kPXVb7Yfd0xwcNGZp$u
zrR}~F_E322DcU1Ka~c~qPYWA^wMIJsDa$)I?%at4?u|4Q{Leqd@GIe9vu_nO_zNd%
zB-dbmKqvL!CmrK=aSNMm6AkrTar~76`%RCApI-}z;}z3pKvG5Q@}Q3h8`mnfjukJ>
z_X9*0<mb3A=5FR@+$sB-=KA$Ra|U7Ob$KmiwUUeT@pJ1*gDOr>I7u5qPsHB3Iz_Eq
zbXnbiM^?v(25#g*IquXg493S72XiA+AN0VDYzD8$0&kao`Xca*k``L|LbsZo)eE*e
zh_mz^Xc1W^15fs|s>>m^FAQ32-8HpuTOL{bdyS_|2RX&e0$_``&nc<E1FD52;Q%(i
zE5p_UnzdNDr4mhp_pu3=?ms@Dian_C8@LsV*&~LjjeOqWw!L9R5gL)Gp;esW(slaS
z0%08CY#@;u%iV}Ji@OKC;+F=fOxZ!A|2`t6Knd-U05Nc312WP;KJHGv&H-1>>D+G1
z;p}!*7SeX-TO|JWrOq;%&Sdh)Xo~8yCLxP^rq89}CfYu*YTnvK;yXl`W*7h{W&<R$
zEPtdYlZ)YdS7ufI!OE-mQu_)d8wMabK+A2~4?PnrGO2Vrn}V284Pd*;B6)|>PWwUt
z8*q-#0ZDVju{uQ&cjH2wfEJ-ApY4!OO4sP6cS+|FzKINz+yGFZ2h@mT;xU{{qwRC4
z*Dv=2oGY|%f;@!T`^VQHW7zYk{mH@v{^^6eR&QUyOk)JhK^&d%NAYRMl3*p9tI$)H
zCeeVMr!BkQ&ikhUW<-UV<_bLE`KSg>h^xc%>%TjyKYvdF+Wvq<rb~}Wwub<lIoEKP
zFLG3YyD$P>2z1McI1mX%h;~r6br%BDn_~D?^xel8qG4R_6mK#Lu0%=eUw*(#i##^<
z7401Y^_a58CyPgbj+Q-)M<9Hbf!SiIgyZ86deKT^l=h~#L+xOBR`6)cNdIodBfcBE
zXWeU5wwL8%v)it%IPrs<1tO|~jG6-mZnAdxfP>gfLZgF>zog~l!XdlJz^$6fEa_MG
zC4<8~2Qa+FRN9cB6R`&9XY7`*HdNU;O(V?%K#mPCT{(Z76W&=UlLA%!L*ATlP6K`5
z!5VsN2KOT5q;EUBXcVzr77tKEE)Za;HJMq`w+{c)rzFWgC455G4jdDC0BVY9_|V|Z
zeBuE#O#8k+L5vApfx+D+9txL_2#uEO>lFhmaOH^j41lF?-`5WPeDc5PjIg3M;pb72
z?qA0#_vI1*O+zbNwI%g{8;7(h&L<ThwZgV*bwH(U`n}PL7!W!;_2wbmKnlD(*dF#=
z0<wHzkz)V{51PK|+18>>5H&tn@>d*ge+hs2@p*UV-G-RcuLZJaa)RCRQlso*8)Szo
z4z0E0P6hhz$iXw>s*gGL%gG>njt-u1DNEBZoc+*^)#{D;b;COV45e<Nn0Tr0jem&w
zQQi`kytQrN_S-LQ%ZS2d?CpEpzdVYXTiDxhcc2i6f!$!LF&aRj9|A-)?Pd5cdEh_L
z^YC<Nn`%h|#0$+z1DsLz^F1HoR3O8lGH$aY`xxays07_LUD$O?6v3Xzy1z(o`0zXR
z+VaIQgAH23j@a308sWC)sFT`+cE>}``crGC*4yqKTBM0FedzG3yhTjGhbE})O*(-C
zGEg3U`kG2sZzm-z+&hD5|E%4c@&Ph)`xJcd=ivF5rXGWa7j3>?XUlD8A<S=^9*~le
z2}RN`=slmL!qku?YZ5J_i9CP)JaPdV1<_P-WOq&)5BMY>LruS4i@^M?B|^6jY>G|M
zF>;YwUPshPU_?w5O%ZfnFFG8T<6W>2!q)BDFu(JMYQAC|5y++oDBoQj_cg8kH3niT
zZ!&yAEYRQyrvfji6+*q&!n-!!=lAew5-@Q^Zsi3id;SdCAXCN$|Jn$6Ur;0Ey`K+;
zgbQ6SvNDRDE@r3XGW3BeueC@Dh*Qw@oO&10ZJqWu5RckZ5qa5=gD<Nd+9{r$jjI=-
z(^CPuIFpFi9@q_+J|qpAylhGJpH@0$AaQ6}!mU1n-Rte^=R3s2QO$K<viWZ~xdwd8
zR-Sp5rcLd69o^mJEZRE~G)(}>APkhzOAI76tv4=p{q8MdevfPky-lu3CY&9AJ-_(;
zzR;65zPMq~c9Cq`3t->TX4$n93jo8XfRhc$Y79B-J%IUn9l&r|<%Hxc^w4QXfg)6^
zk&RVaT6zvmk6i}6%$P%=6|093_oR1;6-xA*B?I#ERjf1Ye8&f^g(u3CeU`%6LZn{-
zp`-2|1B-}Y>jggiS$8s${3iZik76RLNQijPXCBBd+V{mFF1#N&$WBA{7;|aEg!Z8B
z{ly8(zkZp2!M&Tm!uNm!Z**{JA2#OXZas=?F*R^zlK~A6jfr`3wis1WK)i7@ndX6u
z$%GPfwl?H!<Pcew2R+9P7U1CI7Ye}i8MYG80OW~Ix9#oAoF6<_n`6tlEe3M(j5_&0
zlVfYHhzzrFuPk~?b;#YrnvBSPE{c-(eUs)r+h=y`qv^pb^*mV&VF~3Ppt^j2EUGO^
z(}aPw8NIOo{FO5MzE4$<1rru%N1BH#^uYx@`ixVCcxkyGg5+Rr+>`b;Cq3u|QvgW~
zJ#k3_c~H@VH&)qT9$Kho-kuR*@_vjCm2{2)H>B%PkM*Nq*)Ln*;E~Z_vxs{~8>K%H
zUE{;46>eJ{S)^$*yFdS{%v(&S<{?cGWMG*<qC;ZXBR~e+zZjj&2|SLx%SI&Nd$DMe
zUl1Uo3B(A*=aO3~fSs9tcvHEtk<S=lo_xe=icIO*uR%;pGq8@efJmYcdOr%+-2v$;
zlwAgBuVitNF5P1}$FnE224cxMdr5R<N-yqMQa>JXA+VT}>CmJ1xCy*v_>x%qq|fa-
z`2V`f7}m2p;#BOm?*cI;;+@jVuLGev0@xwne(0*+uZ5~C8E$}(&BI-L@J)WAZB~Db
zwuwm$0gBYRX5@gnRY({~*i>w?&{IwJ6v{$R&widtkV7+91k7p1imYKYY7B=K*`E|U
zso`et(L32+;U&4N<$6>s-iEy%vxrAV_#3e4-b?pR3Etjy0k-ntMH;r$*sxy#P%#N8
z5(X7g2plxg_&P(aaryT9W7xDmanpT31^6@ERHP0c6-a2k#W=-UFxZd5^4thiy}>RK
z$IpDOvqP$DG6N@D1n6K3+uK4l&}bmDqfOSDO&VB1T3447XYYw6)T@rB-}(4;<3Mdx
ztOmZJ`E7fC6UFj_($t6c)aZ|~P~y4@;=k7V9P>+y==&Be&=qz7T9P_@QzCG`*2jF*
z_u!xS`LY00|3he0Oa?p=^uf-O9s#pxF+GeyQ2q@AOLLu;lTbt>r$1(m?3lu)um~s{
zf>-<(fC#9anMo4_SK>zb6~PHCXV&REDj!LUN5^l|kMUZ?Q}V48JLD3LtJkrw^$O6S
z!90Ek<u#S+3aV@GEgiOHD43DW7odO-Rr516J@YOqpmhHkFJB|L`;X$~eRV^C_mNx(
zEyq+QGJ$uLuxYR`P+>0E33KOqpb1rw7=9b_Cq+xHI-J`Y&^FJ)PW|^0Ze`C-C+UyR
z6`mUtbc+9R^}~a!l4M0x9(*|q)@l;1=ldwG4UBRBr<b*+3w!W_36d_n31!#e+2m8C
zDbbw{av;JsqT2@YRX|YvX9v=Ud6S`o_^Qdt^h$|DKu|-9uBgx47&yRs1B^V7oCr7+
zM_@{)?F|b)pQ^tBsyX5bA%8UXdB=b&FRD{G^h?<*g$J0tf4Hs9Jg3PLRGPZ#w)(ww
zIZg%cn>oSrJROKC_C5UOzvL|%V;p>YTj7YAiupUpAid0aYP>2h>@Z|}Byxquk;#J@
zM17bJmTPQ2^pC&p7!F!O4$VmzI(e_r22{8CH@sPpu~}vlf%Jqdpe8sLgIZX85`i(^
z!L6XKdPFTl_1=pY=dLWms}3fJnspvN{I&Yx=G+*jRv*o5TxGw^uUxSB9L3QR;kb%d
zEi`j~SFJuIzg%yEzX7B7wIRAJzfbj}9PnD!iv~$MLiyIP<So(^wa7myW@(?^`nsF!
zE|^w?{qdEQV?x6UG_{2Zl=kavu|pX4X%S>_MDs&r({b0vIkY_v>K*onRKg^%j}5jl
zrDs+H<NW();f-`whhjLybz_9QDWAhD=RODMOqcwqs<zAe(YhF)@?(JXD9w<ZPv=fE
zmhius7~DUPzUoPn-S&#7fuiMw<bD0rBguhKoFifm5D?{;l&=9{XFCmg{&B%-DC3KG
z5-?D!ATd$&rp#KG>T$Lsud#(*3+4Y?hfm7@s$)rd4-h9A$Q8&My<-lk9nq4Xx_h<S
zpR5`Ba`q_1mD5&E2k^y|I`$ro7+fy4R@BsY%ez0gnz$23clWi`313!<hF`)e7ETi_
zfG(?UdBpaLg-o!u5E$sT-U(rc$4^SZD+S3bdZjd;KTB4QKdx`MvP6JOjUe>0ubf-$
zQZ{RTe+l5nX~HbRu&uC=4|FJqDF>wa!pQUpDC-*GYa(~()=Uk~exbsr*8OMPCZE1`
z90d?13fhs02IU_7&sSbNO3jg-0dmRr#%H8<uHSDY&0VZL#(aT%m29?Lgs+<F+tDbJ
zzqN-yIN+OmTwJBWb9w<lD_Y1fP3o*H9>DLp-0y-E>YHC(;OAGT`d@tY8j#(j08OY&
zF%TP<pbl(r&adss&qTv<RE3W_XNv4nqXLCa1la9K#d)v0S*@+F&O|{)I5DongKW0q
zudbw<sCWROAq!2?%D6R-47=ZU{nmDx{l#!=`#FtMSghWac6R3LQi(o`l9Ko(d+|Fv
zcWO|D|B9mG-<L&)dw!IyPW0#^&?oqbL!Abbq<{-r$Q1;B4P9a5ZK5<IKfiYatSbAn
zE`_VG2KvoWG(d_K2bz)zfX`*ztGAKQ9#j)fFCoE3d9O_vzyb^cwA-1Uah=x~>0SOG
zYwsP8b^rH`N8~^_l$nuLLLowAiwYSbWY3b3StNTVGLnR>C@Ykal@(={NJa=D*_D-f
zzupJ>c3$`6cVFjy-M@dX^ZcGyaU7q|`~4cv*Yo*&If9R(fkE?|txE7B?YtT=TY)r#
zz0u{BMRvmlcnsFxxy_j=n5(!wBnLRP?KlY669%qcyb$-17d!93zl$JR`eY?D|F8ej
zTcse7akbQ$dL;QSY#YvjNy1wXc9p9mP4|^Zoif1L@ZKvbR;=Nl1ibk_Pr!zQz!Qju
z!>s#W&fo^JN(9{M^05LV(Hk=f8B)AbPGh0#yg}2znNo!55>wMLLL=BmQ~?ZwnDa<1
z$avDey6jQAVu=C!jFhANit1<1F1vwqx9ewDXC&m)-u_A-y*=@JZGaX6S*COe5Q!&b
zCH|42c7bOE6c(O|7f2k{&V<toDy%F{Pgb&W3PIV6@Jz&g!F^-NKi_!i|K%Hlf_EPS
z1JAejNiP$@(6k#A;HS@=(QBU4zL=xK3Z&K;fcSc6U)YHN0RpXf0v%+2jHsUa^ewP5
zi{dkqJa(Gx@ZmSQLP0O=%M|*4tr`g!uXvWao$9G!jUc21B=6=+ZuU*>-y71`m5eYY
z)R?DTyI<02{+A|_{}G75W}dbfMEG4rdydiW@1-!&OmO?Kg8~w&ysZHKr;n-lANMiF
zM|>c~LR)qRl@+KL1*R`^*fqFN?J!X>=m$<&6EF>&EfyJ&QUo>bz_6OS(TJ*LK;=<$
zb#khJ?<B1H+<Q<`f0ca_K3~u5^-?C@uHgSnSO)`7rj<V!<{wNaO0x?P=l$l-gdtF3
z(Kn><9aXCOF9&^t8VcrDd7LujS@)rDZt}+4-6t}RGMElT5U2xd8Vy6R7>_>%wA24M
zTkmC0uB8IN<uFK>y50b@<p{DDe*4uLh(3#VoMLvPRRPeMr)~s`r{gS#a}7?3g5B4}
z^XtzGRyxZ8LJKapE{X14Fi$F9gn`QBIl;i(0*g&TdVxuAjsfDC($neXs(W7Ei9E2`
zU~z?S4{as>5C9OB1<eot8QoI&(ES!&JkcOibi%X_qb^gFsSC~3;{Z`{D0DML6U?yu
zMLTn1RS+@%59h>zl(E3v07cvzvBbUf0u2DFrGcmZrx(!|k@q@K`8u@FpdabpAbmTJ
zz7yWJZ{G-5vi)q6qj1ywqRTK46R;vQ{6^xNMO^|x={?3iY|wlprNBrhUFHShr!O^5
zdl29s#!loB;rXSp|MOb=B%^Dc*dGt`mtHtq=CE^Vh5?tokSQ%}TAwxNlgIey?tIht
zLjMonv>O&=iGvy^DmL~CQwOrhS-+W3Zq3z<2Wo&Q;FIYI7rg}`fEs8>Qh5ZM0?U7u
zxr>zGVXPc3m4D3%eYly;!TB`(e&NA*RuBKZvuY~bF-?h!Z?53sNs;*fob)-^bLGE?
zZmU@U04_<wfE-Hm4$R-FUG%TUz|NhkNze9;MI4a0d;I^8VSnI0Ixw9H1;U%i6W}GE
zgB;~zkt>V_Rj^+y$L<DDv@BqBcLBEP7?e^9%Pt&n=={|2bC%ERp7A#|^AHuz$A&zd
zfUOADIt%V0xIl&J?DC^hjeGRhZH(0)gsA8^Kdycb09bG_I1BQ%`hIu!dr1PX&s3*F
zpyIX4Idx(GNHRGyaCR$T2gRrB6}qLcgSd#FNJ#1^#^2Aoe3P6=OdRH4_arC(#{rs7
zaS-AY^812ZntzuE<`b8&axfIkvB<sn@t_F}OzA-H^d!4I?c9FQtUb(a!7!w_hAnrh
zJ{|GfWF!&0_9-+{U}C}XLh;#8$>9NbEQ7bDk5n?dIu3mNYbaNpgO4O@AD`(h83qp-
zOmd--+W-{Olfpb0|9dDjUq<ujEDM(@P!RUt7{&!A7*rxvAucc-`xBUq{^x;-GatCY
znMPHWGYY|g88mur{5lx^2C5K^rzvD8*NVepZssD&48@e|VarJeRhr&5lM{u!w~*=)
zJk!)I(xlYr`_SN%2@$@+eoRr<?#I6cnE$$nw~iq$uN(M<JGT~J`5;SkR419oY`2d+
z0xQ;)pB{oBB#DT}lDA4$?<Ra~0Er6C7?y!y5_dW8{<)lBp7B2oLsa9iF7g(*=$Adu
zWFev+AGC8=m~@8m(*w|_Bd$aFVf<pL=v>Opf+7D~6>H$wQ95x%`_mEn&zTie%?}#2
zsGBJH)1;5QVy4>{EAv+)G-!sv&b^9GPiWECp?V_JAhW=4KPXLq+|+;dUBX9LwFOG@
z2Vi54tCUat0YMG_t{6RO(cT{9<NM29%Gp@b2TDwx%GH@ujj&DfH-KE*9Z9t{-T>{*
z^EARlbCvuyPvZ8Y@yK47j`Qs$3#<RfM2p3Skk_d4D8L){fGw{>hWup!+P=Y1myCZ9
z#xS`|v_R=Y6O66O8RTXzr@DDk9K%jpcTzrFo&I6$q&UF%-J%buTe5BfdpwTr-Gz6J
zP<C{o`tz?Rl7Ee&EKwTG5qw*+eh9S-x^NgX3=z*5F?PSCK(9Y_F+eo>nBFIJYWb`F
zyi%^eUTO6#eqa8#8>SjJs?WPWNtE2PwWZlx8^3~YK7?nA`&T9Z)RTn$2+iBeiTlPx
zb>9EzS72<3gafb?FYDKM_w&cf@xte|Evue`ITD)m77oL{Q8`4+<ffB7HW4=k3#sly
zV{HAoW+O?hL?L0Z^WY2n^)EgQH3F?kJTQ%FNaB&p50tLLq@81-GNfoHsIK`tW+_9D
zaPK_b>xFW3S#Ntly?hhA_>$0WEsv+r@>ll(@<ll|SQp6l>rWS)@HZoMjklW2-IG(;
zgQ&0Y@(S0;_Qjby>%A1fd_2JA&;V`V!o1T@__fk)&}{^mw8_x@6Y>wJhCZ*z=;rD~
zcAhy>;sUb?(T5jRpOs%ij)1`bDV(|3rHQ~uST#_0$+9Zdm)X#3-R0RbT5*BKI-1FT
zmq7S}Ot2AXZhNYG(`?bY?wk^N<BRi8eOr$N{IrZ%61)FEyyvl>B+VXg>8pe-8#g0*
ze~q2|>$10S!vWB}eAi(W*FU|e-0P_J3i(;UdV{Z(X1IzPnE9hVM9lUt9<UL#@a%wG
zPJgpi=gN19O#Oazl0$6p$*YK11(mI#kq;H2t5GZ)#i>os>Q&i965P9}0bvnsV2xw4
z(O;hf%U@Xy<V6k;+<~RT{U<eW74(sCB#7eR^<EVqvW|cy;sa=GPEARH4>+vgX@WND
z*(zFkj@H+_u{>6&OTRS-yTF4~B6|;<IE#Exo{>>=L|(A>KQD-<D>NxIjBwEPfU0H@
zJr$xlu5^`<(dHvx0bv+vnOI&s_x<D|tMLiEf=Xt(*2o|KVW6^rcpH@7ht5k`ZpFdh
zVh}i>hPXyAVwV7umyc+PDa}~2<TVQ!7R^6V!}6a|V=z#nYJZS293EN$jmFzsKYvgY
z4v&iN5ob1dN}NZiaN%m%05SPi*8RW#gYv4(m5fh;c5CHTu4KW&D^{&<2vq_S{`~kd
zbYjgwX?_%V6j2Nqf~Hw^gh0n$<aWV=R^e*{Iku{g#I(>HrW9Nc(W;%(r~Qi(DKECT
zhPiV7&aGIgm-#*Opg@N6s1|=gsN4F%C<&$BS_NNEw!OQ2-0>}MNf;0Rb&D~V;kneh
zEsEZ_48_e!jSmlgv%`~V18l<$mdDLMax2lwP-BJm1lUH*%rZg*q1)sj{>5MQqvL=z
zo_kG>j)D(A?O@<5elL7ix0SxaZ-?kQw}L2t<gQuL^Cxuo&~eX_lJOJoMKlMDAf!?g
zI?J^L-oFAH#+3!<3kE70jzfoTuLu>x?#WKS$BfzA8noGx*Yf9JjfgpNW(RYypaZZj
zIRb6ky_B2#46U{YjI$snle#l$qU%LneLdP4dMrNB{lsUXVd!rxKK2px<geapJJw9o
z$o3dyD&~!J^z+Z5HNVAbT->$WC_&KnuTx`e2~Vd*+&#4aAS(T|pMmxS;OMs=^_q}r
zN?>Eix%x;pfL38b2J`1kO{VuU7`k&?+%n^|%R#heKz<-SnuVK4`rn6Z7SQRuI$~aG
z^`0~I#Fi<C_Yi@r(UlUe<7?jV&t)!CZxnsbBEs40h}u!}zuuN~H77pIha=(iH1&Q0
zlan-%#3B+nvc$OdY9S`~DhzaJu#sT6tXtvwe5$`(`=yaBazzJxg9Fi+rcXJKN1(RD
z5jZB+rCEX?14D~_i(bGGK@kwY5wu;Yn8aR9hFrn0#W^fCHu~smOU{Nd4_aT9OO>ZD
z(tk)vFmI9$mTTkm!&_FaR?Z`mJY{v@FQd5XI(S;Ku0QF@i~x+DBekwRpp~=<SaBkM
zkoWD5$g1u8wGn3}^%8dOCiz$~qnT!PA>^dXy_uvF)umQ^8Eej4b}HUR@PNi?KSf-N
zUJ789s!w2nGh<ALPS6!Px;_7}D-NW@Wuhd>elce1m^rY4(0$qeYz^e#my&By;Lf6?
zq-=kA&BzS^wL{$8O*fa(b~i+FL`s=UPcEXBr&p8Kceo@|chX^3*o87@E8P<N6xgs-
zU*?7Q=>?ZnCyObyTO7k&(e&>jJ>Td3yo8T_q_p=~jAuWOIxc2_=J43LbZ=clid$~)
z3V}8OFBT|YV57wWff2WW8FqYyO6pgI%>u_5@p`6jP9lsAkf=N5OT3v(EdVHvLa7eI
zMVeL#gR+mWW~<zOJz2~idBA5B19Nk$Uk|u6o_r*4o^}AdrXv(k_d^p&dAE?N1`v-f
zWj1zp`QO?0r;YfU*5SuS0mlMyf50|CvejVu@p+*P>4!Qi>Q#1miKiyICTyiT)Zoy{
zaOJbiy@WjLr5Ee+42H@-fe#w*vK-2QcfPuRm4D&OvNSI^3S9Uc<N*XhIpsLq`^RwM
z*>#g<e&0LZXy0KdUHX07u)T_OuCi^mj`cI%EUDb}tD?uz+g1a3r?yS)<lcYi*7maU
zgT@;vGF^^N2zCJv^wLX>OX?!c8B)EHaCVF0NCFU16&P=5`QK0A8#e=nV{^v>R4PCF
zoZf$L`gl-Ba^a%}^3^jaEg)Ms^ES)Y<?)?JT0wpG0i*|pDW*ZyI<<2Xw}n$fgtqA+
z56#nCZ>z|g9^RJ6yTUCZqc<htN3!3DS_z|lD^~KD<5@<T7rS_Y?%e6S_vomlhd9u*
zDi(XKQ<*zgBapnb%OfU#H??<iF4^UlSXQttjDlIvWM)?U8U!t#iA&GnCovn`$c$gn
z@&bKRTd%rEz1Po1&P)70S64yd77e63y`^kL;I|pjf;Hk(a*;MYqF+_iLId)5Xd1zP
zEqy(%gk&h+Nb_tB^u27L&}n~@$SSvIm7Dq6HQ0)BF;|}px!l0;W%|HLAYh&{?4Qcp
z*ciFZnHT|oUkiN8CLGGMQseG{(_kZz{w+Z7CFWiv1x&79H}Tb}f?hX3$g)lIVnAnZ
zJ%F6<{QU`@FI*U+hXKgP#97!dIk)_vXZ^#I(KLLM)RhZKR-jTj1I-2t%)F?Sq|<w!
z%5uXx3T`ALT2x)%gxc-UYv1ZG5E<5NO+FRDIlaAdZ8s^tCC!eEtmkhT*|LGO2g~@B
zw&>cxaVCHNT_V!~btfG6;c5`f6(~|iZb#5Sb^b&?+zgj!#BmhM#L$OgTYqLNj00*q
z@!(?Td7&62sQ3xWMjc^|TWi2(N(VE^Pht~K7^s^f+EeW!)I>JBc_+dHJ6rh|t14H}
z&R5t{SLji^VB6K%DhfLAh<%3?yJ;94`^US2yX+1(daEurS4!m)^yiI9IW{I2TO8t!
z!UiR)?%6$e$NPg3#hPRLi7hK3?-~NwRC*YqYZUVK8*TXc?R{Rqd0y?VmI{RIw}Z~B
zpR4ZeTQrZS2~3uy1q!%ozvpEJ#0Ijvd4o1&Fh@%3iR>~Im?;5lC#{X{XXg9En=9{q
zw(jVPEDmZIczHIjodI1j&aiyYx4f))mExe}QJ|eryG&0urrIrfeGL;qE7XVIoh^Eb
ze_9he3{9DYl_4_NM6G8uuR-tQm8||718!%)MQH+y9m>I@zI*c4vOUD1@PhtJ52bnw
zr(3=pj;Zuq9wJh&p(dhWE~F~Gr74n~U9Ci|Y<f^OGx7a|6FCh*tvbB(U+=|a$j6f8
z-boimyohGhG<Ig8g+eTI`gJ8_2G+ypan+K4G|G%0o%f8Imss%~nR4EKZ114}9d~Na
zw~kl}FcBF|;D-o`Rm0mjB9+|dyv4=eF3vNl)I^F9Hhr_OEb`pCY<77%+)*SEd=^)>
zp`L=BP~g0wgm}>qyvwdmJi#ou<*g$ddSqV$U}v4@10?OYe4jY|L5GQ)T-GNtk1|hR
zHOjMNW?gLZdA3Z%ha0+8rgq<#OncmGn3ula$PbXl5Yzo$VDXO?-FCatCqY_&Qqq$U
z6`I9jm>_z^K*g$*c_IC%{1bO)nw(@$pVi0Vl2PCl;_f1eh7DlxtOUwJedcdDRZv7X
zHAJ#o05gm?{7!Iin_bS0>k@HLB0_fl<Vy@bGWx|sSI=l_Mgzcinv_dzy#yx6J#5@1
z!*Bs-vc#seyRsOBZR{~Tg_?j(J(-{C>9BTpsc=jZwvEHV_+tr@1K_(0^TI<_Yd`OF
ztml$+{xUEKI;FSv;YoUubga(Gg&IaEr0xNPtQ7hAEf^6T^GDmq%E|~t$4_=c@dj0q
z)DYx&FLKe{zKpKq%V*(TACzgwxsXdY5Zk{tY&dVFrV<>#=v%Aubn8Ygqe5A7bnoXk
zu10jl8t-o-nADSYF#PAbiZapPw{~#159GsUAZp|s^MKgc%cpUPG$AJ={tBY2zRHvj
zkO5r#%enJ!Qz6_0XL*6W=%k8^#>Z<_>z2TGRIiFF0u%&gY;Xh8dYffbHw}=4tK;sw
zwt6-CA`?QxM%w$+-+%P@x!P!5GxA}@4xD4>ZFGS}0RDdSjmEn?sre@+eT-D&wsisg
z5z^oYfvlF9!!r;k6+(5%dc9FN!(My&%QZl0(?r`L(ynq$TssPd`cb$rH*QuGY<zUl
zhnZ-{N^t=93AwP=>*^wXQfR+AsLnnQYd*_Pt_E1|XuTN)lNbk*?f_u@REID9QPWc(
zbdqanV&vgfpa4z2dXx~@iTkK+l5@)qj)7u01|)OuI_;1>MFlL6Qd!F8$1Kbue$?d@
zx~5YC0_{UOUWJ)r#L>4WctB4AoT4Ycv!)9nai?iTd~H1n-Q&Z{vsGpwSz2>ypvR6>
zJ5`^72)&Gnvc#30vMSyTi?Vi#e%EE8=Q(jxY>V6jWhkHo#Oz^x`v*N(Rd=KMAI}Rj
zcjotoTKak1|JwdPb4As#pJD{jIE6oNOAB2-_cTGx&I^pz;+YFcop7=`%BJn=D^<0O
z*4c7ab`Pr=ghD$r^wakH`VT`<#}$RPo}{cq)iNogrb3aogiH~5v5M8fc|=d}w>=Lq
zX)@CN39xRET;0cE;HYJ#p#x2hvDeJRdLw{5*#-<*EIv>qG{F<l-?IqluqyI6_!3$?
zZRLe_3sG@hI87Se4Q8g-ry&2P!?>^PI_FMN-SXlC1n5jC{x!h|A2dGMMMW0PH0ov$
z81vKP7Sta-B~e%NVjs3qPr>Idkt;H(LOY4jGK);ZS7p~Fa^7V5;sMxq2)wXAY8jO}
zrqf8K@Kn4^k)_g#O<hM$o5%=&W+T~`Tx(bA&3ESkO4002T~Ez@r1j-pEm@}y2`%r`
z93!l;Ek)y)rkG>dLN{RR0NRp^GVhMZ&}rb!LYX^V-tq2zm+d7nV*lIed<zN45bG$2
zXMwO6a#J#^tpQTml|}F(p%Hq^bqtO}+UrU+1gWhFg3I0SZxMnl4L;wz!%_K8$<d0-
zrTz&hNcTnWS@4G-{iiyYj)Uqv_OK**F`R`ICQLl;B3seGweC<pyZi5Jjosw~)R#tp
z0<OJkd<-6<<RU5E6msjbnPsJQFDk$U9C->hKRFW)_B2ru5ysC}XT3J{20T}7C$DEb
zqnkePV=7&*eG=U6Zx@L8sjS;P!fzifjgmP?v|E}5%1L(#%tyBlimjGX0Wt+4%shVG
zJc69dDwc=u7#_Urzq^l+Cj;vP{97?HksY9t{Ao%pd31U~y@3}j8w?=oedkwZvH?oO
zpFl*Kh^ufM4WWs%A8VmEaD#cm2;6-tA<4^kNDf`;EzG>NxxPeRuC@Fc=1q)ES`Z>g
z)1b&BUq$|~gn9sWh)tD4C`M{80l+GcdCO=m!;L7aCi>iz`%XT9s3l3tNK?(fKtAl8
zfm60U(Psor-h+~%m&R%m!cND#{_Agn1fLuBrE^tnG~1UHmJ=NX9!jyq$7{A(fx9iB
z1gl|2J9?>&P0GPL+W@8KOtlnW+vm{NC?OLHu##xI|Je?l7qV+^mH@ZQc!U1GzxbZ$
z;?Hv#Z)>&F6d{IZntn~DLT??Pyc`!MDroE%-m1j}R;(9sHam#_lYd)DlcDoGb+I`;
z=K9yKU*!ppm0Y^ZEWr)lKm{`wMG`ZNtQRcRs|dsZl%8bGw0eCrpZH>^bJ+5?fcSRn
zCd1#qYM&)HQ$kdR4<Uo)Z=aOGswDAlfeV<dYN)x>po}#PM{CZw2HK`w{P4tC(v1TE
z_-ckXcnI_gDreH{ZX(}k3pfwtt6=8!6<d+ao&M1XY&%gG!DV=@_xe2S*=_>ikj)(q
zZf?cOl?fj8t_ZGU6AEkKes??0X?67TwW|9-%Il`J9wG&HTIYRQZq2i(vg4lTPe)2+
z<oze=zL-Pf7zquVfs4y}cCMc-txG@Cmf2EqJ^3Vv-0vSZdo$jo6n%uXYICE!yXqzE
zV$G+KHr8_+sBnu5uze^pxp75f;oy<|o!Y=hEzBn=veN~QW|11jXEn-Z$>zr-Jy*cH
zh2!wyOCdSo1fn`zI>d6gto;4I%gULoC@b$hPPTWO8lH!`Yiip25>#-1gVZ!}>|1*t
z<INzAMItyQe6|13b`vNK0>w852K8ThU#%Q^2Z>tm<+L_3LY=UrX}$nH!qZ*Hj(tuP
z&tgzfe86k_EXz(-6dE^|Qy(veLc2WNM+WSWsr1h>oOT`g!G-RrSK8jz?81VhiMk;L
zzQ@4)PVYs4`UxN+&EGr%#oT@9!`+bIl|6Ni8hAGJ<eT`R#SQf)%%`p^htRm(`Z7>t
zt&K=vub=0xK=l*><&_c`f-&87xM;tz=BC~ir=4fmIqms9{ed}%Xc1q9TT{lMy!VU$
z7jU@Nt-Mt_lMk4a7K*Pt8Vmra@1stOhyhfeCrZl86KxX9rKcA70Mi<ShXr8(^#+$&
z0wDFm?q!+qONv-ElYOASXzPTkBmpWf4ehbkK&jrnX^rqh(?$DB9OQ2FLOpmRzg=tj
z%NuCN<t1jPsy-iTsCh_ZF~kP>`1fT{l>U0pdRaPR_j|m9h=RGJzQc&3=DL(isD;>}
zBhloEug^9Bi4Zmk)6@&x=uVQHAjeI;?6DAmna&>h2MNKCB*RG17Ph;)IB%|flR^rs
zPqwSnF8f1Q4r#f34Y25WVQ~jMNLpWf|6m4m>(lFsS(iSB$-q$V$M{LXivqPm!;u|b
z=}XbC9EU=_^4;FkJKy;<<XrY9597GqcM3c?4BgN|H5R*lxO`YJ92^{Hp{wbV@B+db
z+U1!SE?A5A&X*byc9Ap*C`fK6>YxQ_;Hz3o&CV8&zOz0I-42MlB?Z8%Wp#r0r8d}4
zs!IM&d%XcDUXvV!#E9BVRx;37ekN_Moq4H8C=9i~v^GeQ_F4fpkgIG^z>Fl(8=cd0
zj%H9iU4V5C{pu5LbO#US@hDlz85Cq&-}sE`Y(#PX%!YGKhM8`eKxY?BjalGFciwj|
z8wwZRZS}%s@*>LPufJz~D_mJ+QH??U5wZ}>R*dkw>lenqAJfG`wD9HR!uyP7FkLF#
zBzF~%^ko!RPrdsAah5)NEL}*NIlvV5LzDzw=oMMwtrVpQ)s-vrvW9p3;##KCwz<Oc
zMF_E#3u!{H;I%F=+u%ARS3Wd?c_p<Q>5cvrSTNO7t5V7IrH3%~_>m)kNysBUs@)}Q
zrLzrEknU~$^#j>OzW44!lxza)w;~D1do&A5c`0RS=9%w$zIYA0jgJVSGVAx`r3-Dh
zzql6ruN4^9B!~Oqv0#t-7*{b9OjdlS-TIyE7!FQO>mkAkg)^$i%OCmE152s7LvF9u
zd)<8BvP8WX=C|K|WrS%Q(p_tR(Zc!^Y-skQ9tg}FTx`<{k0X}o<g=H~S`(k1WfSfj
zC+WAz(J#G_cyofw;F!=|xH@MMroGmwZFC=p{&a@q*?ruE3MLKASIUY%1AGozjt`)N
z`~Z8R(S6Tt0-;D6G`eLkK>3dRom?!gJ_f}4C#2a^WGC!1Y_tc_t4#IfhFZX_U$-76
z5WskIvEZ^jYEY(`=sqKAKy4%;<sA0H$NL;Xtx!YFc*kSIu8)1LFW`;c-scj^yFEKu
z5qwj>47fL*+;Q|4!Py47oo*?@HFB7wd2&A4s)h+}p>aSGfBRVysFjuRf(&q#kR2K-
zrF_QwBlg4P&)Q4LP>=TVy|{!0&i^66SPE^##8n;&$cx>Wu{riWPAD5e7AORPs}=PE
zPsRjY5!F#0H4yj@&0srbD8__LYoGJ9Uz_kOb!1;=B{$gAy>Ot5+ql=wd=RrN%MzPM
zR7oebjcprW2Oq15r_L&W`&cCq^0+dL4CaJEGu!}uDC4_+U4v_@8ew9jylOS%NY*Tu
z#6_Lab{E=8F-=h*M^YWT^zqULh=>}XzR3A}71@rb${uYmLzcIDD427s6$pu)r@Ga(
zW7-29*I~t|vlimC!#(S~55iJd603#we3UshD?q44d)WaAc}?}2;n+gwV_e+aW<eG|
z<=`&)+Mz1d32$8{LQYX^iu4TRyHHW^B%*UNq`SYlgDJ;bR~17IYvg8y+q4Q;G+@fZ
zje;NA0CYcnuU-vTIn<&{W^!r)Pr_k<c29KtJ9rO15xChFYd~}3Uo!x2@Arg2P~(}{
zbEsLsq?Ta%4SjR*`qzVAYpm&t_xCN_DUOj}-<Po>ucjIU)~`yAAYz5Ru$@`pjamTP
zFfpBnbh?69$&F$*lBOO^zMLVrzgW3?dZRP7@I+P&U&hyDCxTEGY);p1_JmVrTerLV
z2>vs@moWg&lryOS3S|yS47rjP=c@B8v+2kXt^<j>afqtZ#Q>Ur=h4b>RiS0H3lO|9
zoNdDyMBWVr-ZCKMWSEE(gqfV<>Y{DGp->)o1z2RV$gR%#fl8?n)IwI7Q4?>WerW=$
z><b<h*}Jt$hJm2Ybp0)iPJjoBjCfGn3A4HfXqbzJm^iP3?$O72pZBJn1F<C|zBH^v
zV1XTHH{2>0PhpTd&Z#~^;Dd|_0w`eij`Z=*r0o&3m(z4t46F|1;S0l2q+)fdyF{ik
zoz)l4a2@rby#iLd36dUSb4>2t>>|@{7<r+sQ0L20y;{psvznqeK)bSIkm?H;l5ejc
zIe_ec#l-41d?2I8a3;FeioF5Xw`KszkU<a{x<T_%G;^U#X2LQSibK6-{bxIYXL1Hs
zE;qnhcGG_HqWi}B>@yqFeDurTf#kP3n;&LelI1?IQg~CE*}eORw!PnR=9SYd-S{;i
z3t1HHklOAy_N)0gBe@1hw?$v|7!ZmeH)cqpY+w+tz)-3E^@n_ZEponxp-q`5J3IR*
zJnJZtybX^^z7-9o`no#UMdh3O9$E0A-qGM8Cxd(lRV4h($*e4fUs}n@7F@9T8aSdY
zFES-~kzWnW-g%T-6`%|n{g&jV(wf4>{CH#00}SF-CUY(oDQb2PLkvxrc}Zp27$py4
z0B-Q0y}qiyWD)OkLfxsvrbnQe%Q}49TSy)*>K++#uJ8GLjJP+L0|uQfZYG+rYj(>{
z@>=eCuYe4>=V-bE?$zukU0w;4xaLV+S7#H*Sf=PK0dax#ttCB`2{{ch^Y#cZ5i(hr
z2xle}<t(WN&szsvC{zI8mo`wL*mTPUZo5&$-tvLL`8mJ@4uRJQgEDd#1;TX{uYuSD
zKQz0#?i4mkr={LsSAcfvT+E4W{`xV`_aa!+Ezb03cdCn=%RIQN0oDmMo3&K3w{-=>
zT&C%iAbG=d8X_y+DN)^1p?MJ`Ft&?a=4+T`WtKJ=U@@E|+oG9ZeenU(q;hdu0n*WF
zgD)8j8$5uHqB$d6uE09_RW_X=m;oa9O(F{FTgC_H2JdWWo|)+V3FTaw<n(J==TjQp
zu;3<J|J6+`Z8K;Gk@nsCVHvYlS9S3n5|%<T@xuOZtv9#L#p31qK55Nzz1%hz(_I1p
zCmQI74G)Ywxo*t#Iq0fRyUx`R72V{PW>2rzw%yZ?jC7GrO~2K7$PojsU&#z~6&beL
zczY^n<6{fJ?B)2k0QKMhUq%<oM}Y}(I8nSjC4MMSLdd+9aq95hZRln%N?m@+{uF^V
ztFi}K+PCayhakcRqzrk}VYklv)(fLksooe1glP4|RU=po!FG4b@M;0&5{F^wMPz)d
zXJeo(5(UHVnzCn}Od9L$R!8`JDir!EW_|arP63s&ltHsS!+ELk6yMvk3y$xSh-8=;
zAF*Rk`$$(Z6Z|;weTRm(mu>KlaR8WQqP}(}6b4&J#zsEm1>w&nEtdz{y2q1F6CX2w
z_FCtCZxZim08>qXd*!rUz((xCV+AtnhXwIbFtz2HxOT6-Y{&1+dJGI99%vi7$)j@p
zK>3OugK7-Kvu02t^MP<a26iQsiX$c^(p)trh!!Ad*UDBf;_!rE5Gu0*b0xqB|1$nH
zjfIWPZ$Z|HjpA!$#Off#Jd$>S%HnKbWz`-zrzj5Dvl$V(nt_@ZZigsv%lYjac5jt?
zDEm;hB)97FX65D2t9?#05=#TG-uJBbE!})Ik+YCfl(dku)Z71w`O{?zKztz+Sr?>V
zlx|=Z68mT&kDsz(8Td6HBnaw1XrfjLu#Wag$0c0vcz4Y_XB`=Q_KSV^3f3{0j+%^h
z7am5M<qoCsYr{3=5+mU$E{8hC;cpdY99ZvN1T0u70v*~n@AY5sh9;GoZ~Zs>MDH$U
zy2S8MMgkV}w|C*={^%~BHUbp!!A)3z2j-Z(suBy)y5#po0Swc38@-RbFZEm?wTFi7
zTQR+XwrnXQA_Ew36UU<EuV^m!Aus8TQ^+Tw;g{JI4Q(PKO%F%9Lr#t;HXQbwdB(b#
zb@624)BDe5SmX#&7nWG$Y8<dTpTXfwHL*zmV5Z6dw^*8K0C5m4UTS?p$ho2+1?E`g
z<6~~t2Y^F4?qYqj9m+lWpErLYBkCNT;~iRHk__So$>Ys&^w4w9Pj<bxY(C2xy~`BM
z2D<Z2=p7ymnQ6M3LEK60S^>Tm>$>=So1B1MUrB^ZN+|sT{rSUUPl)#)?ymI$hMO(^
zz2YV;p+BRnbWL<RG0Vtc>-RGaHSxb>HD@U2dL@MY8T;a`wI$d`2DuFr<j^DZ4kykx
zdH?nvEmc{P@RZefuqeTK7Wu8mD;-Ll&l~!K;AFSNq%4uiFbVFFi_2A8Q|y)49Vq>e
z_kj0)ACl#xL6+sMm4Fu5ckuCOQ1keni0s@lMZj3DT1N?9Rxv9GU->{L32ctUaqUyP
zxqxqz>LCnhi*d;3FIV*0Jr{mS(*IaIeEWHz-^ja%=M$@{p6FjlSL=Jnx!X_L5_9_N
ztc>OGSDl>$Fnq^*e=P~DoQ?*5ac?zpHmY1-)Qdk)sxtr7GtqVUHWkD4DcHXW&bl8b
zf}v<z=g`)72zcfJ>!3$q(*B|cEMOw3@b^Gfa}O$!o?eSI+>&c}FIJ}tp*j?>{&Y0{
z*rkAx5pyIrLOfr`m3b?@moO`N@Wper)$&eSgkjPL^Pjb(d<{TZH)9ASxp!2*q|H+4
znRkTTRp*T65V#C;V>$5PTisZ>moTo=%nY*RGp$A0iIk4^)O_!7H=_D2L3jSQuZ^T6
z{nr>4Pu!|w^gSHOwsqE60@0noR=zeU;B}@uGJ|{g)%yu57OUI#t6}iBufx{Om^<Nv
ztes)go_tkCCfI!5a-18K0ZQvx!5L^K8<}qOv6i{9w<7{L0QTBxtsedFrmgE#y_V$^
z-t`>4{QeD%MrB`*#g(o*>6Z2Q&&m7;SO-n;?KUOm=GutKcJ3Y>7IP@44XCicfq%w!
z_X%VIG@Ae18?;{w(`f`FS953vi@$bohcbw$XzJI*!K{%F-Z|xZ*olCBSJX^`pmhB(
z#ulv;P}%&#?U$uGWo-)wQ@J6Zo(7;}@*;gtii8W9zd9S{eRZ+(?SL*R9c+1O#)6w7
zS3(cpLxob+=K3|gHE!|Dudo>*xs^FL*_W9@e9rq>8T4StSU5fmpWp=vo<Cy-)AFkw
zFMtZs4veU)lVNa^1isa?$L|z=TZ;Z4-$*#aqB90v4~{?A0|VEC(v`N%!eAU<Q(qR{
zC+QYq*J&S<79T9cvA^OJY!w3-y~uW4x6C;ce6yueyQ^SP%!j-X=9Bg;or}=HL;wz#
z)=dyOAtRaR{Of?j#tR97MZagqIW<{S$>|5yO-u9jN8+<gnNQ$Jv+N=0TXV<y4YeKG
z`8CX(SVbKfIZrUdA8q7JQWk`&vP%i(@fQjifXfCQFn@rArWZY>qYmtyg3W5;0h=)Q
zQv#xfMz<WjW^6uSAC6Pq$MqHf6X$~f`MgJ#nDmODsMFZlcwt)`+YUXn)<pm2_jE4u
z4oNv;uCy!TsVR_&0#7O=lNw~(0h1qiNBj8GS4H-r`v3$(<Xv+i^eY1CK%w|FpNIiB
zs{L^f>d4PoSjYUK<cN9<id4k^K2jNpOnv9aI$`3JL$;Fu9oE=@ZudB^OJiBvq5P5n
z^zeOR_s#Y0ejQtz?6eatwl0kb0n(3s2eKypY_g2sNYCXYp=K)!SmYsMcCQ%<D_P;r
zRvsYEMg!F(ASYb=v0hj3tj8;*!(N3FW8Ea*R3j+Aoo3{||Bm*T4rUj=A?M(yFMuS)
zFYmnZU@{-IT9N~|bI)nE3<8yxlp)r$UPq*?3YhD~!mvtLM0iyjnkLIG*^CVVwh^?E
zFjK$u#Tp<4#gbP)<!}Mcjy$VBPo<C`h!)&od$T|uaayDd;NMQmFCZKG@C9C<?IPH}
zTmnA6-P{W;0_u&8)oUL%G`)u*dPTxe_2IFPPeT~i7P^d{oEL-=u5=A7>V@?RNt;aI
zHx%35nC`Wxvle{v7<>Eu6#;nDq2uo7z&N-tY&a`BcOSJIG~W+e?T!Jq+XS*_uzBCd
zO2otUQ#WjvEcrl!gFj=1DBtN(X1u(9GdYmclwR)?lqJs{t{24}%jMWz42!qK-DFZu
zy)xnR+E_~!k$Zd}>l935wc@jNLHrL_m+_yVmAa%$yW;REm>w5e+Svw+hytzZl_Yjf
zH|{+Ts^XVDR}kqu0h^cr6pjBbgZK)LtwLR~o(DwijzT&WXpS_8x%BDvTW}A40%~&(
zK<uzEin8A_Uq7Cg(mht{5VkxsY#=cDb^7=j-5g^Txw;hD#){+N0lQcRiSX^#B<5b1
z+@{=#aNQ`v8Hc3-n1JX!y5}t`K?f-S8!ZbM8=S}6WMGg+eWce@Zs7qXE0If<hxOi?
zOn^;(f-i6&H3Kp~M+QtVJ*B>tftYS!g}h94T?i=@1;A#L$_4e-5XR`bin+Mo(`)Js
zgW~W$j0X>NLduGU;cibH^b&dsp>!q%Rx$wsh+PLFS4DN&rnzRmss<!dLn3czEkDPN
z7rDZP+Z%aD(qrW;xQo~>)DSK_?%KR*#2TS#FB`y!-$<5(_vM4}zHudaZlb+(*duXO
znU9*0O<?&<nG*RyNt@!AmT)iIS&}+&_mckK-^&h$#rZuGKoGM4v6qZQsrKoX8i``p
zz^yl(ChzTHa>;*9<9%71l#_hgR;ee7mnXp=D(-j*7e4<4O;6Tj25zzIQ&%veun_Dc
zRt0m{0}-t1_N#(#wA=`A>Bw?b)GsDMPkNki9-8T6s8d=$WRc+A(G`5{Xdmk}539|%
zD{me*FGn6C`4&xkFyoL#$hXFl{dfJPY1jwnzxW*u3?JJ0wGk@#DgM;H2Kp|YS^{7W
zFGny44RIidgJ-L*1=A9|?EDEO-syTO*>tsa3(W=@v&@Un;?_Xy%vILDq^yaNHtW)H
z9BE*M?1n7T&@LFVl%PSsc{2wfMMT1klR%r^LZT9ICP&ctQZ0ZP93}M`<op18h|j_L
zierEk2(Q4w2)CIAwj3X9Ov-Lugx{|O9d)a<t#$fci&}qTw3boMzvj|YHqzer&}Z|a
zZNLbTQz&{P;5@Eot%2OwD7~sPOes(vJYoGw84OEk`0tPyTxl^VbB+WaRs8kVLYtlo
z$g!_uEbzVeuPgk-^XX1{DYAqg+?i{>f7#r4$!dv&FpsaoISL$q1AJ++i#5}`E&Lhr
zIhI@huPd7GWbW50egvWwX?B&rEups2(eU4T-*xPGp8|deX6KN8z06BbjJxv*oG1i*
zV`nG&tJufG)RFqY<fT!G@0B_EWq?sq8L;?PP`ah!Q=s(^Fj5b(Q$Sj?WU;B)V=w#I
zRz@48fu|G+3%5?Hjfm_fNT^Uq5ci6&SZbZwdwOV!H9cf3>cP9)C!>#sGz1lx|Dc&A
zNPyHjn?rYSC-^)>sg*TBn_BK9M92(R>5{bJT{X3YpNdmfnlmE&RytefS3K3Oa;YYY
zr?QdNq!QG;&6x8cVM&WYiyfglRepAxH6+;-Gt`c-VJIrOvQXy--a<D>$~%#E7h(Di
z(;L7E3l3q*C^5CaIi3<^4h^I5FTgWH&1~N|5mH-g%7nr_M?TZ?v(zB_W-od8;Tt%q
zT<tr9(JUJh9Mtpbw=|tr3gG{8bMjbrV+<b{dnSmvD8f=E(}Ex#mJ#gTy^A^0yI6^?
zy%RtooHo@le7iv#&@Z2$XC>19G0cbyf47vq$=oFqE0Wv#yYruZdkbu_uvj^)^-j`t
z<$%!bX?fqpWUb{8J1(C4{6Yok+%5Bwom7bWxm`Z4toGAIT}=S|w+tbtL)aWU>PyNS
z0X#2xyNVwQqTtV`y9w2K<LCn;yO`G&nvAmYZ1|u|@_f^!yW7IEL$&(0_33_PA-kE2
z536jht?ag64tAs+iKIPPq7y>fG@5_8PljcK_~efzmJP1svO5_+fepPn04`Ef$yRc0
zZ$=rKy-SqP0|Eo%Vm@4Cg$5bI{hHoGrQZ_orKP3iuRpmM2^w2^P(DMrm&g<noIo)#
zm>+6Ome%5w0~uFpL82&6NH3cX`BZoQz$!F##9B<xr$NPy7TSF7Jt*effVIu1?*JOc
zL|0tjoYg{}o5(|*RWXNQe`FPhO!%@zTPt$E00)8{?e~BhJbs$ekzYNc(HGY$r2YF=
z;Tf5GFPSO1Og=bi;NIJC#IDKO<%gQ+PCW=E!2P#(`%x|QD>BmWPl!eFcQ~LLLPHu*
z#*6dPC$yywf*qzDEI-jFXg~BopAE2v`yj_BP&V;`!L`-?TP0(0)~8q5EE1$G^C<7#
z1FKch_|cAoOBUjopu1~c=|7u-XK0lo@kN<_1~XvHkKsRDcaJE#a{SWH`w&oKNywM|
zQt2`J?$m5F<UWOu^=9D`mNq2-y}M^kzg`f5m!%o5lzkz981-#HA3K|!`K%HW;ZOK0
zXF*VDv-pvOJM}4k5V)@?m({W7R{T^%Rv)9#e<`=VIvxmB7T@cx9%PRJsP!GgZ$zAr
zIH{S=sw$p68v!ekKhim6eC?Z~?+bsIt*4yhkxAF^YH5H6@@-+<>#^qhJe)WRw*6}|
z#~zCZtRvVMD(e61il@BPM73PDg%xvhr(Z~*gd$aHe-m7qv8YHKf>G^;Z|H99Dk<Q&
zRmg6~I%#oA1hOG7h+@|7U0TnKmO!<dduy@hGEy4omD>}p4_g7ncoeKI6!!OjOmSNe
z;o>qQ5O4}}9ZLv)9sE+HG4=kc;HOf*Gg@5nstV5KIwWIC_H8`5ED=Q*^RZc(Pi*y9
za9t-H8hc;qtF<Jc|G5VZ7H;<>Kq=%s29S5sVSL~SRaQ_Ag>Wu72<MgzlOQ)p-E=lQ
zM5on#cV!Gxoq%!mF1gqtC`ME}!`}3Ru_^mvf~eqO*(GQul;GXUdo75GYQpIHf&Ptt
zm%wCGDAr_!h^;id5K0XY2HHI&SN!|{k7kB3Mzq+vO9k@n;moje@{FBJP(!i7K&Ghd
z`RKwWBY85-eA)G<G3XU0#iJLf_Vx~6<rV%F=reM(EGxkSZBP8u#G8t6&uzM0Jnvhv
zWLDRMH|B16Pu7TZTwU{B;a?*opBWYlPeH;;{*$mMaS7|;ze!m8eIWjyg+2`Fvsi<W
z?~^VxU@$P_H}^P+IwM{Lc<a8hACfgI2)1Im$BR$@N<mrd-ch+PeMgi&tv$@$=)0o-
zO4idYWwJg>N|jus@9Tbi-9TxDO=fd?>Ah6j#VXzz;<Upf*zvQ3`;htsk%WOPPS+I<
zRGdf{>D?w_OA|rn+yE!BL;8u74IsH~7ji=kre96q`5)>EVo%i%LKneWIn?XK47{vU
zpV9>o5dmp87H7Zk!9@NZsOFRaElz_x{E$XM^CH+;@!y)%(Tr8E->qr<1HwB0n9Qou
zwF?{2|3{%pwi4)s<`{8qhV?UFI@MVl<~c-e1x|c9cKB1JX4??R9V38OLARs{v<<5k
zD;Q*r#98N3QkUf)wygq?^TV%lpl79ShDESMXyH(;hyS`Jy?Fvqtq0Jy`#i;LU9RB$
z`0(1y+1``+{hP$ap6pA?U4czNE1kfK;C<VZD3H2@cnS$c%=1J8FiS!8`#h&ls6Kn5
zqGC)<M_2RGrfBTDfrM*@*%xHoHZ~4KiHNX~zw4+syT~39Sw*B9M)?)jpTftf7R8Um
zh@R|XNW`T}79OlL9Lj>w2tSc;0U3SlgGKH(7|{}uuH2h&TD2-p2H~Rj!TrXs6T><A
zj?7=N?e4Inbi(+mZg(tLA_NoFaBSSb{F}qc-~%OycJIq;d@C+ztuKK}2XN?yt05<=
zfDwu~hCP*l>AazpJ-_>*Gv(P?q2b1k%jVUSlcV3lg4#R78Nw#@b<XbT*OPPn)-n_5
zeoL068U7fQ)<nU0IDip8@Zb@^UKs0d3*wT%$up?<NZj)WeK9P58wko#r|GQA{*3zd
zcec9qx#M`kQ@>q*`pSln1xidGI3AQtoV@)C=S1Q9NhV1vE$5IP<8RmkUiz>o!<FDQ
z{})`6!3W5%%^t9b0gn9ho+7pEi^p_<^axatv<#)@<WwvixF~3{>CXea&g-@D`zaN-
z=8<P|nG`ZC5?#TZDv6PvdFy9B*@*6|)o?zqeUXoC_r}Yz5#<J&{=47F@Ew#{B)lkw
z>2}NQ<W?3%$l<~Ju9TsgC?WtCZW{k}xVc#<GG(=dkS6@?%z;2!-uWLRgaPl$_%WER
z{~An+QU@Ka8a=^Ior=pL#_YJZ5b7vi?lLSd*%nUEkSnI;4%8D1xcipz2m`cz*>4%s
zcv^*}-MY?flcP~kV}pIO$Mb~aH{PNgqwI|N>06N;Bhqy>hw4gNQwi80k<0(!x7kVL
zjEB7={9a2tY?bG}aAH;Q|1}!$*7S8^pm(fX+iA7!7GODg*Q9KcihgjuE^VN<FY#6Z
zs*0dy4~l5KKX5<p4s&4uxR>Xxn3L6L7((qbAYEv*+FA^8DN4xt6RHxF_0k?9eF+3M
zOJRhLTfiLB@}cqw0b8kp_)YP1iHQ{|!ZT~99N+R5**NPuxRuEXH72fm=}+Lj#5R->
zu>E$AaQV*fxbuw@&^PWl{bTDJ!}0yah;Hk@0@rWmKp3J!?{v@3JEcJHbSN!=_@Rpl
z9e?Z=0NjPcbG!H-)2#vIGXmW}G+LiH2J9oI`hi!SS}>E1KB6v*3}^u#y$4f1cOL1~
zZ7c;nC4P+fT=i~bL+`dc6oib|!Ia_%^qP@SXFl}LEuRrXV!Sk$N2bhkUs}NS%myF(
zK;Q9pj^Qh@68cdLCar#B@PkIXd;6uHvTv>g1AVrRa{Z(uFcFKk>A^bKKR7W^!0l7G
z!JGiQX0uaG`gTf}64X}ECJ*m}mnTib2~SHzb=SW?6OM2ueBn&^|2Y$(=uDi~;U0Ry
zt4v6)h5J2&dg$1X>qV$z#JRZ64Y++{La03OL{3NDyZ5*)yQ)l$G6?7dy3Y3o$wc}F
z^9FCz>?BA`9VK3WUpdT~k>>$dQ}SdFkn}G?Y&X5@JXr06AfAU7tpI7t&s1%C#vhB4
z1<j@fShFZ-yn5FI?wC}3hT>6jR%Waby+Wq%Zq=u0Y`!0TDW#%wH~4(<@6JP`&xN&G
ze(?yaxA71R$4g)JqHrF;h}JEw35so<Bk4lOU|8GXd321RVL{NR=)aE*IS?C8qH{#D
zbB@9Rg`WRlo=(X<S<7aHO+A47vTJA2)md-gKXBX&s>U}D?uc9)U<{iXP7Q|(f!uaR
z05J%^^8@0^&gB}U#Rp1wG_tydA$JFr1rm<vU+=L%wkG6xKon3Yg{^B(jzzT~%pLhr
z^*F<9yd5?q6eu4Sz#qz`bQ<Mjeelvl{wEFB!H4X$di?|Wh2q^=FYZVT!Ut2sAKG9h
z7>5VIQp2|swN$V=P}7F2{=U)Y5L5x49x5eoiE#f<a~XV;CE6)rc2a@8?@my0J)De{
zXlblh>pi5!5ZZ5BN5Tx>ru!EPk23*e?g#ckFSK87J0A|BZ0{kUPU+uRHmAgOf#7%$
zMuj>KedRf71U=1DR~#XT2Tcna99DofQ+@oHWI3W@BMLZF+aH!0SswfT>6)0?lZT0Y
z!h|D(1(4ss0b!dft1sMvIyefBspq>I;6fc<>T?QrYuQa9&I0^P{$_JFcI?D~<2P#B
z0X2SeHc>X6-D%h~`xJl<;@-i2cl;g5pl_C4mpEy}Q(X1=;wiU3uOxyJA7ptE(C;?`
z1Aqe#JZL*OfKxA0w<$0jZnoWJ|KA0pGHnP(I_MCQ?;Iiubckv=<muQZYrPsD7_p6E
zhrY|=5_Tmk0%9uhXal6R>L4hQ!$@W7$UMy#gYm8eq)LF=Ep*mLVEcT-gTk8OgXgIL
zL1B)ORru-XAuvLZ2Bt}GY7%VK>$^GmOjz$KVgJ<QrCIz;K-l4Q(H;58!0Iz-T~`s%
z++5XL$yQ1l9o**U-YWJ*$8e@H)7PD85Rl<~O3a44akjX2{>C$k+{b&j3OXFZu=TIO
zLQrX#O6*QqJ=4H?f;Q)eQnmF`H;mrCmqLl5%6T_`E6lpUfpKDT_$W5l9w@<d!jxP+
z{-)UWJ3cAv?UKSyCo$yjlG0GM5fJOgsiNSEe#syUIxx=mS1NBT07%ya3sgL$$oa^m
z1CE2m0XuNWe_hb2!~gTChC^_MeqgRtd_4V76At7D!y~Y`Z5GdkVap4clyj*ids`J)
z&Raio`#JJFE)6hygw=HvyD!;s8I%?39=<1xHz{OVnY#HN>(FqYZ*lV5YMgfT;`yd@
ziCvfA9N(qP-;ltJQn3PCI$vfB0r&&Hcx$8L<Nm|^w>}YmkBHKj0aZgK<??(2!1Yh}
zpXuJZL1IvBmEJvhmC4oibs{<A@t}<JQcT-FOyE4=wofQ;zQLQU-Ha!xV-$B%+=oFI
z^bE`~eNx!{LUwZk0DJ$`ObP6%8-B$a4NI*LE646KnmtcOx$}{*oCG)|MIa+smIWS1
z@-%Ye02Lw$qv;ypkgr+mTL%ENLT(+n#0Mum_AOc~0^+}`F_%HG!aXJ*k{~^J;P@t?
z)~J!Kg3>$j0CuVF^cafbERL5VN54kh=!0T#2t%;J?4J!sfRTP`CprBIa<FJk)YXLM
z!W$UPDjwDuj^9SVI(|WDNm5}n*_w+_qEW2aaE8lZJ(;_hYaL_ib&@v9t(e840Y$bY
zNzzktO9NL=F|5<c_r5q<LKQ+;<f?u-vw5z+m;<-`E2{^d@Av#+!f6jn(wa&fzNW$S
za*;&wuTS;xFjg`4Xu=V{q5?#IbDw>6Id1+Y(slt>DA;iwjQ%pUvSW~`Cyq8Y)^dCj
ziBRJ0UqexQdH;2OhA3hs>4&sObqyv8nYoBdNds+ZpM|{rtQk4BpeUbx3XNKh5|msv
z_mTNeCft4=w{X?C3jM!^0$-JxsJ=ek!g}S!k!`q1{487;BF;m8tj~E1{p4C|-udlA
z&#4NH-{h(sodsrKve^JA&sdt$tpHZF)YHWDQl1sh@%WYZxBT@a=sD2p&EFt$X?a{=
z(R>)qz+k%7TS#~rb*xa=MML;`s0FOy7(`10$}W%B`oe?p^-mz1YRGsGN^(R;vjGP^
zL)_)b1-n9<{<6-a_s+#j550JB_R(jfD8|yC>HXdHG~ebqa77J6-jN|wz7m*HTwt%3
zc>Hl(ZL7lj>ysJcikZt8gqo77m!8>*D)!fM4HlqR%XW1j$6b2K!_|Qj2UQK##tjGZ
zpQb<%Jz+w%jj=GZSC_S!?T@-`uS5I9KPVQRy%*$o{h`Q!xCgXeP)%LrEuRvA)uWO|
z*WZghmF^4u2s!-g9-G5^q7rEp1wN1kUBZk%NYZK>f+jm(r*ZoQpxbU9D1Agd5Ssi(
z=-Lmlva;q(W@<bI#e;#(0*v*dVQNGtiv<WpXwt6i`(eZ|Z+&5@iD5f_{6pb7+E`F|
z%MFsaxQ|vmMdN-m&!{P4w_1@9*;dh3itimMaVN4~uZWYBe8<N(A+~>i<mdrz^lNu+
z75dj|B7*i`eI5YDzwwYixVL$Um2G5jmCJnBcaE)~ljel~s-2OQD6fUU0rwZZyD&gL
zu@;Q_;aX8?qIxDUR_@KW2khHV%x=H2!MO0y`0vXGba6ta<SC`bVb}11{2vKB5xXl_
zGT8w!EH1R};;kflKR$dH@q^wl4LXxwfkDv)2rSOb@=lx>sMz|jPJ^>0`%nD9`}2u6
z!mB~hnFj1WVMZV!qP5OFHZ_FQSC(Pjm0xSIW3GOwIYjge<=1WF*8!st{^Wx9I7AGV
z{z<?XVXqJMBt?_A$i7kM_8$IYsb=pY*S*q=zTHXloHqUNJ;D$$?|is&R5WDNf1M-T
zzYgxkRW(tF5%TY<njolZ1W;9TW~Zv5gsMhDfRfz5ri@?wbRA=DLCRrC;#*MhJA$y`
zDP13Q4o8u)(B$ndJU(1h;wuPIuz@FY&wZwp^MJvJGjupwFCcGc0q?0Tto~0reZlJk
z9QkIT&y6<m<b`5tVPcJlJ!qgGJFVqy0<DMR2YK3u#e*$~dJINKLMtS>9WM;uDkn)A
z08Kn0b;XZD5^K<5)n2p!+gcJ+$@ZM`8MWSwt`QVI+x`&fRU9MB2kDoqghrYU$WP&2
zfb453>+P;^yhi2e|MtYN2tdI9p3eaP*b{Q<J$xZVW)k;m@+Zf#2DRzuZ(oJ(d|Ih*
z-v0+enX(5jrPJ9*6h%MerA3w{flO5_^IeX%zYT5dh=1hkg-JN+AiJ{_3i!X5zy|vW
zsMYv|DSlt3`W}w1WQUaRvREooguqtZ<pa?aXEcD$g_{|;bQws7CPr)@Vc=JtB0#>L
zFZ&$tj2GQA6w!h|n0gl(AU-g>nF(i2%5rrww;b)7w803(zDaHYngp3lAaLG$uZF-L
zUO~&22<T>dEGekSw?w%Dr<uXVnF|;r7|kCHzSSaIq_wk-j=@bw;v5f|-;-JT6yc@C
z!&q<l9XMNO!*(r&UU)3Ww3WR6Qz5uM!VU3R|8<MDY6UIYU1-q=F6>P1Vqw%kr|Rw3
zU%;!(aM+aGN-$Oa7B&`EGZkL#Y!Z%z&)fhbFB@T!4Yqva8hk6zj2;^Yo)uzi+Z4O<
zAht}!kMpyx^S9g=I<-68>81_XlD#GuW~IRe&;a$~I|W|#L7Z{H%`a!<_nm-jpiIN7
zPv<ZPl&(UMq4s?;?0N;$&F4J;*lqBG$tJRO--~w}Dsn}DUef_Hs5x&tI^vW?05E1j
z&yh5axq8*rk@OqF=YIA?yL>1dX}oU$gtNepx%x-q`rBy%yxD^{L!XWpeDxoZ4mA%a
z;7C@8N}&D($07RfP^_iTff8f~0mV9(wgW`60*ZCqf7ZS$hm?JHNUWTB!j(}ke6%-y
z{B=E8Tp&h>J|ws(FwohgI7%9B3?`QCRbE~UK8KWI&id^k`^N)JFg*3-rC0A$4mD_}
zfKd<%Vu#}vhiLwM^ay!O#<*;LLqEuJ(g6*iq(r`zX!xud%NhYt`B7jPBIh9Q+h30C
z)0ta0q+cJP5oIu=47ycX&`fH;7juA`^t8jgGa(-4G`g&jYN=%Z+o`&bJ5}`m?Wywa
z;4|1ndp5yd<#%kbm)z3VK#BCQSU<1#qHH*z(`T6tO&G`SxtNSN-Hw$T$+<#|hj9(U
ziQi#ygr)S#!Y72TgXm3JPuvnlkNnK^PcJL0<S})0h}ddy+JClv*$5lE(x8sa6iNF}
z`BTU019Jx@$P(tzyN!**RkOn|8rs(oJ~>vTjC{CJ*?!ozw=fdYXSWSiALraqyM%z}
z%*CZsijRZv-JBYN?P!%c9e;ORIVS{74=H}1U<jeh%J$9pPi6$efssL~bC=iQA4gRy
z+>*8n&UL55=KdH_|2p68sLrBN<s23vH>JW|W!J4?8Z25jQ;~9c>-9-{!$sRgytxTb
z$QMSou?R`SLa10wRg=L5=mq3J!3Y?aAQZpFu>$uX3`Ng~@jmUk8@Z@d-he@xaznRK
z>hFLUnG>~bQ8`ftucP!qwG<k{ehx-9+69)aFM4XJk!t|0fpa8itUkxJKqk>VafgZt
zf0lRPkcQ*{^P_hczNg!75YT^EH_g-(tE}IzAtOyqAg&6hA~A^#|Gz#3A5C})5hMw8
z7h#+xx9rTcA0KyP<djwXT5g0y$c@nt<Kd8t;d|&{u#aHNf7QKbSd2bkO8)%RjqxpF
z5KA(S8SDnfVjZ1`m4^<eKXKR(*AeT2X)g@Ao4P-y8VQAQJQ8<n#~whEc(f;;TQdXO
zX!8B)<Kt-5@!|2Au)I779w!anKK--|v%V;_>J_$wvZF}}a0nEk-qbk7A?E_E?(vmI
zaHT}ce!=b`qW1m3EBACbO~R|t?;ivE@s~aA8fbA8%rXxQV|y=yyT<8ywu%@;%(T9h
z$`syV{r1Y+B&`+e52*n^gsj#-%l`*W(7i34u`mG|?Gex;-Nnc~5_2&JsY`M3%|~f`
zvw>w73xv`ltndqHr=v5IrWy*9TEvcA<bPVRTwQ;m^7KGU*{~}VUh;7C94UEo|74wQ
zw4&MtDvgr6C*_#n*|XX(j+kvdbU#(q!~Tr7#2oy|a0OBBe$2u?B}*~+23Em&fg3qr
zfL}LG>D16CTK!=w`uem0!7i5m=73m?i^-d@-N%DM9ljo!-KRF>1B2N{s1(BTf{{7u
z*t<v87=D5?&*GOslZ@8Y(Rdu`sBm*~gZY1`<=gOs4^T4z(tmT1JAyok{?{B(V;W%e
ze+-KFD<mensV~L~ScFupOk}@Nprt0oa8P@f04ikBmMjA@d@O&^T4(qCou6pDh4ujf
z-wDZQ74+cmf%>lE#~X$hp6fTkcGh_>cX6_<5p2hUCAr3DzYK+P_wfUB3k|d=SD|n*
zv5tdfH-2Vt<kNs@KxsRy9!b28Z^Qo;ZK#8+%LVhZ6LnPhOrs5vAYGzo1NZ+30AiY;
zQwz-k88E#WSt03~q37>WpPb*H!KbeD*Sk}#FN7X{s|WS~Cv-6T*L!nb6dk5^YWpF9
zjLzz*wRN&hW0nvz(P!@LTL@JW0y!mn<pOMhBz8#bGRJ@7wH9XF#Q;qXVO~BKp`~dL
zwKbC6b<ptcw|lh;Hpz|9Qr0pB_T%B_J=$;Y`Yj<sE=S`wTj@wT(baR`O-s#~NY<gB
zwZaZ2svq#q*k)q=CpM|>M-X<0-+w!B!Yu7mR0#5qkW*;1*eG!YJ_r@osbQ_un`k>e
zDFA}hf)TpL{JM<)>o34@-}E@7En!e%#_HxPEW(^q5sc}c9cIsqS3~txhauuI)kzcX
z(cEhce0NKOj24(l90qlSV7BDOvf1k|Pzs&b%TR`IZ2`4F)TtZG=P7HRcY-qx*bcM9
zTZ@9r*bECHWlM!^QF{m{=q$H(u1#;P1)3$=G$1oMG$S91>zM+Fj~H5`TOY7``M{J!
zCIqm!=J>OwWn^yvtjb3422m~-dTdGy>axesSG>D=_hcU1erOMhi==?MapmZNg0T)o
zGy_Fm&UUXBJ{8858O{PcmP^t7mFeWG>+{R@U)d`3>`LpgU0_geqaPY1^XF;%6k;+k
z#xR7jZJqDct|#eG<euq*lSb7#f4~YDbE#d<9N&nQ6JDN|T)zJD*WHfq>h%nobu#sg
zXQ7Z>g0`z+)#<eVpL%G!lEWrK!X4_Vd$ET#NT>bjtPTvqP<3qTKy$K<Q<j`ux{eT+
zr0WjhJvAy*Xkfkm^jw8aw}vmuFyg6O83qO(NORFwz}wQN5BWvZlX02}zwEac0KAZA
z^X03~W*J#hsMKsyjDLC-k4QWqIzG^M!p03K-hoGIfQyL3P$6Li*;+qqYdsE^B|pyQ
z_h2{h_>O=CBMSCOb@!%UY|G1^(AL&)A1s97t`+FN!2+C1NZVvmb6(br^a1j`L;3(+
zk1m57jUf6Pg$XggZOy~C%<jqv{)EDZBsU~O969X*v-z1VSbFZI+#7#9tQJQ6@@BVP
z6m2(76vLB!BUX2`(`(KvUia#pBy-Nj`43g);7yYkenw{e&tqh9#OGFgYU=0nzZiS(
zc&^*_e>_6=NT^7Z5f#~Fla;-atw<RmMP?Zp$tHV8HkrvL3E4Yalvyf7=I=P)T-WEm
zf8WRB`}zKHU-uu^eP6w=*ZF!K=W#rb=SY_#K7<^unh^f2f_=O?<0WPG3!5;&#C>1;
zu!z`7K*Cj`jcupsUV2w_*7{{gR3-;g>^q}#!j*2%MYz+pKWJ2wpCb*Uctr)5GhBv6
z$w$F~;=h|D&p`O>7-yEpM1V~zu<&K{#UZlInZ!)d1Rohr=)D6NQ8)ab%dqc463Ap^
zn&@sn=IG(U{PWC|VG5kNJE2Qs{sg9)77=NYrX&e?(8Q@85=?qPe0`X5Z+k$Q4W?Ya
z`;|Q#5Ibou6xyTA4*Ne6-ApN5G=qTZgtL2WrvsdjXS<S(p<y>0!H$LJ+yaFT`@@O4
zhN7?F9|L9nWtt%rT4tmsfK-(e4wjoBUTIerY#>4LaYII~`9KrK1CnD<4FD0o<i5?P
zuU|=l`xn(?X(z_W2*4&jLFtsz_9?3Yl8*12Yf#zVmlNouu9lj2obGWZOFu*{uw_<t
zj0q9lC>^rg(!8jVBiQoj+pfH)aFK`<EsVk(rYxb$JF_;z_2PEm=Qp+hYk(tB0OW{@
znR9h>{rkvU@I6;*w^dM<!pv3<P(#vgJGj(Pb1d+wAf+O_GUk*uczUoBo?pLH!3%-`
zq}o>>hid++3&|6#3x<XUmoArhDOD0Mu_nI^Td|4ix{**vs{EGHk2Fh*y^vU(^^01%
z#OA~IsRedNvG4`u;p9mho2dT%5r?n9H8gUm%h9X*;R{`R_FVSCwSNHWR-gX{s4=~u
z_mU=cxwq5ko6Ms!?TpKhK&Xu=w{{7z8T%O4+|IZpb+`Ncw<%(|>L%ZMh%wB#*OOM>
zIa+4LT1mkyh{zsjeud39K>_zCRM#-_6-bf*`3cCD^1!&ODUZx-?bB-rfbvZ1&(%={
z(F{S)Oj!uDNs1x#qU|mn^U@4Kh(pI;MGfzSIsW-{YJIqN+lSH20Ia|oppy2oTuZ54
zQ_NxiMxOA9M-?u2qyZ%@2p_enQie*w&S%+AekT4kQY9l&Bp?wyZG%tOf!YkE*qrZ_
zv*>Fu#baB(Aq{^^W+fY~ImI@oj(R^Qyl~;{xKp~qQ{aN#6IIfKzJjx6EgVEvR5NO~
zd?gUAj>3BnFMNw0oGwu25?%6OB8=`@*Q7qCougs@9d=^}E7@kx1<>kv({*oKXp5hm
zd|3S<>nU*f{@y`G@-z4E1s*%x6M70Vy#c>M=#@*{U?ydg9Ri8VOeJ#KVq5IL0=jEw
zAZJ=TAS*_!qj<yCgN^<a=dTt1;uQs_N7+f2Lixhzq$gVz&eJ|Lc^$2HL+fK7o+CX!
zF>>L>6QNK^vd%M(x8AvN=Z${QxMJ(|P~{g63{*Zarl)zus{9|g8Z+wRaQGu^DC}F9
zBuF>T9J1{nXLj%70{-})dyW5mZN}bYdZX7S^5546E!pKBH20rclA8MouMM4|h)lGH
zjkWr%BUY!ugT3f)lV#ryY|=9T4{I<l$up)OAlb!YJesfe>nK$Kw7J})xOc(Moj#V=
zBK$j1CVNZD7wkdK)WrvIHu!&W2TE_d!m2%}<0?Q#Zym&+?zsUVY?%k1JJyr6<CfgN
z*?GBOG><o|Ys)!DxOHd9ZsDKxP>#0QG<e#SeJ=e8<b^ia-G(;OUzx{a7{`yi0jA{`
zL}5_LREN|e19XlGX#VZc(lUiw#2+6`gto$WZuIzfER>n@-hLg@jkC7W2in*SU^%7I
zn*2k#IfBi_l9J}AW9aa5d7?(x8bZRpA!mm+&_so$b->RiQ5UJ<AY`ifp^r*}VsC0M
zZ1PMD`Ya8nDy_gHI;|Fz%cmE~trwU=AF=2rCBKu5+S}5(8uF;tIf(G@SDykJ#aa=e
zgvY^h(3?A;-uw*LnSZ;0LkQC21=2@S*svI!pA98&qFYov?9Dk<&r|$&b8felVFk>X
z<l;AHnwDT+b>%kct;|HigT~9l6z=V-zTDq};lG4&{e=WSQQGz}{sipfHTiutE>F0;
z#AJ>j%Nk%l^7<H(Ah%~2V3^GkZPHNsxb)yZ9FgK5uj6gPLGe3w1JIePD4nC@wa+1V
zAJtfZH~z?N(tX1xPqV>aBjb9j=h3;k`Pk4<23Q-PSdz{Jab=4PjiK3aey~<bD)6yy
zXa-c<F!a{F-QpXRV289r;E@Ta@jlHrw&x4l@5maKfj`HqgYgqE(xahzjV?xA(hZkN
zX&9WU!-Y>ijcPDGKBnN^x6`?;ea=f`K7?8VT48B&^>D^(=Q1<ZaFDb8)_$=3KE2Ox
z&wq4jSojr^DC-?GA#@tNoE73V<|jp69<^AFA)?5gXhZX=y*^hlc7_;e|KoGv52wBD
zjaHV+-~{(X-V=XR@<gx9l$J&@aa_646M2!Cf*D}oRoNpxS6n!2v>{V(Q0-5^A}XYz
z(EqUeaPj{r^nOt2d!`d6jL7ZfIk`r*6aQH=@euy^nknJQ!fD21Lz$PXkNv8?rFsk~
z*2CGEhkCv#e87l;P1c@K4*GPe9=%Q4$@Ae=6%Hb7L3>l~@EbXUf|Aji|MIi#?a6l!
z@4{AIHsBAIG}7Kd2jH3ZH&==TMPuobc4loh!rNfOvoxd)jgA$_NmjU?1u%xYk~K(C
z0-i`d9;gXH9wo_DE48DWw{tB7qZi-{B8)Gah${rO1N3RFmf+bN>~G^={#tmYS`!7`
z<+irK9k82b{3wW*&vwE9e3DG~yFi}P1DyikG8KXAR>hwDh~^1y^H<c?b_xNhccG8@
zgyp+EI&J^9;&H48;a-+nZ~?2rzg<~95zb+YUgTr{^`%Z5GzMrihC68Oss(^QX+x?F
z3Npu%-T&yu>lO|nL%Yu<$&*cmLzaa#f8u&LfQ*kUR?{mwON3HGIe^=c9JW`;+SV52
z*XNKx8?>*TA1d-#m83Osb!VrlhfEB{=f|kYHmr3z;Mx5<iJ`v%=ncnYpbKmjU(E`3
z79MtN$w>;ZO?~tQpB8g>7_d&QDGr5MzqWH%(=GLOntYSl2gn|^=(|(=m{$a<{8D3<
zN$_>nJfiW2z!+~~gI|lJo<yA0YJj*{Uo8a84}i)Km-Km1bPOQty2{A!`91aZO~!7u
zI<rdHm`n6*UIwY-1Au<MT02@zH`Cfi0*KSd%gn^og0ulX)Lxod!&*OYeIF^~m-L;Q
zDRpaQj&1>@!}7+$>jWUXDS+7$$tWZ(q0qH7<HsaSL6SnqT;1gQ`dfIGi1Jb1Qv_Nb
zIS|B|kML^3&pEN9_>%n&ofTEon=>cUQa;Y(rAhSedCzgE(d=x)rl18};cR#=enN^t
zFbc7DETl##NNz$JRWw3DJ}8S@P_`tC*P%ce<j8Na`#8E>0Bx}kVTM3TK`<<P;Qso{
z)41!A7jF4;Aob6Mr^uEzUteIfEcphI^W>3+@ZEgci$F)ur|6<bn?S3<Jc5|ON%-xp
zU|O010a3rfX!{drtvjqffZ_PbA?5O;7tCVnpGVwHpX}dniOZMcUVfi_v@l2g+Sb0U
zLu>5$cukI9(7rYKOXv+1Z0%EG_L%?i7C=4E#nAKH;cfU|l)RYl+HW|KSohZZq+Ym(
z9p_GpVs^_CZm`)pDZcutvb*NpfketNxE`n##6{RkL~Zv+%U=R3Q$U6l2-!UR6<~+b
zd;?hV1}MI`!~8P1(VsFtzYE<}o_VZ@bMd3^BLn6{9I(!6CKtvZwu0oxt)D!C7S`;<
zx2W{G`M2&2=-bxtRX7I}+jRZ~g`sklm~}5_IUdgWwWXY`m)cgA(C=2m67Ys*&R?s7
zce0OWLiwOYG`>w$kI#2@<Oy{By^u(?l$!kw1cS^C*r&kw3W?UzT?|q*WCXO_&I@MO
z%RsX@3(C_-P<i0#xaKUwT%6=Z>1_hQbh8QmqgXsoKscW7N&lwCq9W0tV4x!CEey+)
zBX7|^&v^RaDNBstL~$6h102Uo+Zo5F2!NzX6yHOTM}o+w%nfo<__wun4DT2e;u_1%
zmxb+#YZa1tN0gTtDK4#tH`_tCc1$260So)13X?)I+kRNn895%i(h@)3EJLj=$69%N
zGr6V)-C`$i32S_u{d}_E#;v!a6~Wy9|N1S#Fo-yu>%`A~oW(4dnmG@<uFSd9Qlw&u
zf{h1Y$`FCzZZ98b9#wURNsHTjMR9@GINaxPcs)z_0?n`e%GItjEtTElY?AE_P+hu2
zJi)|95wgy#gHe$~-ie<}-vBKQWp{KKm0Ron1=_skE}`oM^&`7p+qLf7Oi<4FpAI&{
z&hwfNU7mu#^t^e()Cu--G^qhlq8W5~O73G3T)GlH<T`3$J`%k_V&Ib`nR0LE32I`8
zKcB{-0&J57w7lH;h$w0D-~}@AhJ&^(?@>ZTH+bPFg3jBe5Z;82o?DIgi7`ON&fpYz
zzuXSj{g&bW69CNF%_tM!Lgi3|+$Yg0wqY2=q`lguWTC0n#XlqM104xaSj??$v~QiG
zI#1|NB{YZ?eAv`-v2??_yJtyP8D{%LlxQa_wrRXIrsnhK$?m$2{s66QHDjd>i3+<^
z){wYG2J{4qbx(GCEoI$taDwT?FRQ8=vBo>nzep*oQa(Mz#!(}z6|wwz@8fvzPuRg<
zL#wVd-?)GNxc|9L5bLAj$N*L}(5;V9(A0v!6vKmAToXEbNs4q%H}sIad-L!-Z&dbl
z+|3|yy-v;LH=ks@hKK3feyIW%W(OD;QX6x}xsGj|>XU9Js9x(;MERTX-q1Q3*i=m8
zf&Bn-vjV`P-TWWwDZjqmG(75Pn0vd-Nm*M=?yo`f^(p`Dge4!QV!?8rKb~4EUBa~J
zHJoxlFlT=0A(0_p7oK~&vM6NIBjbpfPx+p*XsZu&hz#~?6!&uLU=ad+?&oK;TKA-I
zPv^1qyIs3Vnwt#MNef7v^H^p+y@B56?PraPS>4T`<h_Xc_0c?Fqaati1)wqw^B3hH
z2KCJcKd;vTDp|wL@x{Z#m;kxVf^`wA!1$vGV?T+1*q$?|^FbYyHYCKZm)VsF?((-n
zdnN$~_RC05d+Jj64L~>Zcx3bj|KALg2CvtbL)AU~;YGe~Z*O03wmo1q23>amA!``7
z+!W%QZi8&4T=XN)T#M%`+sd?OB0U9*@0?|t8n6;7JUs5=!^O;_Dar8fksjU|p;&OT
zW=Hf+kV>{N5{51`Y98Lqs)_Vr$Gk%imqu{K=)U*|OIiK6K;lzz{>ndI0Irpv(?OkI
zM5~4)dREj$3n5St+!T^h#-lu3ItRlw=7#dvKrFe51T)@qgFdI2qibh?$`L}7|4MEO
zXtm+s!}oD;8q{EH@KwPSV={h%GRpw4W*DkD)rkLM@Y-htl-3SN?}0{UsM;dH6+||y
z@qYns=4!&VZhq*4y02kg)3SX?6Tslb%8aWZJy4~1VT0!apK{P2WYn28@5Q$;q#u1*
z-<&TaZm1kgf!>r6tosbBtx_xq=5^{r`cE<hJ6W_ro^w$v8@Tl%<|ay3t+UKEeogZB
zmcrR~<zz9@0cDM_wXct85xLF9?aAug+DMP>466~o(@T#4+@O<OUdow9i40#bYYU(?
zs3ePBhR5|1QD^Nzc#CAK&yI=ju4&GkEEBLEzOwvFkIP|EtH(;*$^`+iv3HY-;@Yc?
zOf9y-l{%&zgzcZeOHIayPCl0P=UkX(uhkd5zwVRI+l4=+`{@_5F^7z)>-JWbx0?yj
z!Drp%A)&W#m<)7dJoTP`4G;c(6BxMR6}Z;SHj)n&oBzSq3%-L=%{X}X8XjT-dQS-w
zAMBQr{&@v7n?a9SMEiWh)x0lhW6mERvpZqfg<<M-a;#1WT+lVKw+v5;4z8UL@&r~M
z$^vgcx@Ox5_*Hu`ALY&!u&XLgJ9o@NH0B-bDlj0WAy|T_YF#=9V!79swZ4^)+C(5`
zQeH1-JpI*2^2je7ur*lJU5C7-lnWqGJF(kn$)r#>1`z2@ea+o)ON3>P!b&?*ioB71
z%|!T$5?}~M@<pdE^y8&Y{gv+$%rgX5j~-~kt>Pa27%VWfit39;f(8_oPz+yMsNXl(
zRv?5NAQybyvqYwPbIhDy<sRRFi&$2`#l;&MkP6sSe1+Nr0=3%A*d9dE^B2ZuAGQM>
zUHptjYV|pp0RI3y%U-JiuuY0xPvv$|#z;6p{nN3CX;*E=?nQa?u-j!yxw|B04W*)y
zIxFTV+bYhVGa?rLg3EiW$(REbF1UQ2tTZoDffF63(@Ut))mM3%V2O;r%2*Gv{l2*f
zWTpZJxpx&#K*99Os&|1FpM;dXHz_II%MUPfo}>)+_fSt>&)JP&$Q~Xc_&9Y{5qk5G
zH^B45^!1?c)w^iZ3SaA{-{G*8{O}_+n-nQO-HF5dhvfhPZ@5OgPE7kJ?L~sJYtL6@
zTg<C8B?hzB51!x~WO+odWVZIV;P?$E)}pQ=0oO@dl|0%Jw9*1-+9=>?$QfZ4Tq5cd
z<E6*tr$_J>*9F+vTFH$z=aN@xxz=*CwjBCP!wF>oVi>i)0XgUKH>Fd)z;qdfs`PEE
z!-mU~<<`^njMu<tYJm(?>-$<bQZgra7j$&s<Gq!;0x1W?XRh{}6X~ecmRSu^yGz(8
zvj?kzM`+)p&7ZeSXTluB&YkgFm^F}%WfQitU))3P6|<m-AYmg!FpD!_G5O=oyl6$i
z<ueV2SVqDwCi2gYcvhVIiL4T1xJ<uTDiPxyd>!%iXf;X`3Q(T~X(^(5rz9H9z+2j$
zM-;72ajZtRjHPU#(0J1F+zPj+=v)(RQbNTIt#-!2>~)7b>D%}H1)E&x=g(r>9G^-G
zi+}X%V7*B0pWq?px(t3But)A<;Vw*bLr5eH0q4?|DqCCwl@+P(&5xAjB4#YUMDjH9
z5kuX-h@5uOUqI+S=MBlSf3pK-6GodzKtRBFRHw%m%uU;1X9I{XbMyL}^;541m*B(u
zyg3Ib2KQ}wMLH=E#~W4O1laD*nm(BHqHg&Y9lsel^$dba;WT}__uv+~wp|D2^?|^c
zYBV1zH0FdN#oKfV`{TySq%oM<o&IwYLv}OMhC1W)@Z0}RM+X${d%rC9+4DiV^VY*J
z#9a8l%Onel9=x}%P-7}T@{c6x(#XbF;4d}K^JISP#zFYB1%yrIYkVMBV@{k(H>SIf
zx<R^v{IVV>8OZXWEH1EPuW~`W?BB)!C31cRT2|@j(iBeC1s$#Y1(c}_ddXf(!q_32
zgdY_%r26AXRqbZTX99$#%K80`yagBoqVFa-H(L_MlkI8b@%NL+vR9b6XRSkvq6pya
z{U>Y8WuWl;Y;8d$wl#nL-S_fQ>yAec3v7BF&LsjxYHjKJNC;mU++sUh`2+O1)i3X7
zoy_Ab@yYr4?GhIBf}OH{0{9XV1a2T?ZT{5IeTaZ>`gqv<fk{TY#cth2VDl=M+53mG
zHF#R7H4?of{FUlHKr*OX5$332DTojcNeEE!%U=7v63;H5N&6|Y*NYuQ6ZFsvT%b1(
z<2tyNCke9bHVq$-GsC`(K=_1v6byU3>J-@GnPB0of!eWA@_xqwZ5;*3zcB+(r1Gm)
z(<uHAmK3_mzVkkWbjY+`B`MH=SXE|j-TBtfl3TeCse177RdAO7LCe&Lw?U5Ff8zV~
zs7uJTsz{WKWNr$Gfl)AUkLLW_DDx9yiU|W!M0Ne|il~3?qX<v9j&6ai3$l>CZGAOU
z=U^c6@J8?+K2LK2Z0F3e;V-_zH6sLpbV9wG#m@(}z`N#@@w=iC-a^hck-g0s{)ZLi
zgR;nIVGNkWHzH7Y+f|1}vl3hyq&l|%4njxX7qSMfd)|lmzV-d}C7y;f{BpcGLg+7>
zea}$EzhR-WsendRxevL{0SN{i%6g)GBNg=609Hd9q8`ply*{+!-U@%o=MzWf9<~H}
zlDGVdJmM(N6RqVE0}OG99t@vOFG=AXHUuxW@>gpw${J=$`TXr0U}$(H;@0JWdX0xi
znHcb!u&{QIkHIC22Av02PWv9POLcuDkhTKk*@|3aKtXJkCKHWpq;Dg48$W&N2hHM1
z>rV3F^0^ohK8r-Ut?g1CQl9Ay#TNm+1rOdSIZdg-#Lfvk%-QJJb1Il8>aT;1IKzx8
zDYQF%*J0?10PQ`%8rr><i=VgDq)6E(QoR0d?Qo!2gZKw^<#7F(Q&Ra1&1Ir-{Qg46
zJRgL$=1bEL!hQ`YdGk3m7#~x4B8=kxN6Tq&W8hBu_&N)pdtUJE{ZaE0@EaH<uX=Ne
z4fDDJEJ!bU47i{3P2^ElhrR01>k-3`Fra**Svk0hjIl95zVvb@F8}feU(q=?>x#Yu
zkWmmukknNr)30F;x2#)AI08u?yOdGRL*!-k`k+4l%%MdV8YPA$a8Ks38R%QZs~M~6
zLU?R{o^IdBB^j~ElMUcacM-;HrJi!2?*9h%$;fh;TPky;ofNP^XW{Qg0J;>rr!_wf
zb%zg*zi{A^QT4Dnk+u`)iSB?2OF8@QRy%O`nZehG-!<!Rcl2j)(Y|dJeS<MF8AjTG
zKTBF8Kx~fWwB5g+Z4=(xp<$$eP<kqkqJZ~+;N~?;){*-^#tz%r=b1QlMGSL3KWVVW
zH(w1!0~}k7JAo|Jp=ftd7yLv%DBye3ZdrrZXT=p>g~PwHqn{L4trASHYX9~XZ#DFX
zH~25nU#{B^*C1yKlFo(!-R(Nyv;Pk3o)wJUpiJ)}*$QK!VjsHu%wktuMB{0X4bAnI
zRKO+x^olXRz<&DyfMqe?dp~uGbcgqBNSCevX-YqDKROYX9!OosUnSNB+h>pOpR@BD
z-gyy@E<KdLPm{0~TPVh<d>Q4~M?U96OVc&oN5JFHH~)<}RlDGNsLhwyPm?PI%<d-0
z?{u@=hZ`lm!=L(>3`73<y9w72bEy`_xY9GrM2yG&b=><BfT~1^Gn-@PG1@;MoqQ<9
zKej-Hxw8{kfmZJPy6AUH<rV%}-NW*L*h>7?!zQOen|)#Qt*yYP!9bh9qFS{`J^iHz
zNH~aM3V9czAA&E)xh<;L{_rQ`cMzkF^8VA!pVKG492=1^UhbiB%Tw?<xn{)=D(>>Q
zjveGS11c@K0M0>R62k7b!!b?>LfLz4e;$1eJDFC#Ap=t@tIJsx&j?L$&eHt&^rm4|
z9ogf9BhA2_Yk%SUZUa%~9jjcjKyPD@Vm?6~U;``}Ea*5vk_oNB{flJuA~zvE*0uyD
zlkt;<9Vg#=AznAR4{N2P%lG-`k`y>NUqU_h_Kl3^&mC$@cCMMAy-Xv?U^=!Bdv7$W
zDt)eyb?qTDu4E5pQ4~;JqSUO5{aiTsda}+?YthqA!6m`OiOwYDs3uO?mLFwjGb<cb
zGcU#gMkjWzUiAD~W$IMSjV&=o(S7%sOFq@k@}+#wEoVy03ygkKzZSQyC0pe9R}!}G
z1JEc-Np*u$zdY~rPL#e4SJG2FV$T^wr$Z<%ti5TmN#J^-a1YMz`YlC3<?BM9Beq_7
z(-c4Htb2jWdlCVhO66Zw_Fu2R0|}bu@Bl)xJ^>0yQ_NvP%v_7Ivf3aUkosQZfo|?5
zw7Mpq)2V=6(8m!=JtsUIdD#1YxUmAp(RCsO^Ujf<#Q`8Pu0yd%m=nm+w0{y-i>6fy
z52}m^G`yWWK`-S7eTo>n-~}36SQqg`(Tk~4KGvLH#=sKoH>~FpY}Y2o0OP-3I#8I6
z6CS{TRC=DbTP7h)ZssBzDAcYj!Mw}~fB@g8Y3_#jds@#J`e|?<(QP$G^Obva0htv=
zJpcL{Q?GLejgJVS7x~?CKJAxoM1aaXjXN+3Zt>CnMgBdtYi7z@8=4=83Tpo7<$iWz
zyuMq9Dq9tGX@2dSj3QmltiNiKK5!%K(@4hiSw(&Mz_LW7+qMWa5%S%(yZ||z@LGmH
z(8#<9Ry2;$;0iKnAQ5#0@k}#IHh6s4R`fW6uX`kz{Ns#$*cpPW$dr_+84Kt;fxqnL
z>9HRuOYA2ZJ5BpM5S7I{(1IYgcnp%pS0t4LTZ8e=@1O6<^p$U|3QW3{xIvbOc})PW
z(W4$2^Vq<LID<B#=em@3LlOj?_u_hmR>EAb#=FT5y7O!OVcDB(PMHe29_E?`gY1+5
zYzKdj22vaxY7-?Q_NF@(P`7SsiPXT(E{u2GAine&-CFECVo@7t)l1^y0iEMMZ5MJZ
zLH9h$S6OrShK22O!uNiPFe*iWt@+8@KL+`c(d7<+2sa{P>p;wS2lNc)x;!DGe_{e5
zrP#cLdd+zmh<Eq-Vx8T6H<9%aA&0MG>a|5yr5F>D8Is4uj;JQ<#=@mFk>LLS#OOt3
ziuLIKOPNYyP4U777;xPA{ksH*lDaY}9~va5<ZA_1JSTX&1a?pxY{D^q4OiqSiaqPT
zrVdSnj&=EzS8cl{qU*x|(1}8N0B%cdrWl(?jzS2(MkEEdUoua<BLXCm?=<_3ypslZ
zBEhTkeox_N99H)W8MWHvlE&BcYFx^GWh8*+pS2?U7jql{H>bW(H@r-3<^m%v@C{J4
z>*M`ICbpTPlpx3@BTV(GTX6mB#XDng_5Xzj8|27t7bak96kp-9w=Y>&K)LDC1z4=i
z%#bsQ3hbxD;gjVbrt2$4CG3Rnv8#`vgn2SjVB>s=@~Fkn+wTZJ988N?-hkrobKTRr
zq6f@3%{T<h5hXPv+K3%?tGu~uTQ<OpWBjnyE(>eZ`6*b@2Za-^WT)H$dX{&*dMA_T
zOIk98{UOS)vkR+IVX=>5Q{P8KM3{Z|nlqbh{_yP=zEI4C772zQu;9AzBQpl;U>5Ns
zXipT@(Yhp7qmZnySfU{Qy8js+TKM2k>H7>p<_Muk7JjN&9nppo&*moD>p%}}ZBC`#
zz4Tp8K`2+^8Q%g=Gk`(5%H`F~uV#p&H7LJyFWG*zTkL0BwKFW-g#b|vz_a^g4`$3x
zU_MQY5OIEm4yf0fT#@(C3biN3fo2Q1h`4ZdAhwie1q0foSsMk6TQvCAfEF(y1R9-#
zE&d!}>W`8v@<hNOn8X>17gV%4g22;!*OB4ts_(;hAQ8I`{6JWX&ePlgzu#LWepXiZ
zVVA@UMo+GJEU_>Lje&d)K-pVxL;JRP3>0w94;jJQ>;cYOh1@)Byp`(JU?f(}=0_1F
zmyFz=M9Ad?d9%Zw$--Y{{)K$1M%IJ*7dHVH)y(>qJpu}=6FoGRKdM(%Ove_NyHlkh
zj5|zg*EhHSq(f)#$W5vrPH&e{|1&7K`(<dM?3Hn|asv6!^H{s~BLJvOee8aIKyv=_
ztR70)1$OMLcz)iOpr!m#xx|4Cf`OkUzI;JD;mS=Dl|sdJdL_#Rm|T2grT{z$G><lP
zj78b9*=M8-@$h$IoYe4C6eIf-?!p4(p&wJS&}A#JMb=~G93?_BEe*cj@Dj!W6!XQ!
z<rGG|i0_n85Bt6h*bXiP+c)`B0H$5|=<J-&(MAHH`G;k&J-q4@Q`^V2RHovIjZ56%
zn!5sbg>Q8;DRY0}$83+=aM$H=1Xc%(F7Y(0o!jtiW{(nw$0gD8M;PozG3zTJ78pv$
zOqi!1wD&BaqEI@K)CRas_5wr9g+uCM+LZ;HN1#}x)%pcXb|^u=B<dHU3qDt$s4DS7
zgHo5!S=t|KgqXkAiF9HjeK(!W3{TSj&Lc8?pJ>!YIS=HW`61Oj;BvX!H{jCTRF*aJ
zp)v1Fu)w?uYY6=S@@yUpB96~Xf9(&q>0FqstIG#8K`|X%^Pz!8yhRjbY%vs_w04JM
z{8?ONq}j;(I~;X8_A?(+UvN}jFre(KHT_767(|-8X@Q?N>rPQUcY>6_py*E;qjr%=
zfAN2uugEh_$=A#b#?a$Fx3Sazg4z*ZURoPv%D-dF0+O~~XmoozG)_@@;{RDN`wUl7
zB3=AhcOhVuo<Dlq*TWcpe)134eQq-wd^z|&1ka@yPxd>^0AuEiy_U=6>~myW%-w1W
zVAI!J20F^6_C>_222&-?8gl|_PA$nsn0d$7Bw1E*qIS13iDTbMhIcmIXqr~)RnX<x
zKlS(FHW?e9PpC8@(@6&Mw@AG9R`XLmrapnRPm4GOHqQ&eYhbh#3T@o$Z9z^ta+Q`;
z@P|S&sH|sMUaJjVB-(FMflJG{jag)mmUr9`fUEsP<NgT4*k<0b-0m7NeNtw^Nm%AO
zLyME&5p*{MUbwQ)7ckQi_+jA-CET6PppzmsiX*xG9PhBNIbHHY>MUB}P(>f~H8lVB
zHJAV2`Wmd=-_IQ$w!`+28POH~;zb#_jOdZ_)TSWpwJ$yX{-dMHu;~QGak~fBQ8S>M
zn`;9r_YM@^v^4w%1>--x(Y4qLpSI5RLYqR2<pZ~c!cTKzcLn*;GH`WUV@`SY>{<OZ
z#cnNdOw%}Z?vGdUS-S?kgZ~|Z_#qitY1q*-Ads7s?(9x(HL@WGdSQ`n#@M=INt{;c
zH0=BYYU?TQp%iwc`$DpRSSclV>8B)iz6X@&5-fa3{(eoAsJgiiWVTy*&P|n10c$Cu
zC)<@~NFtbjS_mzHtO!|H5vDcvkRuR*c*iuahuIl0$F>4(nf%YyC0H=<TsyM9#S*GP
zX+8{o<p%bx%xT8wNYW8nU6Di-`L~14v+(I>PR{Bw^-G%JLVl-2w%Qjcix=N#&Q7h=
zJzFDpTVXIe9|%St28dg3pmg5*0N_bJ-fdNe_C8RP&e2OE86~J`kdK0Y$S;leYEs9x
z5+Lbx0L;KUv!0#4-fmc!4?bdTp}R{wvQ(m_uMuA<eYi%9M>}j+f*sKu3~K?^_GsNS
zJlLglnfC#RN;L~!V{boeQpUdF!xR&S7snBooigNq<+39_6+h>PA7)@b4?X~$9|2O4
zs_(*CYXEBZr&-vAg#(B5-Pb66?sS!vyLe4M5Ye_qi1XC*0Eb($S$FnM4*JC2_4fl7
zFinQzk1@8b51QjK$#KIH@J7ArRVIf^G8~!;InWTDXSX3=RJ=H7itX&qxCC2|G|syg
z)~QWr(c}f!R5<}D3Z<EFTkY_mo%wK<in+UJQIMl%O#@8=Xtp+w4>U&!ygi&XId)Cp
zFc?JBvi5?1E2Q~lQlxsb#d@k2>FF6TSO_dj?|dFy2WyFJ`W#dJqT4>mwu`rL4%y)Z
zK7BN!rAIcS;Xt3!R(jhkLM4*rOWinXdTrL$@I3Rtoz-$M&Q_KPifa8u9FWIc00HS+
z!%3eP-@)BG3h48`)hhj_^iMW#NvCwW?7$qy42i&8%8-r4-}JpS-)I#_1DYO$$9kTt
zi~Ub^ozm5|*=gs8ccN`yKYsbX>M6<6bE~E|_p&@-wu0AZOquQIfhZ=K!U>e}vj8y#
z2rk0E>nBZBQK|>7g?z-!v)wt`7h{DV7hFtT<oN^`?Jy2e)uV;m_;NRem6Tg<fgN7C
z)jd%a8CES1l;?Ou;CJEET}9AlRGSz9--$$W6Smf)Bd`Zmg*+hZl57+2fQx#Xh%oK>
zG&e{D!vPouqL0bk^vmN=fMN;0_){MiAU&QHJ8>ASh)+B|C_n<30u(J}`q=W#El&Pr
zp*eK%Fiz=UZ*g0|{1h1U*pmzO7?AtBCJ|<g-vo9xjWMZ@t4)E@bv~$tajDV&N)JAE
z_VG=CU((fI$(Hj2Vp%+|W1xWXvdR-X4|~{>tE$apSONrZR2(8WY7f>_be^9$s@li2
z(xjfibBII~+uFS?t5f{eF*{I1%B7PbLJHkx9n=@-L|yvHhczYZBs3mWB5f03UL-{g
zkY4o}hpsd+fNW8g`o46Vh^K1p=*3r)qv6hSv1_5?s*a}y*KcPV6`ABD0fmi<>?OJX
zMI>>{T`;?HFg|KxL-KYgKt;kG^DxlqsjGX+`+VtpbkZBohENFjo`xu*$2%-QU|yU(
z@~pyj<rYS8Faczr5xja!-)(`?e-@<Y0{7XMQ@-a7X~Mw6S1<>d#l7;rKti#s&00R#
zK1#W4{J76|s>^xZne)IGsu9uWMdEJ`f{_d)d7S|#EaV|3eW?@s*w(d{1}y&yOfmAT
zU_wn~^@-C}eb4^0ZKyl0_QBpg^p5gTJmOF&RC>y@jO55E#?)*4rHU?1r5B%aW*$-B
zwQEQ5(-lmBVcPtS-b&g%_TtU420c}=rGxX47F&~tjf(FNT!7}LaIglmRn(detZUN!
zByQhMW2`0@l*K)*mBkpkMwAHmZw({`i<@xQ;^b^}p+<5fFBUmeKEdVfP5(mT5=eMG
zlv~y0NBB7R^!xj<k0TYxa96*EPLFd$<=j~%8uE{Q);dk@e;#{B)c;kh*@c61(8mVu
zc3S*8qQXH4npwbuEIa8!qnbdg_W*%pWNkz=kTCio&o8ueb=Q|U(fgj6kMY=gAMgd8
zVc~b5W2I_E4B04I#3M<)?c~HID|4{@MIvtn;E>&c|ClEhG$4~=a7*L9zon$(vz@Q-
ziG51)U@QC+*fRKCEKz)5b`&n={?Y4T14sU%q~22uA`Xt6<G^$h3>+AK1*cE*58hk<
zS$+#~Ahcuyi&Z8}Y;XVFoyV<0><sdF>g19#SL{5n1N=R_p6!!q^|Hy4lVhoe<AlvT
zJ4$TF;Lju(#uSitE{j24#cTG0ZewxPPIj#=SzlF26_4mR&p1!=lN!6q+}xbpNo|#`
zKm8*4wL30g4NlUb6QKYofVI5oQBOPkpYN{FE&id4=<@kfVXk#CnTY53ZE@^VMCopN
zAzNPZqNiA3Cw7X4MXrfR@H^3a|6^TKIQe_aF%8qhK}WD9@qY^O;fn--&;uJk=J-i&
zGykk95=(W}Wi8J$#pWHBSyrGl7$Fnt*OXlU>UoSl6c_-E9cSm?-s_P91U2*S#@{$L
zO%{v6yx2cSW598Bvt9bO3HEB8@VN*vqNn73)j~67T61k0#q#{$aW-kW-aKIvzz(}q
zzb(nM!7+ykAD`mKi1yJ9hL#%$fF&$`44||kQ1Q~ib=pSwO!D`#tD*0a4GHjs#6qa#
zZ_9epfn@I@nzmEhfF`C5`j6Sa!aJ~Y4ux)^J3?-K5<Za9@4(z>%Tj+OvfU3*n`S!n
zY<d9h&RyF*^JDjM?TDD;a!gj;fG#Hes+XD}RC<JedQ6(Z&u;PIvK_u4<Xw2uHF6PN
z5@hkDyCjiLhyt5D{ZG7c>E(-%dM}`~*<j?fd}?+2!CV%8g?0)+Nyx)YdKHDnFTUV{
z1ozAJbB%Tz9u6Pp9;P}(dG<HWHA<AWyzsEaTIY?1s0U}qC^k*m{%4%jIFTNGwO0cF
z7!_@mn3lTQ@8J)#)z4`OJ7m*BbfCWo*7|H0Sl+e*q+&MTo!SclnsqAnFiDSu*!NF5
zAki;u1ueg_s3cqfrrRIgkANT|{h_b>A^}SF<4a$l1-i~lc&e?y7++`lzVZx&N{aB0
zD9iK)<W;S|hKt$?=QXD|!&!Id?zlFutPF`CPN&?@`%4!|2^{tMDR7%S<%A4FBaV}}
z{~OY;j}BtP>?%B7deTii-FFA#ZA4bx2OuQ=`ZY5785sy<Jq3@TUv4_Du`hs0L8BVe
zD+y8>4ufv&%iv{rvJK(RP~Tr?WF%OndGThMX5WD|8-Q#f_m;Mv0)d``UE-OZ)|3pD
zN000DMCn_+B}h$@7{SJ|mmg(GaK)b`S8)$MLyDC-ZhAgcT2R&#_ajrNH?F^yC(5-a
z55vBpTKTP6f74f?Ek0}7O5xF&Sgx~%yB(j;k>^PunIZ`Fk}Uga?!!Ry%eur%LJ)V;
z;G~MXRY=F2l$Gdieo&Xs-t{mttV!%ew#!hD_>{CXRDEKh<g8$)%O+wDt&&v!LF0|5
zj!NpH5Z-K(dL&u%a#K1xye<rcI7T4WK-MviuOVdjCA24s(4hB;t^!>OxiMt`PXh+v
zI)B2~DNlyeqF}T03Ap=HB|ES^pM_Y(p5WBGDA6E`pctC+fe7NlML7;&t@Ebuf)Jg0
zz=qliS8VeOS?q4dEvt)s>iW~wv3DLtx7jQ1UGnYS3A#6Ba?~@6dTN&;rz$x@4ikRK
zFpGTy*M1lP?lCZb2v10`7W4y7tpZ3IGoW4908Z_DG51$+y%$3Yyt}4r$AwVuw)8zC
z<8tfR)G;u#H>>x+(zDxz2L03@ci7m&cyaOaWe6N2!bz>|8;FMO!zW-rGx77^uv!5=
z_PXK75|Rvq;WT%85$140^*kg<vy~G<r5KK}u3_KD&+zE`QAE+g7+GUGU4%bt!pNe{
zh^o`INRNStg4h$MYo{B?{xq6)#gK&v%gNLSVOwhCxUdhyBNOdFl(nxg=l%S@A0pss
z`S5c-z5#w_7_zaEK26(maSC$qUcF`Ad;ASfG0;)9^DLGy?q-5zCWixx*$ZP%d2&tT
zo&BBn3J%{#w819IM1FPRCRh@pNhK+r;=F7NS~H06ZN7?nokXq&(PmwHkrzRoU4qtE
zKi^ZWpC?;4B{P2Eay{DxdmcdYxzC9xZ2f*6EU85-@^DyxCi}SFY1okS&TaDQl_4?<
zQ%}68m`d9~r@^rk6r=I%K2DmB$A<U(ch^o#F5?ln0&Y^ZOuy7I3In;rNfkx_>x&T@
z56Us(1vY_pJbNTA&Z=9jiH1d#F=E%SZDe=dL_0y82oPQQAZ|ld1%eytfpvMAVW#Jm
zz<#h+1EY56F$a}Pn1HecBJvhS_hh^8%%Rd1vqCl%I0=p+K@z0XrwJtKqa{csu{AKi
zhv7DEUecqE^zNS|R8QW}(7QIfIgJUiW80j7P$+K{iQByfMcxpm24#5Y_|FtQx_qX?
z-|khoag<K<C!hXnFduT<nOt69LR4}cQfV_8-vg?LEPH)<`O|djEsRgf+N$oXw8JHK
z@_SN(IJP6<PoV)7!$nzmsn02zz=f};hPzVYf^w>1F*KpzyACj)#YRm;czEd=*exPU
z<5_qXL*<mmk27&_bgHp@l|{x0V-*j4%^Z{*tc{U_473aEzA3GX*=dSVmwvrWEIopw
zl@j=1^-1&xtB_-AZAp1=soL0IRcb};y!+x#?@>qLUL5K#p+)%f3Hi(MP6bRfBLg`<
zRxY+s){zbN?wXBWh|fXmgva4s4key+aAT$8N@_jbUQmQ+UpkPy!9afxKPPMN$3(@0
zS4qmQpsLS-td54z!MC=?&al4J?09Z9R#8Mf-xC^3f{6}HhAQ~8_4uD_*h0^0i0(=F
z;~tz0BmI#ZXM!b2ZzyPBlH4WI32urm#3`gs(=zUuaEkm1{QI-wp%%{A1)w>cZ9y*7
z_qG7_WkCVKj6ozgjm+rUYSn*gAV#BSyQmuE#<3>%Tm(P;GEzBRhc;~%zVG8_&vGKW
za=1)ZBcn=N(`-V}4fcO2hK;ydS9}ro7iwQAPX~m{T{5mU03v++kJyccJ~`PC`tLkl
z&|ILwUu<LHISr@L2|nGwz=wAUAG7~-x!8T%u{BZnEo+DC9k>J$Lb1YxJRM7I#z#V%
zefOcO9QykGbJd*>l@WV1#cX%8^^03uI$tuY2A6t!{1ix2-qd>GL4@lhZg(rQ#|-lr
zaSSe*Xfybsi~^8l6CX#maPS*wd)@BGkj$nmYBbOc<?G+gx`BHSbeDeMc{LbOQ2W*$
zs6C>imS3TU+d~>}z+(n~q%I+vy@G7TC!i6zqTPK=8jl}yONyYgQWLcFksu9TBt1=u
zeRyNxUmXvYCU0pV)n2@YO0z!>RxYa{4;!!BD=C>z%1TPa#zzh8XCYgUd7SkauFL7b
zzY-r>LL}A*k;@$5w`NDwR>t9C;Df3vQGo-O7l-pw$fr4}eI@vyk}2PzKRy$h86xS-
zif6fJpuiCiIolcFq)B>gNB4ApN;nS}rAcD_;y`W&d~13>ux71=#0!@9ZJdUcOpAWB
z(03`neEh9F7FmsETvv^KW7c{c(X>E@0x=B^s)k*5-$(eqTO>mT>WN5)IacrIvGum&
zF%lZR4yS(o;`r-0lU+|Z5?W<YM8}=hcP5-i-@3AFRp54|$nrIY9Mn(b{P<=@0Y4Ro
z9KXs#+mjCA$LbAaHPNuqPy{+fO@u~1>?)Oj+XQNP1Fq$5NUK4#F~0P_BdB|=-rHL8
zHC(*Kh`4x+%4g<^KGO4~X9vN**C#*}NfuBBE9l}~dV$9{3nz(gIn~n_F-6V-6*;h{
z`@slJ#JZ59;6Fv~BrY{wgaj!yg9fm%q4T&#c<?Dv2q~RgOh36wF_4%L_#UU9<_aVb
zMFJFUrTV=M8Zn8>)ppZfS+vG5<{&N1r`mSdL<DA?<x7S~2Cp|WEVfWu(y{*XKyKHw
z(xQxc*Pb6oP6^UQkrW*-POzo9Tij`hZ1yGWf5p;3FXnbLGM7EK(;FA-B5sAOUbcdv
z4Y^5sRciD-bbxl6s6I|u>pm*bbw3lNBGY-=vl}=^V!WMnxahrN!g44y7B$oqz0)^|
zmTJGj4f1u{9-t(sg?`CCuH!|xc?DwQx;WMa>Y*o_h1#%|?pz2Br>x~rLB4xtN+u}Z
z^vv61Qf+@P`qh^;nVtI6qoP~$we;+}?;RT1W-6!lZZo$gG4C>v=1@WBVQZ`=jCggs
z^~yCe`Uj_LSBFkltKK*AX<bfz0FtuCJzZ%}7B}vntbuTOW!wc@20tj<3V?BFyxdS}
zfw)MZfa$@I@GUVfKbL`={p2+l<bpE`8wYt#qHW9vR*i!{JLNN-qKz#in8kDNy8Ph@
zGA=kSkwJ&Cq?}Cw&q0EL8|UDfYB@<0=0+X(SN?}#XNQU>_>(FE;DSEqk!TrJkzdac
z32eknWj&)82>!R6Ms8SJ=K-}*3lb#b%}cbVeBl<?>8DX(Z>$Jd&S&vkd_m~123a%<
z6E#(i)}rIdI@(<j$8!heX~K3elReGG4SZ^~D<3VSC3XSRi51DqkfV=*K>{o{^?=*;
zGAH4&*HvG3fjey{l0RnYmE~ly`Z@sC2Ki8>M~Ekk(}5o=55`-|h8edEag&{35|$Cg
zx({ImX-QYU)nu8F0;jsWQyTZVi6p~Ktb(uxLXC<0*n!S(*H3<zIpTSb#tt%LdnV4_
zFqV5Y9<eG{uY&dC^Juo>IC{I+@Uyn3*>BpzdZ|yu!l6T5m24gL=TVCai4)s3L#hda
zjBqUB(Ae}v<kg-ZhPx{3Y}lqAYI46%)Na%Q;%2$gS<`oO<8r9oW!;IL71F=XA5Tr5
zl8SNBSuo&U`+Gf{?;%TFaNDH+6Rt1<&n$B2OxY`ABQWJpqMi$NrY9lFWNrB~&wz8b
zvpqO)@ZhMct$yyOD+a{ybz33pCT)x4*E&%DlsUV&e2y#xl3N(KkG)nJEQe^gi(sp<
z_D)q*YxM`R4|0#tL3Q4f7~YUjEVUR%$`(eYpLG1z*eX>YWf}b5koZiIDlzy??OcGl
z7>0&-XlO9kxHlX8QeKFs*)4sIX^z6u6S_Qk&~=FTX&bkodkgZ`vw7<(kY0Kg0IZG3
zUq-P)4_~i=I=nxkW=9vfF@IMufbDqum8zT)7{StN1wY+WSKZWTpzk*WZP;6v;#}5e
z*{nVpWpwa;_w%g)kgRg?1*@Ux<vd95%4P`$mjX2a{+3%a3xSre8~JuiLoxquX;V{E
z3ltAzaChp?9dYahLc1WPXum{?3-vJ}^(tQ@;YClwP59$;>ze8I1A92^T|p)obnDZP
zY%D7`suQD6q5ixbc-dHtlx9<_Nsh~LLNfB*WSGF$Qn0xaz>7#&7umo;7qoW?<n5H}
zx_z$ds@q!PH-D97dGsvCuwN;zMT&Pw&^`xB%<c%Mbw)d8Mg^nfVnhCAu{l%Qdl5g_
zqU9!*)g-w;{-tS>EDNrqkRFwzBf(-N`uSvPb|UpYru^bxOF~t_ae0_=kx_Dj47az;
z=G|9nWaf|Ycc2qY>F`vPm+yYO?M4Fg9U(OrL?b1yXCY{r_8Wl<*k`FAn(`^HD=sTK
zO9;z`dhuGx91YJB3?x|CLg`qP-^`#>dUsUNt>vFi=`8jjU8n@<pr8g3ro7J{c1mFa
zvgf`Bo?HzZe3hV@g#%$sjIdH!r7HjkvKRy|>c-d^Wjm|M`;kv~myCVK7Zvo7r5lQN
z^L_M`KS!sK1BGaBZf^3H<PxO9QLFH&^LAhhIKmYwnw^AS4p*qf@SD`G+m^<NEC?$w
zv5oEcEH6%8fDp9*03Xz!@DW8zTEa^KBJ2iyyC7oOkYbw?Q1Od~?(x7-RDu~jztyVE
z*{nP8j))39#aTWvl~j{h*^gf1YqcI*lFtcm?7(b!k2Tjzm=lEoQR-}z4~W{V$lyLu
z+xW3-xHwd3i{GDDOQ|*~mrf<|1?rTlLt%~WWUXGk+0&3y_y2MjDn#y`wT`~L-RZa6
zzfu^fgSYpTib;&0C8Zny0SN;$S$Nan0-yi!brw^J$?X!*VfH^qb{Xg`_vk8uqe;V-
zbmH{HvA+rMUVpy+P*ue(qz8;T&1ZSCw_2b#e)I+;cd3iJ-H}9Hz)xeC4)^qZxLG~v
zq_R7T5*VgRU9EBK?HaSL^x$C&#Ynuu#~hg!{=to`&Sh=j9v<kd^Z0BK*((XKv_cj{
zSY;*vaHL^N%tJ(uPHCo1wh1v?G~8)zh<yw4mrN8B2z+fe000zXm61y=N*B#}^b)c6
zJ|V-bMFNA`C;+(p15}>uxiwQ#hMW~IkRemm4GB0|fGuuN?Zow!<G%hG>BPHw+ciXI
zM4Z%MaQT+V1X4E7>dh}Ne0j@KZTiK4g%=LQi|?(s7Zn>Is?q7UJD6Ygrb-{{;=hO=
zLos#QLN<uoh8oWJ-Nrp(IKx+%Gj109DRwuv@YRGvN$?TuK|;H(2|ZOyX?RL$$|j(R
z{SB5Tr&aWE-E9DiW9)E!1)Jbvx4-r)Oo%_}XgeMn16{;OO6|ppM+@OmufUvSq)+4_
zG<32AVL+&s1vG8x^FCxgw}QvI$Y|c!i3%qn^{tw>+KzNTcSS#TB>X89b<%m_oUbhO
zyyL`8@?!(sd$Of<PX)2DZ7AS+zbq;vd~m({6JuRGDs+K@PgSXdX7%p#prDuViCU4V
zvunN%OqFMV_hPIX-iB7_k%%3#f&g`}AX5u?uCTpQ8X~@T(-llD8wHL-!<JdrGbGlF
zd86E*J8O8WSfNQUOa6m?YBQDE4%~v^TXw;@`A7s?=q8Q=AEu5_u$ru=Te0}aj*&mF
zpuF8KXAo^K#YisbzEz5N2;=9oSuL}zJhsB;fI>9Dax-TN`@w?;dhdV;hAw{@+{ZtS
z`e~FbrM|*8Rm6GV$$DfV@(P1?xVi*U%w<FkFMds{H9>b`mjBd*;%lW?v~~TxT+($N
z^oga0i>=0Bv2kJ@H>ywuS4T~fT7x4;8TmW`H|C^cHfI3}8%MN)EY8;ccN~V5TEHCS
z1XP!Ic0d4mlfZP|Cr~!@DVo!&b5Yy4UpWj`s)}Ie|6!X5O!EpXqQKcR6s!<iqb}=n
zFQ2bbF`|%H^sqA1<4%{$IbtfN;Tp9nb;MJi3f=T!i_0s{&=Ib0RH(E?Gb>W7i?e8X
z67ZZ(R<RVg|6I%yCLJqJu2(DVLZ;}J#JY$BeKhCgEU(8YbKnUj<63E+B;xEw{{>Db
z6i9(D*mF4nnTi6Cp>w568ORh-klT1~qegpb4xp?B?Z0l?8>0i7;?R;HvTh`zwYa%e
zcUAW!IPDc$^*;W6<MYE!^T}tQW>y9WuP5KQ!>rjpsl><ls%vVo>NGX<s-##GWNELG
zg%mNr2q)p(8yH8j)nEDkxx!eZ42WF>+<Oc0Samr<)_@sDA!6uV-F4@=R6qJ~NFY><
zuX_FMHx-!ZQXb)g_q-U`4s77$(Dl3a_<&m_SUTNX9jo4`0}0f~gX%twf?U4~*~b{a
ze=hm>Msb`KCB|xHsWpUl;$bSx7D5{*Sq)s__2Y03rsX)C-6~|ILO%j366$dR1FQMp
zM6(-o7q|VcHzk}j6}w#-PDzmJQ7Hdq!cWKecsA#QUVX4Zjb#rZBrQUL4v?SMl?KyH
z&lB>Q8cke{O;A_XM3g#E@k+|jZ5Y87#)9N<)H@BrhWV^aO!p`{U~!pb9}SnpwfNas
zdpR_VCqACwVuFNEL*ZshRgqirI<)Kz(m?k{e%#rA4jV>md1D}sWwqu_AcGu@Gkvb&
zOZYe0>iK%FD=*r6A2FREskufAAF*dVw#s(nPxDV{4?7l@vihjB>p`ixWJSpY_6+O*
zjhu@~>Jtw7gMsE4JUs4T!>byN=zEK<hpmu6y0dMT>%wAE<p%eIFV`;LD|)P_w**A}
z5i+@cQP`f<iRN4TK62oE0+*`Z&wEtY;yzrk`zGbJ7F(*lC+OH(ah_P8<i}+uQN74_
z6!jwpP;kulWQnCeK9I+uW3S)#los|iKllOU4cH(p3y1eGU~7~QHXdp8^QIr{!GZj;
zU?c2K%lyc*O?nr%pCbs&bzRJN0*|Vx07@WuLXfVEKgl<)c2Wo9wQNy41<|~0*xHAK
z!3H?JDCN|!XGz=J7iUck!F)}%N52st5(s6+Lk?~xdP2E;C`#(%lBQCJHxs^O=I8@_
zY@y-Q>xr0Z6vlw65w_5^#w`EkhbHZEj{Ck_%TKm<6O@b{C7O3;I@QizP5(&CXBoNn
zeROD!3ndf4MEx6jR}t_drv9ZbN}y=me|Fp0?UzrM2cE@clz;^#^OU3bpJ7uNvRAHA
z-7#>t$iQZG;sml`7!;~z;np_=wMyWuv0osH+wSTYz%K@f#0t}Q#fwy}J1`FWu7P_g
z5|$8KOwh8|5q7=-<CGP%Q;_&hOfNDV$`9O<h0i6KjLNu=ou)jG5nEU@m_#2LlCLt$
z9?FJUBJ&UNcPY4j0<jBKKXZ_*T)af4)oD8xcKbzU{_7(}&VBISbae1(yMvjB%=nU)
zIK9%vBdOT|Y|l`CjUe3$HHq<KIvZo_GUGG*ut}Z;QOY2Kw)eJC>JilAjr82J-e?~S
zmUTB^Oi!f{boi08mQ<lln#lVk<apdG+K}I7`R9IX^l6;JyL~0<I?SO88QJ>sgxL7c
zt^-o4rtpyNpnx2-B~aXiVFWPov;66ByoJ$8+R`ri0C8?GocY+Nhb`oE;U=c{kf=XA
zxsy(x?NZ>=@J#&D$g+S%$Nyclu<Xf)7(EUVqW$<CDEe*Sbf7`iGXGrP1&iv930wtR
zum6m^k|R@BoxN_Q9#&c9u*8BAHOx*4*{4I7_c4?oNHkg&AtJy1`JeZP_IzxH4d++D
z)%cfni_M-x8hm#sKQK8jZ+bnx4R3?l*S41v3SvV8LP)di>yBb?ku!QP%#r8VulR1H
zwZT+5ApG0`$C?sa2-?e3!df0&E8W4Vy6~_i2t99jKMm8IJ3hPfE3x=-#|fdb#|{B>
z`=iiDt5T01&#w(08C0XB%4NM<^Jm2BH~2d%B1%!e1jRp3kKs2C+=9$VaN<S3bQcPi
z&|mOIgl=llVxkaK*`YAa>OelGXe&x2f!N9Vl8%5~?2oF(VG|MMZ9U?J!%yE32Y2sG
z>u2;uyuBOxscu{)UTe`fx7Ftewj=(}CnkRE4wzLW=Jmk*1rAYyG&P)bSrEq_zx~D7
z4;Wq1tzYp-iK3+;<BVJhrsc4vLHjsQI_ni~Qu1`T|K#g0a3*nkcz04Pc+$b*?>tf-
zKdBtC0?(OJ*1KO@BR+;E+0VfEWRIM70lpNi`Z6<vEEh45P+V;qwjd-6>7icmFFcNQ
zv#uo|Iox(Sp3`OwSk7Ot3=whe4~S!ciCtSdt}qBWM#ENtq$>EVxXnFqG8;+<MhTF}
zD5S-nhV&Gf$~o`vZjjs=_)7!A#89x(7DOlkV?KcJ+~AUdT#FbJNpP0Q;Q--DT-AjG
zJe-c|m9a!ay_3RkNF2i!nwLs-F59|#I_^g1)5wYn!G-avx71CZM@Sd+i)hfOgPKB3
z6KL&Fz|H&C4d9LVlAX1Ib;bPX+d`n5wh1f1mCp*hU7qx!TOsvfn$Z~>O_%MzSbK>c
z!%rEc4mt5MI7kss?c8o298ug8tIsdshydC)IoyUEe8HF_8on&SaIDIwV?dSR*h8C0
z`@+NTDZ})+1S!XN!8pu7ZBm8yHw3dLuF(%X7;sOnN@B}e{3^v3oTm$;gDg{%7BH(#
z-%;-e_zdZ;dkA{nv?;lfo^;q4Rw;jhw=RQwtHSMXu|ee{1$fv@;L#cd>RZ~R`LBj*
z%H+EMW>rVb!t@*kguhe0T1SpinKdM1@;V{v#i=-Fijcj<Ya?x>KSlEB3e#ozXLTR5
zz#S-6ba9~o?L+!A`THY2x#1?a04lvU-y@BX1NA$)<Zqz0N;fR-U(&*AuGVz*bwt;W
ztY@Sr-C7OX(W~h@@iJW>b49UX#QtuPv0s+c$^|6e<Ld(FFcT%F;K9BUbwVIEaprDn
zZf=9?^eeE1egnFewQ{8Y?xhuz9?Y^-;ms-BfIkI(>-D&#bahPbEcaiEcln9?zu&jA
zi!UuO7Wo+=pZTnh<>l+XmWy5={_z6Pf*<A4fn9xem{@83KaK~TQ8^)8Q#XwIUQ)I+
z@SednWIr6E79`GvdMgSCn`4$0!yE@!)$qg}ItF>HX17oe^&29uj*woJV><TF$fhH%
z{k;s_7mA<mKWUvOb9{;6vIBWT+L?zIIDgfFXT^<*&023?5JP%8ApbJnhb*t&lDn$t
zBLO=AD@k?$vP7!lA(#;A*qU|jOOSqLx}*)GO(@{ZzHYVr{Y(RBdk7zqNY%i#A;3?Z
z%GasZ-}6wd-GUa}C43s%>Abi(IPFdx9dBaV8w7SirPsq<#?IBpJXTcgQGpiH(X$l3
z77iB)zlgg16xLs-v9Vx+V!US`=cgulbkcQ2RQ>70qW>v2>6|IGv}%+u_M!PzZHAe&
z1rl??c)2;^TMZ_X9hfCG5woC$>KF4qb^P-iC6mdXx>5$xyPOj{HjTw?NAt2wXTp2G
zgr#9cwF$b)PO4KZbDB!qGmx{v!CbqpYES=PCnNEQ5a)OcaIOfmzp2Y*#_itC|MnC_
zTYZH`wZYxI+-7`H$B3J8O6#E6^wyNSj`#gss5G9$GFo?$`|BMxo0fjzRw@ZxD{%#c
zR?@p9Qz-)6;YWp#HBN8UW7}OVFwUrPd47H4E=m^xa@|mJp4h4ggb1LO6!LEA+GTHf
zM#w*1lSO+M?wVDV-|xggpkOgto_{NukdO<OEYc&8jBr7G$7Dx+7=#Mx(1M_o`UZjD
zu+SD*J`LJQ<jNTO+7?9-^b2J^UM++*0EWy4-iP@#GgMfLH~?EW$}@8uhXFUM<Q_W=
zxMdPR=6ICAEmNXc_0c9G?(uA2gj1P%Jv0zO%F1sSL31sw1;K3p+(bvP_Hp}n7wAP@
zP7+pw9YA((HBb7^P8JK!6F@?dz{T~h={UCMxF5&$$yWWZjTzU7EQtyNqLVq8RR=4X
zE3Wrm_zrD?1#bi_fABqztQ=#M_(uL;Cj;LM0US+3xvbg)HE8|u140@}DGu%)K{FGa
z5n^m19Cq0!=r?)ZqmUP}U>ROz36*~175P@*gUj~qo2n%;UJvgl<xt&H0Wbn)b@F^>
zP7M$uyV8>0yzH@Gn^%7a@*tkJZ=aK)4Q12LiPXxwsjlULxBjdEh^Q`qG7S`zlp;l(
zf1ga+c`5wZ1}zWF9`Q&H3{*TY#e@V5AelJ4sIv+HjD<oQ65ujG<IoQJk8OuQ$Wm>o
zc3xde`;V==vn_yn8bn6O=0ZV5<<G=dU7a}rAdYBxUJyBTQ?mXUw0M;FPCO`sj%Q$8
zZiYgFApmCwR$vk2-ICF7HU{8SYVy|&w}Dfmfy=R$7@_q)=VFWHFj)ykR%x{tyQ86y
z!#J(qTb;QN5Rc`?-oUP{yZR)L(@<*Ws-=<1=pNDU5m)`Mg%=gRT`WqO4Jy2nF86}F
z_h)S*b^o`jfWX>gl}~mnSG>~Cc}O%7z*_ZzVH?5W#Bta|sx!mgv+-EKlWuI%FTvAn
zv&vs>@*yh}H92NaI)jl1ycq1S-tpi&Y52o2b3~Hx$~SME!0s>RM-MKaRHU~95zDSt
z25euD^_o+8pFY4QhB+Q&U7{PvnbYthaCBH;+KxBth||_0_k*ChOdZ1<caVFAgZ%?-
z?^1vE`HMOQys(K<YoJGhQvB|e-@wsOBBZGmDuM(AZ`5j>%E{6xOafoY#?Wt8sQ=J2
zjOK*4PYk}W97kI2oylg<E3hMnPAGbtBbSScii%<K{r?zy4{)yA{(YPe@<GWgWQOcS
z%E*?L?2+t3D1=Bxc1FpbnHd=s3T3Bllobh)%HEmT|JU`Q=eh6azQ4cY|9yJy<M|%P
z-F@eJU)O7#uk$=#4)YIN_i`Mj$i(Yzz<b#+;|ZB)vO0<yt{*ogRwOF0MP8QJ^i}#f
zh;Ykg!qZ0UOOsr%<yG?(f!BXZU&0Prt(6A=N*)D-d|~JcJbZrBs?O3hmtiPp5`YQh
z6$32+FVw0)p`c2wfAFnUb$kD_#(2VS_*sDu`?ouUL<7VY?eq;M26P*?<&U_`zjlD@
zBF8F^;J>zr;j;h!%h1!{reuu@Tt9VBmV9Pbp#^HWaQr){{-)&U!o}UwsXJu>g|h<3
z4~<YE7H3n2P7<O+^>Q*da`w20K?gMSG38thJ2wE#I&igeGd!z`?;I3N;7Y&oDryv<
zxp*)GZj8Eg-4bR5us!F5g7eb$Z&ZlcK^hrqi<{;)c$Ac1zi$9d0LvTm2kc<XlL(~t
z8(ga<=RDGVw>)0gAv6jgmVI{g4oHfE@VIHaOFRn@=sr`^u@6f>qr~a~zLI%y$h{h#
z#<Y~mIy?;{LLgYQ-NcS%su`|dUUKW~{hVuV3>%`Y0B{_)xhS~X>Oct|i~+kwO<Kmu
z>`3VtKz4J^KLaWkc;*3JEGG=9nST8E(ea^~e4EIp-^~Ew=S5~@S1Cfyie`z`e14EB
zsh8`weOEL&mc=n8D4PxgF#oz7;dS1;|1?xPCmM~1UPBL{q`oXJdk9@D_8M})E4jm3
zum^{O|3I@Zb=Hv3Lx(Bd<-=D)Cv|kWtn~4SHwU8`8jzD|JP7Z{@(OArK$wI$tSV|(
z6o~Y+@Ub<C;YXZ3M?czb0-zj+T%w{L1Lh!n=>wgDQPW+xZfcM`hT-K0n_m7g2Df{k
z-DUZLdxjIWzpsVIa%tYKsw_0CO$<+08A<E*+JtdC0?m*aMonyaV-ZiEeM9{oaF7P1
z8}H5|N;Vi|ggOp!;5bwP5QL~HQ4ktILFmNeriugrM{bjTNd!uM`0-eR-P)JJDXAXj
zytzPhn+K-u8Mx6H$9{-}R)B{H+qPx;HLvkH4L8vOJbW8jca)+v&4^Z@U0ip-&DRHr
z##z#u@7}!m-6U`1hmmnOC2)1iJ;X$<irGT-O6hy+wUoUZ+q^0tPtBh<c4jF5^?F*w
zLPd#{6up608l!PLFeq-24Q~KTg?j@qLH_^$2HHr@Gwr?s#4FE%q3D|3<zlpB1>xpV
z37><DWyztOB@ME8n-#zyxjGaZT!^d*5f?cp_v;N9%_H-9o>*%LJ%}WsP*rRRmU}qn
zQbVFkHt^Vlfu$eUn3b*ty=I%<_!bAc`z~M<!|?QTp}rb&D?}`vBwFZ53?f_{X(4Hs
zkJ^et(oR9}zN;ukVvOo#t<g=UeMKj>;tgbI^n7KyfbH}GGlakaTrgndaN64ND1fk+
zzy1bhL{VTAs0yU_O~?DB9bbXj(s=L0Ln!JSXI<#6O<?|cfMRWJDMgMDt&QW&KYnKu
z^X=-IZt%@A-I$oGYkj_*oywt+ccVp5=|xQpFFY8q(<PPp`ivyuGNsdfNajbUMDB(E
z0U36xyFnC&LSB{g!K-lSZg&zy1=V`rUBUEM>mNf7l0V!+=&~=L3$Fb<owys601~)s
z)12gPSxofpQZ@7{0(UxP$zht2fM9AF_30&O=3_*QeW`!dZEtxYbLG9mu-}0juKLNG
z<bFd~!d1vJh=4kfB_v&n=f!)iCRjUkZ5vWXE->qYZ+`P$xVy1A>noab!ttx2^!COz
zg?lUSe7D!lk(sx9jLjyX)bXHO@_a^cUK}8S!m!Hi-|uCtHh%x|*HbVf?4TKBNItUf
z)M4oNS?)76+p(c+X~@(dM|?Ol@aTmZ^jInl2n}w($iOTdhHr;E5-~N|{_bV+V;gU%
zHA6{HPqhle*xv&jlFJ@JGn@z!c1`!gtR!F$F(3+Rjq(MF$@~|~NcxU1B?cD5#az}y
zkAO62?@;UiIe}AJDOlQEwJ@%c$+kUn{*;4`ofenE1t_i%29x=l`+q(M#U+@<!4QsL
zOcF<`y<GzAWE{~}n&ZO>$Py!8xd9V^G3+WD!$Ms(6iNR)gAK)9I>Ys8^1;mKM|23p
z0icxw8B0ON1ElyyGr895N_=Qn2e?4dJ%yp_G*~6;SFy!r99mzQO@yk7!R29I81jMf
zI-I-`F?8rSJ@H;TRwR~le|ljPIuIndSGafi3IO?g?I!48A{31(unzsN?*TJzKt6wV
zZ~gQf>JVpdSZ|Y;RS|kU4q&~cyPESA&LFW`$rZS|$a(Vrem$n}dX&-Y@yETMsWJai
z1K$^bJ}b|UwzAy#PQwXMiZ#I8z297T2M1tM;<$0+sJP<Mla!%*{h{lh9%x2NY6Qar
zsqOT;Fy{mCWr5bzS$n}P@{1P)1kE9ESy_To-Uon(W|&RpO!GmNxdGBv>4ju<NpLL9
z4=aQ)xukc0rvECerq&z89+)Ib*!DJg_o3(7&uZ-CBzev|`{d(|a|ihvnVyf5<|J`?
zq@672jET*nBBVIR2{<gzNz%VpWmvJ(@D0hwLs{?1v4*hDaZk|gHZYdKN*4QoUSnZ4
zw9~*a?=~==?;?y_<Ai2Er;|PEzfTI^mzRn_+@!)u4vg9yuqv4W`_L{Q8HX2UU6Iy6
z&$DR&7bMfwS>J6wAhI0KnLLCLV1NUfp)#|EX=<-0Am2;<>D{2=pH>#*$zWdY)YY_`
z{s6sw!nv+?h&!l+mbBI_q7g9C>)h|8bCj1g&_#=RKn?Ol`L)oyce`Tr>+hgl-0*tG
z(ms|Hwa5`A&zS+4gB@{5m$O4*c&H3pm|+{C(7^dJ7DAq4s2I-Z=zxH_={amSLj4xk
zU0T8b80^u9Gh6q=Y#zc%`z2`Jl<=zw#K#^Gy0&p{cmbd@@lE^Vs3){Ax<v2D%1$n=
zd<B=Q(?p%R{$%GVLQncbdEa!rRgxV|J0No^_l;kxgHJB|h3Kex9lv|OTjGJ#Ggre&
z&(<J@9PE@#A6h7fiq>VY9^b)ymn$=uESU5Lu7Vz!;gl(Sda2X>E0>-aShPnf4}V)&
zF)}U)>vg$wH~zVVe$4#aYxlUByeae4dG~FF=i}oOen1cqWP<<cl{6?Iyi$9zRh$F{
zm6*(#QwsHVD_?+%6Yfmu#0UKBWAsG0RFHMLenPdI&?Pj+|L2cSRv)S>7B*}>^Pvz|
zzx_s*96Y+ykk9`b&*fq$CKACLfWAYxmkxE7hfebAKVuYkQ2`o4E-bpv3Xsa@N-AV|
z^$6(Wz+7iVC2gGcOH#1Ene16MLnioV(<OUOsTvFF<b<9q&?j=9qs?Tq_6LtC9cl2K
zi)-w?4z2^`nfrL$3%L?u!<{x*8BZk>rU*x_EL%oOp+%kx90k`tK-r(4S2+R1-SYqe
zD0V$gcUgrtp#er=<(6X}i{0SLg$#R_)XgZdMo=cU1?de6LxtWf5k3OlODAa%Z&V=p
zS*XwtZ=wo)oJF9bc7msEBIFJf8^=xC0shSevWE*K%1>@u6#godp^=e^FjNbz8jVHr
zb#Z|fRNpw=@~b-c2*c?mh$Ku%=x~astBUkpHt}gk*iZ*$q^LeK9~JWOM;1SMg=H9S
zFX0lZlKPF=h=!cjWg?T72~1__`z9bn<}N<&d0f;xyOs(MdNw!~xKaEEVUp{j!@EPo
z`IEsZu=k;;ln-_IbRTMWYW}WZy`2wC(J}LU#)8nLq$TDa?gJlI+x@_geo=<myiSUI
zV@6AI4v*OMI6UWYrc(y6LZKidYq>ChBHx10pC9TPK$Hs`;6)Sgj5LBPhJiQ=ODrV6
zjnJ||pjSOFHnn?Dz@?1LV2$drQo5d+Hpt|67<hm2VB{g?6<mzc6j)B%`SUD(m{67h
z`g)gW`9jl}s$KdA9s@lY6d{4}r!#ed8h-&}!v3*OyCV&T81Bo59xLJbk5OaYScy%O
z{K%_<X3_Vli@d0Q86d%%d+-+XmLY;5AOs}q?=z-`<bbQ#xk=|%oDW1VZMT6`ew3&j
z@o6hAKwt>Il#jO`s7OMfwBXGwSpmx?bD#5Dkl`n*1#U{Y429;CJcL2+r-`(a4)@Tq
zo({{?*Q~F-km8>eSYw|eTy=TZn13`r_w;xB3l4Vx<_o3`3}mMmMgY<(-r(ZjYD@e?
zGE}l5H}%R5{;cZNiSJY}>kD-xY#CVEp*9gGlJfvwUiQ=WBuf2$(>ob8aN6`8egIHO
zh)Kpuetd_FdhPT@R^lavc(6wv%ge7#ha3qjIBZoGY!v#fK%1G1!;PkbC3k19^c^~m
z;4Y!vIrJ$UBaojLaTn-osGHIj{n)wRWziZ+ypDWr_M6AM9`fd;pG<Dg!f3)XKD)Kr
z>U&#a09ay@iqc@diO5Gb#*aIT&CSVpFQ}ZgVF$rLXJnL~EVfjRwWA>2RQqWC3E?JU
zS(<cdOKH+mX}XX+&XET3MFmDEqOSKhYP}4r%Z<?(0><)tOzc3!CIXulp`cpuYt}&+
z4fnaC)O?ox54nm;aTQRI(jZiCMkPVOzVjkCd~lY&r^cfcAp5tm4~p}>6>rnDi&}%H
z5Q4~BpLO9@`&{?+eicNL6`)aK$bQVB2S54AkIWPCMA~XEC4UJQUbE&gICh1i+)%gE
zF80i0alXqiNjeT6E=fCk?{3|-Q;(wY7y4hKyBs9Z^>_MLpf=TaZ#P6mvp)rnVk%3{
zHYG`JeFgxf+aH~8rBc<)g@i$zM4-e6aG91-L_B{h;Jl|#%bwF~3)<col`+B#0BYXx
z{vRN^i3UA=6Re4PxM_O?{;z-I@+fe;AE=Y}!#2jk8eZGiGF!j)3s5KkQ@(tV6k<^<
zpaYjZy4GN7s#yErd4BzY%K?*M@vUR6i_ZuPQXerXsNdGt6mJFwHX@@n8$-PWwc#Ba
zZmoK)li;%UKr`d$Wt~ZcZ^M6sH`kP)b<crbx0l0siJj!Y|1se3cOrIaGe#iB58q+{
z((jfVKX*fQ3s?xA)&Gq+s$h?<9-m8rY3I;-Xdti(Gh{`g3@hBU4tp0_K$H(Q4k+tp
zn4)#uvm6ef^_n&sB)<86SNOH|K%jWEH6s!5o*o@{5regXEE%hr=OM3sCKTtb6WUFA
z)uO!%2!eyMonh+1Ku!=5hKJdv9^CYc{`0~9miT0ycG-i2Ysc+#ff#VolN?v+4m$1P
zP2ZU`7J`IkXXFOU@W`t!aqMx?!r!A;kD>LGBciMTp*BahOq*NeqM3dzC`;o&6KlH-
zp^%Fg8$dds1!E5$^ST<?O5U;fL<4A-NJ|19J_=q8RuDB$6G^MAS_S}y9@_M<MV;vU
z9G78`gdpw41t@=>!iuVo%2LkrEvIHXdzlU%v=Li^qW-pLZ?Y%pz4_)bI^O!XFwi;-
zQI5L2sP{o{(H1hyEDO7@ZO<AN0Wg!R(-+!k3;+aE1{$OwuF$IIfuhw2Be9|R3VMo}
zD@&88PtjiCBQVtiuJ_NyvVdtKBbHUlP*;Cs;0zL>!<!=FGH|2owK0fJi=5a#9dX<5
z4KXBZCd^9^5##`5_lm)}r!$BE?DWF~f9bsIt5-MYd-f~89%#rSDs32pWMz^5;lkBB
z1AgIkr9Vn*j=tr?L<<4}GPELO6Z-dV5BISw{>-5Uww^t{4ikALwYRvhcXW4FXd=y;
zmAMKz;QCWtC_wH-(>2J+h?)Va;|U<OlBW5!2iw#ZAC6O@#O96_6Nm~dIxvmT^r$Xk
zqR60b1RQf=FZ?5-Zn82n+Be^!9@tw(2S2-!|L@;71jcenTN(p!Jq!cr5Xl>0O$!?&
zgu2drZ3R8qLa1QIRiE0nANBxbSgH@WKWAyDUY5qvzQ^XX5^mpHujXGJzwxC{FU!e*
z-VM}ubtb>y$9E?{xw_in5xbwPEx--f$D+!J1cpn#fB6e&1#xP)K(+UD|8F34ud{O=
zu~VBkb|_PA?lvCA@xUO%_{c!^^d&;Ng-e~>f$%c~nbjA-rw_B;H)gTqrngcyU%mtT
zB>6o=4Lfqg!cWY`!?Yi511CuOQ`hawjyr9^am%U@p;a5>An*8|gq&$$toAw<;4cyF
zjQ+)8D0|Sj71+7Nq#T^)iO*$v1k4}H&@X?EuRBp4KIIxR)Ixc1`B{#=9~yYsv-Cl#
zl>latyv@hxzx=qnPzzjApupoQ&kkoLR^r6#j304*Mv3W8SyEElEG8QTcbXMnLGP;o
zKBFr+Oo*sz4*U6~0HMALD0kKLP`A;aXhPxn{y`^Cs0%oHF%pJ1nYx%Bd+M7XgZYdu
z)R>2y;$f<YgfGM<5der(%R;|_mU0E>)hkz%j7Gw_eGx#87|2Nt+B8q{c-V*A1A05e
zmsQi1yRPTQR}Q1VGxG8QRwDG8%^!SU9gd1~H66HDscW#gY-1QPd$Y~uEi7z}a93%u
ze=^V*alX}-5jH;d3pwPnfnuW=;Ev@r9iR#RJ+GjXAKLAyO(aI367Pt}?!o;{&&Uu}
zQkMLgJ3TR02QUMU%#Ux&@yMI^Hvrn(=go2DA>ilb>75|X{f@d8X7>mNX_yB7o)QPY
zYF3I+KVvwvQ5u1rr{H5y`x%yR0ZW+D$qhW8;$VS=os84oiepgn%lfR1b5#O>5RLmn
z*!%S`5y*jwxeI7TTD1>}*s_9Pw(A)&IU_eJ+CBqx$_tSC{E+3_^O$a`SnGzxw5OLQ
z$4f+~B4Wg?zubxE@H$|#$5~4jMd)RXZ1B#10*9debv0GuUrxb~bGHgot2(h558cV)
z2qR7cP|vhm2c24zx@X*&<bG+k`h;I#kt%%e0N{!{mBJuIbcNYLl)XMlG+T)to*3&Q
z#*u6&lT6W5mKch}sucZVm}e$}ai+xJ1lorHs(fq(BmCThz8J*o_ucr(;)MG$L5B6G
z3<G_}ppS0?@^4<P;P$A@c6$t;UOp>yo3IL?bh77$Td3@-aw~cfx9UfI9w(;+3Uwr&
z@Aat5vig|HGv%OKjE{!#X2v}sc$j6>i4_Z$@5s#lo}PgAixW}4%(Vms*#Bz;Z)9RZ
zHdr#{XW#2J@Z(+qIRugtjZFRiiyim1NhWz+ZPiWKnR<mpnoq)`5I3rxExmVzHWxj-
zQPR_wai?IgJP3B&ya-<T@})kVITda^SSwUBiaFIcjc6UjXmhrKkGuycrLred#m}Q;
zK!W*8-7obKwwz|-To+*DDS|B|$3Z)c7TBW=4#Obf`P5`BBHpT-L~BfV)LcELJI_Ax
zu2)NsjxTGxfnbL6fW5i)kDnJYJO>hL=<oIcaVR9$l#vFXP#z_Sv5(@1|K(=LT9NEv
zP9_p{Q|T5TsBf4`9NVptcV5U|-aN8nBWdjlMtC8^q8PCa4Lo9-MeJiJ&lZu4AYG$(
z{kKWdvwdEvZ1tn^*ETIyZW>&BoJ9lwwn1-y6*3eoJDZppMQo$cc31p!Xo(5ho`E^4
z3@2T|Gxv&2umLz}zQ_Gfpb+jKMS|GlNctSQ64{Nm;X!(ig)}Yv@jA^>6GNbC7ZbT@
zy<`FbUEC$J>xP0(<7cuKw=Giad)!~H!2tTn2-B*SMfE(@!O0C)@`+AJ7b3EbPucza
zZb2bx7^<rO&rpGf(dPNX2sl>e6(9r|<8NodG9{FsdbB=i|JD7J^=vg;j7$IpkN{$u
zdMJbJW2H8)E5`e%h>gQvetu?u6%yRfU&xFQ7qE{ozkNV}!6|;11VMy?CMcFNfy9a?
zASon|a3GT55d;<^X3^<dY2dd}9Xga!=(r@k0o@*b2MBiAKo2OAlUsiO$hbaic+it2
zb_<7qqU!DKZPcT>UuW>g@M#Y>;Fo%|z1Sj|cN#z7TXDu`yt}H#s%i1L>w>sl9;6b9
z#^7hH6b`7VGLFQ-DE}xS40hw^qnYm%T{bI(mH=7ivy0;jdB0&4Jw}MV65J_ijri-6
zp3`-<PGAQIxoCReen-Q^_*l09oXMzCK^w_jWm{@_|H}D~<+pX<^!?TezcRPIgc3?$
z>I+EtR~J8N5)?Ut?iv}+!Bm#Kf*-|@2eQCjw+25mQEEe6j>j5VKxGFL4A%2yw$UcF
zp85}qK~W-%#<y-6r`$EbFOVC-c~f`p+=)dM7_`q^U>a2mI{O>T)@5iRJF@CVLL+c9
ziPpZpOw8&u^W((07klXIS>-L#%s8p@ll{6{W_ITdgx^+hg$Bkr_2%e!d4?EK%GR?$
z=H1Xfj1!dp<Gk#4&#PmGPkHtU?iMljZ;GIpBFoBz3gU&pTGx`EVJsvQ+iKt|-Cr89
zg(MaU*usa(c{rp)x$<1-C4IjB&E`Zm?-_I>W@w$R>;#&nK}6997o7bSRfujZqqVn!
z6%MMN6(?Z@*}_#Y#X$eM`qX;(+mP39?+Z}L=*2$*?FC@l>3ZY5`T79Y?j7H4VjE#l
zG>j-iFl@Em23C<(L*nOmFSa+Ydce+}Prd4%ei)mh94fP6##Y<En(-6Lq`k7&Nt<O-
z5)<eWK|nzTA^GynSuN1t{pJHO^o7pQztv;1_0M->vb&79bk_{YJLvW{C^G$Vuh9-C
zUtpQa#aDl731!#)aZ-#}(7WVpNaMC>T-uV{dJ8adS_ypVgZE0GtxG?wG|+~coC1_|
zeYoPJJd3y)Xl^j;YMLW9<%!HqKVw>R!&w}tw@s$@P%Zjl0HTgQHB-S}uVaOyDy#6I
z>IBzIrc0d~KH9px+b5+v%%X`y6U^_VQ~}wXhv9G&P%=kVb@@_6uPzh1X|3NR3eN|^
z+9Q)n9UR-IIau6ZceJR959Y&+(jIS3hz=xN)_EAZk!@GtCnID0B`CAdr8lP|QzQ5l
z6cKTdv*mti)1j~RzvZj{-WPmY_zen`kgnr;4=^|5POGG(G#~xMvaC6~Z;x$Sn*5SH
z^u`zgzr%!1SeO_>;AMc(LeeBJoYS>zy)n?j+P3u0tUnqtQshCu*@f^c74GfPN{PV&
z8Dz|fDZ2bFoW2A0WB4^|!yhIbgYp+bJCYg>pv>XN<Bvd`cabbtxgQSL07!KN!R|1%
zJOj1eX(HRXasVbz1CGT3$j-tPc!xdR09SFUg9k@d7tFLhUTn;(36#8sX(cl0(t-Wa
z$*>l3s3ooM6yM|#qKLG*!?jlloyD%N6*e+F;`<B)qK+rds*}8X71TGJPVF6Iq{0t@
zgg>~;=lsXpE83-|R9qUPJ9HTrtLSQba1%JENHok^eVOjHCT(Qgl=M7rlw{B%lCK<U
z-s9+Q6|tZ*x~p@adhptP@hkt$@%|0ePCJ#(%jHZ@`9(khIACCie{QLsV1EO&<vVHY
z7%7;#>#Qw~Tb{Kd!+jtNfA9uU16sG#aOay!$Ns8Dv}N?>$S-%aeG@dxZ=-I)+Y|;y
zGH}B7E$Jca#xPRm==L{K=EJIL0#3g>tp!3~H<pqVxCe{n4u8ATv=EuN_m}o#851ST
z<DHtsrCWQElyid?;7W6z`Tv<e>=eU;#=E^hbp~U`&Y_6;VB2KTmGc+DDM0rxclaIY
z5dtLeN_7Ap`Dag^Ll~@>E}B5C-TFVCT&M|t-X7NyAYj_>hdW~ccn7o73EbnFtvNpE
zLIW2%TNdX3!+;3KfTi<K@@Ke9j-gTkk`82UTRED=fZb+D;oIW6{$}VVQZ0bF&P&}R
zpjDWMA-pPpF;aGqz6h3V|K=P&?tCY;&gtV;ptVyF`Q0<2Ro|FM5ICq5z%%OSXAGU7
z)0NhC$A5WEINdjzt)&VJW9&ovslx%Nun38Q6b^wBl$N+W7BOzb5EBf<*?mcPu_S{O
ztaLpP@4Ik$?B@TG$54kMkQvz3-EQGE2kmbhT!}p<>hJ`t!9Nby4`S(v;9{_|D@TYN
zdAc*vE2#(6JQprSy354EUD7uc)YLy!$*imat8szarx_V}yhTnF=b^E*5@BoA^^4CW
zTxMQXTRYo(cy#N`-a=+RQg@%HIU-&a`Lz1tSrOsyN=Prgf@Zs&4eCAD#p5=B+x(-<
zvB=nsz;$)c$}d%O(E%Cq^Gvo2?#vG1d@J@Z+fVTqK06Iv5hFW3W57bT2+-7bt^;_$
z{A&abWya4V#9fE8`wa?o6alrSbE1VJB@Vd~9W|+-a5_)Q>Q}FhV1enzUlZG0I3}C_
zAR%Zv`N4N?UhnCJ*Lw;;=WEmV;Y^c+J`?D|XlUDRzcuquBNkM7(bzZiYIu&=d2HF0
z$IfK(fS0D&n{qT{4b{we4Xn>Z2&QIqkN$@`z4Hhx<aZu{MHY3M)|;KYBQN-A5ALTv
zzWenrdHM_A2P`<{(-FTwaZ$*-vR2ryy3&z*XSB3`SbmgRZ4n?fVejwIE&IfaIsw(c
zL<s=T-Ej%}<>FwDYHSV^bol3l(5~U}+{8a20x96;Q3*%A6P1o}&H-Jw2_$l5PkFUk
zU~UII%L=a;SdoG_Jg7^~Q%AerfU9#w?y+RpBSYSj>Bxoo-u1pAV2mYvK4fd{yPl%b
zYSW|==tV|s)cI=_%_V<Xh*0=rk<(4UsYKm>OU2l*oqNE>(eJn^2c`d65QeXZGXSve
z??h_f=PNgI)5~zJjF<2HkK?yWd7iQK;bvBUumg{4euu7S#E&=cZ%Vj1`cYpN<(2wj
z%<8qbY7eX4=6v70#eOoApj#zzzs^0D5_nS<!jJv_TmH!|lEeYH0Zd+#%ioeUh1Z0;
z$eEy>ywsO;+(rZV5uxBp!#>Dcs8?_pR^E~BG?WkgJh!!Fp#7=a_;~s=E}onv%f|!o
z&!nvhc4-*^|KGam=xhMcVxs|3Qv=Zo*D5&e#K5+n8_&)^(U2ulW|@0=vNFc$Q(sl@
zX0hs9yz#!m)QwsrxoxsaZW|?EVpBjC6m1lA0Ry9hQZJxA8_hQU-dv*pH|G1->_0ra
zTwFvM?C_VvQSk-wm+t?_bFi`c+@$byI+85wh=S&XeixszRgn=6cZe3(I6VG{ZsFz9
zNO5&bld?Hg8=IKo8E8;ah)9RF%w<FNtSFkv5*I}~GAEbyhXdIv{9fo^j}IGk$}jM!
z0{~IQqHQqst8RkjWEg=XsL(~6)<2O4!qE{(LdSO_kA+bt2zy+>-}!M(Jh%_sf?eD?
zg{|rUK*WP`DalcgNS6%yZlyiyJ!~Xl-}n89;u*Qx@xsBwW3FhU;1E3<59DHbiI`8a
z^%yvvKW1S6Q^J7F(jSR}Sg+K}Bxt~y=x{W|3O1+dcO18ieWiG=)rs>HYnm)4N&5V>
zYwkCW`(>{@aT`aFrzl|ObZNOGbX6Tr<|TtNnjx%^j^Q&g3}DA74to7;zeHK$bT`~j
zV~7R=EL!3>!;SXPc(wZP=vIzl#W?)}2e{!zdDi;Z!Bp@(cx5o!3L(9rH-L?Qjt4ey
z@j<}N5Q-MS1%po*;2cLIubknoph}=rsmy$>RPP=&or5}12X-L<`y`6%Jkj$)P-NSr
z#1A67c!9vY>8LBJ-GL9e8#<hmw~hpEnG;nO>(sGcak=gcgJZX62OcZ1K1Ix^sjcCu
ze^z_=gj2hkjxXOvpE17NDfV*v@}YmHDA6A*ZVr2TV_GQr#Y8A%K!L6&o>In23}21K
z^gVMpcN+ERlk$gWz3CP{FY^E<wUl!b>f#JP(y=)c05VuH7zwAluHSxVzkNw^Y~<r=
z&&`Z3lZQBd$>}1$Pr&NwfN{f;CxM5)jD((y^j=qO&B~*|8H5JwL|fJn)XIYXz=%WB
z-dY_)MO3b04KjDhdzvS}_Av?s2OiG?O;KBhm{>+9&=y57xY=XOIlQeoxUQaO^+maw
zgGanZj#2pi+lG{Bz9y;Rh11^O#2}2n3A`476*Kf(As}`qSO2$|1!LbGM2GhE5jXZ}
z6thA)M{>|x)SyM0DE{AD`QE>N3RSwE;Q4r|?%TQ}&v_2xd+Y4hAv?E<hn?M}HBjiP
z<~O)i%;zq{kFkQK`9LlBcxz_Hj~s&g@d4X{6Hp0Vi}*|T70or+36qPmu5Rtx0a-JU
zp7B7#V8W_cR)aV*H6RxHbDt6f&5wc&Z~pFm^3NoMb>qB{$`(7@L5H#m@m{iT>rtwV
zuNmU00gFEt7!412=V1$Q1IQP7K#rnT9!Yo&V_q!7mVIeD;n;hpDE?nU!c(`-xSSjR
zI->E$Iv2s;*4`oq{X@Z0gf*)wHg-p?_4g(M8@L<e_XgJAR@ixl28&c~=)jgqr(~|H
zjuc7on?F94{5|$qCr3zYaKE6CkZ<X8ZSs83@dHD_Nj9$|$Q5)d<MV{Iv3}%o)H<-R
zvBG4qsH4*AZ1*Wsj{_#`racZK641`ph)ez@AvRDtdEyb9_CkYiR{WwE;xF2F2>?aS
z3q{yx3|=s3)(HF)+2IBl7b*Ey&s3=ae@WExurV&$%8_})tXt=W&rMA!kS6zzqE^GR
zL)Zwpzvs_hLEGHGX1Rg;)OsL|;qoVE51Se0N*}3)+X4LPQff!&2Y!(k=pA;Oi(q|t
z-5&P}`DykkfDawK!T^mCn!nGGaj&1y9;CuZS@os2e_7uV*Ft>hcJD;<EJD5#>u(;(
z3m-3>FI_78rmrs$Ypmp6Fy-zeYEy>OU44~g$jY7$Uu|1Fj?X0j^Mf{IL=Q}@aYDTE
z6&e?_*BiE!-6}Fs4OH6bg7w?|1t{{te@^JsmJ15BiKVr5SP*VbFLW{JMUT{K-R~b{
zcHnkplR^tjBQE4iV+UjC%2#oyvB}aK1f~ztcml^}@oosoo8Nj`a(TB83LXW{)~~5q
z7&(L=5^X6>Yd-_jAghiXRikG>$5ugv3*J{gv6zfvY2I0%s%|U0$=kyp82H#F`q_qW
zkCqha9PAtdeZW!H>@e)<M^Ic*_LZm7{Yxg44gF&<GjZeLeoNe;8TMOt0zQi+<cdo5
zHvQ~)N+84grphyx3OXlifo-$1>Ap3m>xaJ@?hGWHesYHvojg9G(D#iu^L02avGu~h
z!hNY@xS`G~)+jFJ)DMhWrayjqYrY{@h{Z^7bT{}_G%(_$b4;`S%Q>3i)7IbrnEf`_
zVAwmRJuyo|LJd-Ye`?Cw_n#1nkYDZooTP*mP#1#)Jb~~J{sseYM1BX6er<9C1FqfG
z#HX$KZoa*ixV&l!I7v)g90yEzV`0%yBRFvL!DL)jzX%XqG+VsUaooCz10?D3AS(6h
z=r97wLMKL!5qR&-!pJ-y9On=&CvP$EJj8lFOC})y<-M$l^u_!Z_lD``R?i>EAMA>0
z|E3B1>)hLW+CWW!RRYqW#UeLI&i+|^_*seB@N}+xC4rHKwjJ)oD_kww?14`u<{$LN
zYSnX76<iS6#J%Tlk1u$Y-X!wKbn*E#+^^TR^R9Jgm2Hf{uXEfTBN6mIEgR@LwnU!K
z`S~J0bJO$&8r={niau~rC!P;*n31Z+|3U}#H!+@gA-(QZ3(Pst^T>LFPeSm603ifz
zf>#_T{Nuqa&%(?C7y`%^e}Ptu3Q|QVl|Rf?RRFj{(=O>&F=(OZ0p2_VOzC~=;5)~M
z%yeK5>e&I>>O!Jx$m@vfV|NpBN(6hW<*8m++EKVBb?Us3z%j33wpEK?@bf(8O_DiR
zbKuSQXRmr56?W(^&AoOw=XUU}V!Z*>CicU*JAKX0+y0Yv{kbXr)Fy2`f2~apD<Rl)
z_2NgN)x~OdmyuWHnPBrf*_A6hv&i*|@Fqow-i|g*EXUGL=^hNwMC|0B2Kte+T^L~n
zIA;$yla?kEJXG_xKT8+k_>rwr$>^twzziky6q9`ajg63V;*Q=k?d=pi>If<&HDj2n
zO)iZt0d|)N`*zo3yP`(B&Y8zYvnw-ne1U?zo)C7Avch;M3hX-e7({)A8zCD<ayCsu
zA1yfu3kw@h_W)CMD`15I++x$Pf*dTM^RiZfJDGMi<>bexky_{ZA=Ls<`Z2;TQ#B}|
zYhZ}}`V`sZ0qj}RDs$<r3Ey`I^lYkb%X|F5xn^GRAFDkA4OYQ;OAaeMWdUd_fztKQ
zLKuGZ;s4NVRCapGT!KK*)A>yjX{1*2lGVv%5jj`HSbU7_?92&CCb&{D@a`M+j#_mi
zp3A%fuG#k(QRYNHa;5!%6Vz{FZW_hMuXT}CuAPR7WYxND4GClzlCndJQ+2JEi}Z2P
zFIA9&2d0;HxGImncHnxS!_%wZq{G5U`2CDvo=78TW&(_X=ytX*Fo+z-;ykNXW-~O%
zaoR;JzfqYAa!~_4KJ#8$ygXzFGD@fUWBW03;X?E=g?WHR6^Sm2K|{i}2?mqx7~R|r
z{+x&<)VM({B((~&3BLYn_Z5(e)YK2Pt~A!@zmq1Z`&@s8J4Udy26@JQ_dbsiWR&AX
zV~7A-aauwXE&xd<tHSsE+u%y<>@H$cg8$>{wVl??*DJsqKDyrt%zDHYC*n!*`ATuB
zHg-QQSlqtsk$=XA{!39nuA4-qFwS`4Wn-;+UmSD|=YRV7k+)3jH-ZWF2;uxlcq~0Z
zAzV-knIp&DMb^(a-z{rRxR=HQ<GpUdzmz;Jw4sQsu~YJdn|<60_1=-vri2&T$Y8KJ
z>`?%ti^CeVAuv5iGnv9=BV$>j)Iz}hkT^9^D|W!lYZ|s37iZ+Jw4>*bw&UV60UPHC
z)IDlMD?rS(!)Gfp7xg(f03500V;2JAU||Yh_CN}n`amR@@6-l~HKHWFdH>$4!%#>t
zbIoESpmu(B!V>gS`BqN?&ckIrl_NXz;Xv2ez-3e!)y9-D;;$fEr7h{9y?>tlUptgn
z;B$&Bbg;8Gj@@+Kxv{R@S`?_LRk8RqJ8IP;8Zw+WZG+)Gs4@Z~;4^oWZ|~80)8+9P
zsYZy;?GMSfm9!j4?yqrfJ)c~+k!e5RHITgTL#F<QR1`i%H6=E%p39siW`E+;{>JEa
zXrq@N9wcpiG9`)aRskpDzA#8#T+5^m=-70~${058q)H@4d4{~!Jq!cb7BDh`zOJ?#
zD9++H<}zXGC01)Mz0y@=YHUz!HhqmH3_#utS#TDy23_fv^^t>+B`9(+(742d&4_1<
z8Ij2t>|_N-8Db2BYtwhtE2#AWK?DM}5mgg5i`T?TP6AUd3J8QfWp=S(0OZ+WX=z#R
zM~OuFK-pcuG>S6w3SFOJKEJ#`vJh{m_tYxdUbt|9DMpkFUxKCp&^M#bk9u&sjzC>?
z%wOq0Qu)8tqBXlEf?3+n2jELrQSjWB5B|g!G+MFj23N%eO3yLJ0LsJQ-dl2FnZ^Cg
zuui3YeowOgPPg1LWn<)~>nG&<=~dX37c&xk#|<h3GYJ6A6T4@tD?-L`gj22W!lc4{
z;0t86{myTKisjHxPnV+TqFH$jR#_Oe_*vlK{#P7>&t%D?nd{|P_Tc)RPhK}il5MO&
z=09%ztMV#Lp?hPW+??S+uqu_7zaEgr-FeJvD$G|Kk-`Xkh&(g+!N6y<Gvme&64dJ0
z7@I>Gn*j^S&U}jlfoUVameWAJtpwQ~tO?hFXY(^Hb!H6=@Nz+xGCp!4&ljUAI>Hm4
zUT$2swUpyaWd9*|tA(y>oV+n!s^wjF#ly>}XG~;%tjPKlRjZG-hV1|C8p{jrUgJ%=
zlA%x_paw;sfxpi`ij>1dErE6t90RYCH%H492Mv7@@})j6Hs#ZJPh3|eVl*glr)ks}
z)9;N+2Xsb?X>_WuWc`u%Vy+iP0*0F6mLUnOnS_^a-c)(O-@`%`F8#Ra%;v-`c^Cfb
z{jW}RI9`}=zXt)vPD&mpt^0;?1V|OgB#UNG*CYzNQwAnsyBhnYbpHNawSxe8>v-yy
z?#Cw-rmzJ`2M~Wa15I%H-7I5^A8*1?t+0<xTH1`9S(5t}X>Q$BSh6E?2Kd^NhRFB@
zt;bHEj^l$+-2tpUu#m87G7X*vMrI=zq7Q-L`LUeR3jptsF-uz}dcPRgyhdwiSoq38
zeCRRoMj{?PIq-R6s}O~U&8$>>7^neeUVBry`QN&LRJPqg@Vg3Z1ucY$q;OP($$lX`
zYH}t{QHD95=o?FX>A~V#WlWb}E#;)^D|mizV)=lpt+4KqE*s&H(CA6Tue>nE9p?F;
z!Jkh*_q%7&cm0c}xZBZ}*F*n%Q7zY>$UEZ|Mc>yNq0;~?<5YpDEmF)!J@flSAfCS-
z<}i6M(a3#7bDiU`QzR5!;h|{63AI!+w{hVn0$HImYJzEl?le2l(>JKKs80fin;&*P
zhWHXSAjJ7}fh^Pp*04Uex2GGpS3X;-HW^eC)t%redSxA!H}Zw-XGwtSV(r6Z8Wd;V
z+Wu~U+75!v<T-$IL2}4BahzIh4nq>wDD1kX0WFa0Ts+Zvh2g$yjU@+p&phKNbXHLv
zhlYWt?>%sAgEXLgZ99LB<L|W>E(rTK(EKUDQvc&$0eoUqDj?H*wa?74d?bw*#8<d3
zABgrnDae#y88Vr<qk*aLjQP37SD&<UuIc=0Ds!8^EO`rqVA+u>B6L+Xv7#1#yZ~_5
zHfZPC7GxA2j0x;>q9}?!nW>YT4J~cN3^N02QEm{XGBmIm0r`!|5*KwJpIdSr2WURG
z{M1L+SuV@G<S-`q8bz|eS&)=Dt0Bsc9KQUy52&0u69f8hj{xOeFYKr@R*=|S02@$o
zncuD>n}yW3mql7qzpji*FRBnEtwxC92kQQK<iW%2ImR|^jz;!xRrbpcPLy+aSWb|i
zf4*Rkmpj$y>E82IxI#g(kt(AxouHz|0L27xu8+zvkf>p)fRacOT}t=~{`wa`g55|<
zPDgX;9w;7<UyeiAfINH2N`txZXv*vPkMCkdk30?<NwQ-aj<;!Uuv|HyamD@g17jRA
z`Nz(6W<~=BJfLW(eoy;2lF|Q7j+Nah`&!6S7VE0RAL+S{@6k*A{e`YClD3cll#1~p
z@11W$Jm8U%JB4D<B2nsr2DV-})>><HtliyLfB}m@-53yYn^%8Qe+XFr1~g<45xJnC
z5CqA(j0!yXj@CD3aBEnbhyc|*GT=`GHluvt^mQyhUlQ3{!ffQqIP#1|My0TaXYcA4
zY)1X?Lsj9y*A7p|R~Eisd9%$_sdP(n{@wON?qAN)*S32u>%sAruWjT*2--mO{y#Fe
zzaAbs%R#l>>g!>mNffZ>4!TmQpT_ggr~Xng`sjn1#}`V?(zahRW7nR?ai^+(3%`AC
z$LWeU`0hi^7@^sTnFWynJA5W!n~gIVP+%X{zc*)!D^S}Xl(TQQroA;}cwYkB-GFR)
z@~}Xz>7Xt{6MGeN#4W`RT1;_4Mf<MzuCZXvT)@WmR2s22&D-^D7LuR=whsAOCV6GT
zRZIY@I!#dh?0(XY99~mxthxl2o161~pgCb(oZn%cI9_#6O?j{co^LJnd_U7~)>GXN
z|7oo;b2`<k;V~i=+t=5h+zJLYvX@;Czq`{q+4XSr^cfC(e2Nl9zwKZIq)&=WvEpIi
z*I^78^=CsZ^0EUK@EU<_;TXR*Fg5JJ)#y?m@S6+3&|^f5yhd%X=gw({CxbL;Rr7DE
zBXT;;^1R6zx+-aMf3Z<s5XhW>ZsLH$>16n4woulu_+LwnA|;f?`}be=lx=;-Ph!Qt
zOz@=ju?=M|kOz3dZ}d@M<yqyN7Izj@tY-^uc`x}%zxq?&zRZl*;DN7KJl-VuI8EED
z<3tc+*^E%!>`{Cmw~|coy)YwWVjapXcE-QD5i5I9N7#+oJ!wYzb~mGJ#0NX>i+U$W
z7oUWcRVZDScGdE4!CHKne55c)j`hRCeqkMSU?awgHMocOc5AO1Z9M@G;-o@@$Cm}e
zrYgY?s2zkz)~zqG7}h~X|7}n?bV9Q_O2tl#e)n&2f1p~ogI<Y{bffQ<@9U&jz>l<p
zvh4x)J|>AcSW{se@w2{~FkMnJ=fZ+!N99=FfE!2d@>=}0_PGd79@hF7Bm5N@v#YUZ
zOzjBf;UAf=fcK4f!s#^6&gc2(U$@j`ueN@HRY+}5ux!ME*BQU=!Kav`$2gOymQB|?
z7}YK?^w(^+YH)a7F^Q)-Kc134yfl(9vXw72&LB^I&z+vvDHiWtO$wgdz5c!catfEW
z@p?aFA4Q8*!S*s+wK(No$udil%QRHLt(QDcQWTw@bGK?y7SGabe=Qyh`k?hf_*Y59
zW$R!1e9OPdRIYCfe)G25Tbs```pRHWdLlHm76Y_>m*i9N;Kr$D!V$UtctheMa^dwQ
zd@+1<>wJ_l4WC?uK{@BOhhSN#37uG8Ny)J_(5rn|2H&SPaPQGQFhQK;AU4PBAEHTZ
zJLc~DDHnFDj=p(uLbNC$WzTa#IeedIzk(hPGHoh(tj^~?RAgtt`uV-Y4zAZ*Fzs=d
z`#<iUtMbkPW>&NPNL8S4af+bU@yCPT*-OYQqn5eP>uN`<SJ9_zo3P6VK`Fqpm51l%
zl^0cL&2ax%IhUzI!%J-K%s1(aAJ(sLd#SbI%ahkh6}3rhQmdRj^PsKdwd<^N%}C2g
zLuznBM4^(MPF_dNQO$GUuag&+Eq#?5(%bd@nLZS7Hck<KdAAr@2M0}F;0|&Uay^LE
z8E%A3f}JGa$KXXcs>2vcN=h0A2HvNG6zDO9tFiL%gn<Z82~0P-9#JN|h(yfH&=Uv?
zG_e7p5Vx>!jIec+)w4z>P(AyDTto>L5G8-L`=WQ)NoLq%ew61hx%Tsm`n>H+7Y7QK
zB=*^!+lx6vN`am*raPL3P*&C&^IKYrp8GZdKKg_H^Vi%djHlPTJ#YcZ<*?#(Wj_Z`
z!U<j=`C79b;Ip<V;n3rI=Lc2HzV?XX>VyXp!`B#IojZ|MmZ~@8*aJtN$U1*T9j?@s
zS8LqcW4A(RnCaEIj!G?u!uOcNUed#0PLl9%bKX1P5cbr=>?5_psf0`=Ho40O3R0dr
z-R)FAS5FB<y`&WiTRj4FpB$;?;-Ux$8ROjI;2Br;mfEyxz~C$!^e@e6&+LAtxqGZ!
z`4#cyB=t{1_;d364-Q&X-S~AlS2b2?Suz~GG7An2hy}A?MPfTREo#7S?mWN)?S5=Q
zPn!>d?vbYi7)&S;$5}QGjv$}SrE^Un>Tcv%uLKcfq4^8xT#dAQXo2!ZQM5WX+n8`h
zGCaF=bW^>1rq1ah;|byCKJ0=^cDj0Zvo;F%)~fR^@%^^t$Bu>L)$Mov%;rS=?}1Xb
z-idH=IeecydGI0mZLz)odGW^>{mFzh?LM|#DO(sbJ>z4?zjE$wk%stFLWX8L+8TXa
zl++uep}EfkQBwWaM3K`0*H9HoSFxfenFK+MuH8#c4Ut$5jw8CG&LQJwO*3{zA&kkN
z+B&qmxYZ~0jLr?m&pY2;B)zUTp@^YdFbj$-L@!DfAOG<!M#K91YOxkCXm04Y?K4pA
zs*5nvkGe=F=EMq3s`^ZvX&=1b0?^-I|1`(}=Y10c)3zKCJN9S>mEsN?4{gTkEV-Z{
z_r||pz^IHH6w3j^Rt<x3E1(m}0i8%25ZrFLY=Qf*mPO{q`Vi<^bb)uoee&ez!YT0J
z)V8wX7;cDuJ(2pQehb9>cPh-t!4@VDSb?FjgN8&H(V1F`(~OQ%p}O})R&L)Y_K}|G
zc^c|<>$jaFI{r_NvZZpe2MOnHX5Ml1qeg$KfBUuPNak}JdkU9hV-^8XfpVwF9Ru;O
zbiwuJqs$!>--_D5$7!+$XHMDKoo~vL^xb<|vnn_$P$uHAWjVd&fZ=9t!gC`s3E2j@
z$LaFoteL6xPp8|<Zd{3*4?j9Y*ZYL<1SyFNTb|HU@E-{G2}qoPOMNbq^9%;PIRO$|
z&LKQ};uTNz>B;<D(T#xLpwNRZUA0-4{3je*CY-kT*$_hbgU=J*h<4V~qXEc3TN?in
z?u=P6z*NcB$vvYg$t-!}n0m6X;>FT%NuLcv3N1tEZin<>{%Hr(0#IzQhxOBpe|f_G
z7<^l3Xr%#KREHr<F4`tC<wAMj$;Iun-+>~=b|6w#D^m7DnxWjA?=$SvxBQtr`o$Hs
zikka~m^fd193otVE-X0Z7#z|KKe*bGc=Y;^)&BL@ZxQ@jDa$}O)sRC$jC*^XoeshM
z>L!edG4Cj{bhp#L?f3fal~&j3hwP&h=bBWrtKzQ5ec)<+h|6Ts;%B`Jwdoe5yV7nS
z%8o~On#P?-*w2rbkcuj{V4(s6@ss6P$7`2Jazdlo;*PlS>U(!vvQ;QeJUKTQn?QJM
zq&(-&Q{0`F!dxbUgeI3(r@bz5d7M_zaOl>ZGX#RtqiN4zt4iD@MNbMme_s+bwGshR
z-5f3;X0WwAwN~o|clb&ve0+RZ&L&-Gpl&v_4U4ArN-wEbxeVVHkNoYopqU|^hID{W
z&WqJKeaC>TB7usE>H~6;RRoW>2jJzB;^0CI4D&qjb-i=Z((=})DYV?Hq5X|`<x`7T
zE~llDYik?TFUDWSp4(*4(D9m7{z#@B5=>N9+z@ibY4Q25`C&B{N2z;zFs!Uc{ALM+
z_A$aI_@86H5nB&Svj|@<zIk&R@Xf@zncnuB4@EbZ^!kxVD!p>SYXka3DjCs`{F;-C
z88W^apn}PTS4I7zbkLm|9&~P06h)!olPu%-ESCOPdFvk{>K;mNeqfR(uVN9NCzo%K
z1ARNXeS@U_2Y;!SEkQ+64Hbze#poHzs^*2p>RSb-fg4o|&D7)txbK)gcB1n;IjZEw
zRyEAZiJ+1jJY`8BAauarx-Z-A*JtiXF{wwEBx{-~QOXnZ2Ama2o_8cCmJYygxf-op
zi~ePq;eZK~Q#fh8>({<qUYwn<v<Z~s&SwtBYNfA#12@kWMi%6s*zh41N&16_R~s}M
zx5BLFc`S&~OZ!O~8Wu+Rh88@4FJ7;M7Xx{djnA1VJ?Jx3d%Yn36My^_U-Oc2owv9f
ztgD%Dg>nZCeyF^_E8iAiAgUCxSFt=`k)$Bmm`HF(5#$caNgsGMLM*#UOlr2mgL6W~
zBz+RbJlS}Mr@9KU<u{LKoXK6iD&z?UkuRkD69WBKNT6_8zlpd1pG9-%@=U)PtY&5Z
zb<xJhOcH#{Uc7t#stIv4k74)i%MA_1hV7$V!2*!_30|oxxN|%h8nWtn^}8QSrU^6c
z^UF0->}&1JOy9)vcEef4f5X|>?fUFIk}{v!eQO@A`#k5ru2VC11|6@xO2li<ap#^Q
z!vp@E!snRj?DGgHd>p!)PnrnG5o5(0e%zuY6yc1DN;;eNp6P*C<j)h2L&#c_Iin6M
z1#8{wED$MfP62n7arL$QzC-ZlYUP;R0#LQ(f|(%|r)kZImJlUD1p;p|qqsel*1805
z5x04r_bYi9<!XaF3v)uV-b{>)XJIMFQ^{Lm@Ec<Z83jbHKc(Zq)h7LvKq?|%n(jOI
zqZ`noP{qtQ-(iNAjqKpm<A9u4b#wh$*Bj*lfhj1>^?)&~x(06!=pMOa;L1m}4jjhu
zl&f`5s3KtyfCa}1FW1$1>uYOIEZ>8;H~WJ#V}<gdeJ3@;$!!j)yXr2*th6pGSI>nT
zb+%;7t8&Xj>E6x`mvDd`p55#d<&Wq>{{2@mb^}b&I)QO~E;ue5me%`0mBIOB)p!U#
z)_oWYsojGq<Q1@xOM4R>I!S1lpOHL&icH!6)+69Y@qQ=<H+fDb8J|?Mnj3QH5kmE+
z19*@19a)*Ci@$iX2hyL&HN4DxO(sviXIiXcAG`=Ml>aizUzle<UcajA-4q~hPFA4v
z`R8Mnb9Xw3_WU;blzKU`H@1jvXprVQ$NXV<O@gnwox5g8QMP)YM%L|}@F!U*_lo}a
z0bW86FjjhEc@KJkTFPm-2bk*j5cdFEj8w_t0sg20mcBYN^Z7E=`3p#JL19E1oJxn`
zt?^;_(~%26O6;{q!u7a=+H<(RTJx8NE!t_*GYMS*c!|Exhq>*gL$gY6V-JWAeMpW4
z_Jsy4wY{H|L^si%P5g^c**PCJcb9nS^ZI<}Cq?xdKv~*4F?wQ|P-z6*AoO)f2Cf*T
z&&r#(Z<DXR0H6Lh)II{uB3$+UhcUk9K4;kvre=*gchqU=g_4Gud)*x?UFOVJb^GSq
z@(6P@?{-^Xj!s@G?hmaFO@%D=BW@^-|EH^vg`gtdKoE%MCwcJ1W2nBV25iK1M*!Xk
zTHyqe3|o2%8+ncMI}n3EA}kQ!x2AozNi=c^SJLfavVp?F_Wer-IG187v4fj&2D3Fv
z)sg+VLy2Z?il3T1bXIEQ;!iwQXuY+Q^*m=Be!qifBt$+%>7<XEkLCOyjM#ilRyIYG
zN4TWtT(fq|I4<e98mnUX-uog2xv@_KVpdPnbgN{HDT&DA!)GHZiNgIPH7c?NsBD(}
zbpEorvbq^DAtA-=foNFI6Wci63K*}hp~|@H9u7i$iYSWz1{(O&Qp|B;VC0hplyEKY
zjfXb3H`XsewecqQ6Yzo+tefJ5M3I3+4q$D`Klz*Ls|y^yR<zsy3<No$E~q{@PeuSv
zd<)qD@trzkD9I%x)IPzuPEAj*$}j9cD2?wky6`qKZ(Om)UG$+vTI2;LwGLx_m+Eia
zaTZXKspejK?*w;e&>8xaIWZSK=+FQD=2%%3@Z`ybwRFx>;6;sQB`g8QOAl;q`7YQU
zec|-|6$%X!Ud#MaH|AvXU0^n_2ym4yKzz6Ao-795VnqISc1kivl)t0`cb}K#-v2zW
zL0U{mi*#n!3)5`&Q5KI}AI)=Xj%kvev$Yq~iws1Uc2Zl_xRKb!Kxnr29XqCmWvreH
zeZ`(V_B;1IRrO<R5TkvxKNDdYTZz4wh`<0tuKM=_x#Y8YF$@MvktEN+MbcFoOu7Je
z8?C0>2VY>064OG8+U`=OqGh+vAt}L#4RD9Ka$91vk?c4uei(K2L|>}pNPTt3xkrZW
zYlQshPzDtU>((at)_||iDX<3CI0f)F!8(9{0@e^a6@j$K75r1=2IW$J17aLNN2nPt
zatjL0Pq5WK%)RjGUjCu_0P>a~sA@jJ$Ud}L!6~Eu<?;Pv4;5{=r4?zuq_DZwUlDnF
zjmI-vUgK)j_gJ?{Xe2e|m9jy(0`;{;R760^|9+mqpM>7}4QwY*CKQ>!&;V9!95`b%
zfu~DWdQep0VUyd`z#K?HCccfx&>vzW(+Sw#SyHOC1krB{m<FBTxZ0E?#13v;x;OS8
zZUN__)5^-t%R>Lh6KkXQhphhz%)`gv5|+X0h(fLV0WFfQ{b@NliHqw`s>?1b?=`LZ
z^>ry=Z_>g7ImCh`RszqWh~}NDUScyqr;OcCsF=yvoWJKhN`JpE&StB7Li!W_^$hp;
zemzWED4RcFlOTJWj$R&`fBWPSVtk+mjai9k%R|Q-KtPF}Pn!U~vOIa@)wEnS-=jIF
z4z7$n+gCT11kEdTHZ+?gr(}|TwD>HKe+ofsE9_lzAla3$wm~(r_HBF1=7J*b{dz~r
zdV5Ri7nvkO=jq^R0*L7txXOxvTP=9@#8_lVd7XsR*qy56T>^Htg}Qm<oLn!cuwckc
z1n*Hh^@DfK^=Tv6O?ODjWB!djHA{|KV#Fojz=f)YPK#Ynt7q~3mi^?r4UdpDd61~W
zz0TtHvF&sBE+;g8R~PkuhHIqGx7pHR=}@u1w@uZF@_!@bhM%7xsrX#!HJ=A97s(&g
zzrqLpzmOu6k33#I+u`R_NH|yq##XBI8%35NnxIOt&H>2r>Ff#UP85J4{q(HHm@*%S
zw9O7=Naq@gavc-#b2(WqKx}-@)`+F~8m7oUx%f8JeU)_YUXkRztMrEwPa7|P%gVTJ
zn%B4g$KIYj?rB@jDzv=6J@>fZ5wiE$DQHR*US4H<pR|%vWKe!T0*4~<I@0XHTqoi0
zV^+WOT&YJr`*Y*>v@<%d3oE;LHCBcP84A{q1_bf~j&tr0K5`fPh05g^iP+cL#a6QM
zM8Vo>6}9Y<$!CU;3BR^cah_>|3-|bX-E~6$TZs!`Y_!41f)6;_2@I&2vVdAynLuBO
zL)I(^Z^*Ozj)wWlEYE(+00Y?5U}<o`sMKfE^M`8qvsX~8*nyNM<`Wok=D>qgvwwiq
zybh!<{^_rkv0wm0c)V5A76fptblP>^9=bZ-RjPiX#YrWLjGUZaD_g7%_w+;_7}FOT
z38$n8$X_?c^=k<kCD1I`Y69}udt7PX|2|bbOa#gAtf<#OvU)VYcJj*SaO5IH*YE-`
z%0Us{L@H$cR%BMo+vo+JgS>kAQEx8MUx?m@LPz#+$Azl`bZhM~nX_;IBt6SNNv{px
zA4XF;U~4%Jvjc;HIY-zM&{^{I(A)YS`<kCAev18aI=kXgRR;YhqPD7%d$+lWAI^jx
zB-8WiD6eIiHM2ibxCj3@<g??Xj1{u=38)7&@eAW;9mF7})P&^Bv3@VcS$Rk6^Dh7G
z0<G!yr80D@5V$j&xQ9V-R{Ih~f-bG3M!inMOP%_cLs2)_(i?*{r?G&_S!+U?I|CvH
zovFo@Z@!v<Oz5TVr-#3Piw&7{tO2vfp8);-joY#r(l>44){dN97}*Dw!Fk}=L-Op!
z*4iDQx?hyO>QOJR1Ei?fvk~6l?=ueI6c}P$Zyt*I4b=fRkdxoU@=<nj&NYAU6Y+TK
zG5<7}AXoE{>`&>V3beN`=|A`9R+pEBdXuw-4Fhpq8|6N3M?L8I{wMcY1mXoT4!xAc
zlTE+FV35uQegsBmR8)=~d_Q~Y$B)#$p=6?>)igyQY?gQ^IXRV7GBy(|W^+=0yE8dW
zk9u*5cL-pP2<&6cB*5_hm_u?MNw87I-$ySTWT^U~;rqMVgCd^oc2(X#E2C4sC(P=r
zUCLF>_`O2TwG%9j6{wP>05u*Uh}z5~Rl@~KxxDPKD%KJa#UOiCt2kwE`hfzD6OZoo
zvAd1uQkk$Xbx_Z})`j9mN|9j+wXS|%glFZ+pAj;i%>~r;^sD0;P^EmzRZ~XL9thI7
z9q|O8N_sBl|L|^r6PVkNpRNZ;5Z@rKJ%4d{?)NWCxllInxBlvQ;GBwv#(^K8-_HS+
z+P&6BCYUeZ8t@opYAk?atpmyf&XcDl6)YC&Qba7L70)C;pXjPTq<VHtJEDoEX-Z#a
z<Gu^Rjfh8V*|B5s5bhRxkI#|)Bi!9VN56gugEw!|Su#-Nb_b+F>^!%BiaLIc`qCMi
z$<mdXKP$Y0>8&vfo~#7`L|p{I;uazGP$}mFkzj--S=c%YNQ_!A3I2F8o24dHGmGcp
z(Y9ITS%zAkqpRPZXyZ@xK8jNFxXBs2Q^oVKzjp8Q7n-fUIe9lBnULwha>3x;32iRy
zZ(8&iNcWvOy^>L*6VXK~+rsZ(=BkIyC0EJSr>){Y*sWrEoh-E@3LZnSleP$g4f0h6
zn{VY6*xh9<G-gZe*I^s7I2bv>wdBII3A6irDC-KJ?<WfIdqKf%S|tKy?`=l)rn!U6
zwp{Spd@0c~Ty?e-K&ab^S-b}+5Ix8ikfs<{cny5NE>U9B+2<D04~f|HbGGW+W<U5q
zuk%SXBQE;~k#LU+e(60z#!a7>Qd#%Pv=3m%QqW=d%gmAg<FId`!}em>Ic&2!?_`i*
z^YRoz>)^UPbv{h}E_O|O-|e-d3b>I;=IpPFpH9Kl>2x5%`%YpdeYaA9*{JxyR17qp
z^3l=Ju)$a0ka)O*StWb$0mr(DB_>`;b?39sh4rda=S>GBr}Gk)#;rTb6C6(WAf!b|
z@PWT2O6bSP{d4=W?9jbtWd85HE<pFX^U&N|YILvV=-qJl+7mF{S6a0R^rak-B0w0H
z4N?x?;P%REHbzyh1%IxyKWt>7+?E0&864^hDvKw<;0bSd!WR_NBCuS6uR8ZTc<JZ?
z*&6uf>8pC+M>P(GQJnNm&(eLwy_ANVeJ70OUhwtm4_Y<$2<P6v>Y0#z*yFfFGC|{4
zPCu{H?j(n0u)oRiy;**1m}hbREsSJxVSs{^4PFDfP*w!_M$Aa1{@MfwW!IG%Q)CY^
z+>(}pAs|R=yp%5j^_%PZvV-UEIW`Dzcg%b@-6DUp96YG3t*spg8r;st@UFf53{9>6
zl@9d6R+&I-)`bd%Lw`)J;X4d}0wQ7C*by^=_c!aesu+N=qX1|}tku=ekLVph-*K_@
zp!Q+B`YmN;<tv{fr}7E}S66+&i;RbxJM`(LkC%%{95a5txLFh_nn+Gg{=m-<?|JhO
zW+;g92GbB2Qt0uRn}o*r(2SP4O%?deh0K~7hsXCanT|>dqywU!E0rI+2y-x1<@$XT
z$0+=-)~yBylQUgjz5&1vn0Cor`+9IcdD>0cpX@E|kdn;p)9q6XeD%|w+AP-|vgmo&
zd#q4ud~JJccH0&{!O6>k{~uv*0hQ&tt$|7jh|=BNr4oX4cT0nSQc{9QNefa+H%JR8
zN-7~Kf^-T3Dk-TbEsb}6sB7<a?z#7j!PsN1wfP0V|NZAX=QE$6KyN@A_rG9{6i3h=
zTu0O%;XfpmwbHM-9%LG!+OHJjtI))%xSHqNLO#*5IM%NuLTD;f&aWVtBrFZ#j11Nb
zG+eJw9~*)t&M@qRNQ5KVy1|Gl(CZQ=DjW(oRIxX!!%tKtM3l8eE&y1Jzixmk>@tPa
zxM7HaWyXSZd5vF?lEW-y61Tp7du}&>u4U^yEGF57W)0C$x%`i9tIl1B)YoH{)Xn-_
zLXCl0C&6H>9je5Q-y+Tk(NKYb5%BR-V_wJp_cb$v8w7sO@4@b5xh;YySc|J!)>8Wn
zZi7Ykv_070B*NN60h9sm&!Fqn7y^iM1z1wZr&9WJOJW?Eo53@^2si>VbOA>ffk4>n
zuCjMi@+>a3UH@C07Fk)@(3kNcxQqauJ^?8_)=o!92hLjPRKPDkN{C#zpV|M|#_~Sg
zuR|~m*ukjKHelKw<p-hxrPCOrMo0!2fq<GKvWhKKM~wh*V|hd2U-Qc14XWj=IF*;B
zSXA=XM-uZ|-qMs%YU650j^`*i*x0V67fGM)iMz|mp#xh%#g1z0a;V-@9c?&`PXQBe
zl_XtQe-Mj?BlO&?J{g3NN|1*u5JQ%bebmKGDPSknn?1c%q?)4PYwXsm2EMCO+z;Q(
z0&g6_6O}1xqfC>nXAbZ5BhH>(9EzxCcGsqh-~q&e!>P=XaK49R<z=`h@ZWX8N-yLm
zGpov2Zw3G5%K#K7wXlN=LcGv;<TzEnX$Rnop=)Q}lP|iZUI&aQM93sEX#82nP2pql
zs=@P^S(o1&^WuKSgHeJcBjgK+dmyvtey_(5hZE-+37hNQnYT@O`E1=rVitqHQ`?Q=
z&2didPm!d2iRt*W){B<;yVmoHJM)Wwj&7J5HsNeXS1e!gT*pJOr6C;6_X`;VevMDJ
zeSW^=h}wR(?|YR-FxKVWUZLWPGYpuol=7%wz_Q&zo^;_9^7BAXc8)4!iPQdgjh@rn
zt0;b?{FZZEdjKx%nX;P{RDUr9UW|!5?PcCE=G2fqc$ZhyRCCK*(Yx}3riKnBcYA8m
zqM{G}x+kibtCpBi$Fz6OF%+#6)j+mz;}g=^i05M7ah`K-tm`+w0fQtXVY49^J_ShA
z)h9$<tZFg4SkfkgJB2IFaR;|>n^P0O-K?WkCpy1Z{-2o+ZVlv_!Hx!wd*WqTT%@Z1
z=PSqn-AnP_)cUBArGa_*G|ZmJgw=<!(NGIXM?>3?jf#U)aD9|DplyKYZEGtt)zAsB
zUPizzO8Kn6b5|=i6r$wu@A7+Q+))hSZvmrJZOEB?-14~zwDZOg#^ycLe0kwou~rbm
z$l^}-Of+|%z48fJfWexfC&d2-!I^tOWJtnaHiRNs@uiK{fj0*Bz$5rCE_eHfwSY}4
ztrbl6ZA(<Zbq{>O(HXER)L(o7>A5z*`w@q&6IyUpAQue18e@v0My(<5V;-sjCJWaK
zuCd=bptR$l(sReZ2?!7?{f=R97}sOk>@V&I7B#`kL5-qnOVGv20FS>Ayqz=Dgd`U}
zzN?0Ni*Q#w>h1VB?hiUJahMcYe}|qWfn+FF-O=)<HVJ7kKZeM(`mvP?(+ODnWTwQt
zTE*nO{%FmNw<G&6vXW2m7g;f9)_SJL0XBS@GbkmYPLr=kqo%y7zmC^3r@#IxcI>~J
zWy|&Rw$l#RQl9o>RqtB;<7-8W1rE+MA=KMr9qUReFk*M3T|6DJul`rn9sE3MYJ|pN
zb@R0A@iy~$4(><dKhOCN$3xeX?5+*#Yv_73rCj*VkGkYmsOWtYg}<A=B_>IhKo0Xd
zFExuwVh#*roLvuI!jbz$I74$SFE5W+eH%^e^!}#|8Xq@2K<s)~<=(9wdk6DJ2^l(1
zwKQ{i&8@jKAG6xe!H^T$^`0>Q5!PCN$AV<Q$*n5%sf9<L8($vp;E%BIyBu2|td)NI
zO8QmCo_nf=5tn;0aK;<8R&iVG&IlH&4+|j7jX%#Hoo)yJIh)5sGfSS})a70qZd%F$
zCGiR9L?^%Q1N%&Dqkk76|Ml3sx2fwq0BeRynTm$NCJ28Q>ZuSLf|QLpjPbcaAZQxb
z-p(B_SoQ)CZzNA12V6|k@_gsopy)jTNL<sgQZ(7*3N65MFnPWt#3}D~wA=$wJ3{!}
zZPxN)MR?CS7~brQlyD2Q;OsInGBRJYW0(i~4K+mS-6jKG&n!D4M9K-Q7yMxM3BIlL
zOmDcZ5=@49{0>%%`iYq`3gI*d*M{vCm`ZL=n?Yub9rOWn1-LOu+h5?pFN4ctJ{ZMb
z;U2pr4fPToB^Cx?ldSkQIEQl=z>TU1sQRkIkxadh<z<jQQ0S_dIY8?CLNu%|;}Ft>
znjh~{J*f<C)-vX+CQm1eEvl+N(4H*bKL?PpMY0w)UYbWtp|Ze4i*N~X#L(A3Ha`dg
z4~?%purbM;IrkRZIxVfcaY=D%OrIpyaY<14RCBAXjSj3OIOz_vzhX7Z_zFl?1iHiy
zawyBh(VCGFUn)zgt(yH@%)?~Kr<HgA)W9TFK@Hbh1<InbnWv)AD2~pC>mc9UqI=o|
z=zjm;Ab!hC(VGx!tu{|!eLf6vWwsY8z(m2{O9_l`?+QGf0Tq~A>b*A5`v4-3I0WoD
zm#=A0#_9`}X6n2MidgO-ai#45Z}`;B?!>N}m4_mbzCEX0FjsCnKs$#$iS-rVV(8Jt
zFUUvyRr^#U$T^XaIx(wh#VFd%IyC$~75QX$<UP5Xb%x7Wd&Ee`sOOn?@n0R|%tci0
ze&Z)9BtA!b6+L%PzzYg#r<MKv6*>pbEuiw@E(Jhu!d!>3*AT7DEeDPgMxqc}TEOnF
z_!<&GBKq{_;Vn(%yDqWynq60PYCc)G{;nXxVcDa2el-_RTbWPL?ATV3*M%-ZEK?!e
zzT>?&*4p3-m4;<lbxH_GLZlJayBJq>#+x_FuR^K49tuD=0-?*ZIW`?d_3kS&s>$4O
z%r^<5kYsZ!UBcO=P1u(u-;DdFsEis)V0#evtEsjO51ewUwqaF>b==!ZUX2<hUjwOw
zAMB->L)&%Qasj;752%^jD!gHy-?_$;7y{SKxi7lPw7MT>W-MkOhjH+Z@xfBr>lxk1
z(UjoEMu+B~UAH2`-(7cuQlN{W^XFR2pMlrDAQ{EcTPR%18!>84aTg;b!imb$$x9k3
zyi*?Cx1}B2dt$p{zbctWI`@;}*h^vhoL;FCX}<7Vr(dJ#?oMv#GPtaSdm~%`hWt|@
z98fnRx<*7+@h1c38Y(O9a`YRySzkC^hKou)<6kv(o>mHVC~(<kGM*KK7cLnwupm`%
zIVf<-ocwE0gs`91H&GwZlg~w*S3z4FtZ@CJ@MI#E7F~)y^o(YBz0VFO1_nkDCi0>^
zYLqPED;6zhGOjurY3LE7&Cv8hCL6C1_SKh0k(ZSS!x_b{2eR#*fU$A#E4boXqfHv>
zDS{8t;}fnAzZ4{6-)wQ(#ko@z!svZzuz;8#Y&8vSk{f$sk^vc1mKpf4_DGRqryr0f
z`0rzfje11A@kn?o2l`QyYaOU%aP_nKHyOaLPntWKY?sfxAr(MoUgd$gt%l+$;yvFW
zR16&<ounTH-QOZxTU(2v#gozTPxgYuqR4O<(V-*DlJHYGqK@-SA`62U)*;vgwy`qM
z;79|@Rn{sc3UM(T0oS*#r{GNd*o(=;^TU;jRP>rj;=QI-u#3w>R&+4%J_V??^{^P~
z*YH&1y%k6mDS$P9Mtp{38g^97r*m#NeLU~ouFA;DMv%R~ybGS-IDR|pv$8z}t--$L
zsV;oJ5T4vh$=@T;pv_+shrthf2^jz}h_hr(BC$;tbxlqKO<1P5rFr?$NJ#;P6RnuR
zaMA0%ktqTL@dPK%E)@=q-V5VVUtR#)0ozr^qZPw<L>BwNu*n$MLTSD+zmJ+6bQs)0
zB^gGB>9YL=qC|>4*QTt5LvGM!aPRe^aziC#a^~AzfWgfkY#{YljMPGZo#;M^&o)?9
z`=k~?*Bg3>5P)UG#XP)->{)`E>%H>NPX(r<1(1b=g{ekjoj*@Jm7w&HBMz(xw7{XN
z*G6P|l+<lqPO|v8(c$cT80KF>*?)JQ({f0X(5f^re?sU!!W~RcdKZ#wFH{6)<+l*Y
zgEh?$_O#N0>&;ZE*Lg%`Q(Pp*US+FfewiMkdN-E$(jGPO=2q8BUn1+L1xp&Tu<XPY
zpWi`;!>*~NK!Y5k9NeIACvD)Lz0k14RiPj-j-PrK|4Ev2TH+CWG<NxnZGopYm5zP7
zO3ASW)|+Lqt$x<J`lD~`n#6mcb0n^(qM_kYi;g2Q&^ILq0ty}BjIt_#+#7F`^Sj_@
zoJi$B!~YVjP)iCIyAs0o!~+_AcEZ^tdoZ%j7=ancoLRXV1H7P>S)(VSsBkMeTMQ+k
zo!TpY7si0hbTucVvvVKN<8|byn~eOUJ*ev({D$8q6JlH%Zqw}6PCk#nyw2}a^7VF%
zT`2ll{B;uq7yMMJ_!gNX|M_aMMg`GT9yov*E<j3pPI*^JCjK!TOa!qEE=T1%Su6(z
z%dAn(Q+o6Dew^I)T)3@e(yD)$IZrq7G}Ls#t_9#fxy);Zz5`KxJfRF;Z;%}yw=x%*
z<t&Q%Au}#~0E@s~J6t_7Mt#82dsd>1#<BvNRJfd^_7QicD{Ot02}!{>aVD^hHM)~W
zfGoRAD;L%Ey7A93SUrVeeB*ZH#cd$}1>2|ADyxb~_g*W^BkGAwad+Ta*aPaP2jS_u
zI<uE~d86M&(>QFUU1rnK?VN>8Rd6lV7BS|JFhJT(B0N%u%eAm#daef?kKVi#P`URO
ze2jWsd+q~zi#0IX?m>z+4!X1Po4e!hkL{v}qHeEKN|QRo;fN(8lK$|k;R9R7B@kO3
zuD*Cf(dmX|3Tz~XfF)qx{}vUOJMqnzvB{Pj$?Y6%K^Ql@GOwO<YbxQGpLy!9b$D-#
zK=`_Z%QsQ$TU86MI&Y4BC6d84jvK8t#S*bQPu_k%5RcHp`l_w7bEdthCX6gE6mY=_
zOTc>6`{Ui!thfxzpO4uLdua?OsCSppXfuEhxkNA%-~>yGW(mb8cuOBD(@R`TlX;6f
zbe|17U6|hB!|AT+DVokm)6o^6T68>KD^iE=e;=(_1>b)j2*Z5mFW~5~bd1(UZ^~h1
zAw5JXH*`z}=@$^NK?iGHj(18VjO`d~reNSNBS^jC%6NfJ!gQEBE4Axqchci~e7*HC
z8yco_5w8*A*&`}pmF`z$vuj3EPh*<z5PW8fO-)q$!q`8QMKqONq*Z9anm(9MZLOC5
z@%TY5dJXmr?T}Ll`x$iK2QXdv7IF>Y|I@!B|FBMuLTdGj7{4=HQWBEoV?njzFct%m
zrWsHz<RI@j7kWEgh#~J?L9E6sEt@A|-O6EwcK|b1JF9#LcVIcPL&^fm<`DD{EJ8h<
z0Ki9lLck}HcBh-tCZ4E#a?`Kxa#dhSL2sGK87Vyc%2&8I<NGB=6KUt%Iu0J3o5)me
z_|SlNy|59~2o<8%n&|m1(@pwT%aA2Z5`x=?v%!O7?RtjvrIJ#4*0*Pcv2P(fLmrOV
ztAYav!ND3R+c&ChY{?@M!yS8JVw?1^gD{4pF6UM3ebVE-cCzbQhK8vHB7pW6LR+l$
zb>wIEcU&r=hroDd?&@J=0lg1%-4fo<cXqHvkJxveG{z7=?IB+7O^rTl-Zn4&Oip$P
zkQLCoJ~6|Hondzy723qJC?^Dck_tv5eG{>u!0<Ua5`mMh79UDdx)+wg7ednfZZMC>
zKv(bh;bgk!xA|QEHOoB4-8{;K()Roy7a91u+(NVD=<f2+sTxe8*q%0@tbZz_(^f+|
zn?gu(LCv#y{VFBdWM~=~>Mv;2OG9P6;MI8!0fx-!k%8S@DyM{VvN3wIbR3YsHX_wT
zbbsf8tF!BEOz7}FFQ9vQa%8E2@h-fgNnr*^Xsx9V4P03qYCw_m@)S{QW>Q~K&xU5I
z5T1BiUTaYas$RSdzP4ImyMG6_i0q-JxK^gY_w%q1$#tbmd9rQKFqkW3V$gGEyXO0f
z@{`hPw0-#HKeNH8v`dkP{Lue=t+OGFNT#WZsY`09{6Ss>HCKZ$X!q(0u0W1P8YE_1
zu23G@fVNN(>T~?OCBX!OerjpOw(&;$g1zfDsMRCYTZ1=YK2UYcid?UvBsD8Hb=aBj
zB8nh?FivL@H4jVzRbQwa?sN0{9ngv5w~4f|!OBjyq2==>*i|UuZ$R!!QYI26c`aLo
z@QBK&+L7Y!kIy33kCWVUlYqwtXjVR~Bi_SLVHv&icy@M{FyjSQludbU*przy1UNua
z0Qw4_Ag8MHOT-f6`fCYS3zR6B;G@(lI{;K-BX#zICVXD(S7TR1!#HI=(4F#t`=<kc
zG3JG0nxT@|`)r5OQEO(e4!hVFR=!hwa^SdLJnW*EAG>`=67I<*v@?B}C}h~{c-QnZ
zzF<G8Mx9ne%yn90et^LnK3CF4v_*2l)x7-tr;sgH9eFYFcqR=~z<<Tahce?Pdvkj~
z?tVGCV5tol|7U&}i<x}v?$m`TgPSaeKSqUgsXJIlAV@ib;(#h9EiAB#FYaaaLd)^-
zM{(63jzPo{j3GLh1wgi<>x6q4&z&(nCmaTKMpLEpr>BkW^t55lSz~@~jD{Gu?pwl?
zw&`tA?A{FEk{lT5fPMlQr_Uh&%pooAbV=i`7C1cyP|5Ua?$tFeZ^nzwyGz)yh3i0z
zAkw_upCw-36{7hFyN!i;C<Tx$;P={_CVZJIN$ga)@j5BjJqX|Rwpz~}j@vQIA|Hye
z(YG~VUFx&Ou!s3CT=amk{PwKO0Z7Q1H7e+()YQ%|^rq=r8Bp$nFU1P%PLgkOL1a#O
z9k&bRH<cNTqlUoGZX4n_g0Kzju!(C3%p{uyIHlr7#PoX~u0-&?^6KX-{RI%K9>6OV
zUB28kN?BwRbfWIzki2ZEVfzh{SPJ;=-R@d}1?@1jZp=c_aK{GT`taP|TGDrqA6#Td
zyLoNN#eD=Sv_h^Fv&@&@{iYB9cUrC8>>f02dUbBrT=hJ%u#H{@ujlwo8_NX59?7vV
zTpup49j2H^fMI5QCX?8A(TWIEhSV5)V9LZ_@>rWvcWKyo(Dsg??qc^(@J}R+k$iFA
zJd|2&0?*iY?bf7=56|`AyO*)JxfVVwkt=6H-ouAq2)lgF94_<~$Qg?*nC+tSMe88a
zhqjPYq$6DtcwPkliD?>;-Ixg2<;_USldXeEL(0bG^JaA%*Elb$yd}kbA@fF>;^ACr
zq5tcvD<6kNJXsZtl7S&;2W80|=NUT@()^6D;L}pb6MV`W<rz`1XOdObxzAbJ$Xuaa
zVA%%)clEcK+9(*f&CtUf;kTlnVL=+aWtQ`=pMH4!jeQOOEz!B)`}av|ihxqqD#zAN
z5eJL=KqNDJPT3sm^d(W~;4~`@Hf8wf@A7b9(dBlFv5bclmPsSvecNT9MRlrva}2kW
zr5ECjP&mFF8GSwehfkrp(@)hGMyyx;_J679THL}y`NX!cU-f^WJ${7Cz}A2}SZOZP
zIO@BR)>UR^IUONIu!YfR$Wqc^Z{e(Xh~%-p1k1fH3esESuN+d^Y$8qTJ=DF(9O_c`
z;r0n98@$H_D<s3pcag{A&5&pZy3r3Tb@H1K{Z<GNQdeP(z=N<bqB17}O!ETZxcXEm
zUoo%u;A#AIa!7a&wv!>YeT4asJ0TER4RGnW0xbu}irXD<dN?^gO_9ek6U&1nPzLdU
z?35uUCs%@$&qU6ooy%G;La0A~S3JqEm`b(;Ty}dlj83IDH#e7<fPkT5BI-Oje;`6Y
zQLgmbuqvr&!=8V~j-!aUt;}$8$lVYBbGH1}ZVj5f=F|CowhkGX*63jqSl4n6^+Y4k
zMGln4O5nLnpf`MujMtF}NxqC2ow{V(ulD@PyI-pqs@IC3;UxN%v-PBfh@G`eOtU^_
z`m$ctr9zyXyOx4oLow%N1-mcUPNuo@81TTKT-;K~1}X@juNU@U^?nBR1Nt0M0>^3D
zins97vHnldGL#I|uj7;@4uebbHMYhxB{+@psqCj(I5DY~p(?Q+Wp>)zU1cS=mqc$l
zxIJXbX;3}8HVp2L;O!a&T#i9-_gFBtnLma3Xoza>ej`QV+*cViZ5|6cB#Zp`Dh+-=
z&sjZBfU<@}eH%o{Nw8swMX49e;%E%5lxSuuy>N;b=A6%od#pOUUs0WE#!Iss&YLf*
zfWIz+Ob2zm^7sEU8X&NgF6xo*8W0>(MlM>lVNnV@XSOY>Ff%Zm%7+?$D)E*t|AZAn
zIK~C)$)517GNY;ohWJv{R{}tEK*Fvi<wzQB+nXwl#JY|ncVGkiwC?OszM}h=cfcr~
zAHCBAfh#yTpW7wuIOfZpJ_$)Lo0?8T*qk)5aY<oE@l>~H7@}KPE2@?;%W&=p!MD+~
z@y>Ktts`{ZAD=%^Wt%nnBq51Cqk7_4BJ3+22t^kdv>f)BIBx6XcZABXyamD-OII`a
zDp6jxc?e}mgaMO<kwhPZQ(*a8=bMD%oYCFV*3$Ke`RG#BtgI9aM8L%Lchg}fX8~oc
zb^Z}FbV+Jtg!*{jKoX&zC1!MK=<PK!Zjyp(muU@9yScr(Vz!jAyTx*__Jiv6PwiDx
zr7lN-jRZHVgzI$mmsgB&ssGnA1N1+3E0dQy*cZ5em!c1NPXvp*@#W_z?Zcge6F89t
zAOUOS={t{yKllkp7Ei!v>{{&|8XyRljD}FyQuYvOdli8b!)Z4)OplZCdp*oYxCh-y
zF<_{Rz%v%=yA@x#Ul5FzHP}rgw*Q_ON)&@yt40d%W4Pr=u*Lkdq~113Dlgs0dTRL0
zLB8=C3)zL^a%K$lhDBqP&L<)$pP6n&laqj?3z5_Dn_(#aYnTMx3@|@$vb=0rqd(kU
z$&U<Z@OFeH83!+Kzm0bTiW97kR}wTOa)D$qp=K*eKt;v%+!etEyO@0|_s-zR8a_al
zQrQ6~92W<`#iCf(N{_=0=tF=*GgRR7MhUMsgW?CG4U=VX!S0Gc=!oR1mFi#~WQggb
zz5&100?-S(denCqpA&)6Iv7@t%uln&nv5bfl?XfKypE|Qr=uZZEb|qM)pJZZ1;^}c
zgWUGTQE`C7ya!*l=Weo{xq?vd`Xi%?PFtZ8h5aDuKiW!Zc*$>VWfC@6XAu4hG={XX
zzIc$*vVE39M@Ju5`_dZ$RvSc`gcmWUu$kG}JBfGF20-&L67a$GwfoyTDjsox2Hh{H
z+rQFDhwpQ}UVrmgdCcuXVC;QrlJRXhC<~g6DA$osp_B@}1Vfr=C`}$H9Shh5!Wa5U
zm1db_$}`f#c}n9L$**kVK=JV(>a#=CRCpks#gV0xfmmD9#%NM5{fGV8l4W*b6dvm^
z{U!o?ckZzXKK#LnM3B$PgHYPLFEgH^bmW08&}XnDBpJLgwtjW>aY>&cbf2LQ9;j5}
z(Me>20W@C)!<<AArskZzU1a!MYntAxUiCis7Y=f+JW;_Qz}&s3Ur%^DrkKVg@jR*m
zLV3^MAwN15QT)?quMx5vmeb2CstTvig?7Rd*6+JWTD>GlQsXjRZDIpcTrnhkY%io-
z*MStt?}gw0TxjuyFuk?OV9nu^kjQfrTYg+m9>6$z_T1b#@C?OgyL;#b*>tUQ9_GEg
zA1-uH{`CSV)l_~1hkzQ;dBa*+YDa95+rRGIYqNyy!3dbGRdoHRfHQ>5T%Zb6`MT5>
z23&Ez3K1e~pb7v~uB`L35$;PuD#dLO5f{)Q+9#l7RNV@ze*gwpL*TvnJTfTuenag2
zqQju83eYr{;O@Hq-Cd#K{#_k#63$SeWtmda(*7vdd(v&6jIkMGe3ILnpell7M}4&!
zxD=!P1%~9|{`@izhKn*+TO~MU1$&;|Bf1sh@o95tRC%nr8|Xkhm&(6|Lbq$~gX(`e
zkP8N_cd|#0v-&ApJyxOV=^Tt7$x(5>du}UE=k}xRxM;b{nLtHZ6W~Pnzh_X)PAi9*
z)J7grskvhF{vy=@tZGQ^Nr3r-r6T2@CdO-b4^icDE2v{oo;ETG=w6jy!|43&oIin*
z!VXv_i?#KbU|LATdM3Ra@)r3J`&8ogqI2j!#G{KCwoVF*KUHuhE}9eY+**Q}lH?k<
z*rNH1&hO`-e7u&q;4Rf6jEGu<4@qA9duR0mYaVDKa3ipB_(L~|G)fVDlgP-9f$|Yp
zy1EcBIXMSZLaEH?G5U#f2x|z2sm`z4@IDrUmy^vQ!YF{1(B+Q(8~|tZ88PC~dF2pz
z6EaF}GZYC#nFIYp)`FA=KMW)j+E9?%Sx>74xRPi5m`W=BRLRn0G1irs0E^keBF$`-
z_z6NKq4Zc;u!2LldGXbCRI^LCVj8gWbiPIm1s&SjNxZ^nZ2)=nN}GMZ{^81P{==1H
z3<4zTJ)}Z0vj8$_uH}DkV=lC<KViZ!BHD9LhcZ4PA=K>Ao?-=uF9d%f`~=+*TM2_B
zN?9D?>w>2Dev?XY1v00NMOMESIGCJBb-p2`7~esd!+hc>lK*S9z_7+CSSJ`|r$2M`
zS;WR1B?nNF0_fp9;59xJg2nqDAxdK-|29i?>C5&cXlfbn02`t^?Q+o^Cm@0<P+hfu
zGaADUv7Uf|h197;TSC<W0o=f&@y?;wlE9R-<_jcJniI^$yothK`vNZSjU${>Ydv+x
z^wAQU1g^FInxeLEsVqKz=uvGnCfiXMEq;#y+e>#8%F4jN7(y@wi1FY~$N1O0M2W=K
zQ3HWYMdtvKKTikTA^-z3HDGe!3l;e7t>^Y}8m4AUhuMI#QRqUM?m1lCYQnU~{FbN?
zw9%W1h;-f8XQc0B;u$i}!4^Jc{Bs0ieSkl^)aSt?YhghT#L5Um{FGmCO!~OfU&DBY
zXo8rL1F~1Q2^I$|O`)No1TKWDj)OX#L21A#Hl2$s`8EQ2RH=#z9qmboB})Jf1`9_v
zlLC>O7eE0Jm*^Dl0gz7!0vJ|sMDVA;$lhZEb>irdTyyun^-}Rpqk~atS;t@6tLtCd
zt5(FBX}b-%0We*-6Y=~~hvYLvSsc(}`~$Tb*x1ssZ!+IRpy_vip{j+`p00bR(U)o7
z;VRAX-5(4WvIFAdU#vCuI*omi&Z4%5jkcip>qWtAuU=k=fP8fTW(t%l#03N}Vn1rh
zGenB=?F1>Wo5b>tQ65#qOO-LJrje9-{R^g6k}3e?(A^bE=frPG?Eq7E_~sj@cgu|-
z-Dwkp(tJP+NLkrAc5Fbgp$Bm_f0sW}Ae0LyqpylR2eURDFwP*bX~x%d6T+C%=3`}5
z01O4nK05$*l|)ax^8^RjQZHGb+gY2Q)+3RaWWLQL-!NN6On^yxXWhfAS=#^WhO?zh
zeb503aE|g&4VBKJN+8p<#MB!*TI2`(^Hr>t1N0uC@UtpE1)k&+NSIc3yrt;1HGxnK
z!JK;;KDd?byTF@oI{s-gamR69Pz(4+NtsI+Cm?$uacuCav(tX6g^2eeJfCVpL&Th2
z5(f)F9c0-X%F@OFpNH1VBe3Yz2IOZiaeoi2TXSK*p8CZ%hl~N!W`MD2jvp3X>EKlg
z2(qCnLl`t4v0hb{9`hw-O?lAte%k5UoCjBk@a$ZZ2hhHDJ9k<LFoYqU2Qr`cQHi><
zt3$G+l`a=51Ex0vCLEX-CwgD*jp<J`*^QlAW01Qi{=;);K~0u4Uhps9WY(!TgGT51
zb)?H452$~Na3SV*b}A{;p3H(m!Yrb1nIKDR$oAl(4`GH~ZQ**bxOr{ZdSoD1Lh1E|
zY6+112@)A@BPieXV-y#pngn9Eu`%lwmCJ6g+XpX71y^F|?1kUr2zD;BH%}Sj1ZJ9@
z_}PF-7APmY((|xzK7%qqi1Z%=)#d@O>fC5vy^0FU|8RF@IFyl*ks36>@EsJWVidQL
z00di$)4D&+(2~Iecc4`2L&saA5b7#L=eHfl@pXi4t=8lC;I3uZ#VNu5OY1jnNjs&r
z!lZh?Bl_a0FJ5=gx#uVHx?t)qv0_ZJqZl<xpV<wc%83@0mdZwy+XMbV2FTs@&sUBC
zYPSEu(<H-I2*H&>;&v5HZBWqt{SiA_6{Xnzl=E=w7Z<Z<_?KJAf-&Xu{wjVsut33x
z3x~^IxxcEs6(9-riaMe=q6GLWpnV^&xQ{?7HFf~SBCw&tLvn_VW`HO4TZ!V)NeOsD
zsR5+E$-+rMN%?F~`~XmpV&L=>4j?83fW@VZs}ba5i<A#y$7s7Q!$1LNNzGB`JGl(*
zCU6mcU@Fpl?cXiUuk>4K_(Mm?|3}EjiC+s_R^Z1y^g_kNlvt!nf|{eGV!I&P#Bmwp
zgkTcr&boR(K~#|P5~&>&E^s6gQ`t#`$VJ?kp{#W+3O?JA_a)rP@Xq9&_I2)L-FCbd
zGIj5AHiO?-x4i+Y*{3ZGP?0UqWYQwWD7j=aXvpVH2|GklLA$Nm?x}V@d6Z8@Em^Q2
ziYe2Jw3fL~0GRmI2MmeAJ3Y;W0C7!<hDK3|=8C`#QA4g1ePxwfS^V**ohXtnh@ME$
z8P5{MKM~_;46b<ay6bHGw`Zi%FWz`imj-qD2?F&~A3qSVo>BHa^4s5D(LGqqa0z+*
zMQv&QsYDPL$DIoj*}FB0A$~O8(~a^B2Qx2FqvK~AaIvo(94_vQAyCYTG}Mffp#i(o
zE%twEhP(x|nF3zx7Rs*RpH~Q_s=yJKgucRoVs=N7o7aKjx6ouFYuXSXHc{`*0_F_+
z{7*o+G;U0=4hjWLwrc#;W2}@{5ILz#06W?uz!Ov*-3Z`Bz;ctQ>^_vTqlJ?hPv?Gg
zQnZ!Pk`(3@;62kn7fv8I3pXtF{K5!^9C8S?PvS9;jz(ulC=D+46?&79%viJ-6k3D+
zv!3m^(R7j;h*hdyM+Q%oOR^=5e29cOB28WcjA$0N4^fqEFbO<AE#`QI74g9x38gTp
z00Pg6AB)^kuu6zx-!W76fl^|6IPX~Mzt)|!5GC}NbwKf#b)ZLI4#>(^Fc7R`)*rT#
z=jq`GxlmWB+|m69E}g3wb#&r|+px$>Q6s%L@#utt)HT^((r2u^FLmD@gVSWZj*8>0
zkWTIw6*jZ0oU}5y(O9z$HP{xt_v$gVB!8crKNSVs?Wq!ZVp|pURJ~_6(?_t|&{1m@
ztmyX;{5Rm6>q|0NT`Zq6T$aTqcdoD8-3IHdB2c}smwI$3au_N-O{KgP*dmDzY+Mys
zn4q6>^44)s-Ri?R_K)8r=&9GYZ-i(~P2fsXp^{l;4G-?Kt!!LgVWs#uV$(b#@ghp=
zo6k6Us|3q&R=aqf|KZuJpui%eC4Jwf4Q4U_b+eFX1b#-n;XDMSFJy?Ob%N~I{gBYT
zLY4iKMc8_`F1}*TSXhE#0BMW~FZ1b5ms_GGfu*GtpH|M9jd!&GvB!^3MyxY#0KG~%
z@^<p$tWaviWAN5sPEdgvwNZr?1|lBscD6?#H&C<1W=W<vr<NqgBCrKwmytXgAHOsM
z`}rIoksuVTf=nnzsifS7^c8802u~b>4>Iyv*`gvcA-(uFEt=jSnvemr?xiW~ugM<l
zqR!ExwngVpi6bm2`p)Y4v>`~->gMP;VY)x{>RNKy#NU3z@}GVrs0HW^cAslqWt5-C
zlF`xe-D3g`JjqtbLx?)Wj`Zo*n=Vi$`q<=GFJceWLp!)T5nf|ep8;<2FGBUGYB`E(
z5tOIq`4_%8Xj@S3ILlqYk)5v{RvycHUKxh(7Tb=RAB_~3;>$DVkf&aNZ&)c9nd}_%
z6DtZJD((B-)uYeE=R%^n-U#Y+JVqN6048V|wm$LcYAOA+C+VZS6&svns^}dSk8EXN
zr9$#5+H-KBOQGTL0s{G8LWP>ll2w}^>|$<B5l4x@zvtCj+}K`&r@=75eUA@#fi_#j
zKbw>RVex{ZV7?ioygasZSzFU?e3xxyB8s_8H0=^y)<F%ILZ8FJ#xuI&&*De}8$;)I
zKadY`wc^`h{dfuh{&VGgB%0YxZn*DfYMNP431Z42gp3sZ<03*J0hof>+z{>xR;~?n
z<y>4w#ODZB?kYNP#2$I&R=j9wJr8cv4E`ycxZpa|M6bY5jYw>eg`iNKpDC=nOXQDW
z12h8CfO>0}$=69x%VdGk5?gQsC4^G&76Q`ON6jygCcwfm1!mqG&svYtz|h(eotlW~
zM)d{?&Y7xCwB~zNBzq#;u)<dMTr+k#J#?@D4t2e^HYFZGg_L0ES0J8kg4v@EE9MhY
z;55J;=K)k6c`?!j83g9z)KhZf1fGJ72^tz2i-!q;=1-<yW?Rz(P%c=0goK3P<H(fI
ze0IvHKS9v8_s2OQ`CQZWD&4|v#)@N-J>Q}~IEBXE{}3X`NhgC#86;mpKO>C_*JL5|
zOz%Nsqz=!f`cT$zm9!)q;a~Fuy=VwJ$+n`<8$C3ldmSt5vu|GvcTwzh2>q&Eq$(!m
zF&+e$YV@Q_buSgFX(SF~6}Yb`cMTtk%L+cd8Kfp)Yr!JBzlv>}r1sk59UZJ6WU=iV
zlwog~v8^rquWE(#KieT^D{_$SQywJ-dO`27d@Kw2HgS`Z`=!D-j_1jX_-o80oR%y$
z1}?kA_;28N=@*f`#&zKj-v3XF3}!;TIPoax$T2($4g4H^6M#d;Ec#w5iScuQ-{yhv
zgg~=p@$?>qr1Tf_<@HK4YpOB!ZFoiOLjCoOxE%K+2W@Fow7%Su=B-Ab*p`9j$7cn5
z4+V7!*>A4xqKN(XdJ|Ha2$sQ?aP_bUAfFh+M0(i==PT9cFt*?($%mqZy#&Neb^!ss
zD;jEQ1<<r$;e7SxA;dX5j)T+|0OXX=DZEfbaYU}1<EAVJ({#nI@X+%xhR^Unjz{Mn
zFLz(jQ6}}rtq1%zgOObFc$*0J<XtLa_i=`zV$lvS+4PcI_EFMB$pYaRtn96=6861d
zzDGd*;e5rIUZIU3Sa6!MO4fiQO?j=?Z0ccZ_We(UXCM@?a6sblA&kw=Nj1Pd60q&3
z=Bn?Cjg6&$K3Qejatt~<T^Lu@x>o&Pb9v;*#oni(fHvB1t4~;8V&h{g<b+2XqJK=g
zJ>N+%*Q{&KMovx5ozXs7-9`GWrPzQ0(>a*!;7!7G{khle<;s7HpvvDOxR=rbMhuAA
zoeiy+vine;{9<k6ePr#GR0TT(?p3LqX-!5|Hm!ID)m!QY?(PDJ`Ou}b>31%l?@9X=
z(&e*NL_Dw=taPS5c<FRewniFf2|09y{{QQt<DQ_DSFKT0eMH%=aF;`VhihTD*nOIv
zbWCY_S@C77H^6=URWd5jDd7*IA>;c98j;IE&b%mK>B@7Dr4k>&viCjAu1PksK#0GG
z3bzlTM-HD+Wx;gSnKSUZUrdNv*pf}mwFeMdSzy|}7{4hI{pb-1t16wi&xqrDE4Qk;
z+mhl}+)fz$>edAs!!>_$1|_O8Us%(F3Gv+>t1^!em837U+AOi`XVHBTg|BZ8(a`_c
zsq+Uxw6hi>zRkunegLR2Rz|CSiT)#ms0NnzMKGof!O#;vCKv%iPpaif&uJ1hZt@=M
zbf6slfL4j#bYRU{Er@LNjtG!Eih$FZ+S|&;_izbgGrARSUIa(3$aoG{QSh_}QQhtM
z_A^Mooph%q5CxGzMPQwEX*~p5A<pXaOGgfa3D8nSfl!^054YxggTjMPNC5Nf2%tRZ
z6@j3s!{2@SBBwA`*3^`G<FFY<fs(IYFR1x4VWm+>KNTl>_txA`YB4XZnE@HRWCoaF
zTOYh@u&iv11;>Dr3rX-fQU_a!T;VA%*_n13+6Ha3m{~YFs*!gLRuMaeGR*mXY}O5%
zip#zB&{-mLv$VUsZ*)87p3CD;tyU~-8kzVSRs1+Eq&#F$EKSo?9DPoSa9*4D43wM6
z1z-Fb1DLS?LW{o$i@|z(RJ{!vSY~0-YKw@`#l5^%VYH+CA>zFUk=NSPW~2C;@DoSW
zTmyd1(_B0lkkCkR5fc`RWT3OMy-=s2cWK3#xA_SeiT#faJP3xsI*^#sBb8P{eUMsO
z>*q^i2{C5trN+eXAMhOsVFF~O%v{%o6Djer!j2Sf+TR0UHpcgxZ!;UCKa^e1_W1F6
z<>u3}`1ZCoe9b&1mCYZ~ePv>;1X$!@Qm|DHiO_wwL7sIg0`WojViU{C=_N955Aswy
zgv!9~cd%`<1=m3Mw8N(J?ZTN=IeT3nUs2ZEGuvnfw>IomlzdmNR6O9&DH5YPVU%C3
z(b&(V{1BWct_%ecu_u(0nntF<wUZ#vf9_@EA7C2%LnWEG8-qc<$}@Z&;O>noMO=jb
z2vlN|q!du#T5kB=zRfQVVn9RmU+|IFUqMAh4e7H7HS$dmR^0w|;s?$IeC-__CRvv{
z*woYIps-%-7!|cZK{KDW-ant*W1H<|833dmj0zO<dFKJe{(LqD%%4R7Js?H{5{VZ;
z$WHii8w9BnTgeK7q13goFq`q*`H3otX&ZM|H5b1}j9a6H5hEHk>Or|zxT#VC>mv&d
zr9E5P+je&B@^W%Y@krUz>kQ(urGqAnapUc!cwkD;x+MSl>GhZa2T#mvz2y?c1W}Yn
zjgx*7hrqH?XHZ7oT}k9rS&orks{4JaV&8^i{pVQCWRPana^MCW?~qk|e4IZBQ*!!c
z+h~yqcnm9mrjgCT7KW^QMwyb@@M4V1!#1KTAw<7UI?kEEUybF&wd<MH34Uwqz~EZP
zk9|2^vI{R%FoMn1`!^yW7Z$52?hBRZE7TC`5yl9&myFQUkr*X{$m9%aH`bpD$kj7%
z^;MgL$o-o^E3ZTJ`Yh#2IDY>jC5#ByziGb&TfCMr?B(yFxn)}(BNDb{`EUm?W@Y9H
zl+U;UnQGWk2@jP%R%Qm2@p_Fucgs&N#AX%?2Sx`diNyfs>j5w_7m*EsB7GL0#W}?8
z6n=N?J7?8|d|xYCtVP@-h8c%2p@C~ImUYKr`|j8!y+(JBUCNNpuNi@RY<3qj69w#9
zp!+k}$ah0Z@;_hv+(CdUVKbhG)KHJ77q|!4VTDS>cXnwVIJ{OFCDKpVWk8;_yp$b~
z{305I-H<yE*fNz%lY&4v$$;`*R7+0!+JXVgbuc4V*3t2K`IIA=Ia>+S&~7!ZK;Mck
zJqrai?R^Q)d`l}Mcgs4|IbAEOygFJIOki3v|B&@J|Dk445g5;+Dte8sZ2C)O`x~<g
zG=mW8T7UCH;ael@K7c`w)3kSRhqne~yS;%6G3jYrp(k93QOU<MD{E#femB}){wCit
zMduM`gVq<3e*3FhKyTCfNDY1=jI&_Un}bcnn(7U09v0Ei-r2X>dx)0^e2^Az!j6Ec
zX~L?#3@l?3YI$`nu4dPes7Y3Kr|@ZlogZ^n?`)5yQFrU>b`G3XUxu41bW*=)OcJP%
z%hyAzIi2oTC6qqqA(O%FkCzd3L3|xay+=-uNf{RKakTBmEq(U3=}4vwY)2%|b!R2F
zhJ|TKbw@j3{lo_2yllMw16nvQ)8BMoV!?T-!M@^C+QHZW8snMXyNT_w`FVMrFLA=E
zrZz$n;9?jC_F`sVusa;XUxaq!<qXHb0h;;kORCwDHo48s05*8RqVCHU_6ncnQLn~)
z&c*H-c(Q7^Y=y#UtxA>}`Ock7_L|B~-KTQI`D8F5jHqh1;Nh=J072cW8E*!B#{PS>
zOghtiBMSh~4IdVks5ys`VrhtEdjh@?O6gNzcy4*Ixow%gbxp^JIOtf55J{B31<@Xk
z$%q*o2p1UcY4F?tQW(Y%s%^LGW6@Xi5`%yXaDsI`bJSD_^|%Nv4$|QvE?{NVjNGXe
z0CC^?Tze5Q26;*B#=n0X<cF5`=b*%<t7-Uc+H?a@LmWbp*}d+<2xTZBdeM6O>u3~M
zjc{>%e2J^+=s5WrS&eEqR(h{Ki$x>r^p!YRX8Ks~S=P7t{V}^<F}~Ayc@Sb%p3(ce
z!?Ht%0~HMis%y;*!|X<9!S4eli_3&;*FViv(-B(SEeCVPRFc|LI@}g1m&Gf)A(>R)
zcL|c7U`c%U{S6$lU*%=jN|NWdK`^C&1pOTW#o)KAWPJan;itN6#!bF;7Fmep*N@3h
zmm&t*@IaW$M4G9xpZlFk{h2-|mHM)Fzc6LK8&_tU`-=h<U<%}hzMMk#As4<DpOS(K
z&`DhyReGvBl%V};TGol2dY-77KkQdxXv*YQA~{6a0SFh(Wtnrx2Yry}wEZHYa@m`M
z)8tu0e%0f-)h<LesDsYH;Lqj+_{FShglIU(D?pwIR`%OnY}tb_y)WZq{OX5kopGbo
z7&JEH{>wt@M&4QRjbdZ;vLDY-?z$RrUKfZGy&%i+f0i52;5LYGN4?!-E3=QsfpPZ<
zl#urb>E^jpyz1L(AbsMO3BrU(gxI<0Uy?qb-HvJ~7cBoH=>cpT5Fp~_4HI4_xNzGT
z@qP-`UNcU4vE88m4Vr62J=McHdnh>Nn<BjP!mM6svN<3d$WJSvcarh3@mcJp;54l*
zcl0-Jxek>!GA2ei+jb}l`l|FVKXJi(q!^sLwpS-E->C$Icd&uA*D+B90tsOyhU3?j
zLDm@xQVBDB-x=GUiOORp=ALg;wQK4@WN8Db7vB<J+qn$#1cew{9<0k?<MWB99;gGs
zu^WTAvK;*UgU-e#Y(~@}9jmY$v`#2rMv1yyZWwaKtBP&{?<8ZfS(!9-W@C|kefkD4
z9GCCoNzhMaex&6Cu&k3zacSS>8R#%Trpa_?7)EQYn>PzYoWDsuT~m2)@}b09((CKJ
zu}bwO(!a=ih#>81x?6Nr-=xH=(kkURFdk^Z4MJ=pK>FCn0pN%{G;#DNL!HlT0J*%1
zY?#~@+`u05S0FDWNaW1i<h%mMH=d63=t~w!_3HQY?<D7<NVsg8u*oI;#eiD<S+79T
z&%B;W)z6kXzo<f)*2Umwul@2Hq8(2gTqkO^Ut@av7fL%&ymDo1S<Sdy0TEWQV|XD9
zFv`9AqF<3cFGy0sVwYWDm0#TV*@;V&!K<S?#EIfOp?G+?fq`S(nhd<MpmL;O=ie8A
z@|gX@TmbotV4hy5c#QfZ%!*?}pp~DSr1N5o9&WkXw)(kF_f%Ha%+_-C2pl%r)<n*n
zR<dwxBsM42kSA-IOJX2H35-EQysVglMbU+hy%TJnHx_GHIw!_+_yv{^$*u&aO|oMc
z?WiI1>JOT|v#42!ba7x)!RcS^*-Q`mkEupguzkavZ17G%VAzH%4wtD0y&NeN>2P|G
zuxSWqx7)xdtpdbdqV7|F0TAFAy7%Ff88y%Y%~m7DmTMNEi)G+qwCz;z%K&V*$}3iL
zE&A$DP+bf`!@LYbL|ZT_98}WOjRy_A8q|huNXO1W)!)SIFOk8a%erA-azoe0nXU}&
z#s)e|kT6M;|FI|wGc%U|MLL$C<C||H0hu_rP#$=;<|%BS^@=oTN(AjPv&OP2rhBFB
z{ZxF3ip2d61Rjfd$9tXMN}aj4kds3{dmR)&p{{HiYyB5%)&P5I9k(e331a~eU9cUF
z9Vhee)zhheU#Ree1?H?vyvW9$1p%&kekyRdt@MwC%b06F39;Pmda|%f2fJ9J52BmX
zx+h2b3-Mha6AwxS8Xcc-DDWHvecK$_Eq&oZG3oqmZES@1#DD81uuRzOOAN_EsP#f3
z@}#7gP$9)J2SiCKDrJThj9{uT@dGacDhM*0g}8&TCU#y}9)LzEai48|d6w*>eB`6o
z`Y3ijL_Q?xlg&VuQ}vIJuT-@rE>Zc;BSgl70Z|w+PV(KBx@<^%#Krr*8pQd3y!*J7
z={r{|#}3!(mrHt8YPe4(ju;#BmB&;Vz*mstV!<wQ7Bz_!{g9(T&r<I7?V545xB@gP
zKjl+&wryCJ<Yxrj-txBtCn~ixIG+=-#KmTgJS7YNroP|_n$(P5c^zHtJZU`cBx{f6
zg?XOTMgdB%mWBmz3P-@net{zH)Ro6K^NeQqya)#DYl%F3zbtObnUK27Jf88o#07`<
z!}lhQI{{74PY*B4ja%v4$|9Uu(<jWgg`Y|3(;(|eTu~I>z(5Ve0FaFaMHtYW|Md$|
zS^TzT&xM;1p$`3|LIu*J9>|%c{rX0hk?H}|a$AE7&ZEz+jpWK=GT<ytL%K}P18dH=
z+g;CWA?1rv--ifuisT`iqX3wlIaZ4<(maIa6Gw6RhO$1+A;9A=hWCRl9rq{@u}DWo
z{6(`feH4Zfn@cKuBgac!hXSq>B0gD++_?&dhj%Y@r*k}_l;1a{h=vm4#2-SPNGR*0
ziCB;exQ?9imVY-`MHOMoLse}C3Xh`A2eOir&z}|$foE@aBijXscMphpDiJ>m3A)Tv
zrcdK3AltaH@lrCnG#`L3-tUfZSp)!fo?Mi1|3Io`9o=h*b!sSL`rfIc<P26P5!mk(
z0*ym;oaUjYs;pK@%QKpb+iwd?PSws6mx~Z{lA>1cOSQ37Mi1qnawvS|**k~&5$U_D
zeYHjZ)l&VZy$W0-`s)kU`)6Nz{GcU~Lw6bOG~t>W2s=RB#VS1s!ee5Jpc1@?d#-OE
z2##vdv;Vf<ExeB~L?HA)U3>O2KYuK|h9z7;Nh6%mR8vVQT(0(wfu|rG(sWXnM})7Y
zE{BypdHw?Du$X(4J;6-Ja{K!dSRqkFHqmfZa6OvuON(w*3}NqE6kefGf31HG6)JZG
zLt_!LBT2vV6>%+v%1!E2r_vJ;;cY#*z8uB-WVuhJPT)B_?|1*}^Df_gp6cfZCihdO
zKkWln%@m;=@p7|!j@Tv#W=RI{ifsh|Q(}Rm4On{A_)tH1fsEH9&6x?=0H?!>6PA}l
zR*<y=A^`0I`*)1slL@ibO|iKq`Ie#Mvp)#@N}7i#@yiuqg$BDo>i>_YeFp2wA3yBd
zpKV{Op)g<~&OxXsmxDlb^B!i#K37tMVuP|Q5nfMxDk@bZ8Wc<9_G_^47_6E0=Sm5M
zHMKh6_bHh@S>jiZftIuldp?y_YGoCbJot#{$jOE5n*#Q`49sGI4!?yNL{+dErprN;
zB|e=OlR=jE^8!w%{?V#Dqq?zw)`^(+^`8xw-#xhIHncvd)6cxqBk+^etR58vv87tF
zw3@)E++}teIVGL^?}xnr%NwIWMf<$$?Q}@$(a4vPp@Z$i8WT-~4IHz*3#a56s&ygC
zTWg{Kk?MiZN9?1_>ORS%b+o%{Q<EO3FdpK^vfvXF%NdlK=5DW4*j*#VfB5^VODEiO
zeNSRs_a(jX?iJlDl0;y#BDTSbhXvcp-SFSr$}33bL%)^`<$ngt(>Luu|5~p8r@GXa
zx%-_3v*jPOS1?gQ)hQ_IyH}pp1}FzR$f*}=U!altg8v%<++9lXAfpTO***ESI*I8;
zeC=?3HjG#aRY!(T2SY|CZc2_=k2}z8Xk3e<DX{fWMYkXyi?h8sM~R5bLYe>{qkxlz
zG?r6G#z2%|Lh54;QL|&EeSOC2_Ov;f4fx{R>A#ch>gnORIPt4CiGvHSXNb^A@2t4-
zqZySNkXQJPk_ng;YI)tGqN0cvtVZ~nkA_Dz;N#@@sal;<$Z?sos<k!ZGVY4--Fj1x
z0z2CN3y-?Hi;EaHzZ>j73BhZYOh4h9j`Vi){qe)tZTiu*&2bf1e=;*NxiC3g4DB0q
zJQ!CNUyT&<A$c;o7E_Ui`!|ygy}cIofAK!@t!#3)GwpAF^f5=!oXLs5D`c6q<>K1A
zH$YgSx9~#A_p`L~;!na%9s5@7##Fbtf6vRSc7Vxls-f-(hCzw74qU8pGbHG8b6Ayc
z;&IU4X_k~bernx(j&I#;0j_wxr%`)agS7IT)|<56pfEpluNb@RI#wNYG3<5U8L&_M
zS9Z_eD4%&gi<blAO7pS5d?`G*BB;+_6+C}F$UO))IUvMPG^K>1lnl^r{f8|RFgA`x
zvay`61qTP;1bWJfGVN!fD+X}wan?+FWSBO1T|8vv074d6kDDj9;UXCC6>8iXQJB~h
z0CJdm(YwkO&Bd{Kv`yo-rx(1KJ5o%}>C!wXCliI?GrBua-`~|1^D7a3Db>$g?Os!b
z+=tqCDQ1q>gv+8zaU!J{!-Bgjvh`7^w~UqF<H3`Q<+#nlA)g2N&h2~iy)sG3iPy~!
zXPzt)l55F5p?&^|yDaX*V-v<6v%nh?o@6S>Gs{0m{vAg)U({Z(w832Q-p^GI08n?l
z$>Frd0c~clf$l23g=Y{9!vz7&`ruDeuZs^Egph~$0JR(MqT1n6Aor)Eng@-GQ`dIR
z<zDJaYs+=)%BwTT8L32VP<**MLP(A(fdh~HGR4*Z8Tb=;n)<`~`;pHGi7y&4#NGE^
zjEwuJJChqm8xQNa`rw%8=wyg>?tGP1Q4U@y+c?8uU6ibyc(NPt>oB`v=Yxr}HSn-V
zSs3JiTQS+4lmw(vK#`&{q0DwvH=H|*N6QU|zuTe8zNf>#+(2mcO3D6SR2X{q)%)p6
zqi(qpJAT{Q=oEr3Wz@$V;#~^nkI*ql<I1i(G0ERY-F#+sxY@w_nwR~21ob)24u50p
z+r17f>XPsbqzYRaNGdkFp4~{p;QV{4HQM1x3;bVCx&{VPP>HO2lY75fx+A+97}$rN
zxYjY{_2oi_yn5~F;q*CIP*~WN5p@#3R&$)c#6NskdQC))b;pVQi7NUx3DUDUjc3*f
zUI`Z|qCezp*$UFOec%fBIr8rO>wi+^TkNtc;yT}OD{54`#&<GMVU=K!t4ZyLmfB`{
zzY}~0LxvF~3w?I$;VG>gA0OYwks8PpVFH0$3LpZt)lUiwRRG|yhK*6_CF@+^UgaId
z8#p)1y#(-U+b_{WTpVbAEU-cPsn_TJN%8=)z{@wo7r*e%9l6e94=AYgga;qX==oYz
zHk@?%?|Y{#$x-i=@-VEOcg*vCX!CV&noK#N^Nm`S%Rq%&@25Lc5tJzC+eAtY>5MUt
zV&bAUx}M?X>3CgClWeF%KYK;<iuWFOi$lS{%~cCGlxXh+P5QTxlKu%BIm@5arK$t(
z+@o=VsAdEwF;O2VQE6QOeqhAgS+h;^0UjMRlzUj{n3z341uB65{u3*e{dfH!Q&Lt=
zh{!(Ta0rmhKQ#P$jf<gdCn7Q4lG%08?~*$uJnp-}MJDLy+2L`cu!S=Gua8^h&*Rn$
z(_FMro{zPELx`$Cd<BRrBM>49Rx4(2OZ1+`9|1n61B;n$oGS?9xfuv?2S0%|F9Mz$
z>;ac)A$fba)a>d6td<qD52j($`-CkN>)bgvE(85U99a-S+(okeO35NPvFQ_;aUrov
zCfcvhyxeKc+uQr6v$E=YiFk7}hfY~x11`M|`~IMSWi`3nqn<Ta&qI~rUrnO>*D8*y
zKf~ERe-6%een*or7M$%|HM;-5vptEN?X2s0Jsd3;v)_jhLVa)PL{&I!9qiOMe#lcZ
zHU!;TgwBs|U;2cn6Unao4vkF~#{_=I0Kx=^N}fCI@8fC55Sy6ENhI-^EyKg-qypnr
z2a(YhrT6ImTY`4kJLYJ49@8&fs8ObvC~bMku~B`Xgyig=A3Uu+|2o}JiY71~{-mIa
zpjPvdgHv8W)KTullEbV=F65YM(yL<i06>YJ_nMMC2=&atB!ZA-9t3=lSjAzGR+)3O
zsRe?V_X%P*ClPo(P2`Tv2gaHOa0^NU`jg{ylNpkWn8am3x`0{>^lL*fgP8nkY`L&U
z5w@LCX*NfsD}Yxd`=#>IW*1iT&f)w1s?7toLzY>U9~I~QUY33yZU~+}sIro+Gjlgf
zy`ns?MK1nxR3;EaaMlM<fzeM$kaIrt<;J{@Z{=6!q}HgequD+d{+#fyenzj*%$Z(@
zR<$Lioayj2zm$li*1jX={|M$=dc)3o6W0~rzJ2>%vTAo{%bmp%VV7A$_QA&sSy}ky
zfq|QuwGe*;#-KqMc?@74jud<xQ+1W3i(9+h`b#nL%g1)6MFkH7Sasma`6^C)FlH3E
z8!oDOW|GN*9wiI!ya2^{jQ@^rMg9on9Z!UJ{NW=x+(NO3a!T_S!x1;es|n%LgIpR)
zYHANnKEJv(s-ig<t6ki{?+actZb0_4Cayc4ggg%SAD@lD3FE6i^quY$ptLnfqcNcF
z55Yt$bV?a-rb!RGUZs<&7SL2N&G+oerf2ZO^csWc$Nd!V8!Y<@jJ?0Hd|6+bw{Bj3
zUG9|7nOh#bT2_6gqv2!gOUN3=x_PeGgn%pixoEOc&Rl#vx0*8LyZjl{1_jd3(c*n@
zYF+xhZ;U{nas9N<m`>Xh-druqM_l;k!KP$AZ2{ip*iX*&eM`sVC@A$fpYIHaZ$c{c
zKYfM(zE=EBbY{Jx|EB=$`%{~1dESk~mljvBI7rn9!;YS;RwV6)|4g;blWU=ZhuEN<
zhx`93%1Dj&#v%WHS%^?%Ufkfplnf)LtR~)l*%p>^;IY|*rQ8*25tq(^VgV3*Q3yMm
zzPnW#0NWluKoQ+Pk&=_M!{$B#q|CKRpPc0PpFDgNkU1*0N|+n8+26vd^CVMLnFK^|
z)^i;J-HVfH&8TkbcCo%9O|KR!(qFt?g-*@mh}n((IsIV?Ps_!!YX0xY+4^;koor7b
zCMHf5YHotwIX>2{FyWMzK*ndg)(f~^k6v!B5}G(e%Tf1K3v}XG<OVHv*U2djR5IeM
z{b;T&+Q8}40r(~RYBP`=th?e_{xoYdh;kKR*pGjkH9b<L+y0o6R{ga5UuIa6?rnbd
z&FH+e8F@L)Whm#&K+;o79QgEqv`WV&L@7h}l30(7Q2C!X+v!C&&+(^cThgPrhZaHo
zP+amMZqRd7AQC15S?wl|8xP>u*P1|Qv9Uduv&w(76F*JNyVBvhbuKu|RC;gOM1+Um
z=BnxG4x6Tn!{Ca!MO(ThYM!?4Sn%>8d;YCydwctp4ZrQT%@?bSuN#|xqj_>GTi;IX
zoBY^GICQ%wS_Kybq1zoDdE)e6-R@{bFDgljf+*k2_rcodk>ca|_U26AFAm71A(a!k
z6yOY=MbXU0=E}aXIjBMZ?b|Z=mGA-*v|hIb>>pqs2r{#<v@iKhprBmE`Rw)OQKj>w
z=v5;8tEU~<S=1DouFc`+W|d2w))#zwnsFAYrc-E3*rbD)T|2LLo11vY*p=<G8l!I#
z&Fs@l(p3M?>&k*V=sPB_Xc>W0g3XQ!m4n5@vnUyV4rS_gAamS(WwvNb6O|eGy^_J)
zv=hiY1)x9Oe^)r-4L)IaMznd=ba}P57UR;>Id?zbi2+IQ$p1swd&gt_x9{W8B|=$A
z%E*?zrL3atosnJ03fX&A$SQjlB`ceZY*N_?*;_Vcll?ni(!1{Weg8h+e>|=p4_B`1
z`8uEHIL_ld&g`<4Onm_NB0<oy#3nQM{*dkPmN{wpjc4<*uJ+lN@P@QixyKV75+eN*
z7ost*sD(0=-RjBGd+QLfo@DO%L6jQPld;1$4y?><t^+ojUR=r*9q&5hPeDl&Jsz&g
z@aBxJ>=d3An>X6$g*$@SbT#INuYUISJ%%^XKj-lcI)+|+7IBfq3An~fxAi0<Cz_!F
zvl?p%o0Pa57Bmt!4KPs)Y<c(lxJp+)U;X}BbobD*n6R}26vAFDeh;gDTA$e)=qTrC
zrl#vFs2ttqS^2F%l@;vKtnB1jZ#N7l;f83x;D6iWIq6Z)#a*hVHoJ<h+|XB_KI-G=
zm3y@d8_>W>K9eLAh_M?#S8Rkzm4Cqmn$^<J(!~7<4-GINpov#_GP?6e^?}FMTxPD#
zeHh<xaUt{t$>FKqD6Lvx5T`);TVPOsx}=qV$SxciQ6VHFTTrv?iZn<g@9>;+@K^Y{
zAKiL|?-Daj!jmu0`8F0&r8c~(b2?hrSF2jP?{o*|>~T5}3j7oc<@<VOG)7$|x)>54
zZ%>RVwr7zF>Xe<24Zf-n*oeI@h+8vSmxt@VjZ38GfcA82fja9j^c<e+r*36WJJQ_O
zU=a~8Q7S3wWmWz3IY@aR(q2FWO&8H{5v{q74))u${36=tJ<XZJ*mE4~`Ll8+v5_G<
zKY6+t;W)NzrUp;SiqFnDQiOU#40T?|M}?9O)2f%@$J#Kj)OND|+iwbE@9y_tkP6zW
zZ|jC{-<0YsOL4#NJ%@GiWz^9*YjS7Xl+)TMlEzFCG)Rm>k!|z6s7D%_vYBd71{(po
zm$ROQ1CylTP|@9Ewb8O1D9+x6oyW>q8k{976>6~HDFx;gRsiC7r;4zJTjF?|vs(g?
zT8&G;-@$)x;FYse5tC12;q(O=lYG(Eom8<!3@)~^4UcEL=8#}}QWnp#gHt^q<3LB@
z=N12Ly?BMc5?O)jPg4xw5%y40+^GbkG5oE!I2951p@VqTes@hST?y!7FpBw|UXT3-
zXjwE+mcJZ@BsWlfBqS%da6B;%i7yIYQEIg1BQw{B*biefBK>+=AwtQP-<cY{cFY3P
zHqrVv4Tgl#6P})$lrVIHd`Sr0X<oe`5#4{DuGbdEiM<iZg4Hi6Yoh0SE?u(PPYd08
zM2(JOq5kD!^*fFSil3%`aoN}6q&Ffm<9C{T@a`K(W}gc>cVx@nR`PI%-Lht3>o{FC
zHKNdFv(DUslWm5!?qJi3))vgbCPQ~0nItVS1MW^JY|$!<ylr~Jnrx?5x9?}YKx-Wk
z^%=QO=&(mN`<e{%lW)_9Shp8hxlZcDi=TEL$oqH>uKA}J%$wdY=1m+^E~qA~<1QP1
zOhSm5il2umQ%cWa<}qw~Z{gy+M?4zSmQl+EUsK=hFxb0s+DoKYJoVV<Gqzdi<-DYx
z^StfFrN)GwtlQsKJ)jDEfnoMFoeKCC(I^azzF6<x!@}Se;m_$^VrT0qc>!m>Rm>~5
z=9&!NZaQXByO;N@XgtH-I17ebIh{B|&;W(lHMLVwhi^MdmXHU;ky?}MsT%^m#pc4`
z#B|cc_Plg?WIjxlZ)5#R0%mN<&XlPW3Y>`xIf2iZ2b&`*7)hdE2)TB5`Pzg|LmxnU
zo)pL5-ycLm?4d@mb7=<x2s@Cr!jZng_3PJf7OhoR?RWG*`ogFnxcpe_f^--bUAs6i
z90r@R3kvin3AUDYUdQ>VBtL0*Y)q44Bu<K$?g+*^=SD~F4UQ5>qDrTqwbOWc<ztYk
zE7Q8T1wn&9Q}~PY*hj<tB>)me1-B-~NYKD+-%HZa6N1^c8*<(Pv#o(B%#QnA_)^n+
z`brFp7HsS3?Y{-ZDu}I1MkI68US2Bt+iSC2Fq*^i>tL43YXD#(U637Lg8exjRs%gJ
zT<tQL+6GlSp=7WUL&E=qj+3O`f?%%=$9?Y}I+4GA;L>u;95^(57PGdbG1D$juIS4R
z)gomEXYlOE83@;3d*w<OD)#E6=`ru?C*8ItqgTF&;Q^W%fYf5ZrwyCVwaeBnRsK=6
zWBCS<QDhaAGDy?iu5ij1++C^*i;9{$FCi-Gg)Eor09v(rY?=>9JqLyzqn<Z_29ixZ
zxuYR#!H1P7q3MZ80oUjE`u#e~wd9<aQZxD;WuIXXj!vaOt0CMUQUp0cXIyeL6fQf5
z@$D?P;;9(=Z)EGY9uIy=&!TtlOCf>;fvfXFEEOaORJD4@v63B@R%B*#!OG6$-(e*T
zOjEy~mIozfMu3r4okbi(Yd)g`FESF?%#uL4c|rOF{sO+JD26wUCqqMsqhn*A%r=9y
z)7-`!q5FIhlekp4+8MA{V&ljde8w1Xv>3Jq2uydF+5i4gNchue1cQ{%Ts>z<&YIrF
zWJY$T<~UU4#2o#`_i1guY&tPMScC+UgZwJ>D8DzNN&x_K3g|IG|25KWK$Y#L3gKl&
zl2ju>&;nzAeKWX8L+(H;N&IAAp=mfU1ADUtbBXB)KDciVF@WZyRBi`Ux?4av6Wou}
z()~_awA*Xyx0YCTSYc7#t*19K`arZB7dN%!GZPXYjqP2ZlL|qMf3lF%1TL7j*Ei_^
zamZZXJvQu;&?>$Eq~$`lEc)?)^x^0iS4Md|<lo9=1stmvtVcuq0o$QJU=uoc<^X*8
zj&et<#0w6W$!16JwBJ3bJxaNPD1@%mI~qD{Ty!`*p93tiE2N1DAI9EK(jM?=JW>0Z
z|K$2;pPf2bs@?d>3*Op@rDFI{|NAs?X_wv)Ts7Prd#5e&HCg_W^EZ)GwnI0m(6LMy
ze!13{LxVd-3=o?m!$09SsOrN&ESJ@=AdrZH9{c)EX)=_&hfaMOgR;}|M-tnC1tyg+
z4#AZMngSq`KpCs}`sS%V&>*;#u2>7_8u|05)9ja5j{rbQWr%8sW1-G<Z@HB-DB9!M
z?;x;JBDQ*`C*ur#XMz>WtB>fsi}9y%*lJp3FiH#*Y_T7-#M=&^<Avm2Pycg)#@78-
zPPr3Tw%=fxjxci1#?!y(e+h>A|Bu?H(;6;Kx_SkTwY<2hp5Qzc8>G4q*PjGkdVCei
z>XM0R-G!UPzm8~yxYu)`0N33rn%>`emHdgV!GWuJ^AP{7ja$LrvEH3rqG%6(M2x!6
z3V=&?TH@%x{3<RR{@cY0+{B`*BTDkKBa-L4nPRQT`QWdW9Tli^pG(ZL!yB&gYM#G&
zQD+ng>VO}bgZ8QC1OhPvshLGc;1NRtLVf`LYm}}9kn}0oM{lv6hS9RufQGkb&qDX4
z8M<f5z*4ZBbk2sPTnan_>Vziqm{-`s-Im?T0@LZzitFmxyK!GGIvXnzDS42``jw)<
zv1+e=G{J#nK+j1`6S-mqIE;-SlIrVCY}#{essyYwxKH9ThPj{1B2o70X}FGFP(F(S
zN5OA|dkzhCpAGS<7Q*Bhr>o(P19qe`y4Od5b``r~={*h?v^%c)%D4PVUQaGO4GxdK
z_dV?Ahcl5bOF4bipPxBl3OT(PY%fc4ysC)f%>syQ%fnh~bUu>Nt(qy!V`E58_g4~o
zPIJ_G8jpBUz(!z%lP_CK8o<Vw|MNVfv@DUW@Z~&50H8zmba5VnIdmlZCnxW^T8nx<
z_!|FP3jlVRfJ?=j^*?WunW*FdTc?pa7o-6MUo-A40+IOA>U5hU^V@-@hmjH!b&F7m
z72DtqJ=P8FNb*2pRo#1>2dm0sr%Y?LAq$v_W4{@A41Eee<}Hhwu$HZCc<oaUrtAIu
zhgrZpWC7)^mzQl?Y;lo7af%SB38p%@Kbb_LTAPe_1i`(!dIJ$a02B{z<?0UYyh!&y
zJP%P}^sA?}BM59ao`gMDZ}YPWu7%~6($iDKYh)+f<23*yp-&L>ed?76g}@RwW9z>Q
zt5F7T3)i-0nZ?D8j3U$H=cNo=!*sh+pyd0hcZ3ce*k(xuf**YxD}SFU1j;8WDyp!E
zhzUZ)gbHBpt2l$QsE$z|iSYhU1LuYO`oRFG_nDEh3WkizBaQWxJo4v1U9dXiS2tEX
z3%tW<Dly-8k$px+%|C2cng)f!qEg*IxR-TBU|MEmxYw`Ekf2%WuHeuF$4Gh^42y?<
zR3s3sIfLGWQ4QD=EgRMC^<_1-3)HKAyy3?tne-?;NWf_>Ci9p)0EqRG0)CwsV!a5)
zT6e$BpbN(vDtb+rjGDe7mjPI~Zw;ttvEiI$X#fGsUq<Ca_;bAthpWR`t~N6H>-8d)
z-WSJTBraC1X$Y0(OO?Lo`?$}CWKO|A#Uf~+hs=o&U+rHv9G~M!jhE?w(PC(`ZgA@D
zO63)oMBo48C|Nix`guq%KX@50YGRU}rB|)1S!gn*mXIF?^$F-Qry&MNdSlj;hK6Pm
z;eCeDt+pit(~Wwz{k0#r^|gCb54`{nFEI4Co9}MTmdnf~+o`A8d%^H-*}wn5>}LIZ
zbK{fxd94?{9n03MVJT_Rqc)JfxBc?^FY(cBqsXJ$y2;qZafTOvdCl~!&8v!?43IUs
zk0%vi!1ZS3@P7;@S$t@c#qT%8@&MV1hLOf8$<GUZsix>?&De}>u2&rvLVQn^>NggI
z%j&~Dg{F(v#`ADM@1&J&rq+qiJ1ovIUCw(_v10hRFc{l0GT%$|8g#le@Od|`)x@C-
zQfk=h`eP&M!0)&YHb02oWZXe>Z*lq36WAgcb%Hl%*_4tAl0Hw=FC2B3J2$Qb@Tk_t
z*)P86z(xy&$%@h&x-fJGF~)owBhK=|{y`IPcalNNrT~(>l2&LRmm?y4gy+u9s3o|=
zqGL`jSjAg14y~gG1(7{kM3;xTWKR0}-2&X0VzH-}!v#$C(xIUdK{OeXwAB{g{Vhpk
z1o4>s>}K(LV4}OWk|>DOWN7t~Q2Ue6SU)xF!c<K;3hd9TyFZDM!<lwjr>K5|pmo8~
z@*F7`UP!!Gdlg5EMNq^fD+h>ZZ-MB0ydy?2*ytfitO<zawh=17JfBxC`h|nb%m@zM
zcxjv7n(_RZi?3a8N;W^?s`3?eN7-O4QE+0O#K7uR^O`<BATQ+oZ<nd@a^&mxmFf0o
z;o!U*dmSs9i;RgIub}4mDRFeuMOHJMnc=2Ce}3A6F$8A6knSL8S$4B-!}V!~s6)%c
z^EF%00D*328_8gDJR0e5`H<*_fb3FPq>-MKj7&ISSq&h;FvtfR^|*z+#Ub|`YRWE7
z`XJK3l}Re{-L9mQ`m_1<M-wlu$t4%8`d*=!Iy*W=XWK^kdJ{T+!;s-4R<J@C^L8N*
za!)7S_&rVGPtM*}<-SIH{q(`5I7{bef7HgUL!z23QZN1l3amo|?e0+JKR;=^01VbG
zC`7xyKj|jOi}6|+>OaLlsGpyXg?X58ntM}vUS%Z`*-}wfDf-b@Ssjvu#<d?SX=<Z$
zY4cPTm7*wQ^k?y(4&mnkz=wJE`K{|PZTz=waun4}=|C1;>WY1)Zh)O9>d9Qs852E+
z>&Ltw9>G#)+}P}g!o^RQt#p<KqaqA(fs7CaAjG>Hd$3cX2~-~Fav9bEnUI7bE?R-U
zy=b=)*F~Y(q4E+ftv+?K(|!n!27&Zju~g|h3*ccW7n&%Q*f3|iEe#QD&G<oml)Q_z
zl3AIQT*^H_inj_8Cv0mv1j~(RG-k6kupjQBs4hBS0fa(qH^<3-OfNZivCBFW8?WR!
z*TqSB*-$r$@q%5dzD~LG2(e4MyXVd#cIhXoo5*T>OFax@3CcmZ>=WHuR=iWA=ZMiy
z5St*B;{L+Mm#+;p4kPkC0*%sjD`(qwy`;@IHX8Y&)9V<agsPdzw`b0!okZH%52fxt
z!7`zR%;@G9yGQ>K$iY=aqaVbslrM;!T{N=TG-IWaP8-Rc(rr4Aj+M@OlyYtaWquX*
zC%g49<u=NOp!couvqFmjH8$OfnPPiTi$=!9N(!==r8|79+;{G=hQZY9VV}@#wbgR6
zPaOuL6B0y0-jK*LJ|dnA2?c8RA|u}$GBRgf8~MRIx&KrZX8?ESg1A?FG%gnbt{68&
zlGgNDEHo4g8;5b<9Y`&$<&wxcxv%&0vLKcF52SK$s}eGpjwzY0Ac7n;<`7_xQ&-*z
zg}n)TE0fWBW+R+J)cSY5bF23?VY8CuIMz83m4vT6j4tDO<x*$$<@8OCWO0+1vJ0@)
zAa|1Ft6d^);3){&XX6#$Ujr8{NO=bRna;l>&C=T%q~+cK*4H(6aO!@Gagz_Bp)tyh
z<X6eF<bL{8J|>U?I8Wiezlkor;#2;kOh#E*c`#{z;ceRllr9d;&1Ea^ao8I`=i=>I
zDPS!CP4-4k`Hw3lD+|n=oSZh_g82jL#vTmtK<=j0U)1rfSLF1ph(-AahO!+qf^Tv?
z(c#gQxw$vg*ItDl<7YQT$?^Ll=upco4t%I`@_$u1fuWZr3g(n<U4nS+%Ny=>h|g~#
z`1}rme)Ug*sqg&L;(&ur9pZIU-Cr%&q*o@9c~qN%?M3p!z<>mNMfHnZT*@#SbSWWW
z^D8+<_MvBsxjh=V&M+~Qi#JPg{muBb6_*iT<uqyQq5Aj7vKU5ZwJ))<wz^y+B0$HQ
zK(LM+mNV4<nM7WE8CYX$wa4a3{tTwDvoA!1F1Ou0n{mnk7=R;35ymZEr0y$y=%YKS
z550RO5di=$&>JXu0*i%=0IKbP#qA|4LBw>+kHU^!mqLZE6M+06!o7gn(unnQeg#xL
zNW9i^2FA)R=<FpG-tm$6&(NYHAe$;QqH#UogxL`9-Wx`!Z<xkA3aP-mof2Thg{UY?
z&#^O5SFlw7EiTAKwhha)Cl4vZE0HqX1J%p!@zi;zRe%5@V<~hPeO{IF7G2l_t795*
zTMc#~?;t`yi$4QGM=s+wIFGMm9bVfjSZ>AJF)g<#GR4|{==kapv4lUDA~U;%Z97tP
z9^Hcw(w)SMrYF#Fu$J8Py`^Y>b>;pzK^d&w!qqHjc=fR7x19SbKBvY!7F}$DLPH52
zc|fNGZn5r*p^vE2e6ID0zJwJ_T(B*+L=t++)JS>=j-@QpI1A_}O8vza%)serWRyN|
z9k3{AgcU(_92^I<=unM^fjnQ?_kmQo3rIb_s_u4tu&K2;Y4@&|jq=WFj%^&?UR)m3
zGWtYs7ItqKQgj23L1*5|(C%~={NR1@Q}CHDiu@CsPp^9_m6(;{$6hUrT@|d7^nm%A
zM3htT8WC-4%uNVCJSrbW;l_E}A%%BC;bH6M*AIOjZ=e%V<ET+|xob?ykjN>FD`vZ;
z|1h|7teY6l%YHhl0e?vq;;-Jvbfm-g*L%IZ4lp8ih7w%O%gHViEJlCQl86F(6W>l@
zJ%0_fP>!m=gz1(LUg%RcnBGz3!<UqnCZt2FgFJdD38=QENS65iz0JY84}xYxH+T}r
zz?M58h@*!{iIx;BFgG=hfKVGVG~J5^QejxV*$WNMtEHIhpj>0?r=LL_o>r;VkD_6v
zjx1eM1XNoGP)*9+z=b0UK(+j-NA2LBwovyxT{X_`J86e|Kbs|``hGsDE1CTY!5DAY
z>l&n~<`8c^e&ZP0TX@nqVsBl!JpR~)`N@jK-9d-Yn~4Kc4C<FEVq`sXUfsf;#-v`c
zy{k2b0ZAWHVAe1pNuLqDKR`AAk@U43<~asai|giC+?c$f^BIv#=J7nx!z(?qs|jHR
zUv|3CM`LME&@>fC!B${*+j#_309h3(iN~-n^d*b-JH-{3Bt6*MGUzg&okaVnVLJxe
z<bu%qxgBW3l6cioKBs!$MZkq;wTB)uZz>@6HY@e8V2CVUVW64*(Qfq4eKCh_b424t
zbk*!FcPL4eM`9>~=%6i|wWZH<h|K(*@97q4KWQQFmqm}rx+};-Fv_b;bngmh2pp<6
z{i9ciCFo*JtP9wD%8`Dkcf>6?w2xjnQ31S!{D}~jJ}2ePFPTwzp^ferq_;Xg+EL;h
z&#MA1$6=D=RRSmG#C${OKh6vClCK1c!f8KtrWS3M3?J%UHMu~{W{_w(+QVmxi-xi}
ziU|$w3AlRF)R=V&1jh?QYG4*)GE!YP6voW-pu-(!Whh+<_iXC6F`x<^u#lqU`?YJ=
zR$$0h3VIPkO>c7<?SH)cc3M4u<)ICX4!_aJPs(gz!I3zS3fWh2lzh1QNg$h&pP!Y1
zT}VxD-f~t!3{I>AKRyQ+FhZ**ruz7BV#;W&o?RA44ooVgMc}?JO{;)aP}wNh6oTo%
zR};lU?p?Ycn)_ewIz&~p^s>X;G^XL!RQ21l^99<KcTHRf{g%RR*v}@)e`lz16+iTV
zA<>YqIX5ak{rBzTL%-ngJT#wjK_%s_<AG!4*Eh=Ak;SHG&`?}Qqzta<Pq$=fZ8*B@
zI5miHkjY(OY0?7qT32Ej+5yrqJ+P8>FmkZ34$6b$#g(wS1?h+(Sge-wMkEXZG=pTr
z7}Lhr=*%SVvq($3rjXobjnZ%GwX(cJ+G!J7svE}<2&;$)Jk<BB*~Y&*1XDy0E{6vQ
zU9pVUc1Tau&N}i?DAXBF$_Rgn7V~6(hG(xxpdzjky}$4nzTOP*(b#eAAH6jA%OIYk
zUwUJ@^2vC|sEa+#ep4}r*{M<?B7%t^Ye(ap*Z-Sbn9lH$mh8vsr$w8YTMwCD6q=u7
zGjQQkOtCnB<CwQBJef~7sr_0?IlBmf#lowD+ntOKl6O(ChouqB$3`3<D4R&P4d_9A
zpz|@X@#Tx+ZMcMm_xn1aaG;~3y8)RHTXQsf8~FN6T5H8H0Y)kE6_>U79tDtQ8|&W%
zgdnYCl8NKzL!0YY&4zWioO=XXC0yh*LXa@S<l~1lTg;dtO7^>S#_xmWY`3A&K*eB-
zU2*XJ&h52t)^aCAHfB$B@G)tJSIy9>09+-D6SaUC;xlJ@X=rpkMJlhtr!1j=JamlQ
zMLIe;-JJ=A+nYP2Lw>r8{e)ckjuDJKBuk>AGEfniy^+(yI;REh-JlKcH3BmH<+$>S
zYqq}|8xB5n&p5F)e2^2}|M~HI0B%HuejOU-`jlo@l+?saCiY;O5Vj>K?wgS!i3gO=
zjh04hJM!*`oy5hJ9q9&y*BM%-Lvt{8cN>`pgBG7XbDKt~-E3H)d5=7>$|I4g3qiCf
zJ$}|^v&=UMZ4S~%ulNS}g^4G@ni<ke--D}GH?rIcn2hh@7GdV>E-jmX4rGn%V$OC@
zFSI%acay7zmIEaRIR@`!DmNU-6H&-Y5LmkT>6tE?3{7G2sne$PXiZX=@Z>O@g1<(@
zf5yECC+H=mk0cOYc!zku-zyd3zs{lhBrRQ9(@(t3Mun?I2b@n9u8DV5+@xO%h#04W
zR=;Ik*)~9Xx4@C&a|FUoy%&4w(=R0JpZluMi#Ayw=vuv8P9_}r(#(XP2Rm9xfmg8z
z@4Q6KkE1Y2FGMZ%i_DWspcuwcXx7zaQGtj=?@o&xSTn1rg@uKQOV#9o;B<)vs6i^Q
z<KW=f!XXOS%_zE18h<w2&v+b0M$N`14=gF}jM9COdC%>Dg=I?exRDZ&t34NX*95WK
zu?&L=AA|v*w7N0J^<nx4goL4AA;A@KaKq0j)}eXV(~~$1Ak}m9MI>(SLENl;=3eFv
zP)85MO=|?88hXS#kMp2K%k!x}J{l^E4!OM9DKZd9erKik@ZZHnFuk5$47G4&pAjCU
zx0h~wXMA4LB;V8Xi`XU_=dl3<W_D1KufdwQ1SxF-&;7sPl0$nGRrM~aP7FQy`9qzi
z;Jr$x%C8yP+JQMU$I-NaK#OpAV%9j0=o0ZaUxI!rm%%689w#_Ec~~G2Yc90&Ff8a5
zK)5zPm+q9v;R!Bzv4BYUipvV<6p6VIK0y}uEkYD+D_xIo?*)CA&iU^{@v&HUK+cE@
zwonBz`R=<;ETTv`Ec|G-l@3Ury%n=F)0FoWQ4?|HoIi_{`Zx+~uO^wGcf^e21K(dX
z0sW+ZBcYJ+O}49FMfkIS+aB7D6*9aFUx^@=3TbKbr9v#_oc2*dFihu@6YH_oH}`@b
zSCP)|LMrn6F4jn+pPnu{4iP+b<e)6wRfBadI;vfWCRu|+iVVuD3wTgo@g=xjm$(=Y
z2G}R6@z<Mw9pQ2BuU~m)*FRa=Oj!AysPqR0xY|u&(pOS?b`QPC|Do`m3HwFilcPbV
zX!~FfjCqB~{?FHgcAauL>w{;3U`9)R#V%xlq_=9Ni5d7Dy@L)?>`)ve2=1+QB9zJ7
zMP`vm(*)#C4J2+?dsJpt0L>G@sV?ab{_HCmf5G3ho>(UK^hNiSVYknOcDhQ+J0MsS
zkZ2p;$gkz%XYbPwJJ~J=Su4_f0LbihO;(DZ(Q^MciutYA?|aiONDibw@Z`C10Rw-O
zx8Rt`!d){cZB-N~A3;0*!Xc;nf;<}4AFODG_9g9rJFqhQoEhcQ2n-!E-8^=hh}P*Y
zGlEVd9m)<L65Z1?cK%-108-|d`M|;Q`{<7zf(*{?&jC~C_17Z2!OMHKaw+#xIOKMc
zxls%MH0Noa9X03AdxHc=1Lzs*e3O%tBUVrtm7WByT?0(4YwsAbxWP(4MDHc2i_DY%
ze(yyfA~*2*KT_U0Vr+7{SA(2KBj99{fKaryKW)d3NIu(?YjS_sU3c|WBnt&C_X)nx
zfHA7LO*y}9Yye2A_SbSM!H_nl)&Mq|B35L>e~Qp`$S-AJDiygw3fCcvjt;#z67mT)
z;OqQgi_lbu^DVQLF@y|84royP#tP@|%V8w_!G!YSYm6ica9ypU7p}V8-uz;9Us=sp
zq9^>NXxQZ$VN%?hZ@f|!5)&7%)d-`z-7`F-_x}|;;1U~PK?h7Kx3Ri0)U$98?c3AO
zZRrw%1b1*zH@&?+9u=dqE?!5pQ8^mLH_CKdFvkT4Co6k~&ii3nJ;0peQs~yOXRRcI
zii!%MU?E)#t<n<aCXxbR<3)hBjXa<q432}=ekJ8|npTI?4qcqB^)CmR`STSbc(S!L
zSF2_fJrUpLdHS4pF@#qLvljs0(9-Yjs4H;#ZY}rl+@{@e({H=H{1n?Km(IUQ6cU=J
z0T#)#tCdL3h5?;d5?_qqM|*#L$0fDV@kv_JE9TK;mnjy%vt@LZrvUpmUqa_3EkUl?
zC!U3PNOav-llaip&O+0t>8;HDKd$lc)y*>+tMjJL-@l3QQfd-RS6VINSu*i5`I<sL
zK&@|o6lcRJtRab;{lIy@6YH^$kAGcgtdu$6l{Q26J<hxybnWkfVf*5>CCv93K?S7@
zupHSymc|m$VYwjN(7C?TCR8*rd|K6RHeP<61iaPWv=@O1Gkm{n{z@t@FJ}xw!tlv<
zYfi$%m?Rhct|q^x=TvfZUV&b!C7*MF2D{pMC!QFACNF1xfJ9C05ZX{Hrsw-}QT8xC
z*Xu;qayv>djKucbnc>a2s;M5~jrVnIRjZQT%flAR$-4+*p)Uix%=-f0QNjNN$ir~)
zGF;TmEKcz17QevuGvP#?5j%GQ=dI=KD>oEMurvUQI^@wO*Pn+p?TR(-01rAu!Dy%o
zyLg8|exX5zy>ddc&1wv;jxy|f1)f2ZQZ|6+1+eWwiSUNTmNmF=fP4Hd=cIdvYdh4-
zw>C8|&A9Xh-Y!Alo<5U;le_@dCLBit*SjSD+cp-NND8{&f!m2sU%FoW{)Z1SxE<iV
zTJeF~dE0vK8+hCD!>X~o4RiUA+ev45Ff~r>+zz~;pR?lYSC$JV%LmlMJT|+}DolUV
z<&)q=?b46V#>mRZtZM1;z_ETptfY6fZsso|L!x08UEX%Ir{Ormjy+C>BtdB3qj>4{
zfbOa`MSlG&#o50}Zg@A2NN%QxDj*ZgUDyQkGcid6%rvXHS@<PIhGYwP@3z?+HOKJ~
zIGe@wj%dYd^wS-MA*y<L^TgmR`L1y;3sIU&XUSqRg?u*f!BkvZxRve!lxnP*hotBe
zbRnaH67F`|K_3`05>2u+w-f<(cw^MIxyr?NS0Fd^=5yjt$w*v_D=S2vFw%6>JGU)z
zT~qu|NE4k{wDq5rB9nhno^-YT5x;8a@Q~T8@aYp?qLxUbRL+MBwQy;4cB>|VN&H`S
z;^l>negi%Ey9%Y$GFO+X$>0vX!uKs*hn#csd#AsO1J?w{<b`0Y_-UU6F8A{Q)(HPt
zGowVKJNYOqh-%(o?7a^{Los4)vfTpt+R**Io#|3{prmyG&?;CGZ~th8dIMT`1xB0B
ztzwL~1tss{-lrz66XSt0CzuX$^iOekH(yu5H{AZT29`TT5>vDm!lLVXJzVsYNT;Z#
zgs*uA$%_egaJBow_lpwXH`k{ko<Jz;M^9jD>e<E-GdCw9{|8I)qCl^2yKK1B^1^2O
zy{b@|<NRvq=dOv_tLE>3rQ{Wg;LXW>`Yc#8%y1&2Bkk>n9Mr#WPd7pO*O!4j72k-5
zj!VnPg(gmm(4QoEItPO)US2w=!YVd|J0?_cL>EsbcyO6axyVdaz*Cr;>Kf3+4PnF+
zq24^RM%Z|uCXu%0f-EmVI!FL&b@b$xPQ|S`z?$RneDb%dzo}oVcQ8>;Dux-PbE<O+
zr`$VVEh9}NC)@W=zCsGA-V@yoDdaA)Z1WJXCsMhEH`NcLO2(2ZW9ZCXZz|`!dL(mW
z(haHLF5$@01E}D|iV}RmZMNt0iuDW7kAFKDQEu<vY9hlY-2Qc??`T`MTQXlKINpku
zuqx%tZH=JO$AW6+{N3~CQ0Tt+OKWcWPp)xc{eJInyK7-nq!sNKTF-wR$@$5+o?+9C
zy<9Dbx18ylL=6wRre5?7YRpcu^FNLQ5>t+8j+IWuVLdKRD(v=7tFujL^Evd@>atl%
zBD4I-=DM?)0bU7*Vx|dp3r&EmLI^{G6LrautWNRlv$tF>`?@-JXB-t+OVb{$6B^7s
zH1(D*b#6otm&6i=DW;tdH;kY{d4=RQhRgj#apJ=Ju_rpNa3SU0hquRaLlc72@36wD
zAg~{t$~uOiPDJiHiElRv!}UL3EE{z?;Yl<?o5hJ<R~Y|N;wo2KZJ?TcJ&h8D%~Ip~
z0lhPy20e-hIg(PZlYsvJr*9!gF-Z(lDV0=biM5qF#zw_ai?lvFTaWSGQ(It@Kq<hh
z2;;Z$B>XlWLP<%<5C9Syz)A)`K@h$VJQL%=rh;6^ac&|Ds&(Cf_Spm!&k1Ny*o@Yo
z0Ut(eV=h%s-9yd*5QOn}N&C5E#`SZ9Srxc}Ink9d_7{*=PcXws>MPrp0ff(khN>WZ
zq97YJgG8nDedoC%;_F7ALsH*=RMnLWrjLaB4chm(v{1jjY}99ja0)D!gwGd>_TPUA
zq*D2Na^e}fCq*So9bm!ZWQ`Q_89HNIJ7zK^8<AZ<d^ZS|0wGExlsYVp>ouLL-gFOX
z|I2dh&M+#|D)tK_rt>EC8>5YbyUnI!!F>6$0KXy-2|Y-@Y?rYBgUB7b$TUR!bpPef
zz#<H2c)_TXjH04qg7@bbPBBm<nJ%sbB1^&DkJz(c--K&9cSrW<y2|Og%%-d*0fkHE
z{{8#itx=j!9d*6ht$;4%aYbqMT~4N^b}C>kA0HlKEy2vAi}BHx)q4)4FlA$X{GMSk
ziK4EpT8{QCzaMBjD^Im0E|zSVAP@VC%4nt4>Bi}Kc@wV3%WXU%-JxS~Yd08%Pd?C{
zc|lZ^MN{Ma%6=+L61)?|Z|~%q{ue#fD-OLJPcJNbcqCiu?$GWyqq6u!HJ%PbNl8YG
z9&a&~J|lAT%&wfnMK8m4auNPw|G@6QpehV7v!P2$I{TH=ZIWZjDX9$#s}uzFH!N~o
z`IHxeFwnA(%uKT_FL4_f+kn%5azIBK^N<2IA%ek32Ldu<DIi^@z~~znto1@Dr14T_
zW*LL_Jvs=mR~qT}_;CXWk)(jQ6{JJbuq)wfhN=wGEvw1~{r(j|Vy4rLeb6Uw%mZK0
zusLBBD47{ffWz2T-eOddg-r2FFZzDfownX~t}Ccrb|Azb#CK2wHbsM<7{k*3?JiHL
z$<N<D>^+nF)RkOa0rnOdUmH{f(P+h&s|;pHHHSgGWcMQ-?(4sjMnzOH<vT`T07QMr
zd{g!ig?fVPo}6F@T=yR9F9PUQ*ia6ubI#G|>!2yLVRi>UZ}`s(3w&d}my)5PrUR=)
zCz*9r_da8Psq@!qHri}b$H`+Thikxo<jVB!)r;B#rhr_JR05~P%1uMi0t<_YiJ6aS
zj<`gF(6AxqN$cys%=`@NQBojTg2ibBKxu=4ZCa4El2kaR1RdI*VxYoF!m2Cnc_Mn*
zkZ4^l-Bh-oV>tOoOiB$ZO~&4wrjI1y+dL8UJ&{Zr#}|xD(JXu4uB{P1KY>1&mxjir
z1D}i@ebL=kQdlodPOX2rjW4A8)O%k4K=}$1I5$^jGwP7|W79rI(=9(0YC3OhX?ot7
zsctYi`H$ZMtiF1o{322eXEp+901KCqnq5R!;R=t#0UBO;E9P>X3Y5cB&-rf#d!0v5
zNWl9#6VCVVc4J<kXk>ahZ9(&cI^X!XI2MsM*IQ&HMWy?FdHA@vXpu+O$fS_2*$M<U
z*4zC>88xuu$#CxL8(p6sD2{J~*ir<WPQR)rP=jnhX)_Qyi;+ew;x}(fCkngMLf)m8
zr9ov?xgKRG7|X1eg^V0_!19owxu~u3Wrz^>AVQ4KI$BwjPWv{Mt;wZ*A{q4+`qoB3
z!&7{We}UB1;~eA&P7Ft8M=Q+vv78)RDbvMAF;*z~%>yM>Dty)c$kpqS`lj7{=+8;|
z?`50>@YjxT$!<Spne?rE-D|DC+<gUMZ^iucpgsLu+LUBxRlRKKTJIZ%6X66M^Z}2{
z%}6_Q$`ByyFlPUbn19)xq`XvQ0+DB=%5=`T^RH!^u~+UsEAyqGc?7882ugBdG(P%{
z84QwgsMLzeUgQAvNm2ouWCa??BHkdp9}v*J-Q$7i8-SL9G)MwxwsgKfqAe{UqpJoA
zbPZm!S3g+a*t_tp;2X^DRlp3N?#A72f(Iub)Fps$B?_Ci=9DYOT>}Pa$HvsSj*x}5
z^_YeSJM6wam27{U?KEF6RfsUzXPn^6%t4K^a3*o4#pjrz(o^z^w`;j<9z^<&AzjR@
z@g+AleDCKeEuB#%MHYa8*$|U(D%p9jg>t^(ihL9S$0G;~tsF8z1pmD0&$-mJpf~d;
zhFuE8J0}cuqR_3nIGM=;Dql+G<B;brA_wv!e9TOU%OO6)VB5}6Ms*;~8U@T=@IWbm
zMTN!1&CRXEnde9;G<*S4iBhpAkeuOi&h#hbe@R!Ofy)>i!K}{qS`wJj)JS>zT*Px&
zHLB<O^7uz17zsueR#s%h6bKH7vh1Z=Mt?K5iF-q?IAr$G$n4;g@>gf76L`Ql8TGBK
zFl3adxeE7j^|QYwRQ%d#lWx2SzZ^1nyR0L7aw@Coc&FJ-K|1Y&jF_tEd97Yw51xfu
zXn}QfmW+#4ZX|weLFlUZpDy62lzwY&+T^;z{+SoHmn9|JkC7hSPv&`}($pE!_tfhZ
zG{#J3X74=sSJ8X;PkIu>pwe`ARQH8FKb{F49L{*SyEs;sdXD)K9AFzdt<BP5qmv_F
z6clrfV7f8~3pT;CgFagXG}GTTAzLhG2TF3YbSwK+qw>ELnXBI}vorl<wmkw&SA@1_
z2p-OI^0qn%rqjSJMUu69tbBVm8LLb|p|+b1Q9=nsw|`#kRsQ(M@YBorl<To~_buZ#
zSm?pX5V}IUw%3IX9)>hAC<ouxl!LXgcfVM&%`D~T-|suoQFhJg$AwE>I&e@!X<leC
z+VJ9N9J&tYLHJXonfIsRGeNa96jeQ${n68ZxU@b1M&L6msp$&tH45I8=0u+&01OHl
zzyI@we?U~M7zO>lwlqWN70Gfh!8{<BrxW*^GZc5@IJseJr9sQ<a5w`p9$B^}*reNV
z98YjGAPHue3P4^P36hyC2G_Ri-@>BCMwrRr4$$_D<f3aA>-JOj1Y_a?k$WKL>pdr7
z+})WsX(Gui8MFz6UeJxCq@;o$cl{>?nol<yn?E(DkP-E2NUl1yMSQAtvbB$je<2pw
z%0<?wcX!$@+twmUgrm2U#3AMmOxO{}4eMufkhX1vq>j&yOH;<Dlu=?_W?!SBm+Mo8
z)6$Z^dJQ`KzL2-ka0}aZKajNZ`BT<-NqDEk%QvmD<GEy|Irsp*%J6O?;~PhdlSSo8
z?kH$T3MJ9E*g^XBUk0Wx=AAro%h_x^ccpEMIW=Jab2qWk7qOmLvliU5NC)v^s2IRt
zz*WXTfoRD^NDb_7tN=5)8Rj|`HQOxi%zSypW!gboB6sJ`ofXjE_7B6NE~)n?V(2Jg
zprdPqTg(nkp_c5p*=7(mU^WnGSo_NLa(}Zss@y6)%P4xy=q_A1?&S$iJ25QqUR0^x
z*8mnq!x-M13USN-A&$1WVNrt2=VzWUavSjPkHdbTn|_J^#cf467#w+^q<VS1LcNo!
zSM03S|Ko=%@PE2}w|2Igy~A#np_O9iukSnrx(^PK<8uz;^H?T~&^A1kKl?A<;i2EE
zQ3H;S4L#%j^#jz$swB)4XF@U9d84t70=CMS*P&3;hfBxN`<R0a?EP5{LNSD`((xHb
zt>o2=Wj_iR$+PUb3u>FNu{9bLY=HA>O1E+A;aXy`4q<Mp8N~)`Yjg7#30sAvS6lis
zU(P`6Hx4kp8*<_-#J1xKr^UpM!35tp`5`n8<J&9Q)6astn0=cTYUMt5`b6%}1oC0C
z)i~d&{oYH2b_0&$>2Zi*qL2+uhrul9+-WZ;DgSZ9OV>u<cGVkxUeTc3P)sxH3%fD7
zUZte``fkh>@-Iy2SD?9|#Nbr+Z#*1%e_o-evQ38IRzEvZb*p@GR*w(hTB@Cw6>vBK
zh(j!#MMI%JBpxiZFnM{RbotW-9fDvAAx&UgPXf+F1&tQX;`^6C9^6(H=0;=4gj{Zc
zs^=op+vfL1)2FQW(h#3a$ctC1TSUl};z@JqJlRX<5jzuI#4r{IIc*FO!{D6qR)W-M
z5q;6aR<cUN@%<(_PF_7SBBc7L<%2bb$wM6;{Lu=<<A!B<g^(UaCFnkckb4{pJPaDW
zH`O0E;Y7!A*dUvWQMd2Py7kIVDRxa;?2RVdtyW8u3urokBc$V8eeu6MjE1)UY-%{y
zUh|{WU@peICvF@SrH~G{6moJrK;h$4g&voH$qfx<ebA@u$kh{3zfbA5<Sn@Tfhfmb
z0SAVh!Y)|8tIh6*B|l8C1O#a0+tL8$Rd5+UyFcWryQ(M?jTPm1f^W=F#nyS%vEh{K
zRH7L57j^KjYwKx_D42y2t>E73ef{k&?U=N5?fIoUxpTx~MjNBxIFh6(<B~I^A)05P
zWZo7NK(ffEBKHzTjl#XsRvQZod0yPSS@M?;fGKoMFx1>GdT+~nxdPElJ|yA%Q2jCa
z`b`W<o&@F)cn%EuQ2amF>dz}&x(B=hstIoEBm%a}8LvJ~&g+*IWzx2d=+z@)3dhh-
z5CW$3ivZzufYNoIBqkN{sK~4Z5j|;8h1HoKW*PL@`Cec%DXrcG1k)y@g|1mr_kO_&
zulYUYGjj^FUtTmt>V+}_;!h!Y-)b8b!3;A5p6%^UfOpQw>QD9fO*F6fYs>DM<w}~8
zDy8ZlH)4nGCOgu%WMPE54TB*v1YE8n`e3p_`m8?3pCjl)MSQrs=w`22HEhQ;aQ3@;
zW+Yx@W7`d>Qsv$8n2qhV^hn5Uio#MU|4I=5IfHU6x60+@-NaQ}%eNvAj|Vnjf}2=L
z*$Nt5ux<SLof&GPozFs{H2S9AAlcUdtLEkdmVbMM?#kP}Kx!I%P<JXP1C)8fA3sK&
zR4l;l9nHT}PfEH|E389J{y@)ScDSm5&As!@76t?lgl-JIjJgEqWkB&*h4bNR7=lNK
zJ)rxz`NDnhwaZ6Hdf&ZqDtp;cfA39B7+eIV+Kwy>)MfU<)>}l3P@mC`&OTQzwSJ0u
z6b^o>i%N}q&UVlp@R{Xz=TvTJf~H^a$0TYpJq%@)P9&TfevGPiAL;^LjvBV@K9+by
zL}_aJ$Uy2fR1hurQJU5h20&!`uOs3pujzv5a`!#E%|?gnTfZ>jePrTub+ad5oCW0#
z4K;vD>TfgLL*2iK^lN8fn1#6mW-}XWA5|H(OUub2<JqN7%So9pU~=q<&2m5&r5}?1
znhww{n`KPB8tzZ3z;#YenA+=jWsgMHE4EU!+ladWFf#zEOOlwab*e3!J^zRNqjd?n
z8vzkf1V(PXk1|}4Fp?(hu5}H3CvyumC-|073<x6)Z|iywpKulvERm#x)mz18P9FZu
z92|;=V-7I#u-8Em9HVgYW9)}JD``7M>$z$6lx{CFK2EL=D=BA)=OL^xV6ei;6#1A^
z{`a|v*2tsF@I4Q4j3O(LZ9uAv+UC|2?KgL${QX7o^TLn$lP&NJd<gIgM#(1#UIJ-Q
z>p&+Ewy~Lh;aC>wtf@2y0*vofEkH9Fsi?}kv)?`F+Y0=&mwG>%$e4&1FZW-blpLus
zKWh{E3nMk=VXsEC78POJ1E3L@PKy4ipLy!R0k2l}5kGSG(2q1Begs!`H3}&@S5#!Z
zz>mBIKhlKw5edo7UzbQj0g|fW)=Q0K6}AO;vnL{MAx#>iy^XhJek4kjk1M8R#}%o0
zBscRpj-d(>e{!~I4W?rMT*0GP`k3CCQg2p1WL7uy$*8ju{}}qtEOoFmr@=)3@LCNU
zM47y1St15>zB5mrJPCm*^HOT7_9P!<_NnMc>vbq44PizuCj+#-!+@L)XHpz8w`(6v
zVC&FaTuH)%mtDB3&T|jfjCxWgoLtt{Ezlmk>%X#B#>X`+beNOPrx<YBA8wXc^)oAL
z-i%oGz%-kKSo`XSO3g7@<V7Px$Qa~fV~D*hseCwg7-?A;;l3|2L>itR*1DZC1(r`*
zt)I9Hwd4;f-YolIStmZ__tz!IU+x%k9BV~_3@I=9zlU;=4v&WRX!AOQA^0^G5~Bx-
zFc3R=R=L<vsrlngQ7jD|IB>PZJL6(CpU^?#xLELDchwRdW|BxIZ=$RyH-{V(kulJm
zIanhe29x<^a1u-VC4NTj&+2*cm@kSN+Kl*iypz9#FaYO~f)N4T3yHzET2#)SLPKG*
zJ?INX0Gz`Jb79%0?L4vABy@8j?}$L!F~Q(WzGB5$vM|F@7^OwkfN#ABsVQq`M+YvK
zvhT0Q>t&DBb_3c>jbw=MGs1+mx~e>2hprx`Y`9iYN%&S_htjR#X;-1;CDI+c<0mhj
zk8vlO$I{RQSmG#n|CdMkA}P96vEzTTx&grLV3)7O@vfHXEPFq)vn|rrfG6m477|uu
zkQNQ(rXGM#d!t#D0c)80ct0StqynQr5XJnW8Z~kq*rkR*Z<|$xRD>OlP{E^1uF_&}
zgPZH~7S071!G!AEVmoua+s|@lJ0Bw1S*;WutsMAZ1r^IHkepq|YSFe)tQve!fY0hw
zJr6YsGOE1j_%f9Wj)&(14+j<z&I4VO@kWCgC?&DcEIT+q@^DHXex<;ZAq};gC+T7z
zE#A8~P&XzUvD5Tis9>+=#*p7IS5je+M~3&AQx5xtC_L1CdhjFha+k;cS3Z(Xz4URL
zXyr$b-!0Pttl$%v#G%i0Evd%n0?*&M90WCK#Tf{~(TcYq6SBp|J<lTvvJP{6KVi9C
zBkXui)-JU{Sn|lC!`bfa_{?{ptX5#w#ne+4dG_>+py=te#Sp;X;{5Ejnz6awN$9lh
zq=?NU0SF<y77`#v^L!R6Dr|<|MpNq?n&1B{4k-5Bfjxm~fs9=gnBRAhTeEm?!CO=f
z?K93>vpnE^BOR~KEhA0MxPz-HH-0_p`)r6}9QiR?TgI4p(v1MXwWCKySj$h>&k7GZ
zoT@YW@D1n)PiRk<DV#X|<px+vSzhO*|Mn(cUOuQC+LVTGL>{55)>3B!n2Z@W?bBJ>
zhE%RjpO6NF%|mkdg|8eOP#DV9j6HzmaxxR;i&f!>sLAS7D}s!6fcVy6h73%(^H}uX
z+E^S}EUvuv=)|lPDbn9T=C$JD;;Keqw4E!!@Tgx_+MQj1A02XtG=w9*_|0$;ouTx>
zP34F{PM(qy8$Yn6$v!$hRtQ%Lhw@6^@3DW|QeLZ5(ZP`|<<2QLYJd|yZ!UZ;cndMA
zk>{K+U(OMIjDI%9yz1Pc@PrpHTr<TbXWCe?+V4SSX*Mrv@@|(;w7rhDhk0a}16z`q
zm(tfKK)CVvUu59KTpBx%X5aq9345ucg!NpthqYXKnZ_na{cP}tC1~PgV$B1lZu^}A
z6J|jp6gUT;1vX_(&^*{)Ges;0MhEC5vLz_pZvius8rC~C<Xb>%9eRYWu#d-KKrK^^
z^>(32C^8YAz1tMY42lq}dTz@;nc97x06XN^XR;s)a?NiYWm>9m;{0xviRU3?zouxm
zR44*~sPQ{M=z2FlHFSoC>gAl|bPDlS={rcc%Geac{PR6Fh(_tjfAQhrjWd8s`@L^P
zYb3*-HcqZ&*Omw2lphg~7>MjByO9DcP-sqC|JWfp=)}uR$n@+d(`^6C=|CA3jzg;}
z7h2D-OWia*;@~tmk{_`4UAOu4D?(@*;lmZ9OAWn<pI21_g>N#T?kg~=J4uSHGdBXV
z0U-fF=<CnW;xYQUG$zmlt6C>S4z`{#R&(9X)fKv3?vO2;AlS!h`e?=IO_*R3pk4wS
z{I0bu#;N%qKL;6TibMo*(2J1J1)D{8l$meCT|WAZ9Hliu5U(;%G4T42cfkZIS5gqY
zd;7`XXXfQqiB!hAt8#^hXY>g^s@T}Hx8Z3Y{;%aew$V9F`o@HQG@e)2C;|riQj>7q
zUOsiY!r$%w689L|DU{%zXGm03YAxdo`tLvar>2Bgw%9D&%kL2CrS0jeW3SoHl3k>%
z8q{vmRw+E@`W6Q!$5|mJW`CT;0}ny_<*l(otcYh-y*reFj7TvA&9RB}t1!YtD&%7K
z)J6dVRJfx+lc`OGx6?weWG^U4FHa%bTEx@SVzxU6NrV5~bs%GabU7WsbN`n;oB?}q
z{d`sFkZv%8WaI}w!&_=ap1Kr5IZF&8Eoo#T;pq~amhQv0VnzteT*9S>pZ6Aw;k)3@
zhktiy5OY*Khc+2E&Y-(@gH2nWNtQcxV6!$pLvqowNS5F7M9^Pptt>sV0n7AjdMMa<
z*-Z*ULh8oG#@PdHp94>j{me6JK6?oS9ya6<2nAMAnLJ&oz(%;%TR=&qyL|a_E>kLt
zh#i-1LSmH|VeY*+Gh)f;DB_8a6!t}PPd)xAM4<NOFlfBCWmYF8N9~gBm|U2ycM}?<
zFp+WeJ2f6i@l!3p01gBm+#ZIT)C59wxCGmt_8)E3VIhLWO^!eI-9lPw6XW^|?N{z6
z;W|5}J-_03BmVU2?BuKXOzGSl-qkm?^r(e@NY{sql>hwkQ$l|%&VbAHc9!RPZY5?Q
zK2Ov|Ye~)$c9%`Ji>uTq5-e##4x_801bFcp;W;kDCy4oXxkFKC1bu_?2bU+ZONPVX
zDoHDcxprs8Tb4;x9qjLD6LGkKumCcJBjB(Qli50Wmcn_00P4x72$Qb3uh6I|_uOWD
znN9=_Z=uK6>f*;0UDcsyPGNuf*RAogp|}8)!-$qrGZ8aY?C{%<;xmR$9g2(n@+G|*
zfP>!`+<JWY{9nd<vHO!nOkjdeby;v%BKb2??whb@;CmZ==Kyu=qY)`p^Cb|1z)dJs
z`K(=$p-DN+N9JzM>pZt{epoGy0v&2`p(KmcIAplwpGWfh)h^>pZI#`hkI$8!OI)Z|
zYwbMaAKuEX9`tn>NqRSTiG<Ko@n;50-$K`z-aQK@*`q-jH^4?o@{<8L7#rxgH-PdI
zcR<I%#$1;S@UbV=qVfr02V-v4c1Fe4R6JDJQ|$7LdO^TLRElti#kuWBcLf?c^z3cJ
zteiV+CKTaeW9Czy@sg#=9}xn_pAXu?{*P&)JO5A9N<d7j&6&aM(6rp4s2lb43={Y{
z@^ysuQ-X9zE%@A3>mMRd#}+**Oc!IXnoB<W+*R>LnjCc>F{A-G0kl7Kpg%axPXMiW
z0kkU65GgeLOb<g+4HE1nA<Ca!HxPv&YovM{EWHEVyrlyS1iL0-+l_b9-~G9~veL{x
zorh2?k&dU$dd3dBr0k0;s=7{NK4k0M*yH2l6EOY|O@7ru67ml_kX{3c^w|798&6}J
zGo2@o9TU-(62EqfRfsmQC*g}=4uHukpKMR1-1#1cAQ688&J@WL8Of-bWzHnS&#F60
znaVLO=Lc_nw>0@v_SH-w=d*g;Yw7S%?Fb-TTv?#CRjlu#%%-?Zz%KSeiY#bJ#;|Qj
zN9IR)et^jPq*ExCOK>tW^32!%SA~#%XfnNEGG`LA?y`TfDw9adSZUGW4Hi+n`NS(2
zIhk*b40fm%YDMTvN=a{T_E0%24zr#*&(mN(14&*aqN-{UZ`0mTVAPxpyJRdU9kg%n
z?2yJTjS1^q_Z;)S(otkVr?Z`^^4o<RLtDkESnaxYyI?t)-dlEooUU~~x*Isoe@g$n
zTu4oN9e?<fAyzIb0vSWPu>u$7`hA@Kl`0x*uKV3=r3;E80y$}%HkDosRK+wG-@A%G
zMgQ|R{(9v`^wVygGqzl?yW|Cn#?@pKFE_y?p9mI;`?WJzC(+EYAc;P|DtQJ<hL?to
zVT2?k(G=ejJ@+qx&PyxzFvxl;4ZE$%m3#-SmoVVK$$&7pR_P11|43}zT{E#)s{K|I
zxLYyB|9Y^gg#XV63sm}!v2(v}v?wN8Ilav}A(z4+QtnU8%a9W`MX_qpa&oq3Zr^4O
zZK2Bat!gemt%DE*SWljJOO4Y@z7g=GVx~fEv^a`*CG!(-1Rr3BG<etjb_{>qJMwDr
zkb9VavigZlFehFp#fywaZpSV!2+X)l5Y`^a<JQ3lvaC1cSj|8<>A?B@ya!_}F7fiJ
z9?o7wvDywGY{(S72ZR(9Kh!M!B4tiWbyzhy2-yD!m>Rozi=E19!<-gGh5yqFWKk`B
zR>w4I{1hIC;)c-(o?Jc6{|qq0h?OsHF}NH)L3h+bTmyu-m$bG^Q=G7w0k+*|BO;aX
z?-MOv_8tfG4-wRpO=Kk`61WxqHF*;AjjeLc#;o_C_qYN`!CeZDGylsiTnq`UsTU!O
z+!`ToGzt_71Xa)`BvWn7(|1M3=S4pOO^Zjh3{S?W^clP-CJd1dG>dM9hWzu(RN%l@
zdJUd6!VX?T&|jWC16e-tq`K&e>xYjYrC`k*QWIKsA=5ud{|t0DVCp(Id0%AX$8o_7
zroC@JpYfQTms2F5dW9kkU{>uTCU&$ZCb)pj`10b#vmvk$<s!c5E)8`6aV<91Mse8_
zFY0vBnOBTVl|5?*gYU%YsbZzLN?KYR6C}8SNI$aqyp7b|hI_M$pMif->OwO=S)N;5
zK2T4tyerL+h%OQ2Kl#TKql?urN_X%18%1f`Y3Dl7RdN&>>DO@L{JApqNIUIY^iewv
z_41exnIL<HQbR4{w!T}gJSS$GuKfIRUD@9G^`<LEhLgQl3iH0$xa1kr=)~M=?Y0gf
zL-QmYC2LPq3(Wst<!0}t2W#R{F7)g@W8ZE&oUXGjwW?aXzg|_5^X=0|FSzkZA;%Np
zGCz1+1EeN9AUDmyEIvCA#kc0B!NI|b4JfG0z(4zJpSj-)#W>8=zJ^^2<=^K(G%CUM
zhd@cO=&;lN9&23XC|+~gx%0~ML?`H(@^vF7OQrsl0{EBFRfY$$wni*V46hsP@OF)y
z|ELibNc+bv1L%J`@hH+}k-Q9fWV{;gLl%^9#&&68GMOS!Ws}MIlZ(;417`~#9RnOZ
z<}?XcUbXng84jnsE3kg19GiaPQF<R26rTU{62d=gPI$*AT6LNy?wN_*GfvC8C47sU
z@`_$)<ilyS-p{%OH~nh#(9hWXY<}nnMN@bU%tozkRv+lV+|%aehoE_6G5D_CK;(5D
z$N<>SrbIK*(9~rgaJLD*hSIXM2xI({r)Ph5J2!tXS5bLxc5Y^WPca?8)Zb<>`Xyi1
z-t%s(ykX42+B=sb=2;*C9zXu{_Fu8n2R!l5k6gb12k&D<a0b&DLPa5u+T-t@PH27p
z#h^Ao==`@bzKd7uhpA6b)`_h-C8A#Rggh*j|EbM!;sCLO>h8ey|7iy_AqlbBXbRb1
z#td#dALpPTb6;Wq=G+#tKjPcIU2QCVkUNt2jn^U3t+%lK?wUZ~;sobyJ`&k|+vJoT
z$4z<T%%1AGs2}=Ifk)qoVw})F&*}|*H7#$1m4(c3mJ9*cl8KI<-lgjRW)+nH-W%NZ
z_4S<t{T(F$wg&r;n;NWAO4x5sL`t{u9Hf>!v{Ww?5_^>UjJ>x7k>mWGlhP5!tPJ04
zJ#cXKe|pJc+S-w4a(ujF+?wp4-w}NSHB#EG`(y}h7g~?8@PS5SbM>h(zhMOkI61t1
zTOZx*Xu2DJ!N~R$rmCILtohpalhPTK77UoB1WQ_Ip9%IHX3t&z=Rf?XWc3m~b~4#g
zIJvehhs)yTNqZVz57EwwpJ``9InK+oO5eGa#t|TYF1Nyip=1zSVb{4Ov2uTb<gBK=
zQe>S1-!c&zii}#63e%4TRVR8;17qXIP$sP4A{TVxNPKzS2Rq=&6QB$FtqjP!+XCWe
zK3w6P#z(y4(VKTCj^oZltwiX+nZl&HrG<CF9l_F{pQV@!4fM(;RirPc1<()2CNJgT
zH`|88NYas!UwnWH8Zj*w$(Lo|YGIdEt9GCh5-Btq9pxnY{VT`Ns_~(us@9-CaX7q%
zuMf7qRPi#6c|xD;1~?c8hqa$S%gJ3_Hd1%feEK*~&=^l*zdGNU`z8oS(V5Ci6qKVR
zSYBL^_+NR=#Zeg;EEb^c4*Z&Yt<_Dy@GdlWDX|nAYVpqcQ&HnLpo8AKIuNW@(oD{n
z#=x$apu4B7Dwcf_{ruOxVzoWb1#G_-nfE+wO&j{$*ceK%S%)XmGBN^#b(Om-&07QB
zEA#{FF~OLrx!un$>@x-3=^DICQ^#xnz*hAco0Y2c%0GsKg-Wl9*yJ<cFBxEs9eQ^)
zZmz<YGirkRPjP(=odE#g>1341ugjSy?{)ojLUte%9EB7dg@@V($G(TzE_v4vb;A@6
zQcnYm2M)yLe%C!vyP96M;?=ssC;pH`ej6;-Mf{8CnS7LALyT4A;^ppEGOAPg(RPj#
zfz;OXOJEMhIXJJcUo=6Tz&-a8IyIB6J&eOkFb<Ll^Fv#!3NkXCM$tN{tXjnrC8SoL
z1y^N(+<&vZz1{Nr4GS`1Hw|Rv9B=_42VR52VdUd0PXFOCAp9M??pyG>^^{<ML?>u7
z-prFo>;Ab&SP?h=_??Lex*+xX?pz(IHav@2RlO^JJxzFqmL1FgA7$?y&vo1WkMm9;
zWfros_sZUT%SvP<H0%|kY$9c3&&Zw$MVX<Jk&(Suk<5&Y?BDUy)s^ng_5I^_Kl;P{
zc+}<nKF{Mk#`Aa{&m)WR#V@N`K0KA=AjCk_;8Y&yuHF4MKhD4R{<&@@tAOIg7>BGi
z`+V?XB*S6JSkx2xa5SG3*8ZbX0_SVv8}kHrt|sq{BlmE^Pb`cl5aRr7KO}IWaEUA+
z#X#!BYAq;>REw%tRniFThDuqAu-*feYGh-5VNe@Zcsf|K+1&t)N=0404-TLJRy7d!
z_i}A|?seOV3p?HCxKaD9QpjkqFqWf28G1<n909%_N_*O_Ygb;BJt2A}v-}5H!~evb
zB-T1LksPaKs$UPv8u258gW5Ehbf^y$VxozIH~L71*XD9+6BFJ&8|Y1^<@1u#?#<J0
zezxxnsoR(1@G!;#-yAveKgOs>W8y*+n7OUxHrGzimMwz|K{gz|J4Gq!spGyT1WFnj
zS3t=_Mn<LwNK!D^!fHPUC}uZEf_oY~fR)9$2p9CR_&Prj&R}=kZclcqN9}A_My2_x
zN|t%zLC+Vzx&u4+2jAaF)R^!NANdf`dJ`R;7w>3X%O*NEeQ?e2AFuGJu?hXi-})&7
z6|+a0(y3pbLWL15_0zR6xTkc@*0&{D=ftn=|0H>{l``InWt1Mzf1}0P;~-n4#O(ti
zrJ^}Bw&6w8r`rzvuSU$_`TOXg*aRqhI`Z<ev#h6Cr??|zaI?1vNzrpof^W!UdtZ)^
zrHRNp&ZY-*UIL~8Kdk<K1v)WM*t9)agvU|ztgte_DFL=e%~yVnHH6JtAE?JbbLY<Y
zk-_KU@dCJ%Ej`Imx)zNRo3tdlr1VMJ3`-<U;2_z$BL2t}P}d#LZXM0s3U<-sX@OAE
zd++oO_GWoxzrgRL{p3l9D0OLq*S}l>w=v=pLf>#qnaHKb-#6DV-iY7WCvMYU4&;)J
zWz}7@I?HD&l@4C48jRZ&l|;AQ9;tNXSpQf+w62MK;<NDgPK?Bk51r0BlEz;i`iz9J
zpLt*hyUw3U1yK9U;SmvOh+q@*v`Mu>2x%Gs1M@9hwXMTXMO_OEVK$+9Q-04HgmcJX
z1?BK}*tHEbYk8#jHFW2;x%q9%Y6|DsdUwP9SdB8Po8{TKmT6X61hL;3K?d@l&muK}
zO(Bw)_r1BrnU=0`t@%T;zJB+2Yg0EW!atvuS~D6FkTVRj|BPTvW5Y$2*eahCChcUN
zo)|1G5O;KPT5la`+ctT@f7~IM+&($+Y)n^eUcRDseVP!{28|{eL{`;|h6}%c@xM<I
zag{%H;bd#p=|L-K7x9V`*pO`{)B&gY&fDW8<zIut!mbz?7+f0$?Gt{JI<&IA<ANc<
zXDVG5^I`Wx#bUl;Sz#=vQ9ei}R=&dAZgBchX7OFHPx1~Oduh@$lwNvb68YJg`c^<L
z|GC^cT3;3?LCbwwYfc3!S4HjQe(r|!goqRW{slDOcTfXq5BL-R`X{Z#NaFc9lGu@(
z_7c1SYfJa3a;k;po+`_2o7uvy2g`%E*R6EAgsu<-4k%HrP4Heu(KrM_?wMn<4D72v
zM)mJcPcXdHMN>^bJs03`|50wC<}e!8h=Y#drwWJh_SIM`F}mS#9$wx%%?xoC(2&TX
zYy`eqA*>MnS-PWcD+Xd`*Q6hwWz$U>DgeF6d7z<O+FE_PvKCn6HS+vIi9q=|y;rxF
z?%dvXtuoNfqQm=h={`6|FFm2*zb`!rx%8a(68{_k0dfHSU5FDv4!~B~{bp5=P-gWD
zY2&@poXo)OYVTeBm|JHWL-Rp-U;qP#k|e<kPDT9*hDUz#*+WU%ilu(;Ze^S5pp`I~
z!~~n#k1@l*(qjV)uzu70K&k@=WKk4h?`$7b-zW9u<m8$L&YwG{Z(;oQ0gS-$bG1Wr
zc+}<JP=He(S~?y!GiI*KBHq1AK6-N7XqI)tA?#WBv7>Uc9t_K^BWiBmow&vUYy#>*
z^Le*9$m`Dx2SUL5_Y=l*fd4ncA^HXVfH9nLWVX3foK*^yhtL@(=3M>l#exNjP%qE>
zHAb(>)}%e0Z@v@O;mbDG)+r>N&UolAjpo~p?Enb;*`a^0{HPfeBSzon7IhjtermXE
zaRJ{CxjJc7US6Knc+DH!^AcArE+b58yXut&nZ+DyJx~Y~aaqV+TXQZOg;kh{8uL;g
zu@ZShcMzt!L0a<8=r`LB6hw0hR$uDm0s;f{A|6xx`9{*zCc9q4;`fk3!#I<@eSNCj
zQq};&msR`Eh?+wsT)B{E0O@(eR-~vAhx{WyHY@x-K{6B?q2lXJO&>cai=KCHT5q_|
zS#9q_5L2n)kd9&1tbgllLx}MP3|{(x0PVL2J^Ib-&{2|hLa79wiUe%wImqSMvb@tR
zSFoXb*nS3Vi?<&}!ts|;poNfDShH>Qb*g;4ptt<?4OqS`19D8)&b1~9Ys2J?^E^2j
zSw4iodz9}NO#<zYI=Zk@#L-pSxjiHPIy$5$Aw~G7<oTIZ$Ll0Gy9Yu4oV&_Fw$&hY
zL*!NZKZuu9st{y5Eb4)tzm6@Q4vBk(x4_}wOa+JkE!{2C+-$xxuHUj^UPi0RxZy))
z*ru{(z(dW$M9@zL!d|ML8RP#*Asiajgai6XVIdD+QYw@eEBtpQjf5VAN>O8h){Qib
zcd+rrL|&s<g&{=?n7G=y4cn$u?Q6Cey+y;v#02e2ra4FKwxZw3v{=IB@(T-oTpgkR
zb8%Q#eeWys^xjzGNp5oBS~hD4nQYH5uq=Fc6GG9ycj4fNT(9X_(qo6$Ta38zA3<?z
zRom&{)6$p|(uc!x%;(cc`Ypar%gVdHw9t6suP5$keM8<qP%k<HG!E*7X@8gENB+ae
zuUJ4T^F8Z2EPO0E_@)?N@_Cprc9l|Hm&V+npcz%5ARtHsKu;eV8ym@9r>3X#0kBDg
zF?yJ_%Z80w`9Rf`zq2{K@sIx+;MsRC3hS;N?tA%n&3wgVHA!vTWCd51F&=`9fxyET
z^d!nhKF=~eJl?1N1HTe-pQYnS^44<&p6*2iH1K@c*g2_BChQs7x9|A#W|XZY*TwvN
zOqj~iW0t3|F~lImL_h=SiiL~+2^yH9TKdm->~R$c*k1m|AJSJYN~jjBVFFoPHbn2F
z3eJ#|zNrLFO-&s@OfHZQCa!3!0P&A)*lc(?P3CynM!LXbD)D?|w-3<SFO`_J1<g&D
zEn3!|?-Y-^0<`U2rn4k+mHD0=qn@@`-<tiCe;i&Ub5z}1x`eq4Zij$@sNINu48N^X
zFi&XQ{XeJz{&E)F^v_FCzm)x8o&qss^%{0C<P0$6AF_=+{Ht61L(9VipOV$ScfPoF
zC$u~-O-8mQA?eXFbj38mlKp6H{u8n@c~kdteTRQFfIpjI-|#(Y=$nkZ|MQM7qkwFs
z3z_+=JM+3__dvi52u{|qhKAgBNbEs}^sIm1xg}>D$cbvC;knD)PhXn~OZ<zF`J$1;
zi<w7B=vBXu)PH=w<1)vW|IN;-37~kjkDmFJW%}43-g0+8+;Rvw=%GEIX9J7NmU%-j
z_w+iG*@v9;pCwi<J-8d-559`lv5R_wu}(Oh9FzQyk390z08*?Q4k!mn)!}G1(M-g}
zW7AHYM*$-VNlEWaJ>1b|jEss}2Mx&WM*54Sw_MZ<zk7SVf{<<@v-U++H#n%H=WUx#
z?|Kx;zkwG>erJScDy1qKXKsl^L`=36=Y*d1*1aVznEAnSv-1xq2J@>30vjiG@9<Ln
z^pVW-%6Nzju7Ylm@R`@+?##09Q&fA4sveV!S>A4mS+tt{Zav@9y_8Fg7wn9M5#y6{
zun$Su{g3yX|9HSZ;f>lPunup2W8Qbx`l70*R|aY5Y=2MbPYjUEGe^9B!u-2jd@L1T
zz;C2rtFo4QyRo3tIpfRiPsI(5s$5R1MCJVun12&6%#q@*XCbUs&8w_7r(p2(iT|H7
z^yT8d*Uso!d^eDRh*B~&&!E2Xe~PTTZ17IUR6763%f_)WtZ4$ZG8ATCG*bOI`4S)x
zp8X4x>%c@uW+}5yQi9)y4k4s_c3C20mY9n}DTStmw7A|2;-)(tl`gr@(T(PY4=1XN
zaU33#E~KDM|5s5yfr0qqLr3TfPa_abuN#`xRzP7G0`)Pkt*<1DV{Z%<hGAJm^wXz>
zAt52v`U-M#xgJ}qL60730i3m3<52`jXS?|3Rep;O>LLy_p?JR(oXpHj3(!xoE6Ifw
z6$laM6{43U-XXYO;t|(S^P)4>61PS|{YvTcTdH=Ee=Y{&_2H|2T77Xg!TYQOuXV$U
zxlW&xtU>(Qr@wad{W{iY11QW?)Ea$%|6gi79AQ#<3aB~u_fJM&#~n8MKpsMiD{AzC
zV!`W9)lc@(jU}f=suNp)$T5KPK93N~{|yNL{<J`%SLsqq<#9XI@|bX~R!GB|wUvPf
z3TSb>$g66-!{T_kp&L7vT^}asi7f$bH-P<puVKyP&4b8=T%)Sj@VK*KPvyKH4yPtS
z?`O`Q)rrJkldoe!m5sSQLlmKi=RvcweW#T?rQ76g)}D*prdWx}IsNJJ5#5d-`5UjC
zH(%X8{$vm=@b^cAe;oDArVBIJ^(TF74)r7^pHA>kQjO7n{Q9qI@+<OCgoXcc9QaT@
z9msZMDib}22^o8))#1WPvN_f-#hV=|csE}-(Ta0D%P`!(_c&RWS`dYmL|Z5Pg@v^G
zzqs;Jb&LppRcb|%!L)O!>23B`1`gC~BxVY@EfM8pXU{<b)(;?O<du3FQPJU@`z;;a
zcN2ZJT0cX_@(T1GU@@fL0}F~EpH0;4>}=)FFTt<wu3Rs>A3{q9velZf-x;7!+xK^6
zR_b!DSj3&yn6PlO%Va!t@v@$UQtAtyBRpkHEDDFc!eP3530=HxTzY*I3m>7a5OB<_
z?I@p{nH}C!fSSdxl%~!GtUR8>#>>N-bE^1G<Au{sXK!=wD_CvLZ+Q3jhLWV%ZgMxl
z`^dq9H1*^O`y&D5U(3KRR-V{-wX@a{Y^~NZR(ZQXVbk{dU#8Is6Gm&WL1+&8{9RxM
z6c<z`aaLewBPgm}%h%3%OxI9Pjk;n28_NOIit*Z8zfz`5mhx6`+@t$BN4UVt5-0vI
zTfGJ^W>dd*=M$g(T-o`&BEQmW74YPLE#Z`Jvy;H={STr5I51E9qd&5uI&YMleB$?C
zfWMkV4PRbe?|8#ux`l6pV3`MJH~VF9A+R=A;R5i%K1}ktopZe>>wUuOu>0uHcdmjs
z|DDR_2;%i`1%SB7P3x-x6xVo)!et>5Vi;Ga!4&O_O~M*Lid4~coT>J<wq8&r&u7uj
zp4w||zHoq@b^O6ZuNeqW?XMSk!y-KCdy?Xw5mP#vA~X(UzOdDA4qiwHG*m`53JqSV
zz<Lz~(Op?^f5N{sycny&BvzV)z`6dF5q}rz4p^ibiSrSNhnP;87Z9p?`AkVac*5sa
z>>rQo1?mx1XD3DJenWi|hF}6e`??=n?CUj|jur*pshu~)Rn)jSEFGIta#eWt>HL?q
zrPr@khwnlZZ6|;>fv`0)()8!(fB)on1u09bHd-mr5uad^T^RSGq@;x3Y5HOocs+>j
z*QmaR*fqpW!7RiT5ZOCJMm85BeIdw)5Vo__09-+qTy60R?7{#9l7OgrUh@YZ<BzYa
zMr4}}3Wn4CQxEPMjZL32bc}*|&VN1uwGam5r@>4-;-%u7FA*H`o<}J|h0>AfCn-w=
z@aMCl_>~1n4FXW7DIm#2v1lJaaY^ue*D<cb&J}0(U?KnVAfZ!>eNxHoMYI05Zl1!T
zc0fQ0jr%f0M*{Sp!z9M>D054p^1AojQh%85k<O~E+QmOvp3PN~$%ZwsVQnCvb~S`<
zj8rc^n-zpuU{8f?%M$Z^G4(8|9-GP+dPY`&(d28U`PFRCu>qjXIl)-FQbt#~nV|Er
ziRWOp{`iWBv8Jl)olv&F`yCQQLJsuW;*oJkJv;77o@Au2jajxlIDq}Dz46Nfo<lvY
zQ5Du783-Av|5Xw|e*wJq0o#&B0EO4kU8%5g&JD2$)BSO|V8g+1e9N1g3YEF?>t8Bg
zK;*R{hFhOLPA&eQr@-mUP$+}jFuu{|_Ug`A$Gd(qxMnMaB<PJumXMu=$ku;<vw4jJ
z*05z0#7Dgb+|CWuo{p_5xI5K9oO><>n{SZLp!e?N`7x?5f_u1?c2?v<v*wE;;ig;h
zMDy}i)$^7gOw$UV3En;O3{4{I4%)^t@!}S^St{KJriKZ7vXyl<6<!~)O^0}-p0SZ8
zJ=}IErHS+w4&nxlPYSaeRT*boGkVrqaMiH$y-dsOXG5|)&+Gh#jna-V=fw_*pE;e|
zk#y+DtA7nSd@C`d1nMpF%7gp&>U@upi8$wiUI9ZeuVGyK&+B=hfs1ryu3x`ipn8!J
zV^@<}RP+knQ{(4npWqruYy-;XW~b!zK|w+9uZ-v|okN{Yd4b|84}#K8RhZq@N)+N|
zx$BT{gbFtfpyT6lPRe74i?)_;ATN03%bxwrzGjo@?YV2>7jC?Hd9@!kuyqnzomy`Q
zdqimZ+D*d|@rW~Um8#5@N6zuc7fbh95!_ca);BkJ;G*_yZbbq+`vVg<3QI^EUN&)I
z%kI{X-db<(#R1bcZ6%JRAxmvoU8V|4e-44?p*=oYA0NK2Z`u+M7F}9c-Wh3(-L=w;
zBugQ(yk~c$%*KfO=jQaBsFFMk{f7AR$gb7_n)Ulp*Kr|Kd#<=p7=uK>zX}BM^guUy
z@9IVqc(7lsc*E@ZmB(BBw!ucNCkbfok@&r~se6AVD8TV&JAfbk<}8YUdO818I`@<d
z`cG>Cv=@#T#UIzjjPw&ePB(VKOLv1)Od9);^Bl%%yTfT@*I?_?4qss{M%CnF(X}3J
zjXV%~-sid5@W_V&Wz0UD6)h-e5c$j`a>wprf9+6`R&fHdYVo%-^k6MMaCqQG&viv1
zcIyhKU~q6(y|1zarR3W`e+Qpi6`8+|WZ?Yuu;g)&>=ruYEk7IcieT`f=sf<HDR^6(
z2Sj5-xk*>uv)E$sdZWq&wXW|myJO015aPf@gc>4;h>-4)&X1_;_txOsefs9jKM$_J
zhB?lPDMF*$AI>s7ZbNS;fy>{ruE+YXFpfg7Yv0<2R(y0(q(}R58+1;4VR{%=Ehi>5
z4}K2iB7%(N8OXQ^xu=HUq6cFL*odaPLrdH&cKyojtDe#vr{~DbJ1;8#0rl})Q5v;R
zvcqb2eq7fsp^UWG(1fnLSq_cF5mzFCRB0*Pgum#EzJ!N4J|WFu$6&eXs2;ifHFb`u
z<xW@8HSZEZmj3OcgVBNf%cpwdz0`VtCMlJ<2qbYh?EbtRN-3%f`MI{6TV><hr?6MU
zZ7M>s2NpQp6De9I4Z(d(uv{8j;c|f2aN*LaKUB%z7lvHR4-8@$*3&8~<mHgwsn2&S
zaP>P<wmzC)g=wyNsL}i0-p5pH{UVu`k&&?u3r{r9jd{n0L?qHXOekf4Pg*N+p={!2
zk?k}E=ViI;lYJ8BG)pCs`;nVp_O080e{L?K`sz7-VTp&j9uKE)cP4Ukzxb)Xa3s|P
zv~MJHGWElukP8}08iAY4IbYS1IL4PP<B5yf2QQXG_n7bd<Ln^GXF-cFi8O{o+8+zV
zFD1j*(Yw6Pp#3{7ewj&3Y-GoO&A9-Qna4(CCy+g<|LfcE<xu05Cxyk#uG0kKKvJWi
z)kO19LhIsL0BOM<x|lvgcS#3Wn=vslw=SJHI7LoOjOb2-OysqFVRVI1sdLAXOl^1T
z^79er+>i(rue<9rSZb6bS6@UBFi6JJ9r^yuX?5L8*|Ql-F0sqGU3WBU#HDQs^Z(%p
z{C-AtCje}Ffu0<F_*td7kS`C-)Qa%;!^V5rB$C@7Tn6G<KNn(((9ew+#Mpp{^76Hb
zSN+$tfj|%)<+OdhX=P=_YM?kxz-Ay7C}AnDQ&xP^pg#QAy%n&Tk)C6OdfLY=DA-?^
z7D4i__c>w_0MY>P+QH6GBwu$4$(8GpBO}uZ=oI4dg&kxblMAXtDub7V4IT69Gx0pG
zQPW&QpKsZZ-@(D-$Wgu@*)x-*G=7``mhAj{+VWTnl1XV)%`GtzceZgfgM|t%x3x{C
z^P8Gn&wKop`uBH-53xW9@1km|8TJW5m$((ORfp-h9q<tSk_S4x62oSqR<Clg)Ev&<
z7OFCn9|p#MH3`xrx2yOXEA!!82<8-NzxF6-V&9XeP*)gv<6Z`vug)RQBq1Sz>B$eT
z8rW1{2?M;NLj%i_2(Q7ujr78ikNW#VLko!gYQ>eX+Sl!+kAOO)6Skqk%^`tsI1-Q?
z`g0{|Owm%+l7w0!Uvr?=F~OdRxA*x$BP#zL=66Eg64Y;hNo?`T>_-jx<a$w`xn5KV
z`|n=>md&-i27Qv5s^nN9`thf!0@ykne|OqE7$AACiKPE$Z}AvlE^po0*xa{o&;-Uw
z4CR;AY>%pBcTbDZy{>M6wnI;$?&s1SOXuu5q#p<!VOFTXY<*ybInPP4Qp@g!=X*V&
zAddYEnua+7vmiOqIa1*?IusI$j5okWNF7kg0oA4gSaOmJ12h5KofD_vkdj~EhWr;j
zp1S-vapA}$>s0X=pFd7ie#J>)K7qdu2_dH@^4to-qdVc13+$h-Iw5$w2|VvjNCt`*
zAAV1Mc_~4^B3ApVr>X7f%Iuo2_l|O)l2AK%=DCWij=<xJ$L$c~Gytkp;?qEkKWB*b
z6RG_@I8?-b8Ic%rL?scF;=9kGkN88t`un>@7tM;VV%<8`Z<)VncI&kI#vNcSj8<KN
zi|a(1<|m9`LIJklx#sJaWE<|x3WU5n1?`5dDp*0O#r*U|$v;@_j~zfO)gabfwy?19
zBDYVXgM7)U&oAO%Ge7dNFpBB~**0b#ojc=*HKu{at5`>YWGqem!ovD^lDoxW>Bq0c
zgG!hTp7@%kgIeO?iI_THx&$&L1(bjkuo}E*1?QMWesRbD5LkqcGroc7QafzfJf%23
zCik4^`u|!D3t}}r^vJdoNF~OQ;#`#JnfsP1P2kALRkf!j`sX<ZP5HIbJihf5<eq`v
z9HeFe8=Vt2zJ_)c@~~RLElnC(&RMzevgPC}5}ScC`xJy`k(HD4Z3&Jg0x*rZTF=!s
zsP&6LkvVy!b!V&bymu~8?CorAgk*u=U|w)6<KN8Sh!@1NL+}G<3thFaks+b-hy-zz
zG&6X)^X#yevbz^1%FoN49K+|TUiRDGSR}7TYZWs&^sycW`MP!zH@7IBDJojbO-u)c
zPJKDuvhv4ZF!GMXmu#HUhhgpRUy^1YDgDXze0;DGU<AcV^D7-#riR~)Cnd1B5h#Nz
zEL@TJQ$11Qz%))x#XIZym8!@~+f1z4DnE@wtAF75a@G7yZ@PZEVM`kDP9(@c8LxhC
z36^DyPTx9no4fFNB6Oe)tF+MGF=J>g^MjVgD9}Oe+F*}Z6te!Rui*hNq-M?yBHSmB
z_`KhT<2ymDSA0Dz(!4<M^yj6vXONyFyfJQH0*72AXDjDM=nn9pauyaAcoa^4e;>+q
z-u%<k=dgVoLlC0`PsrCC=o>r11>UrZclCtTDtVPd+aH}0(q{Zq7)@ODRfZrmx#VKN
zQ`ZSu)x5u+s$;~_+`t&2SQg1Y3j9?_nS<fp6?mBu*bSf>o;~U2^a<dRP>J3haTE}%
z3ThbXl+%h~{xWf5l8GD$$uDa5x2=m5$cczDfEa1KKHq=k=FLn1#R_5SMWZq7ysPM#
zP?27-i7zyBCi?7bNc&BVVAHGJVv5&x-u~7n@5)!nl@U=<bdv_<fIq^kG`n}4K}JSK
zees=p6r3L)c5LyHr}pRhMN3W8N-lNI^tZ81E#m>FC=jY%zq4s<C|?@DGkFRofjQg@
z%EUm5FF2(ow3aCayy;RPR#$VOG+qN6&6}9}X~Tv6=b^7Jn$Co-Y@5&X+BB<O`uDPY
zC(aCb3&}$5$wtT4=<Gwh(n%&s-MJ=4f5g=5*dc^pP>uZuqCCR^zdPM_VPq20+JwT|
zI}*Z9Z=gH+Nj<%K1Yyi&+Y>=^)C?r-rJ7+JT{|W^`tnTO$+N68Fi#aQ^u~4uq-_vh
zf47S<tV882pFov)7QNX>grG-U_Bx6j8b&%0{ntSDEZOGL;?OJpyUQl0$pzSJLj|V+
z#j<$JruARxFFx}luq%~xd026}jPzZ{RL;)0f-qS1$8RvZ#9AEK_7V$N!<H~r%EQ|7
zL#8g}c%Qv!!CP%z7GAdgRZHiJ;@97+|JQAgprXdqx*NwAOEEqx_aAS0PS$AfH+40h
z20WBmE+NqW!1#d{1ou=OX5^l1&_qMWs|Z0?eY|MYo<z-S%xpa)+QXelhy8rwcmw$X
zAt|?s4~%tA3(SC8r!^35w?U+G9d`cAf%ZT*;?K9wly!pObX&4$S(A0aYrD~Wh-4K3
zO3KPCm!oOe*i;pEKx?WHX&CP}G*o<h`V5F<XRLDGmLc4fz9@_Twk-v)t;N2X1acLp
z+$)W{8S;0;drBg%c+SGx@`W5V@KWJdkb)@!^|ZKE3tTq!iOTuHZWy;Tyh^e`Ghu-~
zsa)?W@kXW!@1~waJAMCgipd=&R=(1tzj?pasD`Dw(Wp56OHBpx9;FR<N~f;bnEjd2
z!p8>(MugweM1SZub}z!UwO{3CB6bvth<L7e?b@>x0jg5UMf=k?*7Q;uSkJ2yU=EXH
zoH_<djUB)P8=*zktblQ}+P45q2fT?~5O1*nbp`~xMnHZbUup^q3&Z?pBp{PnfMX(^
zg-Q_l))~1MpcS14yv(hYP(fr2oAF|N__)~=WW*ZR#&#RD$DC)o<YAd-0Yr;dZG(H(
z52S8eTTbyxoIl!eJ^EIz`W6`lP(0;(5&m8tAQe1-lt<+0MNqiV{FY@QM%H*V?+`va
z6X^Ib4#icw`_gUm`{FJ8N}34@--U&%r=P^;>pa2Jm&hf%K;X7922mlT_%BMipP@;Y
zsO9$0o#a?*vB9!O2K<j-{oNP+KNYEFQ<??2=%x^)5HE=I-Da|0z4rKK>_9oh_o$by
zOh8voPwB0Z%}#`rg2DsVD}@E{zg?{|m=kUKg1wczzu8dkJN{6s#$(H`O3bc5S4$RI
z{<G8%vw>~#7wEXIIXb?CUA)g)TU&Scn8uL}$gqQ33y@$9fwNmqP;G2OTs=!gRajnL
zUY#QEVx|}Yrb(>)9w>1~&hB^4&W<pjZ}rg2Q0F2r-@C74uWg@%aBboNS>het>@`B)
zrY0o-{tfPccy_ZXv>#l4<Hqdg4+8~Rbf&@`T8>X@GXMSp2u1_0%(CJNT-s)_TN?L)
z7GS`#M|7V&Y4A_`6T=^%CX9L>9_+(^Cy9IzUXoVj0E)*{pSqezbgdVQ6m1$Sx$_d|
z^RAvgORREej3LLA)YOnv(=Zz3&}XvZt~pxs-pByb64>A^ubCFks*4Kfm~^!l3jMey
z#gJvjD$2@Q&d$znN4JJS2cRE%U(aDL{8bQZYO@y%5{8j??xmqh>%s~J!4Zf1GBQ6;
zeLHsc#{;MO>kb1)bfo^2fIjjl<Gpgj8)mlL9EOdtz3Wn83P*q+hO2Qvr&9seg(-Ge
zy1OMtA#g1{m)~M$2}j?+FP5Ju$r9!NyvS$hh&SIS756IN^6$=7SzV@U*V@XaD2Mdj
zm-a-jtSGFNIV1msDr%E#IJ)>eRhwTYt`3DB?5iFaor{*i4Samds=l;WWoIvUl)iG{
z4$ZgaC*-VrnMp{kdy9m-&~pxE<qT61DmjRTIK$yhLUW}bF!woP5k9`lgM)*1whan^
z!Sca|@;e}4)ln05T45y+T6b1Jnqqnqv^NMG67jVTS&*-N5>j0q1bF8Q3-XO4AMsDx
zWD?^u2o2=ggRso-`#LSkR2US9$0KC_0Ayrm7`BO-z^aVZF@fzml^a9WEo2WS3Al|+
z_KE`e?|x*xW%lIY%oo3?ifmci(+ihs|K?uCa~3aYK5z@Umfgz!!yC`;mz}Al5Mdv&
z^<sp2S)j@IkD$wUq%Pca_!z5wQkX*HQ{S=fD{OaKObspf2P?B5Ts>ceA0#wTiCc5A
z1|(O`9Y3w^`41aP=v8y{CbIYO8hE<vHv$6#=Ru3HA0&HoVbM~`YkM;@vui1^;hBw%
zZNAhq$h!LT1M)RMKPn~qX+GS+ne$AdJF&jXd+*<Oa-@O|M37b*oVVQgSlmd?RFMk(
zEfsLPz|*Q0^RuhL&xLI|eem3;NCIk_q)%adJOs>608L38llLQ8eto0hv-2^oyoMIc
z8ef<%P+|VLjcEOeJg}(dP0<#L#wTbR8@B1Xp*P;lsvK#DVNh^ETzzazd*ZNXj_-3=
zKBtQSQz^}9K`!=zhT-$|;l~4hcXW0PDxX<<-rF|njC~tXdJzS#p%bJwV$aio`5=id
z#LSrS>3Kn9kR1~9$=EWJ$B^Ay!p}VW2YNj%%p)Do4F%8PNr{ZIF)}K|8M!0?V<px#
zv>#6L&QFKWe$0qcA;fBzO8n9a7_@dxd1CD)ZA<)Ll_Z+43CYh-(Dm49Q0OM*lqh<j
z;9_raq@(t=G6afvu=cf=f}<WRCIexl7v9CuJPPpngz@sF6W@aHWFAGJ2XXhX6#sQv
zzB>*$A1-5WZs2bpL47B9eLVwtdVmjf%Phfr#iZp#=e^^t9s+k!NB)vY2|Ub!YP8Hl
zpczXAZc@l;F!^%RTYM2mF;!XM7i5oz8Ju)klx)9z(7w*zrpz;sfk^PaS?f}@!+!t%
zJwL2rIem^t4(3>0hp)tjCX2dGfABz74xZA6<*lJGS_>P_E<;3)*n3%0(hKYC-|TG+
z2W*q#eY*!`S8tv+Q$Rq#*CoLqAw+u*DHua0H9Yq~J5vM^^057~cL0g-RK}ktdgLb{
z5gG%i6~YTK6=_p!Y6FKb0O1g_L$$+IhesvGJlT@7x3lTq`Gm8E_+e;hFPP{VO3KCg
zj(XYD*5Cc*Xkm5bWB!3L-QD*S^GuUj8N{Rak0eJw<$jx_Ct-_r_=Ua1nO%h!Pb**i
z84QJ43Iyn^)n`3%NHfQmIyA}aBG17jotKr&u3ZRxF8Wyi0WoZ#J+Sh8us>Fi_~lb!
zdPlfp>{rH!EmQ-)If;>T1SlHoGwVaP?!33&{|sTd$>RdOABaN%GJ9hDK%24x%d7-y
zy!f@r*RV$TmHqgbw$f=*ZYhB3*?K-Pf9Hl(kMi>JXNO#0z9m(K2@7?*={*XNO-5En
z4h{U&hN2ar*d95_-{E2af#G)geF&)WA0pt-bsfb}vaCib_txO2&KFcCIu*~yya0Z)
zLu6f!s;kwPVawO2EPtw<?wQ?9d_BeZr;UMz+QGe*3p|&PtsnU*Gj83t$Xe6j6UY0@
z*Ihz<T@T|w1R3u`;z<eyWO&?}8|T|QU}$1}`1@q+d!0i3eDBx!dAM!)&u)*h0^s_>
z;}recn>yfs_NK{V9!_C7WJvjGP0;Z2YR6r5d<1*+7O#gXL-d865x34*EKk2Od5hZH
zTk__5Js)T6m8YI8a(2!M6ql?)YA9d`M}YK(Ak~o5?oZ=`)8|$s!1(-(23>%_+0^(u
z{W#*LAvxJOUZeL~mBrp|pWMz;(QS6z%A&sQw5q#MBh0y^Am8QvJ4%qp1BbJsmp39y
z>Jq0#Eg+x^3hLURBhf<o=sm%5XkDc-o^?~S&c{}PgN9drbl->4UY~HzGyc2Zz)o~`
zE}74Bbx&P&6_`45-vopHC$%e9LC!Lg@gJ72H$Wkuxp0`?I9btPzPI@F*7>xH5lJyj
z4B6g1ODnTVwd!DciJze__>!EEJf)4{sq^$>wcfP|gsApZOdqn1b=EIxf}@#&V|OP@
zJws`+A%8AsR8NYBJ)tH!phcPse+TLGjk=s$2`QI<6Ego5mNk`v-i3@2!V`gPn%IhK
zU^(SrS26CITb^Nwpgc7(=hgTt3#JHv<i+)XkQYf#?ZPS-FGjB%AVap(XX_nsDA33u
z^3>7t9RIJ#Q;V<+$GK>Tr0OCN`z&T}h%<K`_bdy%w(we}?kC|7*nRqv_N*{Ivu@qp
z)drs&U5lc5mU`XV#1HZ;Umwc_Jcun3;omxPl}$w4UU=Vy-4b|8TDh$HUP_k&7D9W3
zHY8Pt4PyXx94thKjD+Q5hmJBL=_)wN;ff{y2Vzw!Rk(WIo>FhmBGZSe?a$J{QrYsn
zqFH0{I0gc2B?-8Lh^m4A;RacLDNOKxoqoE%v<O|S9`JbKSL#wHdL(uhcu!LZX@cZ8
zXE~ew4m2KmqFJ<^mnWn@G&WYPBYJzV;bN}OWv<~N0bO)tq@z{T@D4<;*&wz;O_-xw
zJ;^2~FJG~E0D2R>FeJ4HnUdWi(w@!;lYoEcNXNm!u@n!%Z{wzRgk)$sE;LkH$Ympv
zcWp3XM)I}EssDn>si}`6Kv6^OtIpC2qwO*>vfECwzekvs575xi4i3_7<ap80q|lH#
z`-No8yJvd)8^UO?#REFrvC{(luVo5aqErav(9w^hXW=lX(;A6%Wj&%b(lU^e$|}=D
zVflBsEBW}7$-FBwuQ_NKZHu<$6@5~@3Aobq$bOZfnR&bCC+nh9kHY1!xf5ZKHz+8G
zuq{x_5asLZd*<9ZWdtL5O_?Ex&rP$;!;&EJ#5Gaf*z4x?-f8Z#6v4K0mwk6%-m~nc
zaYIMPNBi@K0HuA!IJNkFC#`V1pxT1YW7fD;C$IMw1Q=3ifBsy(C^f5(Ni|+^*SA9)
z%Lfh1M=FtC6Sl2vgY3Zb#9`<-<U&cL8zWECh_#%!Jfs(OSM9ECsk-2er=8)#IV=ME
z0ewcQ8*y(Z`SXown0m3XQ#+)LKgJGfzhviGQv;_E3*&&Q(0|LoA>r(RaK%DH<GVW5
zEGzD|mI_-#pQufMr{e>10qE()B#hpDw?5sTtZF;_`s_tuTF{<9fBwenW$!v4bUxSu
z#`|Pt3if$jPL)Cr3JViVym?^W6z;1M%dQOSrl9P@gI!}JM$!_^Wh^1$zQMk=y`6Y#
z#Xbef`ED?*87Ohh?x`>V(K-_V^r)Pio!b2saxKeO5qKzSwEzAyM>OC?pU7BLnQEY@
zmOqtlx^yD1PlD_OI(`R}Z~rMgS?Ve37XpBNk?`9LXws6HUBMrCWlaLqxw^AP85;Qb
zAi(7sAwXH4A$*$aQ@_P@4$bA4CIW3c3W0z9N)!6J#~+OvD6{J0=EhZW&rGRCxnYu5
zHFzJrReU9FjACUbH4Y}mV}~Dif)RsyD)m!<BPI<dqw8|3C)um10!|t-KFZwJ(j95)
zTWI;3U3hP4oGVn?&-Fl3_eh?UO2UJA$ONv$62ik7e0-DG6b4mdw~Yn;pywc1R|*+o
zv{Uzekb>-h;^ev&p+9x?n>Q?sj18@JEzJNw04dJJOb1*dGOSe!**R|&2UD`hopj&;
z4PBCty9H+&*WxGb3-<J1e=>Pd!?@#FEYEszP97-qj;!hDM9zk=X5+wDqQOk)N0S^;
z!NGI8nLXp8U(E<2JPuX2Z`4`dml{Z`;=2>-MsmY3`hhE_awyxkST;YgzYa(<xvr;|
zgn7ZCbZ*Z9lgK?L<73t?-z-k@U!TOM4~|jM`pXR7KX2h7a*9`ctFVN3tM^3q$`=zQ
zXbeK~%_=A^2L@1{@=mDc)tma#lZ>r*l85wLB-mG?q_+f6eugY~c8o#0qS9^MD(E?g
znv{AU>?6kq3%*61rbF@`!6D0nIcGt8yc<TISlHRwb<UNXWMpDTy8weOWEZacKUzP)
zgP5ZRPvqjQZz3SEp7wpTN*iWkXQ7w<?k<yhk`hP)<b5phK93_(?)1)H&h?sj2<4xL
zeDqOEQ9sAW;>8lasH4@CmM7h0%Ir_k^|9sn4>a^=*iuc>s9r#NKt}p%{XxW<tj~F+
zu`1WFpOS=F7RnO~y^4G{+gVgDUXolb{#rfM*QVh)bf$!FlH#x5#m9h8e&AObBA(yO
ze>;oXII%}muX`(v_phtyh(r^ZwNNJa{KpRJ4v2ih?@JhCf2W=5aQ}F_(s_8#<Kf3M
z?Fw!A-!429naW(y7hB(=t;;Q7EWXi*n<(Ka0yt>!Q%P^mp826yYyiC6-`(3|7|zpu
zPDUy0NY7B~@cEtDJKTm4m{(t2U*~^%@&J$pjs7BICOkYmUeRH_Q?#_SGC!+ec2eBq
zXE^MtXL0>LQvZbzz_UzfwFjO&!)&4(r-9(z+pzmud+@%Wp9N%_mtWqT%XL;(R_22d
zJu)E&YM`^H9JE7!&=RzN@`kp79gVT_i09VF4nzAl^018Ishwy0GWmtx)wll-PXf<9
z5jlbBrcN5HOdJ%xW4NA~F5wcWiOO-d7Sb1k34NwXVQciM=_c{rd2rg=Gbe?9tUUY9
zNshPV10S^>oGP*$<=;N`$IHWc^Zs`9W&*F&88=uJV^p{Ddc}s0|Mj=xi0T+A<G9dI
z{Q7L+xQHR71b@es!7aP~RI3cTaSglU0ki<ZV+^p;anYwzMu#6HGUp4TK0OuBfHTEO
z$6)tJjYfn@hN|y5sa)XV;*!RCcPZ&J(3dYmB#w6U=qR%{mKJ;4Iy2D>Q(Jj3^qXe@
zJVEs509NWo`>}b#&d{D(?}O-V5R*d2|8kOkyoD}Q9uRNM1c+lyYI(&OcNKaj^Gz=P
z<r|M2=xpdOFQ67})oo|MA$WQ*-k@Ih6>cXtmX8l77JRGg=Rg)Ca~M;QP*z>gjCeo~
z_99g#a52sqptzzUEoVl$ou@z2wP8jMOVYJblB(1UR`V^9Q~DkCR6Z=e@#PqEBAiKt
zQTxFTw-T3HCsm}`&4?FT89Ryp@@g}W!OCasl)0%7jo;T<3QY!g`|3?sd`G=E^Cvil
zJakgf7AbT-izOT|&4ot|2%@vt63q;jFmdR=thb3Qvmei{zV{<>u>4dUzZ#HdQ$grG
zA(28vT}dewoJdd*kLv?7=^)rKcnP4H=T_s)>NL}evC+|TX=!QB)2-)9UL)i?GSGBM
zXiH!M+GV|``N!zyX|ombPP3c2M-K0Q4pp70xR-j0TTPxO(Ch?FAPv5w!**qwG=7ID
zI+pOm0(?o$<`^~-@w>}X^dyW|h?X}ubHGc18J@c;er7EN7pql|T>c5(Sg%L+*i$OL
z$?nzvj%>;%ALnpN7adBo^kaS~ziH9RqL$PdQpou0<D`<H1~N0^V(Nr``LZq|_z*Kr
zuSTVtK1ZEnn%!M*+{T#k*ZPXnME_8v2NyfluYyN3O`b1J5CyY0!fHuEoi=*xiJE{u
z@(#ly?S~(5FCq>Xv94^!z{j<qFm?I*Nuc$C7y&Le3a?LKe_@ap)Cp2yyfkx_#<1Lh
zX8rr<a6z8fz10p#C<=&!xS!d4XomLJ@fEBM!Ko#3Tw0_p^QA&RYMH!~OojPZ6#lOx
zo+E^(Ug^;(f|Yp%o;tq%7Qh7^^F_6sk;kI0VJ(c;rocFBUfgFB3d?ApArfQ90R?A>
zX{`S@rgivBUA&6ALrJ3kNV(|pQqfE4I@5INCUoq-d~%lvx=okzJX^@tFMm4|f#y%K
zR?aHtM&kI;D_m+RR6JkfE^7#DT#6un(bK>=|EbLMd(Ve`3g2;eWJY1{0&lDbq}EDw
z1#O4qpq;{30>x*9L7APc*$vac&J4vTgA3x1LPN9sj^lKJOK3}s7rV!Eno^{v)oiI5
z^e9=lxx>GY@#78gc<=9C{(^hmlb(sG8))^+7cN9V8$up*7xCF*KQ?$Ilvu)a6qV=h
zN@F%-^Nn4FXu12R{t6yPQd#_tkh<9Wt`^%v!TIZro8y?ZM7ah#7Ni(bQYn!AP`~6X
zkZM<%Qrc$VHhqVI#8WTKhY=Oxz9m=ABS|XXV1XI-Gov@VYq#gG$C@SG|LdizVqMht
z^nFeFxG!@_`2<0r%83h@^nZO6rFkUhdS1jmjd_@Jr7FUGc1-5a7&n>jeK+Jg*lgMM
z&sojgEpgW>v0|UYmtt5oXTS^Vzx<t`?V0ifI7rz3%L>UZ$!M*2%KpzuynNVAt{(>+
zA4tEXxQuHqzCiL=KowP?gmDRK8#_j9Lkg&d{yGc_cSHKxJUl$yYPZy!tCeA~Fz~{x
z7xrJM!sKARla`9)?(#8E5@5^HL!^R0(J>36Qsw|V9!z4X0M9YwO3^-1{l?l_CbYT<
z`pXzU3#Ri&Yo7SadX6R#Jq*Qz;1^`UFA#%Ykj8h^bR_^oo(Ttwk|WLn3PpYXB%Jp$
zfU1Pg>garZ$S}S))~42UYg{V$T-%o?lUjuDD|C#4x>O--UrNdewXpjRml6|~0zci?
zVJ(75<7s$pt>o9XCvngrm4n3#nZQxBln`*nLQyZD_XYJ35s#nTEAD(yEby4MDpfds
zB^*o9e@RW)>z(Nkt&3e>&!E|V`*R{zl8ykD&_zBhVd}rKO(|f&s)@{+##~VRb(e1l
z;5m-zP8~G<SgJiBEDGg+BzPl5*r_2gU7d~B!UZF_X1B?5vBFw*u;lKg?)-{8bqjS?
z9quQlvSM<SBX>Ar8b$$ZAF6Q5(ZAp&(eUL&iswecGQ}`^_s1wuZ(`MdNe|OR%>@mR
zWaj}j$4tOo!)~lv8FtjKt<7qeFbdd@U4-qh+$E2a#k_XoQOAx2eA~?4VxOXBx~WxW
zXEan|9%gxVZn%uAErGYOmC@C2{Gp?B_%ECLFAp<>4_`?jsS^h)Q}EEktN@UIlKhq2
zOOq~abY1VTM~^z2Ex1EU<tA=j;yIn{p4fi1bButl%v&K*lsnp^<X#p-W;_nW&Evi^
zGZti$UShy?juuLPbm=56ZZ;g#Q&`YabotB{IBIpqx-X}p!H9$*W0GQ`feNQ@O3is6
zn`LF3s8fY~S)Orryr%0+ToY}k{-W7TJ?(>mqE5v_;qyA~i_ECCp0-}t<yQzsT)qho
zde(SUcEnX%OOjjCXc#bxXH7eNH;oUg90@!#PRG~Tr|P7sMLga8uf6GF61@`2YMp(m
zr#wiyhlbwgBXP6(wWm~tWGA<Ee9<I76>8_;$}1>jTaAKHY9gPd4A9WCfg_*|JrWj;
zr(#o!i)vigs#9U$8Y({Xd^ekk4fgw_#bINi<boJ9cRE!B$Y7vP9bOdgv))`Sx}g_0
z0u$8cZUS?N&t*Zc#D+Nh(Zh$_594m)FTiZ}tL17*dU}6&&P)S}A*80djs5*`uvz8d
zH*;G^%&nEvDE>+ijusxdiY7mNMfq-w%J+&`Xwl)Fp2I)t%1mBxmf4M_qmG}{)4QFo
zh$={j+}BGS_w)_N8>sGng-&pT{3*U=Z(o>B=X_=9z^_0-44_^l)T{Lp2YPzwQ?Yy3
zyXk*?ZBg8PlQUIqs-3}n3a&wE0$4ehYdr2XJ0j^E)XB=zbjL78lK;e&Id_&TEMFCg
z-1JHAd>XB|zDqrUi$QHRHKNm%5l`MXP+=X?hxiECDyOn}k9x`kPp1bReHd#Zu}R<>
z^J$c&FA{Akg4~E52g;WO+P2ptCEm{7yqr)io@<(Q)h}1p;llZkQ+o$bM(^;5oqDg1
zf>tR}J;l#j4olcA&j+|}Co7T`0*IaXb$xL-Zn`z$hFQrbvU1R`T+Y$aQH6+#is}p%
zl{$nGr5DPS#i<xL6+eDRcGpkBL#r?maa&^s*VIP+o%UelW>=;%(opO+I}--wC;`FA
zqxbOF&pvY)0_XfG%<`G1I=aB`;mhFW7tx8Jp&$B;4?(b^H%rXpMbo>yb}XCG${1uG
ztVGLH==(*JmbapOp3mL#^&M>_yt$Md?84dZBlYMeV_k%##O|`Jii*lhYg(vBav*3X
z_KJJ$3hhChm5GkT#G?vNNjV?GtmOyYa5YF$_}PMq;NK2IYW~*6EH~mcM5a3snI84s
zycFdYm!Tr<mADq54!01EJt3-@5b=s(jdl6~|7(}yTDXWa@1X;qH}^umh1!Gz{VI{Z
z>dCKQ9Eg{Y=)d$y2G=z3dzW8m-0TYP&eNsYsB*WA)m4ws{Fsd#v))xrCG7JbBb&9~
zw0FFm!w?dB3%XyHFy8R)hlRDsO|28e=J=}YcHGzNu}avz9<=-7Cq<A6+Fb-H9iR0a
z^xjfo%sFu|Q~fRlD2NHR*mY%M+m6?We-MBg7a0$})SiV4Qb1DBN6CgBlL2J`v=y=}
z-r@c?>;jHr2yxji1)W4#nM~l{iuD#@Z5@mq-DfWl>ZS7qFsB>l7uJRNLYhxw@4BxO
zoKA2Gx@LuMJ3y$h<F=LO`a9ZZ(x6B<ucn5F<GbzeRv4A6uC37%J2n9vsj@fjeULl`
zR8uI~k8yq?P#mBkfF}hqv<}l{DU?=~lK&P0{N)mbmp+CuuDx@O-Kn*THV+REm)cw!
z&k4jk9~l=%^0htriHC<rG+Vm=G{!RJ48@civD14pQS;EwlZZnQl0kdoOpz8?Nj*V<
zDra=WRV+oJzRMj}3moD=%X(x~x$ZAOj*?w?^e_uEj4;$}H;yJ8LDtC0r0&R|nE5*5
zJYL>yj#h|$id}to#Sa$JNh7=NfnY_5BleKNi^}Wgvgyb83mJL>Z=j1JBbuE6W|j$y
zD4!%Qh>X3wg@X)j7T5uZ&CJ6SiPXMkB%&ZvORAH1DFEt@n8?M?pJ~LLASLk#3jHSO
zIHlw(3r{b|Xf9PVO{NDH<yrL?QbJM@T&|YqKmPF2O%u$&3eBUL(;}A5i1!n-=e3yz
zf%bi^s>g=nG@pC&As$KN@Ipz5Ps9s5We1$iaig!B`^sMJi-CGf_%PI4dM`VLe!u8W
zzFyarfzk%`Q%YbDmT^*0XIg;Yz*Sp@t<aYX2E5a`AoD@A`BYyI=4hNpxO+Dxcn+UP
zU4n4}fl8(kN?X?`)mAZ`KVic4d|XBl!yW<$St$C*lfQfS&H|Vk&J%U$Z$GaCu$Xru
zoJE@vncxCP68dvH8)mxjGgRc2-H+8Aqckscy&zcdV$wQ<_|O$<LB%r3EZu{N-iQr;
z@b&`>bc~VELqB-^JGe-d??a<GC$s{qN(E^jq@TB1x?v(h>?-WlHi4_fCKL$KEa&sJ
zVy-RoO_#W|8%e|%08h+LHAgw`@80|JtsfY_YpY*XSq(~=?<@|*z$!;mt;yPMTuKpy
z1GLXe`WQ(VK_*MEWY8@#I+`9sXW|oKHy@~2B2b2!T4W=Fa)6dl3;O}!<mph-Bo2l!
z0Tnh2R+n}|1P@{SaIg?I0HZO@jz2@0ADuXd1nB?#lX}iKbmx=|Zf!o(RL9(Ubx$p&
zKLwwmrZWagga%-{;DCvyK8nv|$_E(WHDt*JY&s64F31=WDS0GZl>L^R?Oi0+<&ku6
zm$z#~I&CWIYy=ZuU>L5nlq?G9k69q^`f;PFQ%dNaqJ;(Xtxr#r3yo_zg@Q)PZ_DO3
zMhFWz@_#$0y}!FP3(!JC_vM2p{MHIUX;Qlthz{@74SBCm&#Gi|pf23T?zweU=*`7#
zuGMG*-hxXXHSh(rpLv!(6ZeJJa`O5m)mZkl_;}K)6ZjMq7h#*vg*&rd0q}Getd7M{
z(D^J$HU7M9`zE-;G?dc#M(N>cJSGJEL=ih(0!v68tzYgjuN)5bgxw+bde%vh22(qz
z>_$pAQl)6fv5$t_vr7!PaSqknvaOxGihja05Djsy+4L{F2z_Covco*Z(@Y?XRz67c
z?p>kd7#Q(`w!-?rrgV@TEL8`6yhj1LCAV74tcb<O%5T#lFe7`|{B2J=MfMjyvRkWP
zj}zc@djJA+{RPg8`ps5IUpQkK2A0XS2u~x%|KES4cv0S@{uDKV;rGp|I|;BdNe&<3
zEX1=em<!=J^s3PQ!qO677ZM0*_C)G?2Dz3Fw|b<!FY&&2l09{|LvQ{?nj}2)Z-mhG
z;j(z=j~y9G%%*$sp)Tc_Gtv32V{hCaQ7R5d-vn1}wgoOblY}cXV;2#4mN6<(5rRHN
z@guaPE!%ei?<^nIP3e$W>_k|onXQdBemzAna0u4ss!|IJC%|a6w6WI~1K{<oEXgk(
z9xD^P*@S=Hl#PB7YeKZ+Lf?042!Ad%p<3j~S5A@!2b&~KVC&;^1H(aWid3@a1iC9d
z*rsP;q9~4VJmH&7Zr|5bm23}1u2%f$;lpX<!B9E-5rH2G{3$(3Xs{lfVlDgh#e5tZ
zo67un#B%X`mW2HFV`4gEP~IMwl<>ONj;f2Dh33KKbUAEzg|Yg?!J=z(y)CVbF7MXa
zF6T<Yj$6SNDp42qPzvF$Rx`JE@9=gh#`&O?{Zy~O6v)2_U`{jhE{aLRpB8#N&Z|?(
zU(oO_VM8=*s{8*F4Pg_+rOTJ2e|b;53cWoFY>Wkl;@LXN#ES3fZ+a>7H@DoSdx|e{
z1`bKxLb4L-y2s}iqG@7xcXuCV+yPIc$Zpsd3t)Do)3?Vr6QCh_A9|aN31LVrehZ$|
z6*GTA%do>N@f?XdSJbsUi3--NT??5a$NbIvxIIkxBqWNU;+76ql-s2b{dfz&sPaog
z6+R#-N>cDET~~Yvf`(Oz%!h@{909ze^?~$T_{Fhkg^vmt0vxHsT@ctV#X3RPpW@=I
z#N9Po`JT^To-?hf><vH99<5ruz2c#tTOr+(=Awl5b**jY^^dx`57^Z_)dV0Kq=ZQ0
za%h%2n4Ua+ngQ1b9|XmJ`Y(>x8e`ej(c(c$sjOrFeWap(GfYi8f#=o?2yI2v$ez$H
znYj&1(k-D%FuMoy>V*E3Jm&2-W`a)_0CR?Ca8vB;zPr0SQE(Nk$P@d}?$-#b1N|hW
zApYb@`J!0vUB?8Msc>B2Xvq8P=Z@IAC3pMog>lZjKFfK7d$ZlI87^C&KMerHOcaC(
z_C5yl_M|i9<kqGi@~E&eFfcSPUJRiMd~l%x-PgOQzB{}khkr2D>`{j7-fqt`oeQeG
z?NBduQTxVm8DE0F!trl}FJ{+gI`6|LWRK0Vc&3$cf>Oj;>E%^m5<jH85)Q9ofeFpr
zYiBSmdxtQ-dJbhPN%*`OgK-lE4N-@IFjvusF8zg153tHE&-+uF2|Ff4(mk^}rWAO3
zQncfRyelRW^KMv&3?Xr0;uK1yU!b}f8<Hd`BS>Q|#c>&6qKhM77uA&jW)2uNihS&y
zs5FeL*!}dOo{0Az0Rz*M=`VUgQfVUzdHS!!E;mxog>z9yUA{0d8iE!~-cw|p0+LWR
z-A#=l=Val%-^v%czWG`Axd8)~LME$?sa}O61EPQTPGzlIL$VX5Vp2Y~Ke+jxyj&%Q
z)foopN_GL!>jHCbv*}B^T75SHaU6hDW)U%|H|gHnUSm;yq}rq_$N2xO%Y1x#F$BOj
zPUk`m&_F%4jO|ZRa1HMahHt41BxuBJ2^@w3rcOQq=W;~o>&S21dyhqkHK0N*uz!3#
z*~Ra>q)O?l{P$nkV%N^5<C-tuZTtMX15hTPeQYSnANbwB&w#o8Eyep1bfp=7eC6@l
zT~S0D2y(#=&IB%0EXFm}77`(FzI_BfytQ|dG+7t%%fVMwKE~^pTPRRrk^EdKwAi6q
z%WK)2e-_$9%6HI!07NY*iRGP(yX{R0-q5S-J1tTSc!z2c9eP7~_~7ldk6e&sww&=L
zB(r`#RmeR_T;Y5U-tdz5_AZ01q+8fVQb-y=Wm?w>>#HtqDN%lFi?|@)ljiW0{n7^$
zS0WTpP&1LB1ubC>fUJTtIQ5FDSDaj-HCytMBKOfTfrM;W(9;QXYRxSzEj>wdLOeW|
z@NeE-n~Bh;3g#<r^6y!C1?&ke*mB!8w@k`;wcMn8Zyr2lHxzf4P^r~7_GJPH-UXW(
zbD@7$zmJ9_eE+XM?M&d)zwyc=Xo4IR26xXc-=}aXSLl*N*U->Vx!B`HSF_)2;_rw-
zXv~N_qT@MO6^>R@UspaUIbjAwYUplZHdw8N3czE!D(fT2zdEguSMMjse(|RLNgN!y
z$IOnC4X58W2T*FmVk!mC?KK(~1vwNtzC#GaEy~Hjq3?<&tNA#kdrGB@$5KXeD0Hyu
zvpv*Io((j^+>Ht}t+W^34i7(pc@b$Si-`Aw&alX^x7KwwAUJ`fJyZ=Ej#@cOBNeju
zR;RQUaE>R*qK*4*I60Fbk8RI7BoX=U6YQu<<o81sTa3uhwM<BYJ^i1!X5dEX$Zh2N
zjVsc^yc`lr+E{ht$2tD>H@j^pRPmc;=aXsYG}(=bn@?{N_L}w)pw}phfvmw$vFUw>
zFd-I}W*cWHsxMc$*dlmr42NMKdzc0^faOEU_<2fdf3%C1LRX6S$=7BG>v@2i*Ka0q
zbVG9zAsNM*ecpvC><lay$C8gHHjxg}kl3p4p?9Z%B3-N<K(OZ6sE+^@K*IIb+V~{?
z!vE_=sVC5UPrKW7CY_F>w2-c`dNQ;%6IB-6LL$S7P!i~n{F33;hq<YnVid3H(p$8>
zOw70NHR6taNOhJs*-m60a-1ogzVUJ9t2FOUpys394f}yUkB%99NDWO<>@SzUS1kp-
zj=FF|;aP_DLjg28ILkSViPj>K>&~fks7)$H&Hk0oYnSe0Iq0Ztb^2e!r|9Zj9p*nn
ztb6LzsZ0nl&3)cLs6v>NGWr#M>~qCHPcx`=W*L3+WLuD2(elUodW<#do;p3Gk2|K7
z!3buu&#$0{1Ox|SfVG5z>4(tKcaTEC%Ow{Z^<Cw#@~c>5oieE;qX_>}G<gAP1=65$
zQVk8FR>9IIu^(2vF@W5mb%;hPp~bMrhl+_^+u<Tow3D9zeyiV%J?wm%z=@?(#Qpr#
z5vn@+Z!c^HdFIC6ROx#Gd%^;?0h<YQM<M(N=v>Ml!vwL>*uElTdxMdMpTNJ)0&^W|
zZTRv5I5xDn3OCh6ymrf}CEn5!KhFI8J`j1Hl$NMF@J^hSq-L*0;0A`!{Drgrg3{rJ
zG;^=Jv-X;#TK%YWvEh*47uSBZaoCd+L%(vo&o})3hsvAnN#~xha>UQ9^8r<U4SFwc
zaPcO)XrQjJ?=D#1U>On>N#+{EOFwbeNqFeg9E8d8bH2dd_5~&ogUB366m^1PqTCM?
z3?N@%q6__)zd^};(iY33j3;tm%uOSz{ja!2&CVVIR?Rv57O)rkqJT`+8Yr3(b_viH
zvS6olnjXZ^N0fK<I|!7?Q9p4@Q%b5HLe{@%e4__`1?WW)XgapL_THB@Z^K*k@e#$Z
z<Bt$6AZ)-wwKE_tXw_NQ8sCwdDd8sm@ZLB2V3GQ+=HT0FGqb8cP(nlu;o>Ds{CSbV
zbN<trT>&K#<|jr}&m50=5CZvnHhiaqwub@dp`1}X1hkyOQ{mg4+)Uo8)mz=zVBBYC
z)k^o*6Bw=We1b4s*o90Jk=_9`QeyIySA}S4{m7t2;5a_47w%1t3;Zg4If0?*{U3Mr
z@drQQeKLsFqdAZ9|IziAVL`6j8a6DUlpu{#($dnM3eqYF3P?8sk|HfgNOuTQ(g;X{
zL8o-52uLFxN(%3Im~(&MyZ2h}ark4+KWk3pe(rmWYn<0Lrh>c&Yw2Dl;k04@m;3bM
z5N>{PxGspPDc7KJrisWbH^Vo-y+e!Kf6&n0{sxI(mV6lckPN+7Cp`E4-hC@V!*Pcu
zs$XR3dPTw7N%l~&M&e=G?OTSg(8iBzMN;`kLDjLZbgB3BSeGKS`!o8cxN`c(`DXsl
zk3VWpMzp<B@BK_9WyMLm4a6mcx`r^EJO@nye)ju3`YiKs?D$@td1!xi=BTt7cnu3`
zGu1C2{eOK|jLRNR#D$zjPy6vOKcRu>^|_hVFqCFi2U;6D&Ijg|S7<RTqdPi^EoN3v
zkTyJAH!$-}1W`5){Vji~_|O}9s;0hqcj4n=Te>@)ivL31Cnf-oTFgZXMh<kO&$v`>
z@R+0v*i1fz(h>b;1B?RMu4rE8v%pt*C(u)IqkJBr+@SojHZxWY0BsAAz&2;<ADmJ-
zE%#ED)3LXdnDvl912v50?>)W(3n8s4>l?`8kHJt!IAMZJ25ZaS6c0%5{x=!r2O+b1
zDO%h!E~&~LUD2uly7ScRe<xq5-UX+npfQwkg7YT}q%7?V#rQHdaXN+I^A&rG>oL8U
z$dsgh_15Yc{YDBoYy5sTR&A&1m54E?dzYJ`4eNj10L8E+I14g#5dkQiQkX6Y1GyU|
zFhM=Jx#q2!_h9yGT-5&ZfcvkPCM$-DX$IK1`y*MmoiaFmKp-4~zYkf_pLynBrO$>^
zCwK|oZ{$mQ6*eUcn!oiqdFDmcB^2Tv?O0bpO(|mKWr2Y{<pah4lwlm#J$yS(<Z?{N
zE!Xyq^!?3n?XP><ADNcH=gl+a=zBeDqt(%Mie~$}>2N8K%TPI}9nDF^PSEUn$2wzr
zfP0D0<IwI6Q!Lzo0M609di83+8HkTOqt@Wq<mBY&t)D-oAr~XFzrUaO-4qHAC%b;#
zRAC^<b--p5;M}oOI}WpWJYm-~*AA6ucJ{>LGDMIM17e7%$v`oN!cJ%fqyoSj_k)QB
z&4Y%~K9r-;NQ71Tk3R<2eD1_ezgPQx@*M(3M8=w)JX(;Pd8$F`51&Nxga7%!l)M_6
zt{K1wMgwyM*qoYw?qlS~$ryRalNit~ENfmYKI%c~>S&TB?0cu<R!G8<XXafvE1nlh
z?kGy;B@}n8hQt{?hrH$c|MPXNnhvs!ywR<!xEXeym;CE*I@)inlICad`K&^=az@HX
zp;ioR2E3hwk?D@#@G|}H+5%@iN=e(FV(w=M`ruuT$t=A5^;hHj@})OWGg2YSKGkC^
zV{~Zrp=fk;6)E7y6dsq`J^d|L@q1Qx(vdZOE-i2-x4d|<;%OIKvz2-=YKRE>LkleZ
z&ZG(6jr#&rJ%v!r@x9#CQB;g36YPkj^|jMKK_EOQq(Vo#A}==%i1H1fqk0PO|Nq{r
zd?HdS3lKp*z~Du*!h*p51Te%Zt#3YS&|C6qeElc@xS;ph*@<nrVFk3+pn_SMHDf`c
z@C;@X!D?f$1Rnmr|L>oM36N`MturjHp@VzspS#Y53e>VdsCP^+I<OMDBY+VRMOi@(
zJ*Oxrqv*}%GYPuzUt9&RFiHB-H1_HS3@7QK#e4u=Abq$_!>fp!l1yHCX@a(Th-wHI
zd7zkhr9F@9_AClq+@hxJ+}zx#+uq)nXy+V0>U_?F&aq|LP0{0hD?uixtn8=2LCRmJ
z=Uu3W2`GW+b0N{sT}+%-OMg#~9KX2i(6zR;5f-16oA)upo7rR)m1%=<`7&1c?a>N<
zo+<)^KZ3T?S@4_;!qww<4(tc2tyy1qE(EEoFY;S|mp^B-@u=1L;-nDxtv6~ycn<q~
z$2>lBOYls2q3UQ0YKU0Tm(PFJcBe^#6mgE*(nv3~CyI$b^5b<wmOh=O+@ZKin@IKp
zF8?dX&j!gvR1+@g$1H(vP(~D9#ShnXhlj~Z+NC<Q4So8TN4Vbq@AgBgLwH8x8eR&&
zWCa@PrWB}Vn<a|Z=`n>8p!H4e8qz9~{?=$8onR~Zo&$#F1WKhlw<&G8s48ZEN~`C3
zTpGdaP>54JhH4#s#ctUYP%l)^lZ|S?ut~qXI<K2hKb$apgGYtH+WSBh_XB*WwtI9#
zP@l-nHK_O;6s;it5EUZHgHRx9J@}P3jM~{b8$GvqRq|Wt@ZbtoVI-y1NZmMf=MfW2
zW`$LMPr5&doo%@O{-yErbA3cagvN_3Y;Y>*x14#zBlT_tEXn()1!u=({=rtp9R3I?
z@BZTt2`2H!eZ0#fTg7*}&Sw@zwnv<Prfq5b)A@fRvWt@>MVkLbWNHHP<3LmJBXPo&
z(IN_fE|eqgHotYcV=!O$op0^Q>YMk^@EjLTq}A(<IQhRyeYZ?TRyVFdD@H~=dK0!J
z2)Owb7vF}oe9ge&esD$a9vr+Rr>c_@c38kf@C|snWW+eF)GRFWX3+;KEFXOU`_dyn
zrtuC#*uPN$n4NU@YkKMO$V&%E7`^Ha=Ja~4)2f;_q+ihl(DFM^qk|LiDR|X*ONo4n
zf>K1=B5po_pj$HyIQpEJa&K>(zijHI0TGC>i_rSiJD&cL6R2kuSNqNZb|FR2)70@1
z>~%D;CLLNC%*NTdqA2%BM~AWp(l!lx@3T(5Q8(g_0yQ7!-HQ15UPo}t{o4S_ICZYw
zE&K{M*AToiL6_x)%CJK)wyknpcINmXp93a^g_`_dtxxriK!f!AZLmW^`epX_>;b*#
z<%u$CWZVlRE0#@ZInJvqC3e0#b9JzI?QUaZ<30lUfiIepRH<ZHygiKo6uC^e=P^-n
zjEtAY6#pLo_CIipVNz)c4xH3%SpqvA;q(^BGz1*Y_nh124sacpE1ek{6K4qA^49Rs
zGj1|Gtj7Ae&$Al$3pPbernPx4)b#0&V1(kCU~MmRy;_#fhLQ$nrc&+5t#a?+Gh0jj
zg3Bkx{*E8@XfS0_cybkQ;vNAz>CK;O=5>$F2ng%WZtHyFGVS^0u_KEq{H-=4NawS_
zF8y8Tgf3xPJoxvss=Vbx{aevnR<|!r^0xJ_xf2<Ww2;g8pg^Wi{uL|I`t<9|lZY{y
z8rKLxhTS%~vw8O)?94r5z3i!-2h>HiTLr&GG9>X!{NY=)exI$Yg}lMP!&b~I7W=8B
zhky2Z?wKd_Zm?`rn7Obk@pmqzi_oH|3xA7nICRNP>422Msjn{|d<yve=MO3RQrH<0
zEUbXZ+?7Ov31}_1bw;`^90tKF+c?(H!cf6&Gs*WTW`Vo9ukU$EN-|517|hi{VvDsa
zIsl>{=irak$OqH0$fJpxaYy?)a}i*yzD~+8;>$lkkKH86zD4?$?Z0R@pMs@+DpKLz
z!Atuv`gdeo16Y-b50w!C(5@_Ced@f41Io?XZjazh|H?q-g}jY2UDLVdkB(}Dj<m_|
z9UnYWD?5HS$bRxFR~qGbr6&WygJ8ofln^Vhj)2&YsPKb<Y9Kv>2iQTH?%Fw*IWd6A
z%eLvBmGiU{gS}r61ZNP0C=2cO??4F4>+-yDP0d1|*0-TSf`jGsm%h6;$8+REyTJ!d
z_XP7${6^`e7GM)NdhpW9<uOE)0%b2n%{Ssz)~r4#Oy&1AGV2ca$8s5Zmt+a~!mlxd
z$UPUj6Js<N?yv)y;0r}QO}uGB*jSoAhEY(8pXRu9$3UWBxWoP~%#Ve;Yc)%an#>j8
zC2a{XiBs9+H#<t5SBZueiq>YjX5vFtm_MLy3^g+_#IxZGIy28r1D|*3VXcjg(Ey&a
zn9&<{#y@H{Upf=@pNW<DdY5A<A2UWf`^`8ymuGCKtWqeyXY`QWGjNhbr4%oe2_lR}
zGnb~~{}{#~%GXgM7bwrRb34XRk@}Zxze3oS6zLcsykBSt&8Dum%?waBhQH$7J`k|Q
zD;F%qyrKW}3__`z&S84mDbRXHlagu^QZ_RH2Ym!@ziHP~GamYvq>hh_!#Hl-=&RJD
zFs?tmp!D5tr8}i{!^`r!7k9oq#k*+i+xKc)G6CDthCI#ARPnDgQeCJe9vtS0ha#a{
z^Cd#7Ahn}M8P!Q5O7~9*T%1UeBt$Tsst*kcB0$H`Y~Tl#__fPQmO>1=kA~H;{MMKZ
zJ5$#IookWSH)+nDa*Vp%&NWEB*dmzudJpw~W(wqfl*3OGL=V!2M(@77toml{iVp_E
z8m2nOhB^L=Z!DOiCSB{6x{R@d;yw_(gs;y=PK48^q`ODn<>%+u54-O|bDLgy1$3gC
zk7hw0FxK4P98>tMD<QTAJZ!@Ub8Elg-4|JOTZbMIcw_}$A1~74St}XiId3cv@jj!!
zib0-ALYP6Tm5dlmvolwwt`I^YCx(*kF1uS9^<OJg&~KqJx*}VFf_lt2%le8PQ&_Wx
zNyxzwxRZ;qI*J}!xKKDcYCO8p9AVzpe$4)Cl*hh9ms_t;F?bT@C0XTeh+U<nh=c`W
zsFXfdV9dC~;BKx%4C-}1xVf#zt{lPNH@5K+=$+PYeH;d0*gpu<y*z8JlMMSV;gK^i
zKsGTsP+t}$-wSY<CbqRT@|wc4<*8;WMOy<Qw~*zT;|*trS&PPag>Z}n)1Q~7zdg4y
zviF6oh#{a5ZHk^fIYRVE$huqqP@tX<wZ79_3s#5G`t_A>X3|3eNbE(8nX*lrarklf
z9SpUgl<#kFDVSs&4cpz!nVdi9!K6)hAZJm}IA=?k)N^&DfUI=BMx4cQsk_C7>uy)e
z9i~&SCEUb6Il%J|00;Fv!7-8&jXI)si`RYmwHwW>GU~uJ!o2T(c5aDTb=z8%gJ*o3
z^e1)aDZ+4pilNrr2oY3JV4NO*0XpJbn1uGNr+$GIr03Db?**lAnZLr=eJLV(Z|`@4
zy3r1!75zkFRb$j^^PVI|OzCmDW0T1%AtG0)MAY$lh`xtfXpg$AZyG)C)ib6i8#@1}
zO)<nh3+yBG*viOTaN_{;x7e#t6;aetJ^DRQ7E-KK$eq{lop-M_>QjFLhyeTmbruvg
z3hbcD^7WC@%W1?DZcbL&P}D+%)?@wzz2cxlYAzhwn_@2gtoP&n554qYkb~a%7Z$l{
zn*sUnq;eruQDX{Qlyk1j5jZnX()_}|oA6w6Bxi`PU+H0^^>BsubLrtmNl}4}uY1wE
zwuO++@=#Or?Y7Py1jyvV`>u39s5s(Ck2Yj0iHX!O@~h-?cK4%QQ&zp2Pdn;|eFxoI
znr~%Gp+u&`9Ncw5BuJ<)1lX)xn~Z-h=Xl&K2mT;_!a2p!ojux2`BXHnm_TdRr+Dq?
zOVrkH$4gkG^AKz;>bcu-tE|a#{bUzPB%C`WVoQ)FLMX{A1?ufXCU1Ay&d`Ub|Ir!%
zI>S`*XX{;s-8{+27fb!ylE0tpDqjd#SuauDYTH3Y2zkO(p!@Wz<xmFeeSRjp^>OYw
z+$4D#1I5|aE)#N|W#W{tOy}N172)OKdrRNs9j}1SOnTd#KgbziylLxU@N#yj%>FT&
zU8$-+-NkvaK0fdZVFKAh{RM9oM2$Db3z|Sv@qKA&G5J~8yH2pOJw1lU(dPNn!aKdm
z!k017oPRx1gbss-wBaGWJs&PN2B9#u7@oVp+rC6zIlIwOhs=iVMBd9kG5N<!_j*9P
zV|+&~?2uLR^Ff(++>ILUDEc({Ag-E|&96=Fs)fSNv#wCxn)*|>{F*Q_ZhsMezb6f=
zisAFv+a2-Tmhs6mm%Rnx1oA<-XVE~k0h#uAqDFa^Lc4~vQb<p7U`5*;Og%_!dKiU^
znndt+UkT>2^|=bK9vxqKWPv5~YZ$#MAZ6Q(D!hd;4lX!Z51X79WlJX^?F-mDx4B6-
znO9o&N)|;UsUj)(EYg{Y+r~1WY9nc8ur$79J}u?EI>^|;Q)NQTuK5su3?!5wBmS3U
zjxp}V+Y2^Y#@w18btp=A;2MHX+>If94C9TPi9fgSJ*iFi8JCDR#S`EIknwmmKG|~c
zBjDQ2)DMSX^pwBpu`SIiN)b%Tq39vmJ<vkYm?iK9Mk*ca)G<2x4%VOaw7)XVoz%e^
z5(2LRg&$q9^QMXI2S*3%6ULTp!k8;b7l4NtAi!R_)DrizN}6BdX5f_)znJG{EGRxG
z_{*TNQmQ%4K3w5O{o#>>x5J6K`g;t%{9CBdp1n5k&5lIFA?i*MeWEIVj=;Kb)Du<m
z0m?vRFJ6i?JN@bD-8;0W%1NHrljEQgL9hS1u&@xz6aqs8FE6jMx@+qcuule=ko4Yy
zBi(BrObR8>fAqVLQLWCtC`ej=F|y(>=z5u#mgZLZI?{Kf;UPY{HT-?KTO(?pd^mCY
zJA&k&HAVw`_fzC)?gm_KA&_z%x=D!nmhkPD<JXNpX(orWI_#NjuO%H>LK__M?QYVg
z2X|9S$}OSl54;&&5aqDg$!LvVZH9ILPXN2j!&ywMXNVnd04T{-D8MT6DTH&iorOSv
z%~;Rzd?#|EZy1i>{+Cr)rKI^+o#)B=>YdVF7$C6#ISx?i&TLba$PGZ4wh-iq=f$x6
zNIv!z=XEo4GXCi3`I~Z5KgYr{gHmijkEfrwD%wzOF>r_Py!K!^hS(Lq$RIaLaBw$3
z`)Tx)Rp^W(Wsg00cB=STo(h2Z2lCq)WlSdOUrEAbKY2au1IK8Tcm}P*)lb)+&X^m`
zzs5z^k)}D82H^hYeFbut<;QFr59TFU*McoG9>ig7w|RtrU`!ZK)YueN`}nce)gkRi
zo656~gy&_+F<-GocEk$*;+fnsuiHF_n~x*c*17ji%IV*}dXhpF1Uv!CO874|wf;#t
zO=@;b<>IF)-ENi5+9t<J=P-`X+n)=@?pZ6BAR@arHP*`S#`n}(hRWdtktfs3Rwi|V
z?ROqSyI+-0Uw50&WqVfr1q>_C<@a4w5ZEzE%+=uozhjHt+9-$g!f2EL$hH*-0KsAT
zbS7bNk8&s;w*4wWH^a^Ub6grVBDF4SLoU<OR_1C-?XhOeNgE!X7TE~SQm3*?!D0?l
z^3#}`>4swY+DV|~pUkr7J*;(mA-(b*N<Y5mQ=imRBB(?KM)tJ4m}UTmP&b`O|7|OS
zcO~6v=CO6U$N4^h62bE%1vnHJbxDyV#)6!$&VM${`0{D!qo*&LX%_4>Dm(6eljg4g
zydVAFn>UIKvikC#Q9P6V;@g3TN=r*y_G`wOFNoFByW!@^^Pz)!kz3fgMMC$SS!92+
z2|YNP9z9c?_QbL?;c>REWO8+NRfUQ&=v@D+b66pX&L@XOZ2JZD&n<Xp0NV@L%Wzr|
z${8|0O!vm$@$rB0e5Q^A^PVP~-@^Pny~h}9SeFyK1716%XT^jzm`w(Vw-Yfbr$a&a
z?XgBPO1!8EZ4e~$4Ad>U*j;`qpa*)5nt=h0y^O`@mic`3e8Qa^elOhZ9QWz&`j=aJ
z!66|r1*}QSYin#WW=nM=itAF2!0dG_X?f3s(vZ+y&6EkdV)o%P3M4hPM<t!#-rpzg
zDc1SS123vFU*GVzJH~)v)03tdw9`08nw_1VRV|f!Q$0I$qzBkJr>B#;tG2Doq=Sya
zU$@FIu^aN(u_-WInse3FWfOmuX@M|+|MuwQq${4cC)T)O(b_dFu`0wE-;Uo$p<ZS%
zeA_+gs`0ZK18kO`1q%8rGEY(eZS6XoI|ZAOx;fClgVf0=XUe>~ptYVoclG1z`(oFu
z>d=m=Wf4;;z=83P1Vxvvy6bT>XpH7e%`U6lXHrSZatV8`Hnnt@6koU0=rX*2s%0O)
zhJ5a*ed01WT3C7=I*PFIPX-t6HgUrKHQ&49Wle7>>1hsX`^LVp&u+u;e`2Zwijo!h
zUn&~c?%>I|)i5DLI6xriN8E&;3E9t~@07n_z_&xtrv7C`>2K1KYrFeLN|TnKUe<`x
z#rNt2<+db@ii?i+UDe@@7CXCfVaf_#>X}QDPJg0-S<;UpS#8p&ov#kIv4Q0Q4_hZl
zG%y>1cs>TPJ$~M^Et#&FKmgC;uF8PB`O$X>-IPImo80#O(SYV*HgZE4kJ36?vob<o
zSl>!tT=w3`yo>waPW?^vVKN71sakCkm784LO89mtD7DkZpo4!4c@KF>d_W#WQNG>3
z<4y18VDf+gpNmk)Q#duCS9~LS1)qw7IQ`lT`!h=HGhe_L4&RGQ);v*!myAH5-rmV=
zI<zN<>N1SlrPjyyVbnJO6etS>S5~luD>e^UW(@*0$}A-GHX}1LI~WtX*%@fp4S65I
z1`xG4PmnK>P-grbA-h?r4<1T)!Tz`c!@@EVHQ(gsQq%7_%P1%)i1=>k=z>r+P?*1_
zZ-w&RZ_7QkUdJdZ5UwdgTb%^O8U-5|h&!F1d6v@w6_^fcZz52LVw~Ykv~gl$V(4!$
z@>P}<#jGb>THkTG^5&=;5ult+FRpkPt%4d0hQU`pKAIBWUFx7Y;(fXWwd8jOcAAl@
z<dpUwkTlsr=kY@jq^k~(?5m{e%<6fO3nvjN6BGHmtZ}RwifV<vRO#^FavwlRGQUSH
zlLiDB&J{OBoGd<!rJ1^hOGp4OsZ_r3dC*GvgAuTG{$3D{NgLQ#!Ye2@Zg%So$S)-g
zrP94N1s1}DUd;c=0#LmkJ}3@2+A|2f=v*eP?nz@rHeV2D-Vk@=^MtifES@Y>%-ki5
zo;ky0yV6o@7cIlup`FUdcLfPMxurK1^?fb)&G$3$d|+aO|0+V<^H{wKtuE&X0C9iG
zx}1li9iGl#wNJYoHa8(gaIa)Z)DGIMvN7qYaOe=%#)`>4MtrK%nfv`5sE{2dsyeta
zIGkA}Ui^xs3CLzJYM*mFb(^yiFI|BC!C{LewV2<^x<mH1m9a4Ag$v4C@8X5gb2ZfY
zLxj9ACp8Xqc@DMrA+P4s17j_wS{&h(og`)n3B6B$;@z(_q$yCwCARK|AjQ+oaDPL@
zx+E7)`5yXIZO%$NKnXYl&x6gHFi}O2XNQV2*iat%5Wg@mFb#Y(=B~I!7jR!UICuGW
zOPcJO<Nikq4eyA9^AM5PYw3iS%aX(?Ui`@GFQ-Qrcs|7}W(&zrUo3dx)g&O{eLhjf
zXfsjPT8M|CX|Us3nl)l8<AvAN_kJQmh83~iyPS2S5D1l^oe-5=!^Gega^20^zB;a%
z(6w8xZ9L~{cf-r9@J)-f!eJd3hVtymfILmQIjXiN-bx}$+eHoee+D`qnM&NoaJ4Ir
zK0Ir7X)nRpp<kC)54#P{|9`(xt33!?p$!iX%+g1tBpFER={rtAxwoWzL7<vXD!3gA
z4J7txk33$akt9D?)!GFv=)*2EI_f=CO3PG1bTB0MTPXgPgy*=0U;u&8)m`JSIA<)V
zOtP;Se5PH6)OuhmOtcsmlLckUulwoD;XVZ4UA2fJr)?h0C`;zrB&MbMC{~LU;Ds{!
z?aF%yDF=aqnmbJFrA|9@vPM2vNn0b8C}lM{$&1y0faCzi9-||T59$Pz2Cch=r7Sxo
z&hsQVb_+f4p6~=?it<@cRs?CH-(@L{!4C11XRK>;gBe?PnMP^(rz>(V!IFw^lQv_{
z-Zs0q$e2`Y0wX^rB-x|vx@c3EETy5jtFxZ~)AV}h*!$D7d{_%HP<WVjS>dg?D4{}j
zLxWW7*cVIP?6J3&S`te=pVu4uR5cI(f>7faA=&4=Ic1q>IwJfE547QT*kqjS+Nt{b
zH?PyxFfb|VUn{Y|wVkdIVBs$L<7sB|3UESg{EL=@OGbT&PGy5anU>L`as}J-Q{bf#
zk{vToD>Wa~%PWxG_~>tS&7gVpux2IxdD;5+l*=ixb!d=b`T&BcDw4}4KEubf?$m|X
zu)fV6cPQd_VlG_-eN+gk@p}Q%?zwM{4ncxt3RvZzzrKAp=q_ssW!kEx4SRb4GQh>}
zXG5u#5E^E?9oprEsyK6^#!+5k3!K+wLv~U-V?&+kn%1U_<ipgm@!GY0_J#B@s|+wk
z2L(UBK9(vtv*TF{h!MF;{#m&}GpHniTPW|ak0IemHHeT=Ee_b`{eL)-8;CJVS6MC&
zhJ<Ir9nf$%@vw%ciap@zX)8rXskhb}&Qj$3KouQ>5Un?6pU_Dwo|}1{ee0u}u=Vy|
z)1qO6j^+*x`kzev6VG;ODt6gFWE|QIM`F5#xQ}?^y=Zb=t0>EFw?hIjySD#UEvH_M
z(8!(*9$%AVF=Q$yUT=a?g*4^c7CwNpb@9>W-8VnpKmG3S-x-==vM$)XHV7T=lWVwt
zeZ84ciX<>(MBn`HtdW2F6DgXKPMU=hO<7tD^q4jrT=Zeae_9cIt(VSAy+Ag82rHqJ
zHFQ8h@!~C|R|iXzPQU1vVG19zD{p)n%O4CJtwr$P)SjliVY!)(9V;G^i_Aw4ZlzbO
zhGn*&LD4NBS2h<W59tZ?)5ASMlDw|;U(?dkm@5!X0+fB3z0fv@EbIrtHZ1q?xg^*h
zMj#3q{zY_JZ>YqyWTaQ`1W}Z<{(O5>ng_Zn2g8Yg*)|iGSpy!K+UiF2Ob)ofG+3Vw
zBJO#%Cm&;fJpAmsXI3CYP8%po%DrT75&_`h^#}YXec{*P&5Bu%@Mm?I`oj=@rwMP}
zC`i!U!H@9sK9qYnbnL0AsV&9466SJh?u7MYS}ydQocGJz00YI3(k}B>2;kWI3~U`i
z{o@50wojI%!v+}VB_&)jsj=5wW46K7MIy=Oa+Up%X89Y0#aJ#GFpPaAM0+YI)cM<A
z3<msC`@Udfmq?C|*3X~T{mpa6_oulGL3eeOyd`a2oBP23V)I)Hm=Z>h1AB4)X!uY}
z(>@ZPJD5w~`n)COe9z4A)@PFlV6WT94jkri*autd(F$EXch(XF3rc7GlJSWT#$c86
z=he51pPOwjQJ26c@;t1E>9n5cr6xtGee=cL`jH{sNm6;Ux`?ektnNpI_uxU@c<A9-
z4?ut2bQ=!Bb=56*o_O?C%SBMP98|DB0vlQX8ow%pidX=o8?VHeyw@f`y811aCB~vI
zJ4)8RJ&fX3ph(Fqt8LxmW3h5;v}KxEMMT?V^jxi@U7+c?7a{END)BPapL;F+_JDYt
z#+;M64$f=-=F728<b9ag`8oeNU>_{eDK%o^`+Q=dvX@c8EJN&fP5Ir@nz>hM2WnZo
z!fI0Ije3Hc;%*FLw@hc`yISf|ypC6I8R_F#5r#071WI6#a;vbHoq*c;E=k9^+lvzh
zNW5!lX<J+0tNXeGi17PyP&gzo_IaL!qWwpU4jLZvb-U@_2*p7!r2pqTh}5iy2?3F<
zl<eS5A0S)_<UFNOQJA<)e_x(-7SeXS4wgEL$LIL0@wW^9Ir)*v$?Rg3^-inN3nnf@
zuCSvFEPy=YQPK3G7p88Y8H@%n^IIs|1+B+v;y6I#0KcHC%lNW<lj?*EwZr|(HSa;$
z%854SGSghM>mEp*IP#H2D^w8+00X@eY2Q1+Vj4OIEh)NYDgk@wvfc0O<Cn4^<Ss0#
zzFo?$@Et~V`Oxq;1$dU;slvnx9I0+^D}IA45i0uycQ9LrR9Q6bJk+4-g__(QV+UZ+
zM<y(>R<9vPX60KHy?TYkfO_~&83`j#T!#MlGJ=801Ih@{gEk(A&?k$zMzgBt(Qx|?
zFTwnCaQkQTBXfDp@tsv@lw!XzUeH3YvW;hk?{1J!5e22nci6}L>Hj~({*pR+`)w!L
z0fMf~X``P0=%aW=Em@_a6VdwWa4)4;ZliXT_}ynX{gzZuLSB6?PmdXXISa}ur_T8v
zf4F+tlNRajor#KT)1hF@Af4+msmcQTs2Y6Uj0NMYzn1XoQC!Ltw@C8&NzoQLt>^`u
z$B_%KAr5{2c(y-IxxZQjJ+|(E!@Y7THQtnJXb-qQcnj$21ePrZ)W%j$yW`SoCzFI&
zd2!}<KzsT0Z~Ot0OY@@CO#SrgVMF4{>yqZuZ*QE=RWtZU20atIZ}2{B4i?V8FMq$;
z)9L2!pRz42k1rVcj+Z?CYJlAnGLQ-<<d3|BBnwVqPTK83UU3-qt?#2&Y!{CwWEApQ
z4l=w${{l|SyJ;S~L0XSIY~%f@3JI|0XVHNOB$RPRLqmgS{EnH8uv^o{Uaf^OKC!X;
zse3jMk!_DoF;H{~`sW;_o)EPe{9pMAet4-j=#MvQq(iE8G;B;#$Vt;vIYK@l*-1Y8
z_oa#t#=*P4)5#|?q-4^(eg{?G^EXKtwfLkzz0{Z+?e@v^cfJ$B)v)bn;d@;lv-9*U
z?q3YR-!>rc5VSc_$?wJAEw_>zx8L}T6A9HY&b$6>N}<o2;A>)LW|l8>EwE91_aV_g
z`K@0my%lBn-@kwTSFqR_43vx>>$LM|`&m)|h<oe#O#A}gyH756#cd18lxjE~iZx0_
ztmnfL;VvTSSy~~bT)flOG9ZOb$}VpdCF?>72@(Xmg<+gVlqbNPTf!I}@gr#!`ukH?
zC{yAN05zw{b&uVCI0{~wK-26pt-aLwyfJXU(>Q-DW!$myUSnabV>$;@WNdKaV)?^n
z(XAR|GhXpTZsRKucGxh?;9A(y5RV%kokD|#&PRv;mj*$>Z-vL_!0@6it<6jG@y7H1
zx^T>r?I#JGx`C2pHYy%P9e4$Z)c;7}5{suZMSCvB&h0w9mrtZ`m2NeFKLayr0o=sT
zTatg(*V5rlJw$qSXl?T+`N%(=gHA{Bz~)#VeCTt>MnRbiJGQq-UzRf=S9<CjxYDh|
zvX{2tT=RR4BU)pF2KvJ0_no&3eyY0sX8UWY$l+EoN0uxlfuk?KV`={115WC-G$Uyg
zv_&uBarp3$@M}%+ajAxsi};xW2(9w`7$Tl2;;`vM^Q+{<<f^zUC&#|!gYTA=;nleq
zBj&1ac2S7QSCrT<fG;qE`GS&n7qW{8Qix}67OQ3k(7cv}Y4HHhz0Aydf+K2k*mk7$
zqrG}mzgY<Htd9=s)Z?wM&|4-vfhHe@EHx)5K6@_wv+tEw2{dyC>|J9y>*oE?L^jGb
zs-mobs)IR4-1g__6-X@pv#N^q?J8?9Y^h*K-=B2aca-Y-oPPViH1c*&P}yJ%t)x!k
zz-)o<z&!WSh3h$fzxcnMLW|Ne)RV-Nl<ZMB<!pdu-4NBJ$+kt!z1=$0?+O-=Kr;pM
zUr*v??4GmvfEYRmuOdT%a2bk1nQGM4X-bY;q0Oq6!WkbkEnyekFyx|ln7I&%ORg;A
zsyQfh{Ct}!ataTSv+Xot$aO=JROsy0K$%rm32xsk31S1v0AUdaKFi#@hpoyg!)qE^
zL*NKMVFn|}Im!Sa;rBcm5`w(bx!;kB;NB$@NLO48nSjw6ose7L;*OQSKJQ2Q^|w8E
zlSY=1*1iw_vJg1eEVavEnPC?iFL8`23UYl3NKF-a{r$t=Mpknc-t1F`H*{Oa@#;x6
z{hE`qE)MkQ{u>^ebS9r5Hag4YQuk-hQG<Ufycz?Tjp|sIh2(Fd-%l2j2DN~ml&9=g
z9H*}KCxMOr_r7=E1;$Ezb+vXQIxl;Zph?AXIfcC*P1u1vgcj<u{Dr{4X6Wwo<)i4#
z!fSK>3LSr6gfAfnHa@)-uyn35=^#XqXz=v6P<f$kkH<~3#aprS>%Y{d$QS;{a`ll8
znIC+d^YI#?Pv5V#T~91t)}nc|T4~9C`?sVxw(!bPlKf`kOQ_13Sy>gIu^3|b@F`r3
z#U5W&=9E+;{uB|GYwy3p9o^g>6fTg!`ODqRCG1xK`ooRLWcAXl+os^dJQdTGQb3DY
zSSx~;Rn+6h;>wjx7*P#Ako;?MFLvuQwWQb7k~b-A=gHlwJLxUt@Nz~>y&W?|Ff|In
zx1*7em@Z5m$HKU5fMRLY`hw%T!OyFB3x+HrHIB>ok(EKF_JA-v$7KFX{Sr>BAW^H-
zf4iF75%1!BYh%Au1NOf<YhtT0Er#=R>A`C#&V>eVc(WkCkBQUgSXorzqszE{cI?eC
zUKz4K+b$fQw^7Zu@O}FGA3}N#zD*{)5KbGD`~za<6l)^|a-g=vTmOMj9s&Vl1=128
zUqOw%His^Yt2rZ}7c<F7a#d<kvk6nen%Vo~VisC`lY1d#x3N{}vpUN&r9e-~E|cAG
zn`vpi=2}}@i|~v2ZE0ZHYAlsjFre3v^?eUZ>X2zkDLXl&fFX;4_#UXx*DoK|-ik{S
zuu%k~mfA%A`np~7JA!P&J`i4~)h0enPLwW!VLZdL<PkC)ligzvQ`{1EpC)RP-~=6`
zHH1tsHRpKd6Qq2=s@#kQr4Pa9yik?g$-5nY4h8~3QEaW4*z1~K8wZl=AU<W7!S3M9
z@t45Qw@h>vVD%gtPc<D$#+w}F_JKkYbD-*wcWSz2W^Db^G4N^QBYsd_P`B2vC*{d7
z7NMso*P~6a#LE_D`DMA}g$D}R{RJ{IMqNgv?zt%~mUH(D2H2yLb|@IBiEKwOZ9P<8
zTlp*KZC?mg#t^^2D8b3reW)|M{%f7_`3Y(3f)%Tzq~sfnC~AMoeGiouibWUQR)N#)
zTElR{6t*XvZ^o!_VNK`-Rr&-Q39TYC5C7GflKVQZc{=cVu-9QzqDt;Wm6%wd-fc3e
z7gw9`;(SlP3BG%hlw#%fR#Q!l(Eid@ulsC^Y#A-m>u&e+!?`5^rTg%>=$W(1H?K>S
zZ_|NtZ!>!Qclo4|y*oKF(%`XPXFZbO2&JQ>(fe0VK`T#h2=>JpPlAm_%Fs;l6^xdd
zDuIJICK2}*O}ZfJrzcs}B#kcrd4>n+ee|7ma<iWn=L`IPNkNiD%@LgUzpIPA&SV}_
zY^2iRVT{b?M+r~uR*j7syzN7FJH~Q*{gKnzv2hRW0?s!X$14G9f7k=v(uGg?h7EyF
zTl^N~3rnzN&G`))E+ugr2Uw)?hSk{5K-Nd9geU7HyPT4c!$P1%s`$NWr`Fy=HA|Q=
zw7kn)h>mV9t)vLcI-ce4i07cN7c~l^iWFA#=TO*s_i22B9=h<1sIUopUwZZJHle4;
z#y$)9aWh8lMNq3i4HRdQI@yLa03iW^jH$OpIhGNdrf3|%eD?c`3Hmu@<vvvDU>!1c
zTp%21Tjz5AH5~M~SdT+{4`lIdShNzfq(wgB(N}7G%{D7eF<fsCaR|n9>atFHKtE+7
zYpu+d-_cNUlhW6F5}oN|U{*@|X8w<`{xi6V)$2YLGkLxz@+6$K`hX-rB6a$-_=*0@
z)9<gq>MIa(%<SpBrh3|y$d(_M9<wv~ZtN3GFroQm5yq91rTp|0U=<vK@fMMv7S4dx
z5OkGww2kGRej!l;HI~lVxr%8OImLKzis#8Gx|IID%&m+{Lc6dA&9b7pdD^Y8Qm;G_
zN3@hv?J2+%_$>!pSbXIppi!!40#5`w<m@OodBQ#h1&BWKHCZqjP~>{;O>vPfNCsm8
zILMXvUd&JPVv9N@ATW8+`0}$CF*%d`Iv-UldCm~JRD)&X0gA-5<uk8U8y=61TKoBX
zPy>+xM%$~bBOUHfe+SQqt_t!>bCIoUFIbY=Gmc>9&C}(+R#d%GJ>$h&zJnDaGz`K)
z_)W=F)(G6wnosZmIsH9;p*llSXD$k|vh?8*gg+4z64Fhvnbni$dlTz?1f;}FCx!6}
zy9m~G>4qS5yW(#_Tm!`a1#cTyFlbF1v($R)b_Xwof-Gwjj;g`(ZaO(8GmJ;X1pp3Y
zrxML{L?;ht2w!kR4aA;5=I?LD%qjdL^*vn1N+MQynwk$Gm%9@FA(!d!NlEY349K$<
z>3)6e0&3S5yQr<%vdV^9P$rG&qwW=u@$dLy<e#|=A;DAGWtYGHbD9@Ku&}W+V9XIn
znkf2`^Y`kI9N&Y^6)2l8=81i&Bq>cW>kCR8Pl6a%#v46N<O6t5Ps*#raITj;0>o2}
z#96bHDeks@>LEW%d={lfOXmkeW7SwNF9Zq{k<nsMQ_3BO0FFhKn|TKhgt;uZBtois
z2A*b3FUONq98Q(46G4k_jrwv+GWR)AxT@MNo4H`hhuUAJe=V*1%JSE_vL2wWr+3`Z
zV~_42FxXIF$8W%U4}OS}FJC6pemo0&r6n%RhW(6XrDe_z)(q0rjV37e;@i(wA&HTY
z!biHX6mq+chIGP#k3;8EB!seXOkNcwR>G>&dTjS*=8~A1dF_)a5dOxnml3IW_SwZ@
zea5*mbA(jC!XldxF_bcD??&v<f5IP?+>d_S1FCbslRZ7X!2OL1K~(ZVj*EaJ%ZsjV
zz)0QWTm+Iz!M7Liahine`<M~s=BOOV#UF?WX-<z!S`Vtf7NxDaq=~uSQB_rS0y6p8
zP)?ZT;}6E}u>SV^?@Ne@$m!_od$1x%L(?!^=;8sH%0}FVL#5mN3vh4)w=Tb9s(aP#
zg|<#_oOqZjDOru+S3&I!rbkCy82+=~m)csSS;pm>d9DLG=Z@}UXe>N*&i)ITRX&Qr
zSNHo%94RHW1*cSgi_LH6KfPaB7$CaHUn>X|&N2PBj_`k?`)_Z7ogG<p8RhZ*8QfbH
zAkD7CHU!jGTlU({5%>;M@8&^U_uS%s<yYyGRqsoZl$)?3X+e2H6%i7mk9s?2F`vTN
zDcbruWPI%Rif+j-eNVE7Inqy*8ueNP%>r2P-!sbVM+pSCJOriUQ(@TNq2w`vt2uT6
z)Cx$03P%=m53pE|mq-s7wv^F{Y7v961LoI;iHym``t_**L$M|0Rz=Bv8E0YO=2opJ
z06~mEZsnn(BA)-pWcr97y72!G=6(n+BW~Tui);qE=r&7dqpB_fEKL??tA_MZbxD93
zs+Wn&P?ljs#JO*xC%8(mR^JtybCW&^!}u5?Qz?`{e$0RmWe8%ltgkdSNm8a1!V~#;
zWMr=V`^237W0mgCJKDM!u&)UV)#=sQQSQRGWX3tWe?ch|N%lX;TBHsSE8fEcc|@8F
zk@P%{MXK;b+wO}~`q;eY$&`&vsFmM2a!Mgq5w6rcGKuSN_Ar?}x1d?_cA-xwS92qB
zuzO>b0V|xbqe2%)-)8ddRa!$f|3WbBHT0>}3&_P!u@5BY9@NDe!|N;?1V}Io1o!d>
zinOQ=_uR6yea+gFXVO@>e=lxI{X5Ni<d9&dVr{w>2WH+5MO0b{1_s6>SQ|D$ypFrx
zIM)SI4zE=R8lznUde{EH-Y@nZ^@7J7Mv{KX!jDc2c9um$NcL>6=Ytre3IzdW={H}~
zx#fL!BeSG3^FGeN#|ynF--knu@%tH3_$4r$gjKI%{})xwNf|105k1PQxafs6|EOvY
zHEXuF?LMew2j5{av6A+t7weRb7&E%M()<D)CQ_MwODO{k(It!QqWtXg;!jCSk;vP1
zx(0$dVYM1fNDLy*sM&p0NhDW@WREj`6}Fqb1W4{(o|)9E;YL>E;*Ax!n;|E$N{V>x
zy%6{4D8?1uks7uL4Ooaih8jUA1^+V{QRj6ox{_a#wu1|K*VR_xiY=qq$@(SSbp}i^
zPB^XFhVCajA4rCSQ+@J8-%}QWkT~YIwr<OWkUCNl6r~$1<jt&K?d@d-^E6qqvdcV!
zoba;b^g|2s9lv+;2iS791WJ?>Cg5u{(#p4fR=+$hd<uoQ!pLokb=UetfeMc!2V_}H
zKFi|uJ9sS2c8GV6uN*E^O9x!t%tFbqRV@LkhpogLjN98k?;?3Rv;8-ysB9kbuD`uO
zKp?$tmmfQPb~bmG!f}?(ckAhHC85JC=JR!Hwm3Mb4I8##P7j=78X=R5omZg@&w<Fn
z4(#wBHLfv2gJo(C$UTbj&k2nk#Y2MSXwvyF1}I5r$B)vydFu{rZkYaiHet%H3Io}o
z%L#F6kk00Bj?wESkv@F;VwwtL#o!8?zmSzBptt(oX=+|((wQL%QF~!Nz;dMF_O^1~
zkv&Ner&g6;a0H1L?uqKyxiauN2YvW3OvwF@_1toEj>3J+M%do91N2HBxSP$)Nc}>B
zgB3}JeOq4&8^oB-d0(S!QP|Wv20KYUTsiR?k9xEHJYj_!?bI!^SLj168D?q{A~}n8
zS5e()(%Y83sh9k@|5v=&KNeYUQG!F)lNZPK-}y%ZPW575?@RNptE%I2k>5W+-*<}u
z`QUL4AI!()Axb=fgi%$70~KDWbcotU(tJmRecZ1RUGTX5+a-Ey>PR<sQo`=)Qj_yy
zY~^+s;o*9Dp9Pp4GRo&Inya2aBeMC1JrC7~|BAMnY|)P_1`Yl0UOW^q>XCG1zziKp
zm6Y8Jy~jZro>E&n1vh1BW{dk)tvLuplD2`!n4R_OTOa?+n&>~l?`<@p{z08ZI#P=7
z5k0-_{(RUQwbggFb?<=*&}MyFvM0u#8GoG(l2AfO*)<KsD%T`K1YtpE0+N1B$q%2c
zsB<O#V=Q1c;Nx&NZHRx2rL0My@HPC$kcr#SNnGyioCK0?HubIn-eXHhvV6@v2u*!a
zsxm~@$!5OP5=d|5&nySxWZ6J$4D?z1P@{9LhOiXVA_8&z0$a-%B8xKif+b;FR)B*;
zjmS2)DZFp}3_S?`S`;=n_x|G7V4*Bkd<~Dk&3luNW;S%o^Jzfhv>)ZvVv+@j4#*Ru
zSsRNGmf;!Q2jQNco<$$uyrxNic>$AP15A>(mXV+;?)4|ynKCR#oS<}nYea6J2kQgg
zXrYQva-y0H74~AUkqSwyJvWN^`8^YRQ_`uhpdjvd6Hw8^`1L-d{t&T5q5ts&TWO%y
zZ-}jKJa67Tdj)_K{uDg`$YzVJ_aTty9(2J<ag1a%CP26W=w(nY0;Ljh>e|zxTrJsT
zZ|@LHp%cdj*&(Uw|1-9}C{0EYtJ`STn?~PDtG}r!Ak4zWesj9^9q8}ovI@kiA(!N=
zK0dAfsNIvPOGV_e!;^TWM?y9#fEh<S_Y@`3qBs2^lHtGhaX_D(X&V;v4Z>X|Ypz-2
zVsF?SW04aRRzvW-BbqUz$j#N2z_w&>dmF!}dhDAjk&xSK&CZxx5p!lQ;A9PDMm8h#
zy<9LL4uKqD_~Hn>Gz>PR0gucmSrMe^J2}`AOj7#`W+oxX@a2xvRk~|rL`TB>(0lU-
z(abm{F|iC>b_tY&ApH9=PpMr#{89>LEVYR;XXhM4iS<*P)p5F~0|CmI!f+V`B;W*p
z5%vKp{}I@7Od!)*e1v)Jfw5b>M(s_f$STZ|OT(Y+=L1vNhzJP_lPKlBEUuI|bWj!F
zP+6Yme0@L1xJqnkhREZl9OttnQ>xE<|7OlD1Y1Ot5fISIa6kUVE{-Mq>H8#e5!EiD
zg-TgnJhA*3`?(S*93P`p%%hM)wVD)o*&Uv6^PxpGvEFD#9K>?=iD@~^Tb^ci^PlJ_
zKO~w@HX_q4K9V7jY%4tQ{ZIlm!AT4a3(>cYw5aL4=y$*T&>!xcYQ?do`zgWiTyPFD
zGAv+$f|+B!UhNl}ob=%YEym^!SPZO&B~bDT2V-sf&mRFc3E*}RB{s3AuP>shxjAS}
zGX}~fIV2uNBi56kWm9yf{*^wa(9UclTDb_cHW{#{WDs1~ex(iuB&2jp4Y-u)O%Cg3
zaYaHuPmX&36qIH>C01xgkDY#3tS8;gjC)%$=RdiLbRtwqoBDfQ)w)k?__hRDWPT)(
zJiZ8hTTSuS+bRXmU*{7uD1$7(misvV9=AE6#ndP6zaT00S`*rQCh+NiB+);RBpM+U
zQ;pH{>*k%--Ji5oOWQ*(Z{R-21;=qnFAsoAa0VEYmaUE9OId|}jL1d+gw7^k>j7nq
z&~;)v5X}Zi6%#l)#DvItPSaFz-pubaFEqGJ`22(wn{myz1NlOsaf4(IqsQU&F4ogE
ziAX#lgsz#n=G>BykN{3p7EItg{9hL|yIotHXih~yRQ&)ir%Ov90gO(vkeGk)uft85
zKuj{Fymmd}Nc+tgP+_AMYKxM7lu+e&;^2h+if-^DiF$pvfYkThnF7><OPR7yR_4#g
zMXwzu^`1w(V!IO>$RJSqMBQjB-At9=>IuO}{3x^bh-now2Y+z98ppFoBR<-tLHHU3
z^nU_dFnJWt58RLIkAAp2_WoF&a@_nYR{TUyczL4lMEJAorfXf_@>&4;;j3ksE8ML=
zKl|D`fH;}K%8!0mDZg7=hj&-Af@y`JRsUvRYA;BxKVvIr{IH0SV|DyN_s_f5I?eqq
zwQUAkjG~^f?9X>aknWUOMXHG6jiRHJ`!3N#qcCeB><2^+{exg(dJb*)-8ue(A`I_C
zZ*3SH1-3seH&bb5Y3IkZ2!={>SXdV_gqLgbYM-p$NURY3#VMP7KWimKp@r@mW;_-;
z={H%MU!*x)DS3CZqmN#^=}#-y!4xK@o$cNV1R#JkS@>-*vz&|!;wVs=k0MmiguoJk
zQh%62slu;lWrBCUz8xv6fpfEY@pWCw2s(y)M~FfI{$lRnp_vkVE}X&}0kBO0=kL=R
z7w*=Umg`JdW9%zjfX$La@-5u@%ul~$V^)>A0J|0B-#6ZC0>=2^I!||{)<`^5Gc2J!
z#|G=W@)yAn|91uu$xt%aSFk_teB?)YT#cJCrUzg40y+%}IweE;=t~5&b}*VaOun{X
z8qM?C0ezf#WxUiFiCtMFmsR>eTOvHi#4%2}9DU{?k{;Yk;j=;Uf^(Ife1zBs5r;u4
zmOZf!d+TE%ElSseCA`p=mIm9sP2`G}LO`H=4Zdjlq@Y%m@wW{+51VG+3XV^>muMQ;
zl7RD!L@7FSsFRkH`qFuESa5JOWr##0G>qF18~d=_$3H=YnMB9>77Tw0LTK`6w}pC&
zI5cCMhWqc%*b0b6xjmLVUZGDCdGSG_V5I&thgX#{u6E(ahJa9f2XKYsM&Oh^d)u(#
zp3i5Xowa#VaO!KJ*^`!Y^&``(yB3W_Slj0f5FA6~K2b&tOI#I!z=-Y6fp?tra*B&h
z`<0JNs5p*i@a^NdGt5^`0Flwd&*)C>Zl<KDm;wLv)LMyc1|@l?I-~M)sgUcp3(Gw=
z(ZjX2_f6OPlTp!=cV5lJKX*%t^d%jW0UKB@3fh%^xF&28wP}m1tE*|28n)bOYig`H
zBQ83200e$t4XnOKi($_hSee=|WS&^DrA{X+b8D~z=h}*W@D8oRn$>P$ve%J=T7ihH
zQx&%lvCia^gL<kAypJBJ5dUKjU`b08Mu3lvOjdIE^(Ed~Qpe~9IxLv)arG#|aFQDl
z9PjOb*%Y?1sep*a7o6}#81k(7pACOwPMql~vq9Y&d3VK`Nwr9u-)Mlx{WVy4fo=02
zPi>uqVZJgbWt~ej5GX3Do+VFURUj>a|LtBwVj9{6s~q}&QHs35vItoaBupl#C?kXq
z*F-@Z8sGkQm0Ij=-T9e2g(6KsCOSL<NLdyB*mswl0pD|nwzhi3Ru(TUac*_!1F`>0
zu0KMyNf-w-2DecJ!gXMWmBkJu%nc$54@1dR!pMf6I*-pLm5|Mh@jK-P(r%O8)*~(|
zgQTw4?Ro<(=Mm~nv*VBaDIvx(D2!~PL@K=D9<4Qd6ROHHEcn^+uik4pZl10?vK<)4
zKZ23}aEamFQo1{o6cHsfAsjxn0o#olR%q_~%z8C0qu1Oqxc_F9+I$!I;;<NO5qELk
zR(yTrtLeEmavdw!Sf!)vd%SM>eMn$Smw#%}Qj1j*#aMg;>GjLeJi*zr3fw_VcV;Dj
zX8h%p@$YXLoAm{zQecn$f`?WNms$7q<6i~Eg&nL))TBJd^Tt-<ZLZOgzN1U`;xW+U
zE;13&(3R`^-aY;e!=$8pKdGS~Om88dC^1fDT1X$^AMnMvW&!bWuPu_qU*mav3ZFUJ
zt-FGXX%TdB<vqQ)@+>N!^zoc$`@RCs82`l?t+<Qz%+KdV;WN`-%X>Q+@s&7_%E^89
zJtj~WQXYw)eEAcuV6_OREUNy&uEai6jY0kdntAb5+9f>!96x~wd`;z(FW&^9ZP~`V
z+}zwp4<F*g5-ke;M$}%2K@|*+`SR0iPbB7G5w{ttc{ybDqoMjJ{}?EoZH4acoM3!m
z6{?-}bg^mzzoW%mOVB+n*RaOdB+tr%xPH+*=aWVbTGDkc5v>2nx44ALieQ~2_p!gA
zqYB{vd(gOoJU&HfWNgjU+umg{sZ8*OXm_|@rSd2ByvaGqUP*>cr#f~rQvCY;fix>$
zB%(>!p<&27;3locOHbq_1<H9qqHIvo-Bp7QF(<XWY0J7}ZCyAkEyem{)MeVPFXq;@
z?Ck6yQusy~5kr24<*_{D)o^i*Z|~s=r~vp%{8{eVCQJ(L8v5GX+w*^mXy3Sb#f07}
zzhOEP3IwZIF^C~f5_hjsO+L$k$^nUu`&*<@e_Qv947-uh>Q*Z5y~BRi^H@&ZNuwXO
zwZBO@^lCKuZX%C3&9MQ^U=ik2U4kN7jw~x(;ONTHr6fw~^~a}N*T)EFE9odK@5#z4
zIA{GUgeQK?=Fc33Gsgw4$}cT#HEKTkfn`D%#aJ7-?C3kIxoL`|<^vj})zS1Nxz*5B
zMsZkaJs@=GpPc%(6PkV>d0p)MytM`KsBRE-eu(iVv@&Uqa)(Q9bWG+{JapL_a!{R)
z_br?AD_p2R(uIevgDKFel!nLK8wZJx_TTQ|!==UGwToQ9N14u)*zNNS7D|ziTBd$z
zr1s_RZ&yXbG#%kQ*`lUlKEB91on$m=I(n%*P6^?U!e$7xJX~P|&OIGarer10Ps>#)
zXOK)pP<Bm;`!e+5Ma}{6le|^wM_}#K#h3LEu1rHZ#g7a72%1iw4!JbDPg|MG-GcQb
zEqI)VwWl!BzFLksR`b!^&`9YgJZ9>!qr+`?AxL!;iC+fY-JJLh1%64KH@PH+H5J<K
z|Ju;-Z~r`sb-LZIactSnKM}Aq86rJgU$n?eTlV4-iLduNco8n{yu(?DPzZ&a&W7E8
z{SUsuU5{yb>1}pYs5<sHPlt%z|CY|7-6mMMi!)s$eYB@p?UUz%F_8Wge-=%?28Equ
ze2=8yZx>aI82a_94vI0z_~8$ASUHCN*UaD5ZGS}?jvN2?0KzNnsmu#V1Q@x9Q=ZnJ
zKMara9{Rv9QWi!NrMaKhO*PYDdai~9{|{8@xo_|{dTXV93(#AaUVswhm^pKXJ(LH+
zl{1Bu1Z`_}0<<FL)^C`??TjSP6oftfq=G4oU!+<38cI?ck;|r?u>@BGWQbJg;tr>{
zEeWe>X1RPUB~vnWRk85Kj8!g|&tSd(wTP`5L*gpc@C-G+1$TUV54?NPzgc!BHbVI`
zzha8O>_-uQ1~#ZiCC5M>iL^&1Y<7!w?Pe^&r^cC6xJ-JM2b`yR+1!p%TifydlI>Q|
z2KTE4&YhR`HbTEt2nY(G$z7yfL?$9<)0A4^!%Gwt^V6B??C(o{l$u${bycN)D=9%X
z>ZT;=jGPf|DXd3Ss7SWV;vN4qloTj_^A+I4O^sqhzLq)b`v=e(=0Iee%&U+Ynshhl
zvL7xJLzkVhHy2)O#Qi341$~k%MLr_B^#JpS{Zx1;%adNp;lg|AG`}8}XWU%;?IRvW
zFV+*M@%^O?cN98l9Z)Xp5Cg*b{UMSs15Q65)l<Zu0gX35T#WDrm(X-ZMnxIM`d@B`
zHJUb=LUlj?o&YBxUuSD7ca^=nAAV+Q_l@QN+)PMICLdujD{tDWg^+Xhni=?(`Cn{J
zr$Bz%1H_8J+J#3qu8NL|DWoW6lZF-^j<NV(Y<N;3=<(<F{*N~!t5jb&DWmDJs}xt^
zI#iIyLgG^-%|F#)<XbMsfzw@2=!urfXnLekfp2;bWsImMPOw!}16hHwW&6(8;Fl@l
z*xeLbzMZ691uow$`pYrjiVin306Cr5qvS*B=Th!|xcxI)qZsyt3yE<iHn#ZziHdI<
zt8|$(o@?1Op-Z?q@9_su&l+C5#nrXBnWd$txNknGJ$jT6$pQNy_%%Pr8C##ofJA0Q
zDO9|Fzc-$fr19RcbtE^;vyATI@xKA+U27YkV^oeBBSaj@kd-IpRUb&$+ubEZr?w_A
zG|XB5OV7fhpdU71d$!w<s_cSijeGHW7xi7;{d&#kX;40&og0IxS*K*qy*`Zog>8c)
ziaR*eax~)NDZ5KON+9?UQHd7ccEP15sVFmZx+Q&!bdC~u_u*K~oti(eM5SJLtr;q<
zC`}V3)>F`9Th>G>(JNnJ^gBAV%>=^}F6n#<8~QGxRKg?G_Fnri{bn!yZ0Pw()qY!Y
zBwD?r$$-9`ZxVFr_A0TFku){m!4r-GB+R#Ur_z_EUgtLn35^795=u*PAdPx_BG)E{
zA7!ZP?EM?4WY5KNB*WW=yr5fw%akYVSxsEBUY}}A<hcyhvnQ<bb*{9(X1iH5hF;3B
zMEL<M8BupQhcjSUzuU=HVc2{nlh0~HAFdQRu-armA~ktjv34bE3T!)jhmU3sw=4<N
zTz$;-X8|)$_Ou^&>iJv_R%!KVZfXj&LA}}x=t7W<R`=y960kTh!0-BpjzE?VRDi?#
z!Uh^|qW<?*pnvA(Q^@9)DLXZ8h6uWU`;mEWpp2OO0|L~Hj54UQRa*#s0MHK(56>ZF
zRMhMYb`qp7VVPs{EEn!&%sh6I(rEPcRw9?nq?R>+2le$_{ct)M7Zjkwh*NaduK9d-
zb?RF|ah;;I_pd@Oh<pMvBg^Ckdt}T1!`4+mWtDDifsgKxZYco;rMp8Aqy+_~r39oE
z>6DTZmF`j`#Q>yHkp?M|77#%|y8nF`?|-i|vzBW{$6-d!ci#Q({p=?g2|6&5%Ua_@
zW9s`KGDkZ>H^AAXQR{J?TG{+05I_3hq@ge(3a(`6fXev8#6*T+L#B3)a$r~#?2?ff
zk4}v|yT0MrU_!NAbo{dbGODZk&Zl{zEQrf%0j=@f*WkT$F@~7FVw<SQH2U=eJ+LzY
zOX0;4+1h>bCtLkuqo*HbevqB7C7191-7T);B=Q=??|h9kEOT2`$G2DZq;R7Zw25Ut
z#UXra?==ut6DZGJ%dLO8_bP8iYvTSoArf@7xHsFia7mu%S_<8pdB#oTp7Jk24ZOW$
z^FuWD<{>K4@C=;}|8`fL(ibF(<H^?-_CkOattW>Q50vAieGicAzceGX3Tsx4oIAwm
z#Yi`$uZjOx&AU|h)`jya7#W7Kx@8l+TFJZH*R#xMA2T0LyNJq23B}S$YsdUJA@caE
zd)A~MeB2Cit>nYPef+C6cdJjYuCEJ3b-lH|F!ABThv(k!0OSf)+Lxgu{Q<IgC?(y6
zyc3XKp+E30v^Cvmb_!~>vK8byeg~JH>YaB_mUV~Y9-C1^(j8o(9sFt}-lo0St}H;c
z3uHiBIka!pY#Q^5ox6r^w@2@M4z~(?|4yc{|NZNWGi6KLJXLn{?jz#-WhS<ZLmaBr
zd)gOpyu{I7zC(m<X;<bT;G+<%K|KKbaYcau-eJo$=S_ub|M>u2DyF{u$RmRm@3OOb
z?@H8rGq<cRBhRUVB>dq+Dxx}d{qfvO<c+!|dT%SIM+}ohhVvvD_IGx0!sd5k-{r9A
z&QsQc6gd#b-_V2?%NbUM+*%C>W?Fx}er6wHS*k2Q#yj*kbz^lu7(h=_qTiu053)^{
zDG9Br5^Qu1CX8>DhZ>AILvkcx7I5}x|1>^uQgWeziD19fW<(}3HJ?VPwv2mH&ZN2t
z+ifx$N?P=2MWuA&Hr>vo-I;@8wye;Mqj$-FD5kc}v>c;c3r0A^U)wOAq##5nM*>eg
zFE>{!2#XN4ViTW`qYjl9b8}^4!Z^vn77PG#08Y9UIi!MsQxMhcdRX$w&--L9Y5PWu
zEq?OMIZ(OdH4ByH*m@`uXDl@K!4IK_2hbOPhdrxs)Wx@R+Y!7Ald4(?u|YLp?jS)M
zDF`+}H(lC?$$tUSjO}R*ibM+`7#G?K9}Q7K=;cGq7sgj6t>y6!?+D%?aewVA(z?me
zr%Jr2f3A?+olEhcCibRd^u{%luX<3{s<BVD4bZQ&{4gOxGV9EmLodl&%wsR{Ht{25
zZIIBJQCqonHWBkqn{t0>cnW1r3CRj+FRgzHm2S=3tiQ~8N`AV|MLpns<6XjL*qz%#
z;!db`llLu3GHW?J^+qJm2dq4z3SbdLiOI|RIn1q_Ei=CxtveMhiuPajevsnnUvLUl
z1xVEju!|=fTsIrBnR=pV1n3LVlr(Uo=tN??Dlr=?aaty><d{`n>1*b^a+=BOu=q&O
z%DL<(b48a4#D#MyX2J#<*|zr%dNnrK!d8?_!_%70>4}Ppin_tabUfuKYL=&tx0~`c
zQo==w^W=e+c@}7a++sX8aDyxW7b6kKH=P8fvUNuE^GkPvA|Yy0g<qVM4$pfDAQ#H=
zfJ@Lpv{q<ABOlV7TTPn*Fx@_467*3VfN+~vs#w1#50<`V-1|pR!5E-|iT+i=1o#?q
zRaJVzualKozV5a+JeuG=T~%i7c2%wbTuseO?ZH7NKACZ<VH}*NLN@a)IpTt0l*f~I
zWxdoS9xYW^#O@Uziy<)7H&*zKMotE~BV?~2=-%GZ)T@0G1l?1$hg*0cTM56JE63iK
zs$(S>W|#K%>}R}v+?0l@(|M0?kg2y*&hrYVmvKOl9gUKrqNT+LQY3Fo;8s_7J_{LY
zKpEDsb}4bkdluh)V$Dj|o?57447re-UQd#F_?1+VVcPKles}v<>~n)RpT#o?5#r{S
z^~nC(?#ydtd7ayBcNjxWMa7B3cilyvA66iHU<VLAjF6dNsg%$e&;rWLkXi@MC&|fn
zdZA9kPvK_dR8(Rf-j#<>j0?<-3yq-@S^y&&;Rx~?voK1BN{c;42@a0XV!fjH5}Zih
zU*fOtuFsA1<+8fWbzB0nMs_B%mqDU=u-6@QyF+@Cm2>9*($Y0ZAv2cL{OO!u^yqdz
zh&u>&y4+UR7$~4d-}%&Fyv!_2M|g|~E`7$JP<fRenLq^Xd<E0$j}+JB9`nvBtgC#^
zJh<z^JuXbU;eZaEYcA{uF3Fp&7{C@N?(!>P7AYx{`9~575KN&M`9z|j3lpr{9D8$K
z*tBxQ0mBhVElC(Y{`$-CZbwpJWt3g#%NbjFBH%kvp6>mSCLvOuB@e3g&MB8juz8Eo
z24M>O`zA)=VO6Ms?Adpbv1AWcJBs}sN)UhgBM_UsT+*S3g?)=PX+kh2S`YD6q&+tV
z_TQ;HtA#>xKkLPA9oNysJgD;_29`QM_m5;kS71)2x;y=pcf^_-Ri?!mmY*~*=O#>L
zEZVPkQa`)@OZc9&KSvn$m#fry2-9?_^`+nOQM?!suvHLb+VU+~#@Co5VpotG-3VH}
z)?#dW<_mnTx&i2e_)TT}^!P0gFohTEd=JG>^W8A!HChurL7m6Aj`%a>kU_aj7I|?g
z-$q*!k|8f{cu2?)us5ur`W%8*s}J~fQPa=sAjvfgyzrj*Of2ECU|%sLM<lv}00Q+E
zaf6+Cc|&Mkw@`0WI0`KU;ij8S8VCbBokxx(MICocU9DjG{R~(<tq(CJ{B#frJdNT`
z>`(#B8ge5dcIK<IgXgWwtCGy+(1gOEn?ZccE8NR3C2+K?v>>4nNaP9(CU}APymQQ_
zVLoNw?t4S%_=$g2VjkQBkm%Yd%DSW3gex|{Z&$YEy;cnboE*_+I*-7M(0Dg<RSq3=
z?ID0*p%zQ!ZY&OZKC0?S%v79l?B9lEO5FR$Urt{;9C9CdD_!VgyZD6`J+lH$xs?%)
z`AKP<%=x(3v7{d=f<dyx2@&fIqjlb|^y#>_UIoV6vGH5z^Dal#KJyy7@A55k_l&*P
zQ}es0o#+SyPcg}wH)|F(EFL%*^W<;w>9_HxOT}qCy%+*}UuG}vNsLeCbc6Rr(`5WQ
zF@M^_k?OFzkW%e5Azhh{3(<HgB93F=TQ5_&>6{H?bPQv=07LL7@tvo1d~`azFuX&>
zw-e`WU2*#QmOWew*5A>fbY+(0;}z_CX@7OPo=PQ1tXYHd?rZ0%K+|yEK_dKn<!i4?
z0;1DLc8qvesjzQY{F(?WF+=vIz6V|(%}HFC8=LWh>Dib_*=!OU=3#99s*M}5ar9$~
zMX%)fW?1Rz?-aD!si(X@+=HlqZ`~Oa-*0;^4aXr-7)E?_hQ2^niT{4hP~dG%CK4l#
zc$3Dvw%`KXhHRr!IxU{qIGS|s4)1ut0wq+**yHw#QaH@)lw_46qU*aJ1}1ZDpJw#>
z$A&*Kd^;(&=h5xHiYD6E77n?iQaAl3MoA%ZRMCs+oO@4HtJU(zRjUA>8!8}qZgse7
z%rZNPYtOnR-+?{Wf*!fC{tnf$*Sf&0c^lQSjQwP56SM_-03`R@jD13~XD`YsJ*ps6
zy%#ShaddNF3KE?o!g)W5Gly{VnFtcUw9*dik>%FrdMICFS3pzZr!u>F55`?g(wi!U
z>x{dwMNtaz-aSSP;DZlweds!XfUhu+DTm>cIO|lcQ3OsD?3|o*-$u$Uu#sRIE=@mI
zXtW^F(p=aZ&0`B9FA$a4K`OuvH-o*1Cjo{1^!7a_j-k5jj|_~#81mhJa(QKCJ5foF
zHS#hm@mD^u)GE*{_7^86lV&<GXWIU*u$+<x5I)c{-gcK9t<eMhc|@Jxu`jnyZuoMD
zwtPQI_-D-}{8ue+ED6Z54D!liWgkNbiJU)i9VRAbFSNhxW@Riik<8#yraDKeK6$LP
zgeJnx&CPQcuDN(~yxONe8w<V9YxAjiPc>%r0(6ysiE_ol+#E&fAB_4Qlm{pc7C?f-
z<|e6faK*beCEotR_8%_*EV%J<0Xew#`Kk|xD^{(YXiMwCIU15O7Dea`iom!$5KWj)
z^T`qDtVMWU*z{&=3JpV?WX|NHF9S~kX&#@|4c*y252lukZ%pdzVOoS}wbV{4^+YZ?
z@GNB%av(gAET1PWDg&q!tT-$6Dr^p5<GavIEJLBuuOj~e4Op+Y(CWX2{ihuzRp|Xa
zTW<%1#SJU@V<tfi#e-IQgx7emV%Ds33*~Ch>2p}yCdlcP4~=OJzIa|i0X}vu3}M+B
z!o3rc6TniV)DVcCosSvvrJ<q8ztrRdNh%MdWUiK6(-$=PsOs8XPWujH!ASInlajh?
z4wHoM$w+h!r_g;y>)ibw;(1c%!DFB6_gYqdAWS)<U?ns!r_ZAeVNE+Gb3meVDi7xg
zrk6C+Xu<xBMMTflo*_k8>1l-`=|M-@Q`aPzzPs+ql|DbS%J%K-q5JfCWL5L@HCwya
zR;fFtl;qV1vP^dC$Js>A4ZkLuCy6FVb8>QuKsM;MKBrNWu?jV~7d$+<uidze^A<cc
zs$9OQ^&aHM)uw-QFEO(L<Amk0H~swW3&|5^@G5-aifR20gkH^1R}T=-??FTDw(tJ5
z#r*yF__(r_GY0w~3IF~4mS5_)91-7?{x#{*-v?R7Fo`6hpTG|YFe&;S_d#T24!EF@
z&@D0>UdaR>Q{?{UzWO3{GbFbE98T)n)!6Zy2d`MMyDkT8-ZqKAddX)&8|_V^mDsSe
zHiHBYSbuqbwUiO@Nf#}%6wW!n2A&%&*r5pQxU|;6|3resv1>=?CN};uQb)zz7BU}J
z?Cb+UCmWVxB;JCtcuS2>MShhWGGzugmKWfPJz3A@5OPYIcKz0V63r`Vw929`y)aU8
zT?FPPkCo+G+|Ix&4jas;!JOdSwb#yBdKLO>4yrnD-E7T+oUfxZHO<zGd85<ac|vZ-
z1p%mlVI)I9`#FPdRw~{e(`Xh6q>D9O3y{a^B;8qh?^-pVTV)e3fW5J~N>XevxYg}?
zYYWZ&O?m^E1Qy<`_cdZLvuJ@?0yp$gbwsR{es;~ICq^x$MXx{+39zW2Gv)P(-GE{J
z3W?j>dz;+l7wiGNn05OG9+@61<F_}_q=i42)ZKe&>BB4ChV?2S;Kk|N8-gs@=_Dl`
z(X-7#s1s<N6z!!W{aYOD<VSG`wBG;I6Y{^R9Y)M))7+2y)cFuzX*7|v6??KVB()ln
z@Et)TaapIu$%JLLK}!2a@&Xu%G5g+}sPp!0RHzUY`cytOHPs{?V5mtMD7+!dn}mIb
zIg=8>Ge2FX-o}t8YLj#>FhM_S+a0{Uxv$CI^IiQ?Vp3eu*G8`4byiGQZF@@h9x;m|
zt&17@_KKJ_1%|pme=)=@>w6KWO$4QkzF07W-X#$pQc~x4?3K9pU<xe}&rf*p^A$kU
z%Duo=aQitXW>z}EB{v(ImQ~hs9{=0|hJ*uRh*R#0U*6z>HwJEW!US`vMbvZ6LXch4
zU?L4zyU4#{?a4U2Up=1sj<fp^wfDXw`l%Iq?-lnq8zC+*;AEEjtAk*^V78^WZl*WR
z53~8Jg4gdKO$9Xq!LjVqr<(VTE|qBS|9`~l!*-jcbqO-{Ym9V`q3{0q8MTOs_6(=Y
zlY4wsLzUle^&&7gTC$0OZQotJ+Y#0U!|D}ipUD+jau-KmIWes%zc!rIScJ_%KbSza
z#z#*BWUB~#J$A|`v3Rw+Qm{h06r|bg0#ML7LYj*${OwCG5s;o^wxPl3vcD4MECE8F
zs@0657mIfgo4gBpI9Oml8F{gW)(8z)<=y9(EW4Q<1{~BU=hKJ@M#&BD2S|N}L0y&7
z36@SsRQvN`T&8=-UhzR)!gX@yS%7Zr)5&LhiGBl`xvbMR(a*4Za^)IF1A@}ql}(Y<
zspEbPj7Z_~hGxj$Ks==LKgMr^>Xd4*qBc};+b@pCu+o}vb7@4k;=E>0Q*y5^Ds3l|
z!CO^eo~0OEwgR&pAR?i&i-(g@+$|h#1qWd!zxUHCDobdi;|=C+yk76IiAzmtIE+(h
zaZ;XrArGOix+pR?M`(M(1H(i53^z>ts<V6uN|1jBZs`!_4SQNAuj6)}(jLSR@BWS&
zw=vL8MKjK^qP-w<(fW}L^4V}Cevf(1{Tx%V#T4dsbU&&e{RIzaM}1*NAE~AIq9gMw
z&_5mlN7-)w_yV?ao6lo)B~X3ttVG`KfkdRe4_ab0<K0jA9XtE?bbpA8BaKHfSz?Cd
z$LjZA-<_%Wz_&89E-fr!VEW}Tpi#O2iG&Xo0w_vGPe&~cq6I#xL&iprbG0kZkdfBB
z6nL%^Mma|cPIuUB^71-R4`a2ZR-%s1R{*Hm%-WLe4#uAsBbo_uAZ?(=he$LMe-Tm+
z{$bC}j?25dd9h^V9FWu0#aY6V;GDW>_9j8v$H^8+bXUwWC>y#z9b@Cv1bdIt#o_%u
zKq3*N(Vs&6QEjYf<Ifp|WRkmX`k7xStrSS)T~feT9?ZOd*G0U>ZU7tIgA~cEs)PNq
z!DB1FQ}>zA0v^5tOaOws{QPUY-Pli|Ix{~-IF^B7Xj=}}8ihJ}c!=C9f{O1veY@Yz
zDhsd-)K{}Rn#cxGKZ&<LB_TDN6Hjclwi0ax#`@b&iE4h$0S}}thCv-1$;yf)>CDT}
z!rnA=v(Wj9(hgI<fuY0TsxC0-9YWa+sE&@Kz*n1s2rHTAF?hbqp>jYf1UoABX@!dW
znX#BHh!zxFQADmvUI%Ae=GoZgEZN|E-1}QIVvRMA$%1z7DRSxa%sf-nwNdzndbhX8
zLn~k%HFBe(hcr@rHER!=d;Dz00~rk_z_vNPKAV@8-49;?-s~RR$BEh-Q`<O6fs}($
z>1+!dVEo&UL?C*MdJ%Sw57gGycFfJR0HdV`c2=EpGeQ2m!oOhJzkZI@15g+_heSeW
zqp?V(^^FtLa??ii!;;nsSiv=DRq=~ZauU;=bk%0a+pb5Ak*w=~QrFE^B`hd^sHv+9
zXk2`IE9-^RyWLQEd3m?64HwJ2a+{;Xv%10CH=(gHS{pQet@a^cKfhY2x4+;DlD3XR
z=lXM?h#c`q-2m-8?H!T!!kVc=Y)Y}2&6jW9SY|E2ZT^Jvy%bJiJb>$ZV&_mS*zR5U
z_rU-4N`_V6QUA7$pBs8wxR2^XokY)j)3v=^R};W-aG==346^L58QpCEzFEJS?NX=p
zg1y8?Wp;av*R3Jj44BT9mnMxa^CvH(d#!v~^$tjv9%5o-d{KYew-V!Qb^pfZ%UVx%
zju-H5l$V#cm^YgT486`L()GIC=^<Y0#CPkp&wlk})HkR&P1GaG{zLo+&HH>nQcJ45
zJFqN|er=!LnT1=B@LvGj3M6vZ!3GdpF8SYOzIkK)(*`WIM`jt6LmdmN)uh8al-H6Z
z9iDldT6m&>vVI<smlo&AP~4o}bl1hZw>B<IFW=^dcSH^W&Ijd8*ZHoNls1=8MBs_N
z(DAa7v2RD~jhNG$x%MydgFpb0YheNDN|Jf<>I(;P<`Vj8^Pq7s3mxIbGZnvKzcRwp
z2JO!|Yj7tchCG-i2k9GE%3{NF0;1!MD4LRka>oY21;rM`aYRC&u$CQc?toSzY_U9#
z|D^T!v>sTy$l=U$;SrYRL$)+>))9r7MKnQMt+J)%Otjl1hA-_NG549OJMHEf0xA-s
zu<0=HQJ$z8C9GqKp=lp==8<Ub9Kjmj6;d~IITYm-|E8O@_%Uo2(UANXLGdqArJubH
z5sNqn+k8@9f<|<=F=_k=O2~nLO%{)&-@(rqzl}i9U#AJ2OLw1m<#f}WRN6ks;2nnO
zzLOHZtHm-sE7J+RPU(bR7tOTHrP45Ryj`-q3v9M?;ev(<8JMPstQLR;jvFayQr<~@
z{rx3plg~Zm385dT=77C)95X=*Vc)UJDjL6s7N~JCr@~frb#%f*x^}dLn4VZ@6S@XW
zY_FpUsAt&DEy{v|4ACyD^;Lt{p_pMep4iJPlSXs(ZaW=P5#gG-ABM3vCd+SU>yL7p
zPT>67_3FhJ!q={(4Jo|0i=N0+u>U~Z@Y<ELW}nPHE!_vlT_n9b3#Yz$qRb2fnF-BL
zFI6!M3JQLCcL4M?#oHLL43P)?qib%=46d^ca6Y)^@bEUv@TH*d5~d!~7r@fL_g6>q
z&>Qev5K_ApPzkk?x>amF6yik~5T441)JzjN7!M|j)@z(l+pL){yV16@ANKwW)4stE
z^==>P`&S<YVfxZ~z0i9vdD*WY7E+@m)BLYZ00UJnpEH4YZf$A|g!1hJfx^$g99<0I
zUqLQrmDs6`Q5JN%t*^OC*Qe==o5!2Tu?_M#SOT^vRm4bQqfgN(c>s}@<aJyQ=~nSe
zuQl^$#o+1rl6lUwG~l^VbeZ-7toR?)(GAlk@{4-3Wl1*QvHeCe1W=&*lI?KMdfAj{
zs!u2WjF9~x4&!3SX9ggN@t(3eznWDGYmuj)%-u(?JJnfzp1|MEASTAb#56K_rPl_6
zaQt-1VgDAR0I3W%4vq-MC&9585Tvo0n#DLo{#SEt8OX}8ds3z?Msng<)8x*a=Oh6d
zD`0{K&zWi+#~_iUyx@j?putHUMqB~Ro$TgXUU~Vvgvqi|pe*k$5QqIud+fVnKkG15
z=WhQPdWaa_Jd3pXCgD%-$77d4_l%R2=7jgma#C+?c5Ia8&|QEAkiXA6*@pdMN9vJ)
z0?@o|f!x#d(6H_Wh0Ekop)HJfJxItdWh=>yJ@sROTbdD3Q4~beG;K4qr~yQ=vzW?&
z%zQU`cy@s}xM*>K81`VeZU<u+YF;)Ov5yX<^snMezj4uWP0Lj1?~%nTnN(RCZe<7@
zgl<B#3G{9iJ+8-)B6&vCE!JhRvFkJhJO>L|8pwTB0#+)@TfSNJP$-i23EuhD<8>ef
zkVfua!vnbi=lB^B0(P~;On}cX)ttT{XYc3d_Z9#qX2aWtk1JmQRlE=(8Uf<)>=`U7
zIDLS|VBPxV`3nLXeK}eC42)v`P|bgN7yT1(@7KV)_~lGSY;$akRpBjs7zXc-YBAtg
z{_Zv`HR9I!c9@q6wQ~?MSy!oXkP!Gt>S^V5J>q$c<^Axygxch62)3Z(8jX@9R?61P
zn<+YiWhOL$_XPRd;jA7<g3SmHI(FA$;;UC8C$=t-u8<;WhA^d?ruKkeFY$~=UaB^r
z>lhokwx{2Xki)CFrO1MPqo6E2NYAu95JpM-ws*Xe*~eyQ@DGKKtsk}l6WYLg*cxLR
zC<I#0WmRFG!Um>%_vhq+yzOw&n*Zt$V%9FsHVR7u^M}k`ctAD71W%l<sl^?9>{2Q4
zGgkt>XD(2Dq^_qS8#9IKFm%#nw!f-zBwO0d`QOzzcH`E|B?eiHVzwR_sR~}0<Y9fh
zaTmx_YPHIh!Zx;@3zq^ldEG{f^?53r%%GFc1?Gs((~C*<_23SbVptnNjNQ&K?z3$&
z<sJ2M*{cGU8#?P_h7w^PPEu#p;90k2>CC7|w37lx%qx@(@SHY$sk^trQyn9Rh*E<R
zTEGExSEIh<sZ;e<yr6WM)GUHmwuuViI9h#EW%COy58nVVnZ5|iN>26;dNqz%Ed(Ho
zQWy0Qu(PxGg38KP-~OP!VRf=hKO`0xB`D9lC>bfvh9^VJc*`bt=*4orXV-jTQCE?=
zp$$v1$;iWByR&Y$!{}{H>?cT(C~JqC=Cla^p|oy`HY6-e7Z^KS;I1g2gt)hC@$jOo
zsX-ftSrY(#Bc1Xs6RNX1!HBBEy_>3*c>INlM2=`V9i`sfCjS=(>QK1w@eT!|`fpY;
zD8g0-^F7g?=n?#v9L)H}`MaJ19qR^Uymab^LKp#bX%Dt_Nvf_O@A4>T4C%kT%d#Hm
z6OPgSxME`+>65wC@E+r`6{Dfy{>JAX;JR9GfR|k`Mm+e?-m!j}TohS%%NIfun>eDa
z#iL`yAITmZ3et-2v$0FvNq=?6USc>iSw;hwxcWPD%}e()OgN$mr@X%y*4-2GS#88d
z3?*uw&1pDG_iO-Tj*Kp6gnWk|q$B;|Qsk=~FLMG2zwYoS)E)M|Dh+PDz(h8yU~O&&
zRz`tl0fxzXsemanNZm_B7Kq>DqU&r9py@<H6pC1^BmEG2Q<It`$O5$Qjg3*l9-eG8
zE`*Oj_#u7h?yk|ddb(S{m<b2v&h<30W}-^WPPU3W+lV^K<GFQ&E`WL=n+6a@AMC&&
z?&crc?2nfHVaIs^oGO(1FP}k3TDQr8NlO?|B$J*~#~B^n_5AFuE9s>w!8$K5zT2Wu
zq7mU}khesFIgp8RW1~oNV%<>mJ7n7yfD7QLqRxYE$6GIGq%QT?LWCFP(CymLk?Pwi
z`mVl^4KG(4Ayx44wBh|IDQ?6vevxC~h(T6{MC{mLKwM<6d`1LmC<;Rmme%Jc>_cZp
zG_m8A>;3A|QAkn}Nj-)qTiGk4ai`g!8z9kV5yBT+75!YnVNP7vG4V8D=wq!PvTwU(
z5cyVCmK1r@x%yZH%>@9F8;&Bt6jYm>?&PdGTwp*rg`P~g_HGpI9$$FgyS7XZXeeM5
zX?_dTW6tm);z(THJigXH>e<rLqPY0w72oZNcZYsK{BO+fpn1^Wf!jx0`$n~snRhU<
znMpjDI9DpM`}G%K>LIGtGE)}^g8?DT818E3hoMt}B&G-dGIjpiCerVL3ypFwFV-5^
ze+@5JwG>{AODQbmGOxkWMXLo<bra_&P5W#}t~?SSIZPyU?p$XfC1s1X3Ng+oC(=AD
zlfTFL7aW>d0&O>hzFnuAaxkO&<G4Z#Jjf?!{-VtAi`Z*kwy;P8e`XNjzE3rLw)YHH
za*Ut~QPX;5;OEq<Acm7`C7GkWuGt?#*N*@6zk^}bv)6p@Q!_Q)$$@}JO29l+dXw*<
zyKM0FONjky2}?%+wnhnm0p*W$Qx4RhQYM#ZhDP5xf4Q%PK#-J*>=0!6GwLUa_e_!g
zGfH3?p1tk^wU&~|&`Q1Rf;lXW0&zmW8yY2s*NK-c1e^;X_CXTI5s`pi6jkVbn+!st
zCBm3rlM0Uy##9+>P7I~RT!M!MD(V92sE`1G3Q_OR;Ef$&Db86|u!s;~z#6#n+YcCm
zN{rw16^*kG+v9i>$Kg5Tc7$?(us*ik+*4A<6BwcbKCmL8prX1}&gdJn^OQc5oOhCm
zkFWIkMW~8mK5+Du{a!glRUqP1P#qaayHdqSTg)xaRpF9z^uQPvwfAAop=TW$-jRdt
zHml8mx;twu*L7F8Q>B4-K4n@aUHWFiMSy=uxKCC#qM<XGf>!VlWhL%$q*Qz_XNN;P
z=>3DB!02hfXF>RKEE*}c+Q{3#8Dl1W{ROgrb1?omUcg&-nf%$+2MTrvGHp+%Z>F;@
zwAeG;s4ow*EYbo~I*@=S>A6&U9UZmY@V{$<IL|qJ-%{S$+tYzGZPe5QV3v&N5eCyA
zHRrXxDYs>fqd_PL6wIc!B^G<WY60i7a7EY=f;gEt<qc|K2kNi2wl(wdNK7IGkQPE~
zYt(5Y_aQ*yS&~q1C?Q?&%VlWkE2u|Z8&{N#D9mAoK~m+ti)Cp@oW<a<y}pk}{}X0Q
zdgX9DLDk8{G|N$6-gMS%^6%E!GcnYH;SF==e+V)Z`j*ve4rGyhes(SPz{hpOIIIt|
zYKI{%1betG5HoW5j&rN`Xnr=X`PlCM1^lzC+YeyMP5?(tL*X3Jai?3HMETj+z|F`7
zB0tTy&l6>$N}#&M{6gVFe?0#B&KdXR(a-$tMMJfI$KWtR^ZBFklZUpGcMv&S*;H=<
z0{RYL!VcWRwOMnO-5<Ds{!hq-grB2q=dtLu`k^hK+cVd4SH`{G@NRm+)}k*s79Bn2
zKgf%4sqmvYjD$+-&TIhg#af%+Xhid*J|(=n<2(+6nX|Vv>9em=*B2+Jyt|lmB!1`W
zn)J7i%S!9Yp*l~q|5{K&km0J?9ZK!j5H;ZJED6@dt2zx~sUxZtfVl^5qi-2>FrJzv
z7!!7~alN@?DtHW2z~f;$*DU#vLh$csB4(;Z(mF$egFg&12n&BB+=z-IKopIW0i6a{
zI_JdSTg)gtC@Lyy!JgP{kY?hVJ-)q#dJ%@12G&iyURn?r1xFfpY^kWbDymKr(cCSy
zYVz`5rM%<p#yFCh4S98NiE)%$2RA=;5S?1SJwk^UX;%IkXPx(2RiUZuRL9*R&sbuj
ziiY4s&ovk=aLw`e4qOhbgQ$G5cHc}frCS`Of#c)B*Y)*gY`l?YSU1%%g)QFz-^(<$
zDU5XF%OzP^e?5<nUfvlOKH?FQZTLAk40PEm|K#*b2TjQr=EpvdVbyVFtLG3HpJvj#
zP7h^rb}IV^nK?18?*|?*^KSxDzYW6#^IQy0-sHu15mh0|RT};>_K-Hn>oq<tbf%Qy
zHm-x)NBe7(PUcS6l5S)_eRds}l<`)i-<p@#*I8V!gqJC;{e`zYT(00oyDK&fhHEy-
zZ&Yunmu2EGEm&L-x*;%1w)4~?4nFH-+(5Sj`q3Ij?aC)9EadD*2Ct=rI-{p{?wnv0
zc5uQvF<7vLI^w-I<cf~>lWgwD7!D+0PL`0mO?oPV41;4!{?ke(Ax8cB6o&DENee#;
z3#$)$+7QpM&^4VQwQ}v|<Zv0?;p-PmQ|@;9tc**Y<sF%3nT#dGWfeq558CwF;}!8>
z`YM?)>HrIr<99gbk{&9+gu2JsTHY>M=my+7_iUm7acyir?&*~s>0J7Sssn~Kz@tHd
z#oB25K*-)_{6{)SE=|>$vDIHBY;VHdY_E5y&R1t?VDp6G{F4aIGK&zXD>R>HmH?_8
z#n}=K-K_AzU2%%U)P(sWZS{QW@`YjTsJw=i4x`n8`H1Zf5N4{qaVfev+0yrV?3~G`
z42++`RMf=(5CQqmlNsicy9_e}sF?zw=2f6Sy&<MnJ}>c`dAFpSQHLfSE&`VAd*G2<
z$@x}}fUX9dJ+JWldj?IJ;t-Jxmu+L6+xu35AvAo`tuM29;Pt)C!_%2;$p7rM7~0lF
z7I#%gSd+G-2d9yC;3_*G!gnaZqdXZUJ*<OGXI&N$sy7VQS<IQtk7B%#DiE9S|F!yv
z=N8ZL=dZc<PHE6n@FME4Wt`QiOe8A`D}eWMn+#&X;#;Yu18L2eW-OFc#jt4x)lm5G
zsSM1jRGL}JsoqG^^0gd+mii5gZ3^H7@ApKytvwlGek64+g}kUwpkgB>qzPsxMP5|f
zv@0+Ad+59+Nm^)jNHUMYzE}o*2BiT$x?|E}wQyPnzVC5?@CA|bqfdvg%!l)7=i16-
zUY~G=ub$}3p1sNR*~Q`7^_v*u+L3664%rqCO_L|)(YuBp_LPsmVVLlakHTJD&hnG7
zRP7u{nN!Q=riH}tTgrEWq{8A7EptMJNW0OPrp-w!faZ($!}^BAdh*j&{$c#`sViWb
zzeFi6AQGmEfp!(~;NmHicm=e7(Z`&2{lWn?#C<mV<z7gKMdc*fyJbqT<YN<p4j{UG
zV3~p`+9tiFXV^;k5{@N|G2p&*37}glU)ZA<d4+XpVtZZjSHeeFYCraVLB_8h$T4#Y
zBGT{_RaA7H%XY@8O5;YdP^FQC%>dSZ523ojALX*=4wy5hLFnoAaXbmVIJ|?pz-1{z
zdZYHUxb;*SGJewK(HIZY(()EC)+bl4)T%VIemP(uXxv1NO}RE~QuJc{OWk*lLhi4b
zus*~t=P#-hl{5|Pc5<MD(gB-eeDA5+{E<~3pmen?e#T)2jTD+s*-jK|SjT1C`caN>
ztSX+{$R6|r#>YT^_wHR_T&0}e;5QiegE1Z%krEk~$qCaDt71wwTt)hOOQo3+(H!{@
zJZMNbN6#~DSP-*14EUcACtv)xwrO-Tkg7FW-9K%MAN|L)<If2Sy0$FNF3dK96Nq$D
z3|Vs=z6p&D+L|Xh6C2iWsPesPX2)cdDNT0%g_OsGk{HP|39kl!wXW;@c&mic4l_rS
z{v~Xk<>7f4^I@7uQrOAK2_KdExI>-{V8gt`#Kh-n5M8PUv|^3Hy6>$ps-do<HSa(;
zAD8(M4J9Q<_;Z@-L)g%eNOYEofKEe+#Ku$($6g+q5=?2{=V6oL=s)+wCdIRDPMVOv
zb+QheWzOpKx5qsDw*n24yq@VAEsh&A%-?#xbg?49KOd)XlL0&dj1Q<`9E>t}VSEV@
zVgN$>_O-~CP&YQC%;ARa;!r6QM5l~g#Z`NA8uau!w!^8qi5s&;(=&D;@dQp6eu{sD
zNDS5q6m$kPypHb8l7v?)eJAPa64G@e>-<rQSB;@`c1AKw!{tGU%g*T47<v%;cS8Kp
zMFw1CN5#w;kb-+w&}I);DT*&ek2v!|P>U4VM=|ArKLP4=MFbyB7sRIiD+<zo9>J=&
z=IP~$Ml%$mWPQ;ueWcmAI%|Ren11JFvTr>-Zl5Tmt0K{3BYtV79TqFgomXQZJ}Hsy
z`T#Uy`0_Z2-Lu077#S@FmCb>V2?vEUM-wjRP~a=Up6Puj_$3qpX<($-+Pe>gDs0zl
z--JV8A)W^z@Yr+SIAq%%LAx095wa}Vl7B9J0Q?A#n79Dq5Xd5}VI})?{O&vh1OsPO
zru1F**`e?0(96m}zUx1KI9@iLnp3a3FrDOZ=KeERvBmM{!48`_<|M0bzYonL$*1>C
zU4W+}Xg|m+?y>Yl?Vi;;I#4;EoTKfZU-ey2_1}LOSBwg%in=A}GH(0UYvuu2C(TRk
zVrRx=8ScbiyA8T;NvX$>cikU102nJ;@Z}oj$DZCWQUZNv`k0me^>p+$#C<&&Bmvh2
z#41F?Rml>{Ugy1+qL7EZwug;|^t{G$ri&5Oqu$)waryJn1bM}IKL0#M1t-U-pbNY0
zuMcvP=HwXN{e6tIk{S{Q3aTWy<v15(WefP&wcB*V#SH~8TG3q)tt74Lw+J5BLi!2P
zlBYDgzZ}M1YnKHGIs{;1zT5vPu?WjZv#S2m5yvW67PNTK=bQaDCQVU}9OvZZ)CIKo
zJ7`R^zr%7I93LzoGgul_iD2oZR(&6)mrR%!xuITG3*J)B&Cl0?nf^N{j+H+X(FsQ(
z9wrp?mqd2xS93?&4FAai9YB1$?b2vAPJFxnQg}NaQFOM3q3)7T!kRK4>M_q~bza*q
za2ErdIiE!#R0}AW=d4uTVv{|ozL?~Z_WIe*hxNHFR%vI0w>CFgzbFSG??t8%I$Wy=
z4P4zLrA`|*);brzjs^oOUkm>E_X)pN!ay&nHJ~J1yPo!ZclZo8zItv!L1c9^BlGYR
zwBH}lq_R9-$Q2nmHPKEyeB4Z2j3AQ>m#hAibTXc-92kXgsd-QO3IF#eN2=iWk!o;9
z>GWsxjwkRg#;O*am(Cs7xIE;ptZ<I%*C(Zt9~38g{ck$EANxGTyu+-EZWKv*dVf(6
zvjC5_@jNg_h!_<vtHeG!Zgy?)NZ@<7lYPgGbIjz8ixP<u&~B}(M<F6i9%MXss=b=S
z5%~vfNf64cKosQ*9{plPgnqOK!W=o47$qh{egxcUu@E6BeYeMb2$q;Y6?a!w#zGR8
z)}~v}GBYz@e|wG;ya+HLl7jy_>0Av67#-Hq$pX63t~|AkBL4?t99U=XM(#flRvL>h
ziBynyoa5N3_Q61g8w>3U=U+3w)>^_xzK&P7zbJ!|Ko%lifmOzI{CLG%H?UbRgd?&6
zgb~P>66IQnf6ybF^J9A~v-LV(O>RK~mVRE_XAF}*VOmUN3td7wsd0a+x*?sR1>I>J
zyia4Uy>dL4<e+a1ro@m&5AEthw)K_a^5!gIHlFG;0fZFee#br^;|HlhVx^3Zd_<zE
zzYkt-<F)5JUDG!D>igFo)4cgx{a@n99E?_|!43}mU1&1V5DS|IHlqbrNS;0=e-X#<
z67!)g$MO)axS}Gb*1&Gq2Alr<iQ)D|NsjA{2jc^3oQpkzY_Y{jYwV@;-D0<wc48I;
z6FshO3|@EY1v8k=Bp8at5xCcZ(VHCnST*1U%?%C?cF-?dZL}|~OFDOhiIkC16V74(
zg!Q3yFK)He=KA`l2!UGF%a?OOTto7zg!ngy@`qEUZ#{63s~~w=t|uF$%1CN_F|a=k
z-kzfZ7pyGgL%&Yy`1s5AfIobkX?LUO{`qlgpeP;QDyD3I;+*Awt}eDXN&$y7*>$6}
z?)Q}^_$#&Cb0It&2Wf#rCcrK(c7u9(Ar?k^_F7But}4o8?yMIhLPK?shKnEjOB>!-
zCgp3)KU|m)4!_~8Jz807>-(E=OM=A50WE;@0V@chkQLZ8;V}&R>T?1bJ5T{^0Jy8N
z`^;_Lh*Q)9i%tkN&U9c@&mDafQ3^I>lTz}$X;i-u&N(1{5$tD(knhKm-T4elNK-+E
zmSq|PC7CQa7S<<mM!mm@ZC05O29vq(?~na2JMH~Sd=~8GZWiTCY4q;&Lv`%#vO)z`
zr8eozT+Tl4+QlDrKkm(@aB{vZGzm`zG0~<ACRf2#qA>BGm2jddSGB2*j?0K>i+K(d
zl5r)P9!aENSSKeZH`14SdJT{`U6@563TV8r9WgxttP?YgK^gaL%1z(h5a@m#Xf-gf
z2uOmdV4Yu)N23hIGz<wFXJ>AVc@KOC(<dRC>5`Y+7YA6ZbBU1<HK6O`y`t~VCPB(E
z{Odm|RgdQOkkm#fbf4P$!RZd1c<l%4IB!h4Hjv+9usG?@|7bI{=n$UZZRH*&lCtQ(
z)_aZBcl9)W4h-l`VTyUzSS!uKh6kVT2S8r(sn!~kogo*k+Hu0jkO~6|4GLOXHCV+_
z_<VQLNc%AJ6g9n4(0l!6TCU2+cBi=$`Z{58adJ>Y`Yy_J0M64jwlXl3rUn0|))(vf
zzF<jGwsy1JTnYREV^_&ZNSJYi*x1=|Q+|MBW7Vy0#%Ya1zvIIphdN%e#%#UXEiHxN
z8c)8u)h7IS6Ekfd*KDeI%(#<&K$Dr@N{L`u;?Y8IGVj$3PL^Ey=a%io&ba*1pF$P5
zjI=^T!~~~*|Av1AZrKdM_>gSRc{FL}_VHFycr#{MVxR8vF^GBJL@rp6e)rU=j1X6T
z>aVD_?@aj;nLKoGWgtr#Q>Ojf?V3W2cV~*{X-w|~fO~!)_#K?b5rNm|-Thr;DK${<
z=Qkp84=i=e=`eNQ(}3|{HUNbEg@?1VmgaG1s%eA>pqU^~4KX@X-0%&tV7NCEEwTU$
zsDh6PoOA~4#<0vuh67&ud<U4$BwtE?f}k7`(gG@$2OSln{40c0C@%vPuG_!1wz{}2
z?lsO>ytn4PK}t=n3XLjTHT&8ua`w}Gq8R>p_Rpc_oZc#4-<pUok#O=nt74Lz6kFn}
zM}VHL`J3_{dSwoj1(+e0zGCS~Ny^{9e{bRf>LWWB<Noe0E;c{BCM%&i?_-VSCcjr+
zzE*ZN##hSidhc9N#4L-nLsIwqN|8GhAu(<6Y{x=RIYo%r_9+f)U3Jl_<vRLVvtqM4
z*{CLLUtNOW!$8$zTjvD}0sd>T9sVb@!`KH6(ZLjW8NPvT9tjm;Kv7QSFDpi92!FiL
zrvi=<_AJRd#^1Zd%l@tVXDpZ&6dA}un68R?_D)~knQfP=o$mt;znwm}>RmTh;z87X
zm8#JAUw1DG2w;9ur&KHxRGGWy2|2_-9;1hJ0e>(XwmAPRb_;R!f?hZ37cjjeth>{1
zQ*8ehOGQmX)&Z9&js^D3E}W%~98ixa>z5lUsHsIvZZvZ>PUh=BGpGW)@gQpBYpkZ<
zX)0&+T6GRpLD2jUvK2|$Wz(a!$f|Gt_vg<{{JfU_Hu&_bG16K17t11lpVt@P6<Odb
zFQC^mywg}0zL8zfJu73|X1BGj5~8COS8Hw}_ZQLD|1=5-v^#gLDiiV^uE%{@<6ZJG
z0~wkixLv*Han&J&PB;hjt2onF&Gjqf+JN}|Aew;$sgts{=(+|iJ@0fML<maU<cu}Q
zt14x^okjY$KXQfFDOB68m)h^r3SW5gV?c3o?4=4a^hJn1lHF#sDlD=uUXFY<<T|i(
zti-UWcV2aOT~(r3``Mi$yD7Wp7SK-Sz&!)KKNEwfvp=GT^FSdz1lZxweS6AphK`t>
z@x-Lve*`*IXe#1Qzo+8eYRD#rN6yS9D45ulE@kiP4$P7qpaqN=cO~FM5lpznWbx0(
ztc8yhK`V|PVv@3lzaQ$mUB$ETz2TxC8{bt-c>EqQeo!QuSG+<*jHs?f3oZ_zBX1ZZ
z&?E5x`&Nd5H2x8~x2550XO@=^BRYc~4kh;?cz6TeZ8b_nMOt#R{>Fc%HmZOhp1L-x
zQRInd>y?uyY3!OBid_gXHxb{I|M?MqqQ6$XN{adSw^<oA?I@MC{C@f@t9abcu`2><
z&C#N0&}(Qn+Z`dU6pIFHob?Ce$~0>Xy*OdlztT&1M8Qy+A(Rj91)~*K_|P+1wLZN;
z0PnQHb3-$V&T2w14(^CZ5piOSroO=XBb^ThYDKu5S)80^=09;I5Q-p`3)a@7TGM_;
zKnYM|^4ZWfXn`J74%mmRpi60-0+pyjArPL7zQ6Wg;7tFwcUlW0;4U3sA0M{kr@=>_
zZ<7kT%v!CUF@Y0B0boZp9ynrNdkGKLg8T_kKff!|a~zT=92@(xDJ&}<e!=Rs=Sxb;
zbJW8_o=Cc?zmz0!WDHivNyy1(^iVK;<pDI(2i}U8#9Y<9sqfh32<r_(3hrpRMHuXB
zQ$%#3O=j=^>&kE;eNw?1x|ODu`0QTJz4?a(dE9wECwq|Jo!?)BCNm$>Ftmdw_1`*)
zNk`O+_XRy$Gsh3a%T$E|t<c3au;n@e;FZSQ*EJ0=4qgdQdSff}9{!AlaO#!6-pazH
zw5NtzqW|2mmn*uJ9Pv?T@95}==zhdQb$GZv5vUa;T>*Mex2+|k!;)2R;8h_GhB>0%
zPLTBzyfr;Dqx5X`*RO2YPT|A|fLKcj8V>pTE0FyFrb&o3OFIM(X$hY>69f%}CXn2W
z+z~%^aO{->UdgA3^$q;{L|%L3{%ijfKNSWvj1f}V{*VJJu24^&D{09f=OT2CmVV9o
z8wvH6cA`Fw^Ey?wtskoVce2|O_um)!*lyfEr9rFrc^WHUmrltFd3=*BD8rs9$Qo*y
z#)ReYY2n@sDR1lkg}asMBHvO3ZRLITHnIS|F4_~Q0bB>Eh`kp8@PACiD7hf$bUHhl
znGC5Pia2FiGm|u&T<bUE6ggEwLQs9d0s3jiJjEwrr~j=$WhWR%(q6RNfB)HE6*ZRC
zdH)9YZ5hfd`#I!ie|l8t3Adz>6zYF`Xux+6Og2rVy?Ex11kd;<l^zZ>N=uuGcT%uy
zmHKjO_+2)v5TME77HvZe>zTt%lkapqxd1rA^XCg28*abW)>xY7o4_T|5iIc7(_<fv
zbVU-&9K?dQ6~a|v%Tom$CzzCLweQ*4af30%C6X6hI=QL_qhKIl0~{Zt?fcsKmuVU~
zLX@~kj?VadDtVFpt$ZTCY}kbPe{xSumRPod^f#=bVtv{QhHGB3VUHc*&Izvv>C&$E
z6)qI2GMYOz-t`%n@BPY(V|nrj2+35{)pNjtT<h{>0+U*=xZxRj7nh3Fb?{gL{ahY8
zCLReL3bMjuLEI_7WGlC#FNd%24#c)4H|mmiL>@klX7J(#Ksghbz49$`_zB_CU>Z;M
z<UoNyX(5XK5e{<OJYv4do-9xI$Kg4J4p|Q%@$XSG{#$kP_y39_W6baEn(+)dj9C|9
z{6UoZ#pIF7=~}--iq?#YGY7d+?^1+3nlbi`;$HU<1SBR>zyeg}_HbF_oHP}W!R4|M
z(;5Pw@R%4%2ilG8gZ)QWZ)CwEr%@j9@d4&*<55ru&E;lz1#Q9Ot%a3hFc?59EjR>c
zFZ6n?K=<-!04RmM+{3%7;IwS&*VH98EZj<tK7RUc6R$<o-@Xs{r5z{ptJvmn$8X=i
zakPGpK|91i>ja<%5vka0#mT>yla)Xny-8I}NO}+bB8`WqxRWE_w1Tyq(yvhyY`i}n
zfd442@xv=wjW<&>FJltWf<r-H0Y-5#1Mc1l>*oLe=Z=?<e$^3|E(rb4YU+G8VE<vC
z?a%{o&gi<~qxkgc`Q;dETEF|-tY5|V3z!#U4<<vc_o8Fe$Pt=65<G^`LcU+dIty?D
ztIzS_HnK4dGGwKcLu~t^r=|Co1opt;JbHSmZtCqD!V{4>Bx}C9oYOhJ?ggAAtHc6m
z(A=PxCG!?>0o4Ss_r%LxNr`w(8?%YP29UqFnZc>Wi#mop@O-GC^Pec>U*FR&lAU{p
zN_(3<x9@TfeHPEI(fp8gA(-A0BPv2F{rCMl!=!QQ6xuFsZoW*8k^T9e@dR8Ck--<f
z+O+-8vcG=QCi$a<t&%nUMSLO9lbe07UEWqdP_j8FnoP16>yOKH7&y6fVbxJ+cmEDZ
zH5i~jjL)!sAfDPQ7J7H@GMbhBb%~;;gFL+XtU-4?-6xv4HJ7zT8?~n>jJ~o{a_s=A
zzy>;Y#GnocPPxF8uzIw;ww48vF0}lnrwHkUd3nZH%>h#r^x1Q3=0Z}Ifs#bmI^ycO
zG<?CXCll>y1Xk4V&QbRfki{V1=(4S?t@1aBtmoq8?Eqn5KLDyomda5Q7)@X*!z?Lp
ztfq#?&cUGr9S;{jKR-`~gz=<FoJ0p}+g~@;|Nb@89o==%B=dII$_L@7xKh2$nCOU)
zf50=V8=T}b*l162jq4IF>l&oye0r6@V0X4)$W?GYjigwI7Q2%k{Xx_JzlrdZVs>XK
z>5!&+(JX<e`1(i<_fX#qqC#$1NSz>O|1Dgyi3op2Ng`79AH_I8)@4z5G{+T>oHR|f
zmnwnl&^Di-<n_%*fL-c-8Vz}{eUCe-W5pL^ql^?7|FGvE0JUEbg!w4}VMP+;O8vYc
zKymE>ZKQo-SO`-R-%Uq~TiqJ#bDg#m?|hV%5WX)M3X&+Ejw0h7Lu{21HFn}NxQ40_
z3mMVF&fhok-u0<%M3mZ-DOUxC3ZuZ22JNl1x@BMp{gWYx?+C|aUcyBZvrqyqvmaQ(
z)i;*EJl@DA-5QuzrK#m{31)2n<5d6Ez4^Q1fW}>nUK=0hb;R#iTK@`a6cde@x$H-a
zNpgv9wwi`#m$JR~@mQ6!o5rtBd*8}-j2`pw|9#qysweMtc{Z6s3V_hXC^<F_+(DZ^
zqYzt8<buyj!~b?2o<e(ri|}cD3dlly;uEMCb<{=$-lVui?qT51wv)LRf|$f*)~oAW
zTZr5fPQ>vQ+HCFW%K$m$1YlrWN4^p^2P3l<dHhhYynL9Qom~i7nxVavrNpo#1nf$<
zeSUda<M?2m_@tRhcY>h_5$U*r-Xuqm(oWQME=oIfSOflpwjkySEDJJxt7i=E3O0b`
z8hu1eV&MQGZjBMTi|8v6iLB8c|9ye}Gdsx2CZEa(I#^z1_)b8SA9XJey>6iJNzlto
zQ9$A;em5lpQ7nvbM>?KZm36{0KqE9nJ!c-n#>J})n9Op2yn|Y20)DKWr8sKEO~a<y
z^?uKoEX|;$e-zqt0uuhW|Mq7?+CDW8>XS{&A4iKMAgbW0+hmpIfZVn#>lT`nW<X((
z&;tJ4c0T1xJkk{meI8x^+FIsCQ$mq*bC=-i1<<b_6k3(;^}w7(l=Cp-Qts{@aW1BV
zWhG=j<wJN1lDTWsmwRcX-OADuhZP5$n?=<2!J)|(_QuM02FJ?6F(SpNL&4ss7l^ck
zITbIAf>|L2{Kt5DWM(F_92cwK(SgBf8L-!f{+{MrCp48ncY2_f#A{1Y+Cn&f`m7wy
z?f-s@|37Q^-$kz+S`4O-`M@>Xqrf@aLn2Q!cMGkE*;MBDBR+JEh=+HR73Ij5+mp!D
z)!UkKMNub>{PcJh!KST#yV#>F%0C}7TEL}G7B=-=%-RpiO!WwhpRczE|L;?TpTi;!
zkytU1R{gas{GZ<*!Ikvh;$t|PXGAosYp&bhrxFTA1Xwue0uJuhE`%h9fTKX{FU$`T
zw_+Ubl4RXR2S_;^CJ$14{%}m+m4+rPz$g%scG{BzKf7NL+nn{}$rA@9pe?Gyq$3ys
z@!OJ!9>8#{A*sw``5LUN9&3~S#a%|nM4_l#*E?aghX5-ehwXH*4J8GrD;Hq#-a7BM
z-NE5F5f2}VT&8@X5=AWsp|<YTUi~<8y$D!&Z=+C~EYhIND})D{i8q1bqyJ`nYXPt_
z|6G6sNe}Jshdh|hwn2r`yKFj9gxYiI#8lHvxQZ(i=IFHc%#3B@{u^}e{D2pK#EAI2
zqh7Rf_D}L<yJ5jv*OTokVu<<CUZ0@|9cd2#_dundm_2Ax(-!{&XfVRDv_4XQ?`BRE
zSg6o}p>|pAaUO%`Gr=p4e0G$ysv7Rzib6bl+Kx_qo^;voV{U;p2117H=`R?>PraNf
z8pm<@iD}Cc?D_Qo%u~B8NSSs^E)+%->TfUI5&$nk{9J+JmbeKnSJ|EuK0|Q9HTxtB
z+QuIZE_fnH8I)iOK)gZ0)~@Egu|QhB&;+(C71KMco<GMD<KyM?S4<E;R#=46A>c7G
zb;Ck82l^43M}!0f6fGP8yx35b4n*5#Xngq`dFM&n3{-_bp8@jAhP$YD*SrIKWnA9h
zJ)JzT;rI}5Aj|ARg5emHa!|jW$Q&At1eT`nK}iGk*#k!D`>F>QQy(Q6X&4<hf3>i(
zbaGmG&h!olAbe|)x0slis2-l`w10Xs6xD8CyglNlUT#JKo1Vz`1G(tCwWsaL3IF`u
z=usT!9%daDL=)!B+JyZu<}b$5Cu--rx>-o|=byPj4LKuK75g9l;TL|M)0uTEpsc>B
zi3nKIuWMzlEyy~q&=0UFaxP|)dlZ^{eE4f`c0m4BuZjTbJeYMNghDVzlrN(ZN_Yen
zNErY`c1jg~0syJ<`05Y(_#<E$s|a1PU_|AdHP8w+#=Tc)tV*=9&<_Fh4?HZ^4*WR>
z5)H(J5)AHyHL?LE=z(bmWft!6;4Myd^fyVcxD-}<=Q>Yx`m7EcPmB_sVS^mgPFmCx
zt|q{#ZNX%3&mrQtt%kKVE(c0;>f{36f4l(x>~P^V{h&PgNofB2!<@B|>9o@q-riEO
zo5(DzELM<$V#I+O*o%Se2M+SQ>My|5C$D#yry2~E0isndlvJ{3EUt`{cT6u}dt*e3
zmt_VWh26A-hAmYskt=Rf7w8U%f;{&9xQAAk6fXUNL>vcir1pJh_h%Hj-C=s4@-s`k
zRC<{I>>@8sUb}yh@Q79I7O&+kOkoQ9{Pim2yxrysc>ce4H~ie?gg9?Jqg3~|-$2&(
zV20A)HM3TA1@Prnev_{Jv3y6T()q%vbu>7FU%8KFQ>e0uuB97O3K+g}WMBZgsb+@s
zx#33@J#uhO6oQ;`0Ft9=FPNit_xI@t6K?AU1qEShqnc4C-|}^N5O?=beyO4j1xm0k
zKzqn_0bP+Q?D!N0>yEMk7?YRWo-ke*sVv}kY?g$5P9^{kR!R_173K@R($Bq%A`z(o
zK%BYpRuA9Tu#S-C{NodX52;KQ_PLa#EEqV2$$y<5In0L|$ab-%1&6Q4!ouP^c<;JR
zHDON81A@oBkadR&xcDJdKMw*k{0vFO*w}(0_f8F50@|M5nZJOVb9}9Oop{{fe(i6_
z$Dr0k5a$mNtc#i7cGoEF6_?e!ZJ%{V!eTw{dV#mm;(t5uh=X{v0(Mx)^xr>&Nf-3T
zPMrNjU_=<KIWT2lUi@i4CxmGBgWLe~j*NELJ^jNXQ{UlJc4p(Ly($-2f)Rx!aG-9#
z@SXh-hiuq#T);?(f|S&ZqM+WaF$69KD_7GtAM+_p^%c$6=|BkWg*J0TAFuvOnz$0U
zVi?`|ZH;ODHKoRs&X6)~l1bqLj;mGbBWv5*W9})M2BRQ3<jwy3k4$a!%E8nEnTKtk
z2KO&W@#xfGDg75SL*`Ge1IvoVRmlB%a8thW*M;L$3C|eA4!IuzVAG}~3>wqORj8PF
ze52=zSnYkr7LNT;eHYVvS4(rw{xqn_R&|k7*=X%q)XNFX$X?FA@W)l|oY?Sk$CCCy
ziQ=r8Yk#dWv(LdbeU+N1Uw`{!YJxptv;rp~@#nv{82;Qi#`P)k5@(Ld5mKPiX&)VL
z&h-8f^Up=bu}$(A?$bm^dLQ^+jn+gRja-_9ksW%`GdX?WPoE8Z5FMb@Mo*vnT82>x
zw=~^IBRF9?B9%Y8^?H?!960!1P<y786=-_hKDzTed;u3Tvp>LP$P-y3{i(qa-`d*R
z3uWs&Ou_xjokuB;CdQC_0votN0xJbvcT*pQdMb6XKDhU{_vhrk@I|lm9mEx*Ptg95
z^%$mA>0tV$43<O`+WPMTG&kRwH2^y^;FeH>ZbEOTa|l+GE`^#Ur{}O`AZ`K75t(aa
zoyK05GjEuFCziLLQS>`LgUnK)onlI0Z9t(!Hu*y%vTcCPZ^AgDIuhBF{PzbyJ_$(~
zd4~lnVmcw8t?KW-!qVq{G_iQ_;m<?C3{a)|KHDH8ckHXjh5y*|jD<C%c^3f-u7(0e
zFDfVLI~f5VCF6VWFbi#MSP)`T1m@0%G1}s*@eRM>DP@fqMycd?2fx&FMC~2ldMv8}
zKrGxoT5yF4YQ-ZkgtmtMegJeo09K|bu^|LP5X=dKY5VSfJoVHZAy`!_Ondft0%6br
z<DW}FK$q7@&=ID!lQ<K>TPo8gag`6UrT=*E{%TXo(tri$a4VDyvTuJQU!2nH$gTpo
zt2&B`sO5?J48(N=`aMmDMyX;v0)kus7JKpeYMBp$gM(x9WJtED-`t8a_Z%?UJ18)0
z#l8-XX&QbrciZC>IPo9;dEU4=oi{Zv)=3Mf&95mP#R!*@sxs9Z{M&%=8O6IC34@+p
z;@>}qQ|Q9vs+ZLY>RA!vE7u`Lo{F0muAKUYnV?*@qa`aO_rS@`CUMuAA{YUyBf%YT
z@kio`5iTMc8d??jSN+=FepSSJzPUOxv=KtF&CSgV!B;P|_k^|q#-(ZI7SXo}e!GZv
zCf(#lT;;N10ocj43pjECM6Xb{es_X)Nga$OdM5Dd;cu!iW#j`s(F6Ue(wK-3*VUiL
z!!^V^T<r%BXKwbXsxk)$YvqQsu~m`CYG2XPBe^2h8euLWMM0uS+N(f%T53~I>mGS;
ze`Uru$?p0(W=dYR0N{}ZqtIDUs#EoqH_Rz@gh$dwM!t5}-l)B}-XYMva(BGj`eVn^
zQGS{$fpQ=&59N!~M*qjxd%$zufA8ZFA!TN-LJ^_tP1z%R?}k0HNwy-hlt@`6duMOm
z3MC_ZQ}){m8JYj{a#wVJe&7G+@wo4Y`|)Y;e4poeo^ze+y3Q-bQy4H<LOb%WOfvVC
z?&7VPAj59$6_X{yfT!F^+Fba0V=5Q~=}L9Ey?wG1D#O{``5a<m#?lRj<LJ4|;4Wbr
z@lue~6E`i;`{&@|{reacWFG+~@Pt{NcyMs=9xuk$y|B7$H{O{ADWQmqJU*+LPI|K(
zWVAl3jXO<}N4|w5S9E-lS~X*iLG{W2l+L}L=XUaTB<hXH0;i6Co1Y2t=fWkp8suMM
zpY;X1!(adMjmKQ4VeT-G!0H8~Kh1TH?a56$p-?JxCf++qP9EGY7hEl#kxw=dYT!ts
zjjz5<i*pR&B9#2FtQn(n1BQwfyX%9jTH1igzAY>yp`~3^GwPO>mJak-pP$|I+TA7V
zDeHL)O$z8z_R8wR?&C{;3kp%t(1_xSl>}x%+y`JAdMs6M{D56()<A0I+_DsY^EvZ@
zwUm||G4Rg*I&8m8%mjDSaM3*nw{8;QA+E*D+wpEEN8{HXN7fUjtNl9b$tca5`X=F>
zC7=Dl*w~FAI&T#KswPP;<~;RX4(&*m!8f#jhq=|6xcBC%vZyZ4?n>810<2gT`!{78
zrOM466}FhE4FV+&jm<KTMmuyx1i3|B$p(KP7#v-nw8?577SE55Dbd2iv)OjbFYLWC
zZ5xzij()Rae10z$y&m(&X`lRa*vPS)RM+|abJJ_D_6=S$-@LXe4!(<{IjNHD!yUE^
z3B9U;;)kx&1SpJ$b-*BhGnZT0YRU~$2A2yO55vP<F7ok)Vc>wP-9TX@&)4cm;TzC>
zG<N4@CIMe@3mR@i*Zt*3K2eeRU;DtuL=uoWJ@Ec$&IQHI42*tVFRo6W=ne7AP`3H|
zWYI9Zi8r^Kt$%z}W^^1|^YyJg*|$bX6-;gAE=58@BEYd(v)U@;0Rq@!0IDe?OiKAM
z^@=QB0pNmq9m=j&>%9vpV76t-vRTrPraXxIrhMQ@yl2nJCpS~68t+@T_Bnsx3*q->
zI)C&W#mC+~U@dkpp9^~W8Pzj?WjLyfr6GIDlI-Z`j$$_Hr3!MP#tS-kByNwPqcN$t
zmq{9%P0h?@8<=JfrqxTOpz%p_orogWHlB9AlPMbU2Jd6e>$?x1QZ(^e7Ut)xKuZK8
z#88DtU=;o;l0l{->X~gA?2%SaRlGRpyr-zF+zSLe!%vxm?vu~7xo<Z;9?$fIwH|`o
z^M#LC6S>SIfL#4m4V>Vd%-;q-C*K~DnbeDHK2zX#c%cry@>Rz8tTbW2(ftwS<M>*+
zL0SdF|LU*s{Ft=_-fA9zVw>Ln$>TjWIjN}@N}q)ZPi+$D3g*Ce!qv<;A~dwH0{4AZ
zN*#S;_-8qe?955k1nngcH3kuZ*P+)WP<Kbj*^XS~F)6>6TH3C5rWjN@CceEOtH@R2
zT!rPEOO_{(9%j{f<S<7l=@yRsUN}}iM*FgK$<UaEFWM3B@f3O&JFm(-!HRx?8HTm|
zV<90))Ns7Jz*geHcir3<Uv|<P;8LYSUp1Q22$~qZ5}R1iTpN6X2%YQ60P_bym^n3z
zGGzP%rOU8KQ!lh^U0K>3lhf0cMOR=dh?u0Xn>IAS&-*McE*ltM<?}Dad};#mXi#cX
zbEk7jetMjE2=CjGXgTZaD@l)Kgo(Xb>T8%+5ZUzACod~G;?5Tk(8wY@!x8uK(}?A9
zt+4~z$TsYKL~^ONq02L_f`E8V1}uKBoFKlXYU3XbWP?^-acNiv-dYN$E7RlCTI_|7
z{7ptA?)G2h*^Fu$vuw4x`wqhLlc;iJMrL9Z!(j*#grU1uJCKJty?YMCo5Mr)8^At8
z5`!9?7vZF}U+qK*Pm+=LPJK_xp`xNmt%+cq@aiDyDmRpP@E{N)M=I&mkh{r1@*~kq
zpPg|!NT^SpboPI?`8a=lK&B=5zL~vJ5%yAOrEEj1nmTA#nbYBA`k=wwLm6H)7;#l(
zH~lOG>#WwdK&>l_qN0LfNgnOJ`71Fi;BCxiSkV=!^WG@$YA#8+A~DrprwNC<DLJ<6
zllvu2Y>Ag`T}0wZcb@rc%Z1L%=<nD~daZ4)CN{k*OSfjc<35`b^NHgE)^=$3XagMy
zNeWzh!KZ=-kVfrH&DKk=Y&rRaD#-JgHE`w(c-F1l*s-->T0@bW-!(v|KRhv&*t%3>
zeAE<E%%@11_8*g%8nM1!Fo?P7L5+Si8pVe(<~_#q?UKJT|COZ9@?*-&KeGDxsTwY}
zz}5}TgcZCqm`svvGBh+4nS~YS)7|u{D$ajkq5o`ToSrvnpT0=Y2l834g6W19LS7^+
zC_Db{RP3`wu9N!3Rw=-IgC)~%Is~CIbiMFU2f!SM?di|4$#9-D+IG}WpFE*w_$v8f
za{Q5)79=<p(*B$A)>0+UPYC|!z7tHKGdg0gkR7YvxQAA0bPM;BBp%?H9!L^_ehcxD
zS~4~`HRk3AwB=dYKhpylzy&Nc_G<!o5>L(#+|#LNGS^A)tGN+yxuu-TFxX+uOVHc@
zNVJ(?3zylakaCRP!fE)GD4*zHT>UHVaQoI8ZY;?BL-?Q1IZ!-<A){W|dgX_UR&I3m
zKvtZWjIV>lSx$P%>dJ;WMVXsHiysR@!mwS&nKL(^)~U)wZ^QU_wvo;!yJQf!2X5Y}
z8<6E$^318X!JuajoIHBKm9)9JS!=Ou98YBTN2lpacCozjDavXv=7*ad@K4zKe3^Yt
zQ$*pl#G}t5e8c|N#|j|o%pvM*`+R7gWINrv@S;cbGy}mr+^7apjbLgRdt||;$~mYm
zd!g|T*yW-4KE4DSOu1eKI9wHFXJ;R4wBA7Tj;WT}5G(Civ2(Q<R8g09Rn=jR9b{`{
zI{W83i<ke2xn*63&X~duV(zaMXFA8BU_TOFC<u6XWatQGj>Ibxxf*Ah3`&(l-#Sw7
zI~Ft5rhjAa$s%}&jr)bJf?GAQ$bC)GGJB>epT$aTUKyRE`p1*p%0<B#_EMnI0Jo$g
zTelENiwf-Jl0X}XT&^nt@uw9iZW_C<oMf;5p@ujOox~B?bMk9AKB6SFz};BgKYFJ%
z@WF$NtgH&)q>x=CVoc_k)6~l)?s?=JJ~-1ijYfJoZP8QDrLXM0z!OZ9QSTf`yMf_v
zh&(muqLC7Ea>0OuiA-aTf{x++<Ucm~wafvgJ3zdsn19<GryJWQIG1fgQNq?({ypHj
z$IquZ(HjfjM%cJfPo7?Sq#%CeFZ_f-t5c>ndTCU>#1hrk@8+uBfpaCbA<5y$uOaxZ
zSeMI(CqhLcb0l;Rj`2di-5i=1%KgcxHr(FIb=utQlctT>_aF3hX3&hs$<Zh|?AUaj
z-FZ%FvlZYQ8Fudcc_~n-JIpRRwTYy$=9c{vYjaLWXwWoKh^k@wD!5Ka!T&V%cA-)S
z!|G6j^<fzJBD)V2V&i$sv~H*MCvp7e9=u}_rw=lTzuQ2=d=O~)!PF;Dp5*q~^F$&T
zhz(9jRT1PtLFn-LbixeH8HUTO^z{GK_wZMhpr1c4<I+2Cs#kNRwj{{IVE+}BIt|6(
zxQJcQi6a4?p7T521c`8a)|&=;mtH0$ZRaXv?D|pvl{TWG!fJSxf{h{ZS0wtYm~^%V
zY)zyM>lIy+8FZyPdChMCUw~0k@cenD>dtTRA8Ec(u9Vjfux?E~tleF>^WyRie+Bz-
zt~g|A&oye|N?2;YJFD#Tr1B#*Qd5{G+->-&!Vx3QYU2rlZ5Z^5KOnlu%c}+2L(R_A
zb+2#ddqWIT!%iRlAdv^nlm7dRLe|>P4(iUEk7`o`T4!&J&Jm}cjteBrmpzCu_DDoX
z#sjlr^Qj?`d)apfGNkg({o}5h1pe9xt2hv!E*E+mjuJ(69-N7Jq3jz^xr%C&%2G3p
zhUp+_3*74Lu9ADBJ(zOk$lq>7<Zkj+-m7rv#cJoTK;v-<-aYzhj@?459O3_Q1^4UU
zZ*LM&UxNCKBI@#XqA_!-<#rO~Jwq)0yo9KPGg(MInC4>jIyHQ*xA<ZNm#zs=ssIb_
zf!EA{9?`1<(aiX6>~ebv$dHlIAGi=2-@W_c+02YZJnZ!J*!&vQ4K%Hg$_V}aak~BV
z{q5R)r;9$|Td`Jm9KS58-!{Nn%Jn7TR>0qt3zP~z_Z*6Anl5uh)>rqsw)YRqO%R0T
zu{GF)Bee{sB0UxZ*AlE%SsQX{%gE1ff1!ImWE59YN@})ZE*FW#s249#8{^9{Xgg&d
zc@1s)H}1L#6@`4^Ds|Cxf6j1S2=`CMJBCh*O|Gm-xiz7NOE2fuN$fo8axwZnJ@%1b
zf$e?)$#)6iA<2K89~uUV-XxBp=aHO7sM7JM=bq7&7gEzxR~stc<GjCBQDi-#h_d&X
zzF<YzcHDcOyAkNXg$Z8kx-sk&a-@6Pqx(ojJuy8!2h$QYSJ#r&#K^8!KyIrT^T&+~
zrlV)DP8jXzPy=|i%rlUmp8gtW0t<JK*II4Gi@~Zc3KB6}md7zMY3CK9E#MMz_P|<Z
zn*sIJO6V+e!A=6YNF(rdtr+^b%Tud;v<rrKwIV$mDbXU`C0^HxT<@ukVCpbjBk7VM
z*r}}vW-YGXr`!3cZ~<6BNV_li9q4grF1<*L5gm9<COTxJS~Oa3;I79t>>$;#gqeM$
z*)Vxs_kLOnspu|9Z&uGDL0e<`<L_VN>uZY%RZDw`iSu7=HvzVZ->o)621;(#)vi7H
z`F8Ds7NHCyS&L(jM(#-{*b<K`qS~3*{NCV=+{#f`08C2V!=obX{Yg^=*bWS1oNl0I
zzopOH=N46ti;J7@$StTgYoUzc6tbh<7PYsblbf*5ekf8B8X6j(Xw3H{8mi%@NuE28
zXt!#}q=~Q?f}4RSfKa(d-5$~&EwhfHcVi2$e6#Y3ja0vWy3KXi%@%8IVWg{2nx7I4
zBlArD91m~p^$Em$`1%WOIgYCD{$aQ`E@8V%)XC1+LuwQv27Z70JF4p)(j@5TQ)FD)
z;%rQFN42F+uZA!gB)*1<>^Yfl^pOCw$F>=`t0IwTU={qKf0)>LsrY{b^WY&)coar@
zB&j0V$cj~ESm37nZHDUQWfrzipSoAVw@p2#LQ>59cw1gOM~q7>Lw4*g(G7TkpA@K#
z32U8&kvq+9JL0+80!A7M3bCbpV0mv1V}iz%eh}>;u1Tmb$w3O`!@QaI=q_)`(_>Y(
zJpTcWGD0cfBzl-%#$*p=+8nsZ^p;O2%#EORT|1hs^|8rEN4*w`dz~km2oY4F)#(-R
zfAxQ~492@p&}IEFHOno1$#!R{W=^e(FMf$TIw0TQHwGdlyk&_<0Q8X(k3_f9SN(IB
zr`r5|DbS_Yr$I6{14Wh?yF?+yvVU=l$o3<P&8SlgEgWy7lCJKmv{V$n1i0WWH~YK8
zBl1@;`R2sBX1h+Sa6-h~redHptb~#D>ftXS5U+DONSU6uiw-ODgRfNd(cIbyaZiS>
zNts2|xH7t2rcF@SP*lTm%hqp@LCw?Rpq@t8?Q*--31=62?|Q&hDldX<kI8_SW<sK8
zpBx!!YHQQ>^sEXAF%WxX0sA6>AlP~vufbfR(s}kZtkAkER3)=ezAZYYN`SW>B5rA!
zu~@ThcRfM4pLGHA7DKPmy&PE3mkH*I7~|h5xQ?VM@ALWhHtfJ(P7eeF>`NP<vB`za
zEYWwgOW%UdqIPRAQ4+7a<-#^Wel3zqe_p4@8o$0>&`8?}7U`NgIu`pBX+db`Obm+n
zIIqGCsa{_8>|i9t(-#Ho%5H5RU2li+Ca+}<a@S1o+B(0W|3@M-_dV^A*9gU&Z#jMS
zw3nkL&7<7m8|H4x?bW@hU5jU}j@CH7$_Lj!jTuAcXn<h&Yb3ExKYS3R@MG<*0h=SE
zi%tcva!^Fni43EK!1zA?YhIOnE{rY@wdLo?yKfs;_GH*%lN-;I!{fOJZ6FF71*Is&
z6@xC*^g%m`w!teT=MJdwjkqT#+*>TSn$xK+Ozr)qnk8mp$>W>&p6dxeIesu#f@Jfz
zSl`G#Q{Zw}fMb=2^~1kXeVv9MY=4~v)Q#iO7Z_Q95uQ{-TV^zbqb$%ZyU2_^UjgPS
zNGuTN_~{}-F|dO+w3ApwtDD2^6Wv}Ti`TQ*Ii+sF?pS-TkXBTKo5tMz#;e$6=a9~^
zM0rBRzyOM0v)NZcv`->G`(dIPV{_XL>42>mT&-l7m@kI1V>o}-78}gOT)OktGmI>*
zNAXs!eaoKhT_)^@e;!bR$LPd~`Wbf$a|D9oxp?E#urMdG9$r06&lqog4ewB&cq?%0
z??P&xWo%a9TjXmsuR>vMV}U1vJ5?8yF@xmi1vT{7SaWZbZKX2tKNWNqEQka+r9abW
zZzR^=A043=a(eFpilzY~pYsx&v}fGjVUqFJUQ*MnLrjLG|G^SsOQ^x`v%KUt6a)G_
zYjElI6P9dlG-GVITtiMuM|bD0&C$NsSG)xsR^nv<XL?~J1T<B(VfVu|*`(32#$OGb
zKmW`)k@HDKHS7rkx1>9o*EfqiJ8-k&g4J74?{_04@ywqzy}M*IAeGeczPd&}+G<V`
zZ+~~mhX?%LU?dR|Lx5gF085JlKv(%1MhESryEsuw<ad0>r*qnsBdCJ4INVMvb@&>!
zWaB9a{|RirW2Bp=(^rRgCe#$9-CZ{@hVR*ioG-xk|8w$?97c%8>VJ+DzB&Krw;@ea
z-6&?&-3T26wag{0c^B`oogmG7&wTRxM(<vl^+}$&Jl#of&n;YAZ(I5#?&$(X_N$)>
zFN;3fgid7kBU54*P>M8{j}J7Mi-mjf?2g2C7G~<_WQdV8s<^PuUs12y{PtuS^8)D-
zF0hq9|7@4>tiGcVdZhSsXEE`Naa<TEZZ!*Y>IEH|6ws41iu!@R+#dV6Al0c4_>3*!
z>K^<3RY#zw&Sv{+7IOPKJ;dp;IxQ^8(bAAa(a#xK56)%Iuu|-k`>NU1x(IYovbA%y
z2`9ruOvts3V>`+%NfsJ&X19Gpg>5<t!q>5G&gw7CPG1UFxcuit{OU-aZw|?Oy{vA0
zg4Rm&GWV&Tv_4Vj*(2Za*)frW+&hA8iF35;grhV>D{LiHkbi_~?5;iUqVv`<Z+(ia
zd@v;`mkjFT5zA%N_-3z2Mgjdla;UL3j)y}-1{8^Mt<BBqU_|EJKw*xME?p%!=e+7%
zkFRy&k&u$I2ngsxUA5XtxG7T#l!$DYs4$;9_fAYuk>ki?jE(MBR>nUqP<*8emUC&{
zHPX<~cms1^D`1G)WpLZ{sesx>8<=#%nRhhL7ZM97R-OCc&=TxKn>F>6{~go8J(r(&
zoS2vg=rVty6)D+M9KlWR>pRx(VRGLKdwE;{FCJX^r^d%}uV-+QYTbH%-js!{xMaij
zrIckLL#)JY$@F5s;J`m2M_`X$J1?C65-Nm`Wv#ObNz1*dI{!UO!-KP2WVSs0pR=Sv
zV{Ii2XFA7jSc!D(YrCIB2I9>Z>RpE%@^8_MpS$WaLzKrwBS+)mY>}P5<yrGNV)YER
z*V}JiX9?mwc>;eW#0|~Oq>z`-pQnS@1PU;9{7eTfoj)Hp8pb?}38Mi&7-kmUlYSkT
zot-UTX|+Qm1|`Gzx|FoEAl;j47<iDJIa4-I2Wgxu_NBtBM|`v%-YGRkW+U}G7e*eV
z6lqc8H;8Ecd`XZpYmm{=9m3Fg(6-bP90}02=}k=hF1Ev@g3X{z1wed10r60owLFee
z%hp<-0{+ad{fX3yZ4_Q~<N33H-b!&397RvNl$sXaC3GXHobp?pE^5s)AyV0Yy;F(U
zNZXnC{mb<uKN5lX*(3~4hlnzp`a5zoVejz-wq;Rct+*`0*$GNZLy}$<4RcXl{8BOx
z%ALqv2<#Q=$d-=VxrE6p`2h}O4a{^49cKFvB-;GMf_)3KRis1&M`34uY>=w6a&dL`
z$3WRNIXOwwoA>%PJroj_Ku}{xR&l<8cipJL0oX4+r}>7G@I^P1s+8Q_%XsaE(utM$
zUsZkXWyNPV$oSCQvQ52t+($|CPGj+TQVfF2GDs`xnb33tV!qWzNSYpt6k|uTZnOR?
zB%E&W>}0wXpRx#*)y6iBvG=z3hmi_CwMK?JwN=$85=N>bWSqzkk1!wCW=J(jKjBmU
z{jYJ0IFvsBkDN$`4APb!MUfu4Oa~Pz(f2ihZ(9qPTl3SyTP$2MLL1b%8I|*H*!Kgw
z)bk=7TA7#=P@DbCFJL3)KLC~VE*(!;6U~3bn?O1E>Nq~m>W52AjEwwG1q|*^KC8i2
zRy_I%*<%~HbZdG*t@(*f6t=U&^0->hK^<7DRw4a}c0mmh3jzBw9aPkllK?)pu4aYc
z`}@FX4P<h4D}psf=OP30%zZZxdOvF$QH2Q=Jv~%a*-g)75izk*F2e}vx(nz4#dohp
zD5}C%%zU{=JB3u+_@tKaCr&MVD&!r!C#mz$Y$!6-mCZe~z^BG>hUfRmjp_RZN8G<X
zYW6bbLGiblsfg>na%?{<i(li+9m#{6Pm!m3>*+Z!slSRObg4R}>4z_UMADVvcGk3h
z>Yg%u;CIai#ZN6j)0Hf7%Ums}Q;<jXsWf3~L-KOnnf%Ueag&q1UK$PN{;;E^p@Oe2
zk`~v%wcqpwk99QErXqHsM^PDZWJoIf-52Uh4e!U1GmBtp2D+x4I8#{krvX+Mri_;9
zWmSM(R={L}l~kbj(e<s%cUNu?-<^(c>|WJ;cN|w6^>T8`i*)CmSQ;uUce#d=)`1Y)
z9f5mS<)D2Rng!ZG-crpr`xwZ_P9%y;O5R6aWEeA@7#*wT|Mu+V3IdQni=th-nZ?7`
zooN|m7CxJrzV(VnI~5j)|LM*4iTlcu#Y=^s-3V?J7$)Z48#tqIlMv%*vinSk#GsS3
zLe}ALJNT*_r*DngL1yus*OV+OzbUfo#_qa5DzL6*AE(QS(M4A1Voz+>*B`CdVMs<t
zg)u~wKez3xACLcC9z?Re0CdxiB^T{~2eDSX^3R<4P?Pl~5fRD5539}xlarG(E|@Yb
zs3L|9#;O&wsV}3TV~2WF0+?(ZgOfY4FJ5HM&)Yn>e;*l0fWeFw%ypZa@03rHf&)yc
z=U^|ZnON5n;Ad+=?2L}z*My11><?O^6-hkc!1Nfp4W|G_;>4m27bgZaI(*p!G^}nQ
zs0yQ)zb}0d;@-_U?IK*U7!<#-d2^`R^GfcbFLm(mR|Ml#-7c=yOeM)ZHj1BBl^xx=
zQbqqQY^Dyfpg?kp=FvMQe#{SrvGbMM!ENO0${RfE654iE>%<myn#7z97r}y8Tl@)c
zH-p%(cX6Kk)Z>-nR73JTXajBZ<WKf3*!_j9G#_>GZDL19kYCu9jqu!J9C`n0!i3Qn
zEbGB9U>NT^)ZWdTgVk?Mfm~$;jFO+ZKn$9N&wzb&eeGm0wWrAO_cC={NQ4z$imZ>I
z5sHg--CNq<YvYoGyZQ(NS@r_mfI8gSwPd5gtgL?%<c7{jMr6Q%rAlJ-U+;tOKF;st
z_GWzghehUcuXFx?n?zycapk`Az9{VXasC^ZD$jgt;~5rP#9p_>Oj)RE<!ads<ekL7
zoo)3nN(e)+DC7VPcipk>j(USv#cY7Nv*1-%yx~{9HZ8WV3*f;u6a0=g>0a363FGYR
zy#`L<iED9-N6_}MMD&IYfABAlN=!5!KV~KW`8&`-=88I%Sno_%t+sZ;%u%OiVUC1I
zU=tY*`kEsskC_~O5E6Ufl--*k(&IxhSZFzJfgw!xFeIb{8bB9Nq<f6$5R|B>=wyFY
z=PYt+yP+UMN2{mYDV=ulqr>rV%)^TXpZ=n8{<YDZK7B)yyoHx;+7&R5U)Q_G4FvGR
zn^tph`$jDYNDhB1e1&a?<|8UI<dZLJY_=2Kb)!FUmxWz`a_H&o)QmopbjBSIf3AY#
z8+${NxZkbzc0>ur)ES5b`rHCJDrV?gz>5<U&iQwkW+{&RRcQ42V4rE|@PPzaOPbFZ
zP<;i-qCY2yi*0n*0-hl#xS`MP+NA9%3Qo!WooTV=Qm^Wp>*oJ34R@)gx!HZ2yFo<v
ztH@9%7Kjp5V*{ei!iy`nr0bL}9hayx^n?5)$W;e+z-9Na*Y=q5UFXn-McEXjNJnR^
z3F$Hr{MGq3xziEoT08}v%I(ZdbQ&ki@btnzKoaTRv=(%))E)S%0i!#(V))G1e=g=h
zBw<vGC&e+4QnU2D{%im^j%GuJm!O&f<qxMhJn_{FYc8%EsjYAD<d`J0QhNi0Y}Fd~
zF|1x}B+JM>`Ie=79^SBt;ZQ0vva}5KG!%fNPD{tE4x7QVzT)wi0O~tK<vAy||ML;7
znl%_v0?V;hkqrY8nB&XK$})50*-ZN5vbST=d8*V{gWSmjfN|vLuF+!|w;fw1MCKU1
zeToVb_A~A044qnd5HXqXQwS_{*)gzwaWM-98}n01s=8cJ``)GM%C!X}FHA^Y|5&8V
z!`h3}dv*V4UhyV-o^SG?#<*Z(1D&{cYWNv@Lenk!%h!+8WhSqXmr+kv{O^$`*aF~S
zd=vQgis8Io)qLq2Jh7aQov#=eUF@eZ6A7qZ)y(3+S!LK#vM|0$5DHRdI);J1>~;27
z>IVS~xF(H?c+~tVFptk!|5&?sg~;Q`N8#&(y#vk+Ao0owo3H|>1!#9$@Ksxznp8mQ
zqH#s{E7%fqFq8)vk`{`Q?R~#a<Dxfx!Igr2SN!Z43~0k!q9pMElIfw_-<0tJ)-4)e
z<SA@4Wz1I5Z5#8cbZ{HJH=xl@v)>e&wPq?EF*#l{37p}O8YUgu7~jLov`+f_wUc9N
zTlk{kQ=J;{KiL|YY`*%*z=VF}YQP^+SAg#49pMhVol(=JFWS20TMpeuB-AVd(Y_1v
z7?h3o=YX68s6P&>BNv){Ig|>Hdfge!t!-a1<uk$5JLc<PkB|+3vR*@6!JqdR{$Sr-
z3?&$_!{GK!)vbon?@(GIi)ubKdscIP)onb!r}bjPw^nAaLp9Bn9zt?DcGHg)X^4yf
z_|VhKWX-Em+u!P~MR((+pg^ZRiBu*YVEIf#Q79p7bFmCh%spMFRP<?vZ6&_0;P=ao
z*Yne4T%JWv#vRQ@oM?rGxzQi4c(+!v8ZPKbYG1lZb2Q`Z=pfU!-RDucM<NNjc5hY!
z#p#D1&Zp>o+q}F(FQXPjM=eLQPI6{@p-Q)ZIZ*Wm3Q%W)JiCQkB(x3FcsV~x>9sCb
z<Li8#Z6SvTIW?8XmO^<o9xziSem|@-T?VByN1jP5(mnv-9DI^TK@Z_E5h-IO+V>A7
zqhmh7OBL+rXMY`oJ<`m88O{ZQ4n}mp^Os?t<@H2Sk^B>;?<{z2INEL13vb`HDs_|#
z%gPgQOU;7GH)}l+?w_BpF7<boximXmF7g7E_y2gUD3RBy$n1yY(Lp%;5$VhX+0zdJ
zlU`eWA?&pvi~2qsA~NvQSR?Ym8Vd$x7k)Q|;mdSoJ%119hnl<J+y5QkzS|NV!*>Fb
ziO=(t@ojP{s;rCZsWYiB?F9ekEu9@B_CwxXs5}E<f!OOeZy3$dOJJc?;*XTfq>UEi
zSLY)plFis2omxm&7vjB=b$3AO?94%hlNC&D1d5x}w8H!|OjKYgLgJ2~$3PUQ(2w`n
zC*B;}Hs+aJbBVqe*K<(`e?R5vqyx?&7-k~ohXbX}Hp!G>dKUk;yKX62hN1u4;vHO<
z#KI1*BcB>z4IG&KGEqQ_&HTK!UDOoYbtBJpL1aG0$<if_f#I67UfKo0Lj5WwrC{|a
zhL6WXeNKKCH5t9j5E<Y#-4LIEuxY?7MH?E_;S433qkpu$gbCmTAP@Be<8_p@?S7`?
z(vHn;K1*AX(*7?hh~74Ln0P*XXOu5{i3G^+)FDGwh$DZmTnaICJ6v)KV=kC>Ko?W`
z=+(D7=ueIHv-b7hHakYVWVgu5YO^){i%ildV^Y74e)?tH4Y{D!|08YmKr!6#$^~8W
zFU%f;fSGuVx<K*UMq)qJhV_@X2Ie=h03BRZ8`g8z5osi*8Si3Zd-wB!@ALG0RAynD
z8G03dpUpKE=O9TulFK{J#UNvQA$<2OLRLy+NIJ?4+%y`i^E+MvRJk>9%(}?O*W3RJ
z^eZ!QPOV?A^9@hz7!@qto4t$Q(AM1ZC`%5^S{d2Sj!_*n5qk7NNc{($$yzwOD&hBA
zelYpgPV`&}rh8P!XDH_UMv@`%W7m{1F57~*>_hI!o?O&WZe)rB&LMtcLg$CmmfOC;
zx}n$c>gRaH-2181mx84Tj(%AAz)9u9%OO5X^mj59k3Ba1(3I%e{t$E@iMw&*vZ%+&
zwgN#cd6F5S;uf+Z@o^nGFk*3mII*H8UVNK)dK(8Kq*mbK9APl^C&rz9^ZmWuwfVxf
zOepd8#K@(O3>thNpu;#na(M$(ty!$$lt3Z^NK_8~u}s<C&1j#O#CtR{B$xK4;U_VU
zmsDNe!4tz8OpWY!I8Djjly~8NE4>MmAF@*$!0CeciU~5W{-V_+>FX_BuNYFI*ZI!z
zUUOPh)V{=5cFBe5Pg8G#<#&1zm)piM@Vwz{Z(+tFMaR4UP-oDr55Rjp%-2UN2`Kr<
z(c>5@_67SYvQ!e;h9QqU&N8M>q`8%$TNl}Rb@pQcOh(3JQ3PrC`?h%KBu_VEy3^&y
z4K4UGI@-Vd#KFP~C>k&?-hl7-6xnXd)5B*EyYK2)W>#(oc&2IDY9n*{MDHy*Ak%gC
z6mBa)S?CowLx<jda9<jOsgdoR2+{(VrW-dO_>q}sWaZ_N5i97fGqdX>K4AhYHU`+S
zwv$5FFY$0jQ|abgc{uSE;Btrmsa?m8{|OJGyL)Aef(3Vv|F^ZwCP|+iSOHi&XMy0$
z(bff;K+g+x0yl?e+!h7iKo`LOL~z1lp~t_1hv>eHi%3Q!Lxx1vUVC})3!dOp*U;|a
z#crV8YYqLviWIqa!n6G_LCb;#U|LKG?ys3z>@OXOKQxZ-!)~LqKRT7&tm6O*suA5B
z3t*WuGc}zBj;<<<MP0yDGXn-8nmudX5w)!?J-}yD#NF>>i+6UEk2}Izqh)e*hjIid
z{7-D}aC&11lV0|Yu`By};?W>Z!->>g>*|qhv$RQx+Qq7#3U6_$=5Zo|!)J#NeXf+u
zVsqk@g8#-`zsAO?SrfTS{(lt*YtVQ&SX5&W{s#c!m-(^7^SE&7`#zkERri!<B;6O0
z&Mw|7fb0VDsC7MgNa&$nDmGl2lUdr7P|IqQ+xCMmURZ+rH=NV3aXYA@-!X9J%%6kt
zrNICOWm;N-$EOrlzyVL5KJ6k(E(&msY2tFy9Eg_7Vc<RlPGiVB9QP6!^3E4rX1cSq
z`|b317$XDNQvH<9s+b=n82(5JsO-ROuS2lj{wzB7&IM%e%I?l;n|;^SF<I3N<R&%H
zV9nV(598xxFo|5L3I#e{Pc>o0Ap-E7>Q4xDjQ(F4#^@l!#GIx7^Mm{wg3}L2yiU(U
z56QvhRLsmO)6BQ1ZJFUVh~L#SkVudNB-8j{ngQ-{(Uk7h&)dG;-xeF;_1V_i(<=mi
zGv{7GpTb#TVFS=`YpJO{3FFwHn7ed%rFw8rQF6-y0%Lb~3>l(AVH(60Sr6d0^MGZV
zB`9Vlw_#s6EV5x4%`IDj^&*+Dd~{L@<juV@bY2=Ywn5MCO|kd3S+B?TJZro&o1JiR
z$JUzi@@cGUWMFVNstMU+B?#hx0B=V$@j89<h9wQaHmbV~REi5-++Nvv397@W?-Grh
zLA`woi~`C638Nd%|4Bu@zQu>pwf=uZ*B!`w<L!Kul7C}l{`$U7lB-i<`z8f7j*QNm
zf~?xHbL$!D6X|!~p{Gwma5-yLM0TCKx}YzO)!sBObY=u=UWgCMvZbN+!ZAW@92`|J
zZ{&8KQ85G{@gy!TZo)e*wcjs8T_ZXyp6P&%jPcdjr%!jo*3E8kC|oh}f?Y0x<6&H%
zO8379iCKW>=HwuWHu$Lm>k4;aq<T=-hH>fB$B*zRhO=m`&Tb7mMO)A6tr7|Bc<Gq0
z&A`khQHk^2aXqA?5|*F1>^LC{f)UmvrigdvOj7D{N=GsQXxF3QOuFe+;9R=5-a*;i
z2z(vZar}(Ok3FIrFF>sRx_`br=m+Ih*18_ie80nRcZq;<c>%`Dqr;X>;3o0mdA+}S
z6tf3_I~}T@BC{=gT9%&P=5>omhJt`)za1cr_h78AGPk_!>eU@4f;#zI_bZn$E05}}
zpY~ooyDvcae;y||?*csjHBXq>(VyXh&mU6QB1Ec566V8C>3gQQvvU#R0%TJN;36>1
zn;jOYe3HRHYed(MahQdfSw&KE5?>Kko99CnxVFLzS?)N(ESenrFFZ?o-q8YTs+2o5
z391g<=<itD!4uKkEOU(HyXQpVNPki(*~0g#S9&G(-2yp3VsHn0R+uvn+knl0Isl?6
zMFr}g<o;iQwsa6^ZuFAUhxi_R<*yBaHaET39!cc&(w5PssrymZs)l>n`z@u-8}B!z
z38A6-x2bZzDM9Nw@j+n1t1k4pkzI%+p3#XE0gNif%P^W0S$<C(u^)P^s*AqmFnelT
zXa^&nl_kN3h6WZ6j-@iI@rLkDb!8DvqZ**Kz>e{lJK6@nKy5>#9}`cNQNXBV`5#0q
z%ZWY+I&hbzkzp>92<c;f;~3HGhM7Np{22dq(O!8>)*)l@c|uCDKJJiEokB6H`zdW@
zd7(!t_qtjr_aQyYgv@VyDzE*C<~_FA*23|MEZfx+|K&<;9u!PoOxJGzzAOjlb=DmM
z>c_$mRU)50jO|EGr}&p}Xqg!^0))6a_%94awx2TsdTo~ew$6h<K~eEWHFLBPER>@8
zmLH?z<@o2vu)B2Y++VCVbJQww1_|Oi{t0uH!aArT*MI~Pqdk;=BLKvk+BLS><lEsM
zrp<s^Bmwij)~0iEnkmuqHBjhdVut^vxETE`$<uCVG0F{d*;bruAg&k6y2m*x!hQOA
zU;yOg-ab@Z(+%LYQHpxHw<}*dhV8Cz0{j3<UYqBoFHdK#yH3z{JiJK3!hd_=dU4;(
z<ra#Uxlq|$ooq4r{nYu1lj@W%_>SLcR#XV=%QIIW=6dr#{dx^dwZbJvGYS4`R|&9{
z&{_!>xR!uPxi)SEZE3cF*OKhwW4ELx(fC3K<(sds#pBRyy*bY;BA`WhLzt$zr1QhJ
zxb27REM<DEx$NNk;)WP@l?%L3Ugod|Ii5Pq3smU8@M|NZRVXt$!}&&Z;K;H;GY%{U
zb`Je~6F)n^s#g|5SYyF(2*0id2(ZXXLapp?oi2ctL?w=4yI(?@R~Jxj0Ni5*-b=Lg
z?eKDyoeU$tat&5iII`W&>OnW=XzrG1%gbDINT6*`F=P;Nu_k*E`1r<<x<bSs^2hvH
z`(G{4kK8V&ouB;sxr<hVRerMtAj|&qeIG-wIhZDmHn$)DJGeG6AX`nx#YEG(i28kj
z1{SD2lZ#?E+ad<CwN=+7)`D8ELz`kGqE5-mN#A8AQe^H1wdZsdc{KnMUekdI=Tt6y
zhbCn9i30n@!3m~6f2S`QrjhqgQv}tTWV>+Tf)!{~tw4BiS64|XqIn(mEP-8C`pPbe
zHkg{FO-xLzygq_Ue=;nLW_8*X?W|ayc-ShhY-MxU*y+a;gKyxlK0y9>>vv6eu-y&E
z!nj)7fi<zP<{D7&?(Wj=YrdL3v!Cr8Vp%DHhFz^IWf(_li`=>jTP%<crMgk&ye{5j
z;_DvPc1yp>@!@y=M^28w0x7@9En36=mWO@ELZVAd0jvSK-=5w!WpvrJ`#`%e<kYtp
zUvd?HE+}ZGOUdB?4SF`9bDF;1rk(YktKZQYb%p}n6rCx~-XGzgr>CdCrT0DrsNLmE
z00wG(k2nxcvKcvlkph5cFC{;Jv~?Bti_os`Z!dt8C(jTQtJmxYGbOZ_IgEcQWzF>j
zx*RaA3qOU<M>pM=&o=#}S6f@<Gh%etN|MY)A!N<9uqfyl`aUsCgG&UN9HF-F001r4
z1cTuFTXCpdOmD~BTsBB5d!up7>{4fD3zqPm>}WP&&Z1<cg}yh@GOC9+txgGjqqVIE
zKWjyCNv+zWV+VOAb&q%6BK#}O&jV41Drmj^71`0l^%b|rv5%PXS*><dui?t5f1OSJ
zM6gW2>+7pEXU9w^FN_cdZQyBHluBg8hF=Kfx*k(xK=o9>$XS(Vbzr5>1A9p`UDYH0
z#Q2zBT|K^7Hn{#)6xnahrp<M08*9QI6b&^swS=yB=|9>f@M*Mb=f|K3Kv;LCFOiW_
zk;yy%6<g!H$4&1Ji0{IGPDWN%#m+9*oP`?Nnns-Wjg72KOw(nKCs7yjt(R_`PK<lC
z^Ut>m9+k#qO6dodpgTL?(4BV+m-?_t9#imh8N{=AAEl^J(f)go{vUp9p9E4wEWDE7
z_^So-8zUN{{!%-=9TSfQ#OUGH_jvlQyL-;^`?Sx>tMPFnJsA01FSzhmI}v~23A@af
zL|4?h?d!!9EQz-sPvsFPF69R``$Zw4%A15o2BStq{GOZn;HK<3B=L3p+czLe@ejL?
zhg(6pdAA5bsbP32{?oNz-rKuIO<lbh*2%o3M_8FaCh|E*`mF}3*{8R!&o!u*x~i70
z7o-*|*JEQ;B6VxGK%@jr(nCUXc5gc8*FGfou?OD$gAB03=RaZng}K(aK!%t!yNG#U
z>4soQ@Rul6LlRO{+3Nd4D$tnrVN5yw|HKr-K}-?8X8Jp(zGK@*kn=lLLy26Y^CJ09
z;f6&^&v8Gl%fvD_hd6j@>VN@@1Qt;dzEbAgwzz{_22ipxQpc~y{o{M*g(M!D*oRY>
zeKEk-BN-CXF-k&-KY-sn!*_(Q3AoEM@5H{%1cwhx*R-OW&VhOYD$))_8gOC_G<B!}
z>pT~bKQ_bzrK(v`apH1+WYf8DsOsWw-oz>szuwyf&=`u?53XOjrMM`?>8pg;1Zc+9
z=P<C1Q|fg#-(9HQ`Pd`B5v%L&utZAf@=?w#Fv<2XYjfacJ9(yCz3+@QvMXMh+DaVK
za5El$wGtn7m_uMZQFpS-IWc@215Ie{)Za{DekEV0_v5nYug7{tf__LzN06FMTCI`7
z{S(TFSGECG`}ZLBWh+9u*~J^)#(B9*#}V=4bobqg@sM>!r^~aNP-{Omu_Q$!#vo-k
zU><jiK17r2sQnOXUyQML`O4~Q)@cHmpa9t?6Uf<uKxF}-A_;2NjH05hNqz=2Sjv*s
z4XY(%9F6V5Vha?ZQRB<Og|a0)I*`!*WEX&V{hsij`p?%1(*ustwPzpKI~^1iMcr*!
z7%rBYxqqu5&<s_y{cBe4A{ZTTc#pEl$i+DH#RNs?Z?4}IHZ%hu!fG(vum6<=-Jdob
zQkBgCMV<H7YoWh`Nd*a}&Ca;W{fCgCdUXZ#mKAE`y;jHgY6BMASsGxz9hQ5&OYlnm
zQby03*H|Q(_MC#uKabJU7*VaPPtOp2E5m8|rin%*;Qs7s1Hom@u{@vMEJQP>eyg4T
z_cIAvBwbzI3h?VLcFCH?*9Q}gB4DUmv8o|VD2je5L*<a-?&q_oKsu=cMS~d?q$-;m
z12Rc?pan}m{(>q9(K~zvy+dA7U--aC77_5d`rT5oGz0CIX21La-%N7qDsr&6zs)Kt
z_$N$nUT!JNdcyW!$Sww%4h<*-j~rIg$XD;q0Ul%j*{Pp-74M9ts#ydnTeD`*)-KDr
zED4J1#bSfS7GnqW@~cd|_*HN4T<+5Gs~KcAGqJ0F{YsQ(SetM?ej`zCN@YPIqen<?
zLdBXBPrS|?y>!Si6zrzv3O6zS^alP?Cg9KR0gP%M^Hyr0ne|xqfk!KBGl(Z5B0@P=
z>0*V)Mw+4|-XzlYYZ@w9-PYQ$B=bumOK;}kWo?(8FnPdvg&;lU7-))dv_es0xg%Nk
zrt{iWLknLpm4@_m4R)NNB~=R@Q9GQf)>-*2ufAOyM+tY<_c#e-Pa-W}s_vLqQ<Jms
zQOzL1v4fi6SkWoigmD;Y1mA_ACfK$6RMqqK_=68y&@+cp??>l0tE(1#ptD*Xkj=K8
zxqn|azMt(~Y~Q(Z&9LHk*F&^oHvH~HPow-EWEprkD*-c7<W0ae32MGKx>fXW#qut9
zV53K4lh@_}QMTx>0s0##Hc`mC53B!>y}_k&9vh1}KUb$&d3d@PZe?e>H~zR}y8q;k
zURj!q0Fgiuz#Eoz+|)nNUCJBGFe!tnvsEsfKrcQ6N(Q!fR=T?>k4k19;PdsaFt=qH
z+v{<Km(zb-4YWA?DeKPo73<^A^rUfJx7_ZNZWB8H<tKO4t>*qe9~>Nq-xwnPuODkS
zkNWE_`Bo!UX0PXFCLh2SMSaU3JV7YfX1(G(Bb$*|@lD%-cpQby&`|NwNfEX;H9D7`
z8e(ay*RpOa&W;)zB`bRmcnv*<GO*np*nG?%sUzw$B+!DXQN9i5us3DL0;%lUp;=i<
zp1J?0PxjYNDl5}bh`#g%(HFLfZYxsp+O$J1LOAqH^6*kA3C=R6ym>S84=zVn3*V^C
z;Hn#u0LIeX^GK{fal_mK-7Ahn0y>BlgVjZqeyI&T-m7a?81f2_PzmMzpL9y|Ug%BD
zS1eVWS#ud;zVYX!AjnUw+q~Pq<=~PqOOrLdS@jud_^ub8NNuG5Q?4Mzd7t{h;}2S<
zBcTMRSVuZWeI<Xols#F<K;t#SeZAnad1KOu#+mW1Nrvv(8rAy@uOzBq;Vy-8>8aDp
ziIOLZz37-P3S>6ZZ5m;nWc9HnKe1u>@)~E*3u9iM9{7;MB{jtBcp+otw`jBz{tDgv
z@;VSb@&Hovp|LRwTsm8Qfa0ME?t|3>P7F#{ulj+UAvZt&a$bq8)~!<IwvG_vl@=?b
zwlGv?YlJl^U}7c_;<kAOv7Mpg`sX9-moG$}9dm+=_goC~-s&TZKHzh;aBzz0Jivxl
zg2!5suh@!G^HtQ5p|vq?_OOR|sc5{V9AB%@dDZb;`Tr3VBpl$N%4{AF{yIL45*Q@-
ziPR=U<ffNth-GiFWz=s#-Tf0QRK%8(IpR?`UMGd2j{@p3h_dqA3G^RTmMk4_u>Lmk
zlQokD{;wAn|Eygsp}SWckpB*W!cB-~(9!KqLV~@{ocRl(7jMV)glZD81`kE5n+n(=
zlF;`~_WR@L=&7{{Joc-$Wmi1(qP6TeL37nLSdrNo45$3Ht25;oVfq3OZ&n}(1+v}*
z2Z*y_fI$cDCHzdY+`+WU=}X$Ps`!){^{<$UEs|zS-P9$^s_)=+->{(mQx9$C{0XGp
zylMBD=Km{@v=0J_Rl4mzfkXm<WIx?A$iNE0k=Q$B351%OmXYMD<2|D9n+C@VGOK5r
z+DD%V@)HUDm}&2eP8`fy*shD}?VSx+^2v$(4h$~mTDWrV<cXTugjpdc;mtu0yoq7M
z3<-)#>7?QBk(d6-s{OrfN`8Bja4)VWLa?3cXW0ZE2tavg=kiEoC&u!)WNHaFuP2rC
z_GLRlI@0WTe>I5|dBZTHgEWnIZL&zKYCqO%S@~>Yrtx+@XMJkrb5`c?9db7`S?D=l
zu{NuiSE{cYa9l~Q%TW8gc1bAb(L{_*k4$P$y}oFY{&xnlMKVx}#-^r>JdHBPr-A%7
zc)wpx-is<+dT6{hsvRF5h8-C>l%%Y8FqqWXRWWw?aNf>#1S_nZZl@jJj%kIz_*UG7
zJpoNZMko6#Zf*gRc#HC?_-LcJ5HK4~;`qPLzD^87bbrSw-l<i<1!be7qm%NH2FyWG
zT?xZ@lp;LptbBBrbM)0^fLJmZ_^W&tw}otDDMgRPS{(eq_a&T)J@F^p?o)H{a^+9V
zHaEI%<i3-mY|rX1_+WF0zUf@L>}^zX-Uq)Yi;1nQvQm$UH{bpK{lDljl?TJg;kX(6
z-z5j6o`jRP&CF0xq^W{(_vWozE&yC#6ciljG!vjj%R333nPzK}&F9Ti2J0L@_u&R*
z#oxojN|CU|vSs9*=`JXM5sbgm!kILXBoFX@E#4}=OV^|mljcxZXic&`5!K)^(-0AY
zji1r&pb;%QQI~ZLZJ+RLWWfGs5|Q5tdi3PP<YWd|&VnGeP+RaOquWeZ##8Wf$$%P&
z-zPOKt(oqr(9o3ZME<Aul6ale!PJ7FP+-jyqP&{K={I0-D{J1p@Dn`v(U%tJ$Nb>I
z7p}{yf3*!I3M8J%#Zq3@^2p(P4K%}DB^1UJS@lH@f3%5XBz`>lcoq{}u|}3zsnRFn
zo}^di%I9WKZvO_LkoWy8HUeE<#-5=#oG^$J;CKS*M-Y^}hSHueR7ZUvGj_gs>o}S*
zE}YvO1$NbDR_VkXn1z_%41&2P`W@d3Va%6`5<CQ-(EOOC%<B~5Q%#<+Cw{Fld>i5`
zwYJ!3->9k()sWfOnJkfg0Vq{#^%7a>R~<}|>!J976J!1&T(^6zxYfU}+xAHtuu6wb
zdQVqJ!OmxXryRsfdCvUY+!~<ehS|(zB4P6VunD}=lzrPm&Oo;K^rI6SU6eSz1*#+=
zAd()9w9xM8J41N|h0z&BU7_I>Uo-~IR6$l1NjOC95D?1p`zHDJV(tm@=>s=t)uXbJ
zS0^u_Xl!xOD&ArV_i^L;%OgF?zjSFnycP<!oIhPj0?ru{l3AoJ*V2jxA1q``lLhQU
z5CjBQ#QJ&iaicM01_!fU!v+CZ%v=qYKkTM9XUD9exbhM??nmQ;D3&?7WQ+y&Y`2Op
z=VQ&Ke0gwRc2*9Vk0x<daH|^BV{;T6xNt4j;FBk^K?nKB#oOZ9mE55H4pmR10#rG_
zD?Us)Np4V~Z6k@k#!2FL&aCsn01{XB+c$Woa<t%+I{6HZ{38u?!Oc&F;yh3poO}J2
z6qD}!V^)n4=q-AwqY0^9yr)~N5?l$fJ<#0p3Up7y1<v{q9OO%dT;Ss$@3ZW0Ili3Q
z*(M{_%dPq7E%FkTaE>N3@4eAs%F4sdb}+D@nk^eM`@or`{d$oy&PIy)=3(9glF+{l
zH@fyOQonv5_BB74P|o^c(b`ko1)fAPm^K4iR3266CoToguSS>2a4U?I2kyF}XWbce
zI=S|%%Ns87ucM5UOTI$1?q=HywB2(?h*se}Xp|%!H@xcHO}d{`Gz+672IRtjG_@;*
zWVY3(4f(-<Pe{J&Y8@^7T=|W{?YO_c4`II?{nE@!84~?%V3^g-dK{VZb%UBYuV^w{
zI;`y(RU!wUb?jH!iN`>BD|83Po$`C(Emka0SFtGxV+;{MQlOH(pmAkr$p<6_#s$6i
zd$R7VtTzu@kIlAZ{e4s?cU<gBy!D<=ck{AZf8Mlj^RKMTa=97s$yNOjb%E@5^}xyJ
zDnH37ieFXwj(C#K{vKz|W1sn(jE)w)xx+{vRbdC6zT-(P8}>SZ*DXxU;+3TU?Rdrh
z)K?4Ae7QQgIYSKwRWyU|!SDkpl@g(wvV0Ntc_Ecc#|uY?(@Sdx7LAg-yHMX`m6m5(
z7@g7s(>=+6p>R>>YTY_=RJ&)sT>X7iXZ>skOK4(-Zh#oPnD#4Z*0^uJf2c=i-h&1W
zu0pgcVZL<Tu(eIkkEso2M#s%ze;y1Q2DtX4*QNH}-8`N6<lMOQvu6r_sPV6JPAOda
z9Y|*_FtzS{>`ga+dx>J-*C@Ec+Wzf-*=>6VNZj5lncwQ1vwgqWXYWn>z<v?&ECREo
zuWGF)&`B_%SWyjd5rU*<G2M?7QM1tRBWb+Lkx>tVSWS5~tkMfLyEtD^ojgY7uAv+J
z8MSaMy`n40-fJD^R6_B6ppiN!$oV;E`Goa(D68i_bJ_VozY!qTO6m2hY48&#I-q!c
zq;Z|^Q(Yi{ohg~X2+-DV{1nfS*xL0X2TKvx($np7ii^Z=#|bzhEul#8lymrYZY&Uz
z{*X9lK}h6{oru|Hjho9f8N@H7t=6AAOLgP(I5;06C759U3LkSpX-EuCZW}6D@0tG#
zSMGq!1GsK7j}w07E%D1Z27tjL-eD0`D_}$L?D1oJ0|89zD4fSluN3_gUC98dTjV-p
zusdk1@<vLxywxXmZX=`Yem34>75V^oXh*CiC|gPMp6d1N-B4Pr!hgP;?+?>@kDch&
zkO=x%3C&eS`k>6)%rbS}NYsqRwKM$wT<c^fz5`RIe20rjG>Q$;M(4sNySLX9_p8=s
zPvs2jndpg`jeNOIqu2QSm$k{F!L8K?K$P)(rKE^q1s6=>v^eSf-o`hY7kfP?aOukf
zXa04HaA3QAga(&|Zsu8+l8J{m^DMpc1e130OUK{e1cK*cs~r?p9kV8Vzn%jVCqIM<
zTmgDQP`~PosMjQ^E4`EFfgd6LM37+#f?#TOUNvXIK68fDdU-_zN^^l$#-CCj7+uoq
z)F-XrI>Btl9W)nFyo(~Ho$(ufm`nSwL8+-_I#jWExkh|t<3>D*1ZR*u4qyfsaUWp!
zsrjx->MP;K|9%IlaBYSwG#?4C01Z0?ybCfB2NduvOWkmXdj_y6((@Jf8fAe5t@Bm|
zmC5_4Q#pg6-Epf<4}$#~L*)%qp`uTl-jSi+OD3Iw+}U;P_Ls$=1mldnn1uoas+7jF
zO1tW%w`WvG`KN;N)y7<mtuzqJ>p#dvm);;RA+6Oja?O6bPe*I;vkwu*&*g_{_h0=g
z+k9Ivk+pu~67ez-)Cl1OJcF!XZaiP?zv?TA=lUFTYb*l#nnh;S#$I$p&ud!(X<8eB
zF(hQ<y@t=%K1>8TZ*?b8oE(i{v28lR_yx!Fp+ZZ?eY|&=YzMv4pCpTL$P1p+^PwBj
z4X3<+Z$!tp7!uvFmnwn8?eOVf&&-<I|NapD>t?<rgR{num1vOR<vkMH(}DlE$bnlz
z0sz!!%FgQ_b8h;FKU=0ZiS-rg?YlIm_&^*!Vc&Nw=&ndfCt~(b6SFA5N()Uq(^E2<
z$*1sh(lU}eH|w)m&VKq+P2pqw%1^=;?{+2HmY$^-1UwxU^L_EVxIkJv0`3ENg`2oe
z{LgR1@V^G<a`j3J+7;PW!s9n71FQ&SLgYboR}DBz<Iln|6*-mg+T4I}Yx3L?V0OsS
zbPP1I6kSsbdNG2<Ya@oJtkSJ5N+YaCIrWsuKT^L2Ac`;yICeZbGIb-!joa917y1?K
z(7T()7^aqvcr>TnY7l4#2}N_W_vTkD#BfMnPEHj>JkO$|FED(SNMf%~yKS*sg5DX(
zHwYLnfbXe4J1(R17~Wd#-YAiqsIw+UqdK{XI{hy2jObFwKn|AkcF`r0XVk?-9Bkhh
zOm`eeX-Ld04gCMG!Ti|m+lD5$?zV__23dMenY|Mc=>q-Ee}{Q{Ac~|4+GCP?ba+ez
z>3VQZnuGa`M##%=Ejn|30V6~~ehr+#yq-PF0vO@Tp8vN29)dT;RJ_<o`wPS>#`JrR
z*`P(wdglIhpLh1dqY@4MMk*l$(372NmkPfnJEbxO)#61?PBkF?fOIp;YD+BZb9f^5
z=Cv^pE%w4peKNHD{h#UDj@sA@>s*Hmmg#*z0g^GM^Zb7GE-1f~xt*pyuPlIFn#<x)
ze{9z~*?B~G2u5x#cSlr=K0eY8RCZ-_?6bQ`A0$K*Os({A`PR9L2;XW<s|vfXSxdHX
zNk{UiJjmXl*LX-^t1VIn?T#+FqGtJedUe_cNj%(aUt{eJeHm=&p#6PM35w9O_@=G9
zx-Ka+_;}|@hSS8^{Fh|4f6yQ8%XQp>8CbNqkx>F71%#0#D@a-OTeeJvKt4E7<5dZ8
z_A?M`Zf?4F)}*h@<<(k&#cD`!u){!5pw|9m&L{Ekd(GwL<*lyZQ5WYG5gwkCovGct
z)AEsg2bO{rS@*Jo5vwtnMLX8?P;py~X~u+vgL;$SSOF$1k;{3h`}?f4`jF0??5aea
z937Fp!aTME8ia&|-7d{JUS>tJT4Po$G@_n5;3TF6-Uc8-o1(tM(>+(zY|m}v(=gY%
zi5^#KgzJJ*d=QZR{1J?AEMb<&IT8cbdAeBzWF#b(kZ};#yw&QKO_9unYxF?_A*!R=
z=seXN-*q{STTVGxv#O|hXi!*h09X)#fCVByrHXCondf9=0<DJImaKaRIHI<jiJnDM
z3$%Xa@Y$Dp>CGmbI~3+$((?Mp#6$KF>zSXUK4w4KP`m+bkrE((Lu^6fy|-rN!T3;)
zDSkqjqhui7{pig%squ}(upTONVE04Zs&?m$_gOfcrsJB*#JWL%u`+=t?T^J!9uRq|
zb(9pQ=!0Qjj4=!#x8}r4;^nTsCk1W#*wjn5)ORVw|61TQIb1BfIUX$*pU%3KU~;si
zM&tN%$r>rtl=bYD@r$icK4fRFH=g?oVB$w|_v6{CfNK(6uDLCG;~$Lb?ZwJgMm*DG
zzM3KjU<%Fx2#fuk!y0keK+x>4#D;tM%TiI`zLhzzu+u=cjhO7AfGVI4W~CvaK0YD&
z%ad!OURzC@NdubNhUdrYEqK@(Vv@t%8)9O@C+`z0wSx#P-dK`%AOSR&7nBnWYDUv<
zF*=n^JW+sgB|3#NSEk*ShGnR<=SKJUO1*cMWn%Uy8!70*<_1gi*lRX#n%6=uH#T&d
z+(vKt_`Db4kVE;BP_y|BmjXv2!Z3^xGqBqi?RcD4!&9@l3Qw>iXid|A6}yTsix>bV
z2$z8GmxL>64eA+TOuEC)o!6~^Bb4*cbDn7up@N$qPEOBp3JK|FY3I2vb3O?p2amE5
zDxfK6!po*DT=dE1PO&AUFa?N`d2I(KJgccz&1|UXm~(707Twy@?-ZF=VUgF$KwzkS
zXtN-<HHP|%66PMO5^2whjn(@>xF2r|3suz`YC$g&MR#+@8~QQLhI_V}KE$ahDVE>f
z2TwM-TYu1BRehQl>3IVf6dvGX823@})7G6TGMAb5tDtpr+N0H2KWD@s7q)y)2o{@D
zL^Qbe-((CR=eBs<5J5A?eTg-1(cF+&z!UtITGKu+43&5GV~y{Jf=?3)JTEfXskaHz
z3&k~;uuM;Dc_y;3jx(XVl3g|`ZRVYKLkLTJ%}&LvOPdoIu5<w$TiR;>A7O7DRn?;P
zfy$Pxl!ziAf(la7h*AQ&5owU_mJ+0-y9`v2hD}LHgLF5dl7bRShk^)7NH=dTk8(Ws
z8}Ge8&KP%$<2}gUYt8u8RLP7#HumjPsXmMImvZj17iY@5DfwzG%$*16_o6zIrRc&0
zPEY&29pmw>dHv?i44~E_Kd+3!(kl?+-`4;4+>qqw6Y2wjzKg-$J54dTL;XDVUuBWx
zv8mF)@0DO?W*u{Kep8XJplRn{<ZeG-X5?6FD9_&a=<HEBLcxE6S$JQ1o)M?%|NW9Q
z?zI{3l%0Z?WH&u+(e{bB(DP)(6*5?<&|H2fvGQe(($JNXzrtR8F{B!`0{Sn~O*Of5
zm)#jGevBLcuZVZ&7lOU39Ii*_1DYA6PwINE_G#67c}Dufm4V?+G2<_U^LuqFYhrp|
z+WSCv+{w8~B}Ea)376ka>(6@ufTGfa-DM3hi+B!8HZn~<*I&zLi3-I{f&hNiIVZ-^
zjduOLl;PoFmhURVa9YH<7EZl_Ffdy~XU5HBcTGJ5w3xR<9}(S?Qc{{pVm*Z{6G%kf
z<%t^)Kc7+OvmXAo)4lJ#x14z9`x&ypNgBk&2d~ch8GRXXPr8Apn)cQuHZ@V_T*h(0
zIl5kvO7nKlh=MMrl`dFFX^vvBcEr>qU+xRxu1)>!TH%PjwYu8aMZxu9`{z>Hf%M17
z!dX>5J`Qzg*sz@ibFTitymWiuB>P*6-Dj`45;)w~=OaT^UQq>*Eh~P#OaBsL5&{E_
zMf%vq{o;TtZP^qVEpiiI3WpHhTVo!}a+->1*I&GNvD(I(|4rU|OF3<um0WF2D|5co
zh<Y31hQ2}Db(Xqd7?F{h%1`Qj8nme{h-lL$pgZ9#TFtWU6JNU>2wcmiwYDjR1l=ln
zGQvJLwe6Zb35p`m(ey7K5qO>B7<iOINFb&7<n0j|qUpYIUJJGmQUOT9kVTR`mpz&`
zr~>|FbzuYWQ*MyF0X_;RGwx-Ox>nJgmU{op{nV%ifJ|>|Pt*N$gvsdn`ny1PGbb(U
zd(g@6Kqs#R5}XOPwOh)!^ihb}Z4d&<3^=bmB0Y@^UeiL>V1rleGM<tAxnJyjeq2%I
z?%`(rFw71A_i+w4Ntl!GPw9~n(ma(WT_t5zO`}L-fiymI-c&-j!lVms`NtG6SMh75
zBxpiNENCW%E?2eBoAM1vIhmR{?=K5gT!!@TCBg9Mn1SV6RawU%QC%y2{UsB6*<{!0
zyfT%z$)6SkpdL|t0u7ZD1J&qlZEB9qI|_#t^)3G5bP(fx4Zg-|V{-Xg7^qH?8Il+|
zb1*od0(w_uWO1Pxa#K^`*revyQoAGaq^=Omm{R*ww^mcK%ML_dZBWzF6q0Y~9bZ;K
z>`Y8>P$UwLT)PW%Ksp&zHwL)B(;6aCOO^#d#b8_r_m7SXBBiQwl5n!!C|y9fRIoq3
zyQjwlOoUnnsMR;3wK5>EGXWvXps@$fIZbvb#^hw<Up9Vz(KLBiP%johj?5@ew?3z2
zya~J{FDDlXQrmXOCl%spyc@eUVhi^(gSVeu?CGhD88G&c->iq*F`6J>^8o7g+|V4m
zV(qgZusJQ44Ta|UC4f|<KPQ0h#c@^^jYnH&;Y8xS^LAzlZ*qTXfU}PrPYf;k(~(&*
z<-rcvjX@6N$lma5>uh9L*yo;!*=c(?+SCnyL)m5bs}$UV&hP*+viO8<ZQARD?wkpn
z*G}@LR5<l`w$Hv%h!fyn@E!5*Y+Lg7yGkwOMf+ImoR=KCx{&zjw|hn!NA2<6V2Ed{
zBzB+Y`0<l8(C&EU&0P93@&WuKJr2J(IoT;~w3AI=b|iVr$G8qSyi+=jH>>HltM8N$
z;KH&G+W#nsxqa7@=aYJ(g_h=r%2S^)++4T(=NMuys@nuN8oW`sc8iH=^)gyjTbwZ)
zGuGY|aVD;E7G#l^Myx|}_sqgfM2wNg;#b{%rC_k_dum45yqH<IK0}ImW{CECIaO@;
zX_kL^LHNV%1jdf}*lFCTvsceN!bsvd%=SuwhT@yOK1^uL7Ps%J)mM9Pa2=u)mYb{v
zeUkC=*B>L7m23#KD(%uBx3AgAMyM}ilq;3d&8#%Dfil~wGk`$yJs7|>tDRoj7qz=V
zx@^glW|k1&@+la<u}9CjGz?k)?438TtMb?^8it&TB>)s$=U(IsB3gtPp1W%412D<5
zLY4{^UKrz8&f@2@LWolD5u53=%GjD!IAmm;sn3vkh-P*SVhLX>D-cb+L=jhTibu@M
z40l*^*JbRm+-U&m)o|SRo;A4dowrNm)}G7O%by3DWTPlxsh<)Eg~b2RE-;+RMw#4{
znBJNDWco}3W(oRBmp&NnUzUMAzElW6_3mWRqlzL-D{QmQi=a6z<vimq*x6G#J~ynm
zFJCbgr7aYs4EBN|GRbD`YDtL(U7DQw;tFvG3M0V?T;d%^y(lZ`2$MMZGh1)l6}T@~
zbDBII@Dp&TSL`LsdUX%K6hjPgVo_k0xjEaLMH<v3mH5{Vg=t_+npefLs7iyID%Ca9
zrw6bEr6deaq9*{>lX$ZR$LAM8X1M85Bm4zaXlYQ}*WUSfUt8WfGT{89#&FK1-di|J
zhq2ypN|_5a!C}znK8I_Ew|+T%Pg}{y-zJKhT;c^I$1sD;{`-A#G6|dkof#(^u^A8d
zP0Mrd3=itj2Ejk|Xmbrc5Pwl@@?^}V3f~CU1U{xW|M9cnJ@EBWJx~1zM_BhJ%@1_%
zgXIfiPals?hax5@uyUhcCT2XTnSZWWA2(EbmkFYa3Nn-M+OeDztB*DTyR9<oD}yU)
zoS-LItLXZ)wYQq@E%lIqti<Oef}FG|>3cwOMZ8QU%zvajPx?Kxl+vCCo)gxJ!{~ug
zywY~UFS8KI7@xX-{F-f7-ZMi;XX8y4YPzNEmp@mQ-2=~Tn)AYFLKPzzsxnIMe7FGz
zl;|yP)W;Uox~*Eg>;lnYyKUgr^D5&+WF<{KDJZH;8m4Tb^|U)5Qo0L`Bp^BYiAs)|
zTyHeXI!O20z(=lA<P8jr)*?Rk=FlTbqr5>Sc<71hkDqh<Mhl51>eM!4ki{Gh_Kh}3
zN0@}r>uwZLCf|lztb@(Xn6|e)l`~e{b>%xOBd&`2vJmi;<7AyV_GC67-6F6Tt2#nK
zc1R3;<7XKW6|XD|igAYPNLT_44HT=z*-%(Sr{cIxo>m;yt%P1V1+ww(rL0!R@-eUK
zvV`#cXn}L{b1L-`eG4z4w%wiHh4K5vvlnm|cxd*RB}>v%f{z6tL2rhgKz}pn3{RWz
zK;azBbWo9`VX&J5JSNC!jurP_fy|<L>Y-1M#b=eAxbUgvYA)7WUuACRV`}TXZ+1Ux
zhIZ^@ci{&uE`O!MOSEI^>%@<gXpXL~=U3@<#9<e%R0LoC+cNkd+{cYKd0uJ=%Wcxu
zGSTF3Pu;5<nLJ;blhnL6PMk1bIrv>Q5lky#u$=HSp&fS3wQ2mX<4=;jzm#N$Yj@$F
zN4cdn(~b>w1pnLa=LnH^NgA9e3gw>EPeXCW`}!0$lU$4gYktHr{V#%sH7Y|cktwkJ
zDa<5(sG;Mjk_%D;?!_s3LGG~;VRdJ-XM_WH{yg-*5THjWHBZ~qk3>bVC-4Mu$l@`z
zRyqsZ5N_=X#-o*UrZGDSj7wMfZpjtxfdNqH)}Xo#PAu$6_D)v1a-Y=`5~6ZvmBXi}
z-^ERHlUo$+&-7$mt|8-rlwZOlwmTJflU#Wwv}?sw(Z?1`*dX)>aYV6K++W^<3_0`g
z7tecseAg@$JpH`aco(_C_I)TGE~T@w&@=!tcE`-tBYPa)p~d?~L#|?*PP4tO0#e%b
z^HtwV4$<%p+loH&g!5UaR!y_OZ`gW?12&v))HQxH-n$+3TeGY$7djowE&8PI)|c0-
z_PS>JkxX_a>K`Gd{^pTcW_>dz+dzKW^cG4Wv;26@C5Q7FKp3<NuMg8f-ih;!_JcXd
zSeYA2gELATGi%S_fIz@@+*=hcBC_$uo)j<Qz6Y3TF|?7Cq-03+>3zANEA~S1+bO6E
z7xh=GJL@tR8in8#<zUwf;deb5v-Je0;Xgou9bt*Se<ltRPq8a-Os)+!ANs3<%T%vC
z1)Oii^CzR-21@T@Wd<zn(-2ciN^(gy3@zxoJ9=O{I2z6P@#7_{=vhCLn#8adG(}xk
zwrtFJm9_ry$KBQx@1yn?l{>t>^EHYy@?H`|Vu<M^f+9}PiB~uUvHu5GmcemMtXI$1
z3#!YRmbz@(VQVRvpGou6d*`SBnou@GI-VpWo2n6PcO+|RJlET}H$%O0wRl=hliV6}
z1SN3VM>-TQ`r<#&Cf85;Bo<r`vhj(yd?nxZ^8Nf}csl~JX3}|6G?Imu^y=M|Dlpc{
z-_s5>12h`wBe&#3j#Y;pv+=sg5lRVcRORQI%2&3QEC;<Eh+tgY+^umg+&eRn3}8&V
z2$ZiR6t3cy$b0LMoHzxf=DP|yNhKk<$x^;uP{T(8DXRm-`kWL;BKYaL0pr~SpM`FB
z-&+{NF5~Jc%*6Dm=W1NigtHlt-=b{t&dzmn3M(du=d9OT`19vaO)pJrh>zt4&|z9M
z4Q?zVDU`uBs%^O=YR*dGVUX}X)%^+X#kE5}VrTg#X4;-6gH44Hh7_-B!&umE>g}4(
zI#)2=ZKUX>pm>gj|HPA6%e+yY$bl5b@9VIs_fFm;BTB{iqBsWLomqyG&<+8>I<pCI
zAi;DZEYqmzF}Lb!Lg%|5N9+_DSzvwImu&Py3$IziVdxapi3(1@_I!m_dzI%~?EnBF
zGSYjg#WMgGc{A<b52bje!<6^6XdC`|j`)k8i!M`9nFjgSy24Z4=g05j-VOraRb%KY
zYd<&=0WXQNsoNA+Nuri-htr)?kc8S*a2HHFDRr11=B=~Cij6u*b7}>X%4O<XB>qJ7
z^rtU2@TB-woH~0Fm7%O;vibGq+}PLJ)-daN^|c$X-s?ZVs7N1N^I=z<nM%CtjVNtM
z6S!a*A$)J~lQ~A<KV@eUH9HXM_Q+Vjch#@Tda(pr+~A2*d6ol3T3mO?*aVmYs`B&v
zuKpOQR)-r;vpjcPZ_T#M7s}U`jlQS2?#-1{r^51_v-;f^eq5548x7g(gZPsnZyE}z
zd@*<=Ewm)FTi<uas=ja=dcEc2{>CXacH&q*=LMt0ys4k<s*nfEcTgq=f0xNPxtYmY
zIqBAbv)M696p?=u!k9%uSVRmL+q=Qc(ISnhrLBddshTG31=stTJY8a2a;|E^2Chh&
zQ7iYYOq5#4MFB~R-%kHhET4Fd+q|2;xy<v8wkH6Kuu9gqAlkRC*cmfv>8BAKx>}uy
z*B8R%(bNjz{l!`!BTqwHE7-``>Gj4138ASLKU(Jq!i{C=H0j8jaxG4J=R8*w6-nzo
zHB}VQWREA7rrJ3M`chZk1W;6`0;Hl0gq0iiO6eugnqxVynYN6J`o$RK`AOA)b)DlS
zl2(xhD-Vac8*ku6<-P?NC~wjDBhmnqH?NKN$Y+oyDk2cSomZEe?PpoRIp(?keAs=V
zmZNo6u9>Am(llVJ_A1B)uxtp%B?PWX4s(G4E_)#rK_m)l6I9#uAZ@u@oFSjcF;eZ^
zRiJ<N1?p?&A&Su`5LQ0kO0fl#rwlpSjtF}AXNj@Xd}?~SG&l!tqAl;~<9#0IcvbcR
zr9*dyU3N2>-uCN`$tL?=1bb$a6noNsWJykPZ*9`I05{4$Dq-u0C5BgD7>+icYcT|2
z12+H>N&Mw^=>L4Vu$<sU({UNSC31zd-?Nc0iQYm^hc%ZiKdl;jIUhWPg(Qf_Ro-W^
z+r!H+#7Fn43^dlSm7wi1C25W}c`%>G$r-)N6moc)5k-EACgeW)=o^{vQOy#Gy1m_K
zZ0%cy>x0>TLQd*$IRhxw=@<@_LXAvZwr+lFQ)gdjp9_^Semj0di6e0I>g^Kdse?ZQ
z^vmq8<~uvTavTTuCj8}T;ht-+Try+j=)LB%Rl&t=IM966G8IJPGIg_6m=j|MAS-Kb
z4$~PE`6L6sAz@azIZJ8?98%647kV65A8kO9s+?5(Qf@WGi!JHkov&WFwh3V{x<zdI
zb;Dn%7GSMC9S?pw7Qj-1WzP4#1p|93oP?n4u4NDc$41ee4;P?EK?K9xkZ$*I{dlHR
zcEcfm;<I6J-m!ABuiEfyv^QkxEcGhOS6)$kt`F#T_(=xoiq!2YLujbh^3&;~7~)7)
zb@C}HDyldW=iaWY*u}3kLhu8*tp=}5(_480nn>xV-w`jMd=9z}k?V6(Bnv>X(I6fR
z7dl!YW94&v#0$sE%`f1@cB~C}B`#YP6VSi#ORwMrPUFL9&DVtax{*3hcg04_cx(sC
zz(o*#&K7`27YNDOY&oR0nI4NCIeAi+Wc<trT;ps!DPMbv{FXL$mQ2x<+>r$OEcI^D
zqC~YHDwd7I(6I;@MEBWK8mU_*;Mt9mzrNf1G=s$I59Nt;Pv1iXxqy#?JahhzFF{E)
zK(dNmuZAg50u=$1mT?$jfn~@LDmqy(TkC`pFy^uQAwhEp_F23`ZoE0Nwei;O?aPb<
z?!Bw*`~z&A(@nj`jh_;#6=&60e=J`zKHzd*ffZ@*)XZ|HsdANvO}Nt1ZGVgOUa`M#
zE76dg`SsA8saiqd;Bu2n%F9!bRgA;fA7f%akWp8Y#yGmgy*V(2A+&OdYx#bSlE0V{
zhM}A#h&QS<{z<OVsO-tI0O$$5e1Z^%iE&vn<7qf(JHQ|VGH@|Dmo-OMfOd@mK;Rgt
zmE&ITo$p|5uRgWaB~B9$;5b{Y`?`~PCzP;UG`+1<$L%k0^*kKP!zTvp8ALcO3Hf+*
zwIDNFf|ep!)+Y^5kMYE-haBbuP$W74C+CECIm%c~=v^^_&Usg$JT}$ko<xdK*7|U7
z+*qwXgvE7|JQhKVZNg2J;o>%A^8>jo$INWJgS>sK!H|LtJKR*q$`o-nS5k+(d8GQb
zsWN7EUvzdKwkMr?x1s>ykOu#oZnu!t#mgqsHALsH=^!0Eeo|xIi6fJShb8*a_6?Gw
z&L_&=orI@^&3qQ$#5W+zyvX9axn|*J>W(4dtnGx7)<x$}Cp}p$2)`~?%?So3?=gNb
zI5w$(;4P;Op*gePt9y_4_k2|lgn{Z-adZmM{oKrkH=qtXHSMlyNYP^W%1yv_$ItNo
z>O{kqQlZMXqheQxLi|oZ``TQ`X5{+Dmdnu`m46{}28Bb8*US0fkh9>Ooa+vs{Z?G=
zhR+=L?!6zfzSS3=g7(}mQh$G;$M(<9#bh)z%-|Opsuy*>STR$_DKjOqrkQ#=-RI5q
zm!K0oD*XgVAL}Q^J1arcfh#0;sw>do)dPHP2$Jk5((p<we^TRy5^!O80W%6*s|-ut
zPgt5g_Be^)EI_E4_-&o-kl{%KFGz{KQu_C7b`X2tYh8w()(%UQZg=ma>w|MyL@%BT
zp>q^`&jMZZoXE3-O$5ykK-JAhY%o`2k;`YHUXUTwnqR*++poqd>Q78Y_WD9bK#Y;s
z=90<WoK_s4JMm-pPrA-h5S}*$0Ib7ix2dV;l@wXk#EfN@;zZ`!&5`XDaGyP%tzYgg
zGLg}&^Qg2BE87F0rFE`+Y(8x7HK}|{AmwG9?qNQa#b6MlcQClZZN3!0kd3>AZ7vn-
z61zfK4aYhI@_ng=&;g43!Akr2OC>H>DTJn!&2jH7GKyAFLKBKjmOVQ`AL7QK`_cM3
zJM4mkeuD8~#T0-{^lnfvhYYRV$R1Rxl=XU%jptI?c)CRdouX{hV+D?;hTS{=VYf7t
zlVEB%cUa|9rlRP4<@C9^I%k;1#LzTOUg@c*xor2W<SK_^v>pE)0~-HeOJC|dx)9OK
zgljA9D3n_5r&}~f3FE1+;NUppAZ16xdmH<VX{t{^P}b&7t;Iu=SH-1PHa>;#XpH|w
zz~1AQ&B=#eTX5N1Dy)ApZ>1!y_e0^`*c+t-$`)yn4=R3P8vKGIpk!t81t_Kk9LS#|
zKD(|)X<LWFk>-IlHkMcRk7%S;uY~dyON}Q~mBKZWEQncnq}$rss+6ffi*@ks%^>jj
zQP^w@)?*HnZKI|kV-xQZL6Tb#;~WY|{>O)30j<XZ$W)zbQktC5)BO6iq=iMcjN0nu
zBtO1>EP$m?-4_q9ioHz`b{{M@-tPHvDL6P73omXf%-S_&l}c&SOK?e}wYcjVtJu?1
zKV!S<U}#=Vh;v&W8nWI;Ox0KJyl<$>@+<@7BaUb~?9LEB96!(Gixl8gR8%a2n0uf%
zgY1ipJRx!=wtCxHv{(B*0R_j8mg!(J)<A#v>nfO7xHeWzzkQ}<0zwSBQGxvDg<#&H
zyzW(FA7(TH*jWEence5t_{@4W71}IuFk--5EYR6|G6ALD-Cas?h;7J0NzGb-p>;a%
z0n-EDT8@b8(t6(W;#o?(SMCeo8qpkev`WcS)cBi$I)r0`0)<K_aLjH*8uN3fBlp0?
z+!nxD7p)nUG`jwJV>Lx@!?O1V&CwUe%oM)$=zqY|_+&f{g@#?jT6CObxDcM~u3f%I
z;ZCN%@LNX5EoWEPB1bVJ)?|l2wSwj`?h#p;kZ0B5$7~9prmg;j(WJ~pujzXV>dznf
zB|l=v$+^Z)#BSmSp86R4R1gP4?(@_XR)3A(R~v18n#;f^@C=W1%8M75Sy?Wwxxx_4
z7t#bfO`6B%634D;cxfNS2;8-v2%8VyQt_Rg%Yx<Hj(O-j6_n;E2(vZ0*mi!7_(Awq
z%0$rFJdt$@HVqmmmL2b#b5F9UWYR#}uxnj|A1GbzG)HX?<=}ay)O2S%WCYR9#<$Jf
zUW?9IgGW^bhKPFtUOO)7?>H5be8A=`0oQ2~rWVlN*x&u+3yXq3LH|UJ3*ltaQf3k#
zz;Vu<9DUiU;fQr^Xu3wUR1W<^0f#1@f^4Nx81r~}FFyCEo6zu(bsvV<dBhRRrddS)
z?LI}Nd+}Lns0v4#`9HLcQ%K6xv&5~_w=cbWvgH1XSxV^3nFiJk_tO>UPIXW)<PYdc
zs=oB_ZQyy86nmdN<UT=N8Hk3N)p9k0#PW11FTkCizC4*e$8Nw(#75)|L}*T-?WQ}~
zSwak94@(0FpH@<{!Vf0TgGnSF4RboZywJ}F^9IDTn$kKt@kpNcw#xVUuD1B^`00)>
z4hwXgR?%mP=%A3LFCp_y`aP21_L<)_Y7+Co5H|{RWAjtnmlv#~bp1PH27PtP@E^cp
z!~hGDg+;B(fwEIW(ct^uDc8N{@6xZ#ji#qfzVXlbKbz4VgrYIL!`wp!YK~t{vO#+5
z)vp^cB>)wISX-uo1m|10o(y1@a$}h?!mCv4dsDwdgof$2l#KG0?HAWn6<=zV8iQ<t
za<)UtZ_E2Vcofqhb%gTiPl@Wo6RCC7E=!YA*u9FQ`nCfxt>Vyir(k2{5J0?;XRkWg
zrjl{$=|DE{*&&N5WEg>eh8WldsZ4__Yo9d-oec*TtqsBeSpXPVGY(E<8TV<+K;1De
z+bEW;g?V*ne)m093-Xx{F5FtT>+OmoIFyW{xcCXACsr5Yx1DdOC`6O7s!0M^Hyp4y
z<V4flR7>hhJh=#$NVN&yrtF>nR)-`BhzC*6By{ki9ue9;yTgBpj3z_`v%Bxop2)1F
z;6L`6bSl3G%RNael8EuG3FV#`ph^>e_@3Y^36y#jj5wJGJ5vC`Qo7&u&R8B9pj`9g
zPrWI`!%g$4;!B!8oJo@bo4u#Pny>i53Q<P@AB4m?*U1xe1EKlNE^Qfp201>&ojB)u
zmm>_c_iT2q3V3`@F?_e@qHvg5Ka2wJ%n{i}L(bnQ`L;6YAy$Hl>m}v4sm#O8d|y1F
zNZXF1ni_u7Q9t~7^V^RM^3L-NJ==ZZpC8`ee<nGgJ}&vENMMW`N^l1%1P5#hhGdw<
zMr2vF<KxU987brZwS3V6P;}IvR#Fa&R602t<m;!kK&>H>X&15?_4R996fe-x>2~ib
zCbB5hp522eSX<_5BxGeMROh6l<rR?QundfH&6xjs7F5cubrne$o?`(AmQQf~N;yrR
z_;Govor!`&$F{ZFZc19wdMypuy&}i&YTM~+n=pa~#EwVyy!mm-8IHAhZcoB^5Cs?L
zWMD|!wr+NUOnfA!H|)B%mqctinRmqRk}>CYSOBs&_`1uiwAXKYjIK2s3-0gv=l{t9
z&}Sul?NU+B{@#~*zJA|}T|*RNB)E5KB>JuQI5mrIZts2{wJtlsrQ~z>CY>WQ70wv0
z6ZM8*X4BILH$|B&H$KLWzl*#sj<(Na|D><Z<svxoamj#SGHLJHb#U#ze*3m+_0{^j
z)^MhF2q2z>yxY3-H*ewIP@#dL9mX|`D_m}*zZ$z>=KqbW=MTVFfUi?&Csqf%;wFey
zOd!&S($CC{A`5mmN<YC<tZHo4Z({6*_n#p{hB*uwjj;1FY|K|O-<M)AWd8kBwD*MG
z($<Aap>{sGV<_W-LvVbh@UrZYAGL*cBCapo&kj+%v7=PDDt2Hr_(;)TbzLmMyLpvx
zNW66GcdejQ6`DHuBLMzj9{)m79vAlrgS^$qH~!YFn!_xiCBVMr0V!LkL6|dU!qRzn
z;91B3nip9YOJ_A8L!KVla4om(-yjX6=kReM8aNJ-M);;Bl+FYW(r}n-TG2ByKFuf>
zeJ<px6ko5=pDstW3B0K8Qz{r-c?+8uE&&)8Tv^VTeTV^eln8uM#umt{5i<QgP@It|
zrB}xxz^uGbnZ;M{etrelj^F7=H2LHdbiDaD#+8tlsW*2lfA(4HJdR<J5WhcV2lesw
zbS=QQkY)rJY6y%=-L<ZRUE2`uAXY09gvKextq{K=6YEB@C}*$`YG34GpE5TlHL#Dp
z8^Spzjkln~65Z5qW+C(PY4V8<+Gqfw4V+jqb)BN%MujVE*i$gezU!QzzK@Y4?9JEL
zN0xTDo<%g{)h%%F4&tY#r9C0A-37yC3*dCnEWJWsNCa;klR#4i2L$x&@gUuyLyGHa
zJ<pFsKfK9t8^t!!2AzLUjr*0;IHk1X;fH*wJ#H5{q_5ZSDPP;3DwwhI`1Y}A>4#^e
zTt1fIfURT2BZ`iqp>k%T8a*{cSULGc<XEcZy$dO4g5=PDVS5iUe+`ot`9_eowR2Kq
z0(Qj+NE)fUgB)xsvK-(BbSkt*YH|(5bSwTMq;!?rSS0FJIfTN^p~}?FyG_W~tc=Y5
zFn-w?egnP;UVvr>_MBs@4__k?hcT;GNoM?2$2NwSz;#igP76T*FgKj5!+OAy>E2On
zVFbsv5SbB>;GC7f!^*4?kg?pc*>;p#F{BeTnHA<+{pxH}mVhB<DQF9!71epF(1_!;
zw`n#|?VK-D<tbX*hlDX`B?V%R5^@;{{n#z@)?x^`Z_!06FGLHpZ3*AIp93HmL{^W?
z!rLMZL^ydT-YKLL;}$X5e%nB6ni761=pgSm?3%D2V5KJtZlkh-CDMVtvd}I$X~B~V
zZPyReDqPQ%FRyglo(DQFv({C8y^-1n6O9Kl15wcLF)_n78@&>3eefQ66J{n2z*p>K
zo7wP*rJ)bP!*7o0uQe;>sIgaqAy5oGuO@P?s7n27@YUMLpBw#-cJx0}%0~olB?_ea
zMts?x7H~>E1j-8I)onV3<ud%qajVyouqC2=w;qx-aQN%=!3dkSFzGVjG@d)-=|0s;
zl`8d8qvw0tjX#eaO$bVRJ)>oL!IC!(f1Q_|f%BTqyNIgH|0LPpP;SURPAsY%r-z7b
z6ArWPz!kZu^WHlj%+9T<C4mgJa&n!V4O&9WTx2HS+nm<XP8rA#GXZ<eBabKV;O2Rw
z`O;)-iXSUKCz{~Y`%#j#Yo5A7=sU1k_G)LuQ^+^6^iridEsU~*>#<YQF+WcDYaQ<*
zT&U}=yE&7v4{xs$s3z8p6U$Z+-dsh0Rz)v?9CpX{_oOY!0E;idlga<|4rmq-Mc$xx
zJCua<=M-x`#1OPcv3eP*smYV4Pwl9U9+J^P&-Yd%n6#aEg+?e+2q>xMQ-)9!S~xP@
z8gdJT5(hEGod+x?7skDPjWRgeisYUYy&~_YB^e`J@eU4Tzr-m2fV?C5R^+R<YR0xR
zjVD`@;V{XdwtaCEayps4MZ+m*>NM^vsU*;n3{qvibP}exWrWfH{Q@?4;H-lTBO$BG
z89aNMij9-K#h1x0YI}k#<#Kc5Sr)b2OmvTRPx`aX=cRoc(d;2jiicquEJE9ji}~SD
zYOx$6{pY<GjgI$mxLkpYH%X%NW$I;Vv%G%TG^2Gy1fhD3$tC()gVBpIqi|Z0?TnS=
z=BcNDmRJqNr2SspffA;H#dNWok2|QTnM|I(Iju)Q_8+YL+b@ja+tZ%QQ01L*<io2X
zTR|Vf0q=QNm>G~bDU9c;IcMAN0zu^<PcFY|!aE}*inCAe5-XeMl-2q@TgaO;VesW_
zK~)b>WU26^Q1T<kPT|6@8F#9?G{5Ioi3(t%ozLw(+m}EN&4AD%?X(Gg?3qM~o@+4(
zkjde?Y8@^7)ooB6gt+Ydk9JS5u6KYJ9gz&J(T^n7MIlOxA7;`vUh9wC&m843G=pO&
z^O~g?@+VK81g*eObrVTS=qFAmd-1kuq(K^Hw4q*XfNDWPM8v7}^tKw;<x~*NQFcqg
z-6YrvwzMObNVxRIjh7g-G5$9-t+%lS3xTQnJawH0xguzLm#J1QrE;ysBldVv7-Fa(
z5m2>t*3FBMsm1&IQmx5|12swduUoT#TrkSI%a{ND>S?IaHE6p<<K+3m!TW;!hlMh?
z^9L$N6}p^N4CX+*>Ym+Ja0R$IahXQE=-<3Cm&ryJE{-Q^VDqq;{;^438H3d_<)v=_
zp^ofMMfrE%30kykXC+lqt<q@^PE3+~MrrR0(RJ0w{j|~)cc^ykQCfP{)E9U$miwR!
zKoOt`{6)?MKJv-C0L2R5CARk^UhjetkJLV#fm&fnNf-g9R_faBiZWIR-pxV!HHaUq
zPR%*?@0YDp<sjAjdKLhYO5-QcsYb0Q!@l_+lb=(2qct)E<sLSDvU>@#`GdMp**0Nn
zOO*`0U~##}1ga+B{2EvxxStpWiVEpci@f#2!H}ccm+`jQRnewl7tRata85g{JFZ5t
zs3sC%e!~A<8g$^)njPotCEi@iz32}**NRCY2<LGMs-mCmUo4dcD%kY8vfFYspQb#}
z`V8HWi7fb3!NwFh16Y!Dj)nj6bu1)8vxl^Ux*P#dm8zYr*dI%&d24SC9)X9X)++~X
zr!c4Ns{scb#z$4=WL|*;>tpWaAM`jl*9nYQ?K_-XcOJ_r;+>|ky~?2KCBtt5AfpoR
z+%-D$n~t`ML7iu5UsR9K9J2s2yt3Vwd;sqeh8V&)QucU*pM2QAK|uCeM7OyBEQ&U$
zSuIQ0ZIrT<7{-WcDE<uHaIu<&=Dyo4WYc`fG+P-<;5YaE_VM`wxLl-YlI2}@#Q7VX
zTSgAX)~QWlJ%a11)PkPddG39|`t36tuPNv;ev_APyV0Qg52rSMP05nZ00-w>ou&cF
zPXB}8`XSEcedm^7hAlOf>L`=cPlr`K+GRNodc$M%mb_euoO{@@1(65r;vYL2hcTC_
zV+sC(I`9tv!WIs@RY3r@A<MYAQu*aKixW+VxUN$wXccA!bDN!e^gG2Vqwn)Zd*{u6
zs_2`BcM3-u%7czSgqL}^&T3ypKQ<0@2&XuH48>XXQH7|z)M5aj>4Z&qyKw1{UgB0s
zQMemx_!-vqG~m>VAVwg+@dzmT5b|pY7~{(GEN}}a1o$6gNp*bi;Do{;%GXqfq%xUy
zC9nf}++HBhklAo@Ho$*Y=BWjUVS*e$7Oh>r-yC?U-frsYu+xy0R!9J3lHRcyFa5MI
zqx@VDxvK4YH;RBeg)v}&g08IuJNKv1xC^hz3;R+JmRYfa?DW~L0=5eZqaKhPY3RcW
zIOls^ldln;x|p;}oxRHvA_4WNnCOrlV4{jRB*XX3i6HWyQ8s#9DSxn_+ogi9^C^@h
zDj1qw0cUj{j$;M#C<6Yu^CilV2_J7m3RaKnB1%LkauP`aZ!s@<$%v_+Pl~)G!gOv`
zTllJ>>+oGuE*Ptn$96AnrTE$ec9P?Y()c6GGPF*|HQ7nnH~aDk9nQDl;612pfD?4%
zKd;@0E`I+fejM+fFUTreflDsg9yl>e*e9T&V(SmOw!0$jSa$7n-8gzCvz>DlRc7gH
zpwHw!S~?>bfg#?iX?f<;_gnj?Ca8SpIEoh_%?BUNlP*~1v_8b&?u;HSJh88?to4!E
z)U8(KASXSVp<uP^^3qQA&vcbfRp!jD6NaD47h;qy{g<w7`2|X}i1g6HV>x~v=|Lm+
zVy~EAFnSyKfw2&&J;-pSS#l?omw8M$e^8CIRyE|1d_cw)cbvm~7LofP=oi!AiL!%E
z>mK$+xpwy+sMp}?_hqXabw>VHV#4;LCFbF9bn-Z&m3b1pzkMz?w&Xe#rs>=uNBKFY
zA3J7mvTs>*^gh@)e3ZC?Pt#o{oggS_Fz+oJ3yU-o)-KnPz-tvt{;hGBhUbHxTr{kW
zsoNY@+Gq7a=rt!3Fi!~~7@TdDz=<irAR+)J5?=D`Jb=vy8MZ;*Duzbc>UEr62B^hi
zuKZID%EMt{R2MnFd2bJuPT-I{xC61v?Vu9;F;LnCOK~f#2Q!kFQ)JyIKxEM2mA>+|
z{FBjsFKG+HWOP^Bn{6i#Y}FttNGP&Pkl$9hAxM#q!bh~Ipu16@u63iz-M*dn)7B<V
z^{@cYwr%X?!xs=+f=v;ej1>VyCYts+F76Hk0HXEu%tvwV;>Xv!h^bi#BiMOc>|DlF
z_msn3p7)a<CxhA;UL;hAoarOVGayzw4!iI9l)ZDgdbL&aCsc7zn^Y2fR7YuO7^EZK
z&$HeX7f<p1Gl|fQAN4IMxv~lHOAh*!`ueNq*dd#G-CkkuVsex~G=rQl-f6nu5*5=L
zW-?!De9#0)cti<pKzDYv{wywa&G}95+Enn92y@TxQZctLrGJz;5GVU+qIT2Ywtp3V
zd$dET(a4NA=*t32@sTLEziZgNUp4HWQx(Mf8L}ZYjE|V7Ro~N)t=O#0qPu$gfyeZc
zP#<;x6e32#FT3Cj3?l!OuckH!V6SiVOg`g@9e~4A{BYv0c@7qi)pZw}$;Uwayl2l;
zk8-sxH@}s}2Xo*NQW!N&$|m2%sjGw>Sx>^9Z_YHJU^y*nDpT*ZI|&nP2U`Fyvp!mM
zv8*GGPag78!VJql809g#^}*vRS_{KuAL{n6fhiH<xj?1}lbBgcMrScV>TPG8hbQ{Y
zZpN&=5`G>RvnxCE&L+QkFN-PU{#8?Oc`R*w2_8D$wPps9Yb7+?v(%3Iy$kW+$>|c-
zj4LPy@e9$6+L`0)Z#8*fH_9M5(vu`|SV~sb!KSPR(gc<OhA@+4!ol%IgvQ6k+As}+
zh<>fafS4t-$wJ%Ke;M1IJ7REO3oFQt&BQA6WRcxfIJU>BrfZaW_gD_ctc@Mdgw#Sx
z-jg;gCnrlZ8O$o*C=_tUh0sGi`X~QlxA<W(szRneIw|yrt6n=5Mnm5s^YU{e>u>3E
z=f(;&eNJ^?v)>^)oIYMa@efV=U^ffjhYZcVtrS8wBopUBs!>iS^~j^(%~~4g*wQwu
z`s&i=)Z8eaRKmrfAEI!Q?jL~?%>&%BAA)q8vU}Tde&Ws$G}OJcBdYo=@6pnKrX<;4
zQ<Ch)3?ncs$vzoUBbP2+O7S}pFYK;{l>{`rgfEJmS?TrndkM4cmfe+~{NkNt<M|c!
z+-4XmU4Eu0XYciaKCJBlKB>^TZ&@Ralc^}i%tJ)*RfMss7f~2o_ukuC0*b#4R?>+N
zEdrO|J|y9*0^`<`)5eLPRNuovY+OYVej#@8AXk99GPjDWGwmgx%X9{uFd1xLcd#~H
zyPU>XUwdvl133HHcz~gUcNeKW%itz&9mwbIc^7|2r1Jo_**q$OS@&fv{zp&$NSJi2
z)QBgO2D5Yf_$$1fnSeen0!EziZsqZdLtZC<PUQ?d<%fBXm#Qa)T`gNU$q_6c5s9Bc
zKXqem&*@k!aW&7CHu8>%spdTFS3Yw4A)|M4IY;d@pdOdNA#lD?*x|Bujb^Wf945hF
zNEJl5RSRSfK7WWdAxKS4m9(_XZMQ#1=Axb!Qz&$saME_T0YhwW<x4GnqNOn~IW>rj
z?DzLPgD+M7l6`!StAFrinxh-;*F9SWQ%`S97p+&TyA2GvpPTx2l*k_*>@jL5zVY$s
z@h%7$x!7N@{8?a192Cy|=CO|jKkwhm4hiEgDxPu2ffNV#l|sW%>O2%{DHzL4AU*tA
zRwiFzPV4GCY3|7JHMKFoiY5|NKdr&aJ_&Pj1RQLV`*d%u1CZqi5bv}DC~7Qkx4H6h
z-oRc9#O27W5rIuegK?+`Y-m}EFCKOIXPIpy*GY)ZUhSQdd8kR2G&(2%4JZpV9O;_R
zsO=rE)h~An#(Z3;Q<gJU$01Zn6lNEX$eR7>WbL*c(+%W0CosH7p%5G!Zg&BuKNV0H
zt;I-0nX@|vGtML^=}xV#soB)K_g4OFp$o1!mF4FJ416<#oE5@0KuHRx)n(8y5gMj4
zTg4D(*Rt3K#SVX~D>3g8NA7U18<(AEDFh*{03$#Rerh!AbJ@8gu3C=<^hiJ%vQd-C
z%zXh%<i6roxyktbsC|JQ%~7&b){sM@6>>xPIxY*I9ths18gQ2D+p{yCPl`3Pp@2wO
zcC)gaa_Zqdcn?qGT71f52c?1I6-W5@DG3wNg(NMFL?*MC2L9g=*FB?Kt$4=TfMyP)
z7w`Da%&42Zch@%lGH={7bT#+xvC5ZT_<v&?4S2rn+np_h;a{Fzyeb-F6&o2feLg!D
zX=;B}e+*4E^36w&?AcPo<Za;S+rj9VDCm3(m?w_&o-3jBOzvKf)M0a>gn3nHt>5UH
zqpBRrrZ+Pbn&#D|oYn>466`F^oW<lSVnEw0`~0eEnax;&$t8s68U~yUk>yuHp=E$Y
zLoXPgfr-c53$RWtgGa!hxy=<1CciC*?P>6|+1GGQTBs%`d7g-RW?xz=xhWyh4!{`;
z%mw5;l*6Civ|Q2$(WBEf&+WPLlu^4WD@7b?pjy+>xeC)NrcexKcYi$@*Abl$(2q~I
zb`A6)w3H1tAOxBNR`LVNn}1qaP|tmihZ2}S8NKymtF<AB=5o8%Zg0mGI|4$lPhM|z
zPg+sU%C&bkvcmMt1dw=`|2el7;T)qOf-e(yI<B(<)f0wDC4(j*z|*3U9Z<PR$>p-8
z#QhzKZO^YewXq$e58_7bY1!H3awMLGE+eX0#YRC5lErd(qV~wsgmB>|`bZEx<<4Hz
zxJAi%tsK2fu9EMoc6UcUdtH22aUqwwMMQTwkXk^=d0N${9tNM8Ye7&*#$y|Pg9VMH
zxF7%hHg_14{LbWzzg;td1Ov?@Z_B3N%}mo7eQf!edbexcPAg&i=dzx~MTP+JYNlcc
zr}g?xsco41FdxD3Cg8tryJl7yVRFp3hdFC$`1Lb8mZ_qwV{KGtnU(c3S*K*a@|5ba
z9ArcVlKiPHH$QNgWOD}p22v_gi~k*P5LSNjCW4^ZcQXYc2~N3eA(}oY_v*ga`^Z5N
z_qZ7#2gKwqs@d?>{(MVS`ebj-qs}Y)$p9edWIU7=fKw@fWK93~sz4%(Hvnr0VGW%*
z0z#anLobM>y8QgClyVdQ<(x@d#Lvna1xMIR;y`~X)?2~Cw7Ne5*o(z}?}l2I(i^iE
zx!l=Tz4mvV1ppZ5<v3AB6=*?(bY%GNh64a$rg}4ddsq1DT3fGc6$k?q^&hEB!^OWa
zATlg?+cIF8KS}yx;apaId+#>Vb4Fh6WqBp2Bs~HuRtAR8(y^m!o@??v=GL$SuF1_Q
zqg6LRZm!eKJ+uKc_7vQ7aB58%WlJCBf=o3V9ksc(IKF(|YjwES6{c{3sohTTGs!%3
zn#JYm)}}(^4G&$oQ0ADBz3>rO4_QI=%YwX2)31Q}zqzaIHQIT<W7hOMs*V7LlAKeU
zBzSVZ<=2PU$MUvVuLxqr@_Uyz8g(sN>e+VuSaOcoJq`@)&C_p>WW_$Tdpm+C=h$?s
zu6!X@g40W8zR2N|bos=2doS(-looG;BPl^1<~l7wjIfyiSV@Bt^Eb5IDL~;*3NEL}
z01rHkTmNj>29?SWkI>Q#Hknf2&^G9qH>Ttl7U|B99qMW`S!$h}{w|#`T*gFma1R3$
zLO_p<AGd<|(pG=hCkts^m4gsFq>D|!_<PiU;xcOG6m%xJ2J9j9gz4w3dVW_W+G98i
zzk)y!NgzWL6oOQd`$ss;W2|zuN~H;xCgLT~o&DuDiS{QU$rvCZ9lA#whk&!;J|S7|
z>09t6lPApK<1qGyNjH6uGdN`^`zx!4&#8B>RC#8-&-SH8T%8F2<T>LflYlHIHbrvs
z^3GCbhjDOM@R3s<Z-dY0FVNT>(3p4|^o!v5KsefLbYh-*egdLZy>@mwSoXO@erR}W
z*51W7Bg)@zq5H4|q=QZB{1*K#7pB3&!I3n2rs3okiqPvS{dv#ov6Z0Fp45nyIfZvC
zZdnIcU>31H1dmHWCsqG+Su4nGySbzBo>F>r9p5`w_K>rGkr*^S4cCzh-*N~+uaj9Y
zKBzHqa;CBPmWMvN7_?4*{n%>EKCD*AYqZ9u>BRxohO3UVlVSNtgfpU|yaM+@X-d!)
z;zV;l)12{)84FzHi&o-dLo}lb`mybR1ba8nc4vpK`Uz<8gn*c7hX!%%%r6^3<na=)
zzu$jVtp?R`_;(Qq!Ne2#hv)-AViE<{)sUtgES2<=3!pRHB7I1#q4xWU8c=@p=wUI?
zI<JgY`uLCU8sy}Tgc`f~Z#(BQI`eRt{Jm07svytY%=LHioJA#8dRkjT8ax?tUtRv&
z?(QR;j_I9?_`i5OJd&h_a^Kgjd`xz<1{9g9IZ>8FW$!?A1HvCf(|Hej#fA#hzfp*A
z2c$eYMKx=KcexE34*lOxg=;FHc4_MAix?C9I8T;n(J}v=){!y_RZa84+cf@w)|%U-
zznk4ZKOY2MW~JmCXpnkTY;%;Iz+f^kH*Wa#=ILrrD;gi<^OF=*VKNl&(*KmrHhi~Z
z>>|pn7mXo)grjWRPY6~(`d^#}$yx^EA+=K=vcqN7df*lp)}<FUJvw}mDg4<sU01lF
zYyw`)j)!iR#?{1(k<^2<kSMa#@NV+{kmjxN3_Ve4{59l=(pBEKi&dF_T#3KB$}4%K
za5ve|e*>4~gMdRg1Rhl8@gDXUC{Ru~t80DWNABk<?fUlZP7DA5$aHX$l5(b~%k>K&
z!V!Qxv;&v-JX~z|c?I~E03Ege_3qrz_PP7|{W*O%e;+=aRpAdDMjvDMMnr7p3Z=IQ
zkE=(1l#|g;6SxgtO@IihRu;ze%Rhi!Vd{g#`94qTqW~J_U|_C7M$^0}mIuES-h&75
zq};fDdgkV7AZYaj;j!6BP5u56<O}y5o8LZ;vKnv}-@5e-yW&76y{`AwF>T4Bk$Oqo
zeShik`ip^sn;X%b1?J90D~VAC(r>Kduu<2JtMR09{Kt;zyGU9vd%tOs3f~RR&`w<V
z^;(_z{aP8oU^)2u+IHEp88Fvrz|SZZo60GEhfNV=MN{LA_m>c!6ByR##eyGUC}E}(
z545!`hQc>N+rV?%CGc|X%?8~+qLukp%OTvYlKcx++7X2h9`>OGUcY|b0?=)$M&a!$
zPmuC*fL65k_b1Z;Tb41rc>iCYjC<K(0R-|h6pD8^i(`cL+MJ>=Voh>r$7kDGEs)jU
z$(Wg|qRV?gG^^C~chzYL`Yq42b0Y3@lr_G|5rc4XLC-!oQpfzgG9JRwM}zI<{in5X
z5QLN90V;NCgZDN+7%ez(f1qZ{%6WvATIi5%z+EZm8u&iFkV4ClQ%`2Ef`63G$A<x3
zv{gBshKZu)J@XI~gdW{@OY~T452ePEW}{a;lAaEaaW6Jf=aRQfz+({!LWu9@)?iIU
zG;#A^_~9|uK@*U(T|7!m9RkM2ig|;76RT+GajypLh~7%m;mzeF6u%O7`{B}QwQFcP
zl6gd)m@tPSp178@f@FO!mN&GYw~AENmfx=KXKQ!gOU~|2*PLNn4m}7s#!GE@J0itx
zFi5T(`@YQ8d}i4|Re3h$<ll7%PyQFk()A(f6LcbC6z;`EtB8A`w$j1lC)BC)(7fvO
z<09P3eP~~lw+g{QGVo}dfQJWBbLcjQK^WZ)QRJODE#c<tqvtE3uglteOSjuyLu1N2
zPuIkCF{%h>t{9$-9k2=^;NA#hm+*Rg4J^D>!CwyHlGwrvZ<{0=&DHo<?WK8u{t6CG
z(5*&-)J+QbN|VW!P+HIP)gba}G6&9$4GaX28A3$y_Fum*#r%G$P@bj=*I`*$492u=
zvMvpS&St~h{(R@;lcyohiS<(15mSV;@qxdakb$nRZx5+)Q=V}X)oIF58B=0&nlvLL
zqWplrk#r1=2lBvhc}P4+fIx^$LIm-%gE7C_q4-Z9DzP1FU>@!|w3SHe3-?vbNxuNE
z#mdage2q-C<c=aD_D%(tO3k}Irc2zkBvnYipYQL!3>@)~YjAq1zO;47y8Ah9g1NTT
zxw@C}%1<kEfCp6oDGu=8FRXn)tSt@FTF*T}RcG`s_8P?K5?~<6*cEyxg+<_xe->*R
zcwa*xZcdPX)O+946;giB4kee}pz&9rQrsf7{zv2LZ!1VT!->Mv&_ev(uJ`O}bfl^+
zECkylABSdf_C|VKQv45A#|hU&nOc3!wLMy6zX<)VJMZoZoI*zEgM}X2FES9tl*V`>
z9gK@mr@~eQuDD-7B*{<)Btf8_2J4hkzFw`teAL)bQ23Y>GZzsWZu2ogwhI=BCd90F
zQ!ZB794IZN!rNk5Dp_G)^kI9W&#5PD3JKw$BZ)$3?s8PP^$jnacsND#-W*q!d}E_}
zmwVsaUp9}2Kf66rCiQ0Moh*e5`4Q!w`Gp>>9(K-O!cK-WanqbFP@?e%2&NshtvN#?
zy2saN`()PdQN!<%hq8E;5?j%7Fyeu?)EzXWp73D?hcLLfygnc1uvTQayX|01$E?K_
z{C~~RP#dOj$@iU9Mt42(p-h$P-E*D!@cnOrhi~p5WoU6*)R$eP!F_}V$HL3lY@+y|
z50fS&=GQ#%(RA(KHuCGUK8+}0FXcr%^@&7R$!TyN4c0HPBRyuL-B_at=bjaCsFtK3
zhIxBO6~CN*@|3H_OP{qsn#Lde>WI%6vBuT$j)dle3$e&`*|eFL6^|xKAG2Ev#5`~1
zo?D&SOZ{NFKR8@88U!t4ALv5vdkZrk16lMhug?8#$wK5Zq6WsGw_~p8!nF@8Q7bj|
zMZ@0qn*Z7K<Ng#No9f-z(cdMgK1%Aq6#naX_F+P5&`Pgb_rDwCum2TD1~`|7zO+Px
z1!_C^IW|LY%%IeivO8f|u`<=N@U%qI5wg@|zb$o^{8Me%p^&S)x|ymV^WhuK<{O*U
zjpEejKmv?Ln_7Yop{!BIuWckjy5Igi8501L{`pyWo;cn|?JlZhhUkD$4got5nhLHg
zzEXO8Ja|JitW%FSR}R;XD=^CM#QWcbN_-ML;XY_|r*6%Kk+ETV2~y73!K&fVN@JzO
zxKuaSUBx&#tBaApzc@_)zp9?Gmd85WcO?%qlTjJD=&Jnp1p=8TE!{z{|L4UrsDqXK
zYn(D2%J)?Cc9TfV>aH(_w{gJ{dofIc@S}nr+Wcb7r$W8#J0&;!6W$zMz74)quv*%|
z3CG#=AH$W!Y#IfaKhHQ|7-5WZu@#I(2eIQ1LK>jmPr*geFkp~nK&Z#S!Ep<QLIz-R
zLXlG9mW}Gj!q^Y!W~=9OR68$xBvxO44FDz!5MdqG_q?vpCr{)sYz~vvWAAb<7ArnW
zlk^?Ov%F(-IS~&x+(&hxl(dC5y7R&y$Q%qA3bx;CBJP{Yc&ENzFKp1Ws2QLK_G<oH
za;h61=}K|iZjrAWi<T|iRn&4T&$o~q_*W_M>(i+~)NaY{g8vEZhi{fk4|usR&$4iE
zSn-9H-$IPRN=;<!q_Q@_<g*C7H_uT#{3zbnk`D8diEVgqW!3N|-~V<vSl5L!$s?|o
zDF8*Pf~TNCRWL5Xkqa4Mvoh}D=N`O_v=@;}ZIp2SOQvfHvWuW%AxxHp+pWU_L~!nR
z)aA=D30G1^*$}8{SIZ99E+s+hqJ-A9(@fry%;igs67J|d|LS*1Cb>5A4YC+<>l?yU
zn<rm(twoQ9lJmW7YhkkHs?}pTurI}<e@n}shZAWAm(=3?E+B71mY87={p$s?1qlAW
z=(<zUzc);OJ#Gdy@I<|(Qa+4QJHhX<<hL_cv8rZ%>ygt<v9Do8Z*HgRJh_WQ#nS%~
zX$L9jW#g9f#Q5+LaoLOt&2*;)SkHs9ycHb6j^~?k@My+a7C{Rj15cI?>A-=C`i88*
z#^MCGX-71hJK5Q@?4U4-gt$uc{(|H+NcFIK@^op=<P~&-NZ4Jj&LV0&6W~IE<}mp{
zBfY1??BY$z{@y-|?@ir-U#!8_x6t0XrHks;8NFb4bXUns`r~72ox(8(8O?Cd0Zh-v
z<16gxy^i_;-QCrm|7YI8U=ZSx_Qn7EYj#p&v4?i^+t99W_Nbyw;SJZ+8?7};qvWZ{
zHEWb~EC%=8HhtYrpxSwSF;F-l9fta<gXGfW2MS3#VPV~D)f`3;;8cI41o#V6-T;U*
zJ@_uWf6c(ZD|{m(7dW6Pg<NdQV$YJ2zCb${9CcDiEVbwhYiVgAnsS7!Cq&sH;RcFr
zL}H43EG6EqrK$1tw>2)4KeJ|gxHQe1>p>t+GyL14FyUp|?Tt%PP}A^G#=QH>OF6r5
zuhGgw+2>kn`f`GX6$+EGR+T9KwCg0?KMI0o1@GmKb7z<UdZYRunGxItLF#fzd9eK7
z&lNB<%ZTu@D>jw{uLRRo&esar4RYq<d(o22>ccz=_l0dmWfKC!&1KD1NYqY#!w|#y
z_pqN|n!iVFgWck$+~eC2KNpYh0e;bLn2O`cUVZp8Mqo(rEc**Y<}1DsL(+VNJ{Xm3
z^s)f0fNZf=Fh(VzGqyp3ISCCs98{bVXP-mx4%`S4gcq!Pj{&eB!XroQP8$(Uh6H1X
zeb1TA9@qV?$oiIGzXC^tZb%x8eX#HTKKCxnjbWr$FOn@xB7HTd`TUdgY~?>6kY#@)
z{oxFoOX$ya<Y{Tn+u8r?5cKf@@y2oDxw$U~o#Y?pszC_RC>EzP;-Z*scUBl&CnHO}
zto+>Wi2wttRChB_<zkF_TVx9>ndQDp!=wHit~*-l73AbC7-Fz47lEWCOQ*8PdJn=H
z+W{VQ`1Ow2$|V83R|IK`AeF0%YB>mPC3%H{n1M)8+Oz=3WDKz%L0ur7YH||>wHqa$
zb-dx!clHfmtXk^;p}E`fWr6(*D1i-2F>&0Y-{fse=bc}-07(w{rg`r$kW$R^)_!Fy
zecBdn%_ph0Pj|W2@IV_t%lh8i!q0Lr+|p&g7prA@{8@OI*o$NTGcO;)G4suxFVE(b
z1QE@DW|{~7P<bXUb*`NB6>AS_S;=2BA-qs#V9jhp*qem9(tnsIbB??4(n-n~R&5iz
z5F3;)(WEVG3&yOco50Zo=4_U*1nc>LCJ#~u3REK<mr2{H8R$@pSE8d1aKR&Zzia}*
z=jJYIjx@I=Uato-@MCCHcqC{j5Tm4>(gXcq9sqWxjbbGmJ#Yz@zb7f4M&0`D|H=@#
zjU3S_sA_oqu-bU$e6yU5X}LSA^z=}BoE|6H@1KL&z|S3vp}#*mbX&T4sPrOx?V$Sq
zZsIt&@xNBh%4fO%!f?Mnl0Rl9KTHx@FTO_e<eIB@z)WQCvfbW_$WZ;^!Rbn6Z|`iY
z(qINI<U=TM`FY!XsO;r@P!8YVt!OThfBN|++@}<r9LOFCLODGd<70p7Kb|O}V`S;?
zgDdDvg3j|1P=1k+fzmF>j{!&g`dvMF2gljo6OKCIB+VGXk*fCYXAy9|^sb<)-X-nH
zsTLWqX8SPn|M*>9481K-@Le|2+g{?4bE~bDTHAMQ>^04)f5POHPsgXmZ2Xt2WrFFw
zjYW;+sQuf;Ac=udX_$B75uxM%nRotw+qW-<F#c*d%X9Jk6GH(qX|k4q5uBmwZP=3C
zVo$Y#-Mc#Z*{KOKD}x;%k8m4lCN@H&$XX_<zUPI*W;T=X9>SAZ$XHZDzBIq{HjFN$
z6Pdw_W?h5~x)xCUcf5=kJZP7M1c-YTF4KOw6GJeGgt$VS0E4jb$cbSQw2Q(%-1r*j
zsb~Tz02`u3hw*42px|}J2Vd$sSVUVu%_bAb^9)b!cdfaTV6#F1wMJvSL4P_egFU+G
zO|OHpyJoJ!IYjn(3QUc?Zdrd$n7xuP*zKgJTp1--zB-&IW<Jwbm(A>`i2tWr|9XqY
zA$RGh|NNXW)t_(qcfoOig7b_o{Rl3K-S+FSiB#Dxt1_E-%7x2y<`H9ueb3P@tR7_v
z5p`k;?LxE*kn~V1`U46u8+vKFixYWz`uqFYmc#yq4IF(D-V~k}rYBFlL8}EB1d}ml
z2|u%jSSp$zEgZxZ>~@ReB8ccaf>U1)P7l(8$)gx8SRy7^(hu&rfMa;?L;yTgt5-*n
zvB5-o^=Q}+WM(GA_0n<HdD}H+;->uB;#6p%{HG!@*Rs*`?de7ew7-Ah4c(-%UYCKI
z>A<meM@}vlZfwVylSqfEUby|=-=0ehslswILjMsbLD8u|#1O9<kDv^h8y0ZB6T8!(
z1SUJ5YsXEh0An-41<v79JRhDy8ekFUd&u^W;lFo#jqqRV4b5v@u#csnjrhIyY7lS0
z?c29+<obWA>aqUDBH+zDr{$OUsm`PH$z<@8CHt&XpiqHkIcfxplIMdBWLoo+9e*z7
zKKXqtIA(NKLR&SYdA_RiFUSgrh=VVE;Q#eYmqLC&;nK(}|I>ku0QA1{5!-zN?+A|9
zaSi7P)CmN4BQCO?xG{2Zmc5*B)aAuLXj2}3CWgg*?MEo^fI5E_2@H)T?mW0uDh$A-
zk`Ml<CCH4k>}0xX@&EAk-tk<w@B4T}WH*p(vS$>@=v5?Jq^yi?dxXr$dTD6N9?3}d
z4v|gMCOf+#d!%HA?|HqD?)Uxq{r-7BydQUu^1Q}*o#$~L$8lCb&i(=#{J)*BhY@yG
z2_yrx2$@;iaOrbR%Slp3q_ttn!!JjzmCLQqdP?8&{PTK;1FPq4{kL+9yoQK2G7Tt*
zIs;Ge=pH+na`u|kzm9X@Rg_yaxcZ;?(>d5Y&n&%wj+jRW?A}LAhx;_{-TP`jZ&}-^
ziYtMwo7u|G@lEgua`P<}ayCcqA5|R-)g|N$knuS=pB35Y*uwv*4tXTV>Wf(-W{`0|
z?NL^&*hGJ5x<17GqhGO^$Dqf4eeTD9!_XxJowF*zr)RHzbWw2?Ewa&-_mDo%Q3?iA
z|N4p{T3qF-Jj(qqVm@Sk5%AQNysCMKQ6QPSQC0HB+Lgth-tN_DsGHL1he1-NnSEJ8
z!s-1It>u^<Ajeck@F(1^fzcR%?UD+5Dx!Zt%ddpr8lu<{BO%oW<<K800fwVDa^Gcb
zDQTK?Sw&Sm^V>U<Xr+NMg*SC6#ssest+X46=Q9?ZYv)iGW67mW(cSD9Mcb-Ik@c0R
zxyx7YJ*+YF=>PMuKjUbzSAvLb^E+a10-IB9ChIlLJWFRwyjGALZf*-g--%D6uO&OF
za2Sx~?LCK|Zu!E;vI^gI#=|ne*RuHg?qRZn?}d3DiZC|adC?M#_gW#Q_J6nrfBf<R
z7*-ME3C=)W`*r^GY^efMVm*{(R>s(A_qvel13wQH3pCxQy*lCgond{o8_RI{&J%J1
zKh)Gi-{+jNpBxQc5FyzT<XXOG+>_$GCCI&L^K84pLTrXzkPI;eQi96(KU;)am9z&+
ze|;Ksdm3<DqI|J@gbVL{G^Xkh3b5Cbv<UnZeFt}UG75<9mKd&&P!C{`2+pcihsYNK
zpr0;na5yU!0Y2ZQ%s^0(5}a1Hf)?6V5g+TD(<L`iQV2d*ddB{+i2^k@NoLN2r2EI=
zN&hE+u@K4K$ACiQQuzmY7XMwb2enYuwxm2XALi`-x&Qhn_p!{%`#+1HUTba-zSBj!
zAaRQB&k3=>R$TbP;XMmdJ=xCZfA+6{b>sr~_OF0@L*mb&aI)|dTeiJ>m=naQwGJ>v
zJq2V=RtF}Yh+XA{behsXyX_p*(Hc-kA9#^Qg7*NO;)LBMHsLuT;lDrbUP$f*GzH_P
zVLiVLk7DNP2?|(38~h0iOTa1rwXZD%przn}Qm--mPDy7Ls7~(;Q4o^SG+c5wXJ|K{
zQz~|r=+8~BVCvHPXI}2`AG8}Sgu^M(aR790J*@ip_Cf1f<0HqJf1ady@q*TSU{YKC
zKCQWCV%pr6o)$w)oY@<6<mTYuKT7H|_Hboo&cFL#%?x-8p*6_27XLF93G1+5=cRQR
z_{26O)x)s3FlZ*fN_>Hs0>fH>Z9gS(l)|qTLhnexvnVVBq~=`YhmW;9Ld=~SXw7dH
z!VaoK(e@QV@L!FTjsK)#z^z&W4kaHb;X@!kFP}Za#`bz&-!r%y^9`cS3eE*C-=4o@
zD$vxtDx@!Fy=?Az;a(GXR6O_Of!Kl`6l9D8(EJ(vO0FYycOMr;4?L&mqYV`==?YTl
zjn3Bnb5@NASJduBER|b$HpWhSY7|y#pbUNg*-WSBLcKAcF@K+G;=Q=5y!Z0SKZJ-S
z#UY=K)wn)jItwQC`)sVugCp4g=Q`+O84iov@gwp*`8CiE^B7f>A=2;S3xuZnkfNU9
zm#<$rG6vsW|AU!)=6CqLbWNjr6s~lRSNa05c84R8aS1SS{23y?=D>;%2`Sc{W)a5M
zHRv)Z_c+c%4h9XF?lgpL2&)a<v|kTZ_mQ!^OlYX&y``WlpW@h_i38mPL@?Y_S^4hx
z$M~#O{qUezO0A=*&iA(`bZsuePaJTx9_v7`eLnz{hgYKiMUn-vjw+n`j7P-Nr$G<o
zO&MuFWAn(I&~XWPnrAzb+>l2B8J&pS?3H2Yh;m`!v?zeBON^$@&c88t7DE_*0grKN
zqg{GI)Z%CP&}@v`KoCqOT-A)>ai3xQS$etRpFX)+`w!wFQpM~l<D_9MQ~_Jp-$9Cg
zTog`yK~-3R^SC6(X)u^LS{`Ro($Au5^Wn)~Rmj}lpE3J|zTa5r&#YGyLRWw4*)`1s
z1!q@^(L;alH>QdTZTPXo{2d65APt(5e$GRdVlZ6@ntN0fhgIdPn(==#&55QmKM;k5
z2VbBOPz?k_5pE!$Z6B0*DeDRH3+uXODvz!^O0J>!Q~-Tc4oBY30~{!T1ukPPU}$E*
zJeda$Zaqjz^7t(f!yEVt{6C_?J`Z3)8fyjRdZaPwFx6QWqWt$E9;mU4i|*c|Y|`*=
z2veQ#UI~_3{3fp`;q$J2fTn?-qiv-#4E!Pg>QPW##-DUOQaiL*6xlkqz3cze7{vn@
zJ%orE^{UHkFAWO=QfN^;S7F`<=|k7u&O=0cSPTo{k~EdVTEnCmd*B-|hu{VmH}h}g
z-*eq+gE^4x9pw!LK)Q?w0ZKgraPjx!lFNT?A%W+?6E}WkBA-p1<<jY~rx75Ex=&)6
z;@7GQwHsi}(`AzJlZ)N>5|>%waDVWv+gEdMGv{LEdS3Seg0t^!!u4)fy*pm7I)2i0
zmSy`Q6i4pte%1zO&n+;JMS;NZDT5F&`40{BAndub1w3Gm^!#tDoT>#!RPk8B{29^3
z(hjgb)daiB232kV`fNH<J*Y(q;Ip}xGbf=qzIk{R3_6rS(35BPP(J`H?`yydbTYgS
zV$l*W6+Jy4DjXp8t$Gmu_3h3iz<9X~Py{F7#I6iry?h#{D*|;$z_E<XLyR?7YHyhb
zOWt30y+mi<v(IDe#J5OxjqQW0I7GGG%VFp>&nn0-UoN4wuMm3>y~PVB=Ec60RI3kM
zY}_lhP}OLb_+jm$U!DK^Kh-}h9p*YQ@=Lfr)~39@%~j%{eON*rUaFd6_+AO|L;?ao
z1KewV`Z;u*y;r9b)@RyFQ)@@TvpNWf@8$l%&}CiUcL)ss0Z_oF-2S_E;oT$zH#s51
zurq_VKi~;+(%8FqFEVI3_k{Y=4AWd`Vv^DEOrF8=&WE&I^vNI@yQ{OnM2}@SAqkH*
zmWt-Qj}!$6eZSv$<n#5V<kz_mHg%2S)K17Rdh_3TeR=3z?f2l^=A}wi$8+C;b!ERo
zS8Gg|Hf44D>4xWt#rQ0`)ddOEJ5E~o92gdVOzj0B{Lfr)sh1u}NXu8DPn}`w{3Hgj
z2ap02p*${;l!76t2qW4o6V_>JB81|mV44Sj3KToFr|({zWeU(t?yVFET_VeL>GB;s
zBHcf5X2VC&E2~(pr{Zf9bj{wws-sVQNN2Bv0wF)UgUE8QvyW5Ak(hE@>*)Cpa&-k2
z3yCT(4*g|TnVVhJ!<b@Lv#PIZHMCniB8X{G`{>riz2R`pd)o_O=X(7_+Q(-D-hqDx
zUK5)kWy}0nt2jX>5=Fdxf3uT?iWovS!I}P|laI)T&x1hmDw{hDda-}B%+&=R<!8&w
ze*uM~6~_)3dV+y871?q9D{&FNJOl=*f#!`*KH+oxX5@JWe7`*VK^YzpMt^o$)q`x;
zGmMQHzw==3gtQpGgw7AoXBW}kH6ldCL-nkEP=@xJo^AN;i+mAAoD0Q@^KxfRqaw(*
zc93cXUr23$fA6X$*{r3Veq!q<Akbg`Z27-FF&Gv}91}LC!h!71bCr&stB#-~BVrwO
zc&;WN?G-pvWs9CGfted#@_>OJ{pVJyc*}%=%<}L7gxha+!8V9fhj_s4B^bc8D}S1u
zEXH{l@a8ZBu3>;4_{>_QO-7flY;cb6X)4!W@o<8AVPxzSbP9)zwyIp84^g;oT|G&|
z+UNL&m380o>Cz&Q3w{4=o!q5-d+QsG1+d3^)$3a*3!k*664?o|OV8x|^S?|wVpCjR
zn%_nn0Jua8m4cE|E0nJTW4dM6zqgEqHW@wz-Y=xvY@pNYsm%1&p5$2vb&ew>6Y%;z
zJjko1Ald|{H<m3gOdJ^c6wEUG6OZI4RC7F55(nly0IN_0vv@?e<oslP>oF0btp4uf
z0TF@*!H|Mj+NbhfOb<N>I)QT+P4d%<He$izz()Pvehwr-^v)4(870Ir{FX^P2r4N<
zg1i-<pBM87`@ST_9D$ziHFz+Eo5WH$=&5U1VWiNn*+C)T0JZhM)we4?3q8o}&E&X}
z5U>|sFxSHZNpI^sR}o=>8KRqKdaoV)QV$=qWP!7}_?dfF*U)>}4Cnhyw}+C6ANIz{
zmdGP13@mPfioR470r`O?P&VE_YDHFO>%n}@#b|)4si%akS6Qne1v(2XDpxfzGcU`o
z!}B>S99tz1cEe&;X6QM08&SZiMEqysL57JBSQts!XuvI=_%35m05`Y8EjDAkMD5ca
z8t2vJG%&MEH`o+rIedSfCpZ&yUiS5=*}mx#fm$_JH&@q1@wmrDlCxzCs}xRAQ85<i
z!<<OUMBP)SWE|g(#ylWvX8U1cqlE4IC}aO%&(1xDpRf%ikqhsyn17H3y9HtuJnC1*
zKo3Ih1p;?(njM@V(g92JL}Y;kNR_=JMM*P^YpS_V&1VYvR-V(S8caFK{S{w!*wWNA
z@s~*fDj-1?ijy-P_omMM4AQ@FVDP|Nj?ijgMHpxoguS7`nFI~21SYb&C&x2PXd)&j
zJ}<-jWQBKFYtKSUGjL6l3H8rU@-Dpwvc2g*nY*4bNmz0?P{~k^3qyc)=`R%8ep)ZB
z{^k=B67tep9f;+f*#>;p{duXzR04Vhf#5c+0M<tWwd3A<G>r_%-^Ts8KGP?G)Qt@>
zlG<Ose9`ndwzN1I*Iv<9=I-PU!xR!B%w(6$gcLw_R0o!|rZ9OT&u7ECbWKnC7O3Qo
zPj`O==Wren7d`PJPF!Ql$1)%}CMe^BK<I3_5uh~`4vvFLpjc{kzOozAM>I<Mne@g(
zCOnR+YgJ{d*po&VMzseFkt;tEN6uJx435Ba_t78#!V2(-kZPi0b!Jr(YrpfzSdsi3
z`zTvFaJ4qx&j*ZeCf09_<eSz8hd0uBtHny`LUCu+*p;Yw0J$`{^bZ<A3O`F1??oVW
zKzQ;vgvIM(IiaXbhZzQ6!t`6s8eEGVIBHwGEgn!rq}gkS{B!%M50$y)Im>N0^fy~B
z&rAZ@l5Gy_)gq52yYZfchpbmW{+V-&D`f+9@$v^wG)!IR2IvjTm21_;iD)*%(3iAw
zt$;|~BRkC7f)N%#%z7qYx5aZ;z?%#)-O&gi{c0L0L4+X1DIyIB3NnJ$0xCm#C<jF9
zCMT*L!+<dO*a{EX>oIDaKG~|THRq2g;k??}wvEa7x@9mpEzhq1q|K8IAy6ElN=X5j
z_?NF1Dw8Hw^ted5yo>ws3Mi^qnajaW-lS-%^P)olkP>UadGnu$ndvaN_H-hqcU9tK
z4D1FgVy-Me_ot_~pp`_AN@4YOMn$?WCuGp7ZJ$AP7X~$DFX0)DC_b&!c)}5dd=`4(
zy(SW#I7c=<ZZSt}iO)dPP|Ci<3yLzX8(;2mIa8K_iGPCM><dVTz7aJ1iZ}iyf12k{
z)H#9Y)^tzH_&Rla^?V%3{P|=Rx|I0aAj+Sw+D&PBx>JDt{OOL1LoWIJ>OXZS+p>Qj
zTfyU^jT2KCA-)yqtD6JhAnRLKxjPRa2bntIth=JtE5PGZS@4c3wAIZbLa}@%WUR(P
zw&(fwRCeLVRlOmub+OO2JqMoE%<+qyDoj3D!#Mq;&=O9@UqN9`t}0WULYnS?***-5
z13~+n&p;H*0hag4H)<LFzdH0u0DF$UGJrt{{vNA=vSK9la&toy*<dKuI{W9(*|RQ5
ziHS9L_puic4fMNr_rKT;y?GHxb~Gp}{cqovuq(_i^~m*v7_G-bOilK>Kgqb-FAv1q
z-~{sY*Y08<!T=Lu#^B9`wP@iheyK&mNks&jMgm{wMy0ZWzfB`#$PZ!sh4Y83W@M5C
zsnN(K!}2I#)FiE?I>IklazEX~f8}&?21np8Qf)6)p};qgZ1D0fHA4U8>78}3Oi~6Z
zP7L@Oz~_g;J)I&VX2D%jUr9=V@#*rDl)W%t?9K<D4;~j=Du5<f1G8vHUfpv6!rTQ+
zOp!FmV(q?h(r@Pjz8Iv40(NH!D;RpPL-093-5>>`u>?PP5G1Mq$I3z*T7~7NFl@!u
zYW?ze7Z(4x4^m^zpFcwRCuH1dgHX~HgU5I%mu|v;p#7mflQ}@be(qY`1s4*<xWW<7
zK9Y{DMwV|Bf!I5PP`tS5<B7^yOcnYih1@0z`p0$>3Uo=Yvl(~N`^P@Kmk0R834G{I
zhJd}!tNza*SW#$y+$Va{wb<ZUoCp(4e?tyWkD0D71D7BPgjQ<okOb^scpncgA(B_y
zqH#?d=yJ94e}n6}kj%cY`$YKz57H5UTzHTqeW9u%mIJAN86`QP(WjQL!?JLZ{Pp0E
zR6m;&>%K=-FZ+`NjW7hbE|Q5J73qF*$PU}=Y_a3W{spATdV6th{|Mc_eV?_&HaCqg
z@C~e55)k*1K0M{9+hnII(AmC(f)5E~-ur6CW-}$-$h`F*B$;@!`TAagTS)f2vp-M>
zUjob}<UJN=vPy;8Qq&>MC>awx2j_e#$G0T;dG_>QjnP*^@l8gkP+TM(Ygh&ih7u@b
zT*BuKT#Z1jw@|w~4X;0c$!=HsL-2NrTsRCF_8HF)rc1D*Pw*%9*iSW-=kd%0f8I4)
zSbpwsS8@HEtJOB5{+oU7S7J=wXzC67`5%e?L7#O^y=_%_0wsGwthv$M-RaLB!fn+Z
zvRHW_TIAqQWW`L7pYUMb)0c1YSS$PTryQ@v#A^$I#Uu(Z;NGQJBgPA`hr^)Mt3!7F
zz&JwfX=FBwJc9%`;FT`$C3n4-#s`F&a76Opw3$S%dQwY6H<7D@2IWMCfBD5z?YqA~
zh`z@AeErlvFe*axxjcEwRo?23n<D{qKzkINsQwP5e@hb`cwR0w9W~OMge2e~_4zY|
zHr*f(|CO8sPX8%W$ng+Bq?1Yd`|;Cq(wohrKfqW<WZ@+@Ejuwd1-|VCYxAufqc_T+
z_Y(>g{rfYsVd^h8X!$NRj9>qV%<PA5QYN0cO`<0WcAI%H6$(ubaxuF1@S%$62?dsE
z+U%7H-oQ7|a2|mv)q$|HlAjK|t-f~)bZZW*Fm9RjU@#T&QM`cN-(no>i5zPnzjz9n
zB*?T<ZnW^8c(&3T4Gkv|t~v0^r~?K=bn)xZ9mb3n3k-CCJR+n2`T0d2#IFJm--K3a
z1LRjY!DI?pjC59=&4?Exv_p<kdSvBj-9k6jy!^d93(;sX8yamzycb)U&8awgw0GpU
znU{Un_k&y~MO2c@3{;h9nr5snhHjJiLr87!mvZlx>9tW1(m&44W)LTE^}iRG7k7aJ
z9?Zjm_r@aQ1N(<vy=`#8kZERWQKZ#8a$<v4;b?mtcvROw;7kIlf?BU#-z!!`??LKX
za4UE@4GhmNe3@?Z3NUq`+4TiFAceIFOt0gPSp|a%vqP{yF9DxDmFV@x+N_U$aAw>@
z^%}gmwIz0|@8B#H_IA7(wl15%^Q=$s+w~J@o0E?@8aq-v;w>TQwN`lDC~X9{FEI2H
zfKuFb9LBu_gQJ<UbHU_LI<{RqztIF@;!AhuK5>{5kka@LX8uflW8>#nS^aDZF<`1y
zT35HwHmdM(Pv`gH5OwPf^VSu8`8CN%P!0mAT>-4Ju0GJKBx?2koSqYaWq{C-LTDp#
zeJ<YRVYm<sHc^<nw+o+cZ=%cPm!WO+x9T+m)L}aqaHoL}PQ&$u4(*q_u6mp+gS=sS
zR;_#lkrFqkQ*~Y>AV^K*<iM9dq&$|u7gb=SUclnSn|gCa9XilHHfdE89^oJ)wZgtz
z2>otm=t4eI>(;fLx0f~Wrn-9?1D33L`L|0J%zB!73O5@=OgbeoAMrD`gzxzNCohOO
zNX$UwQ_YZf)}OvtIBN@&^?inKuinFRBf#*xdvwq~UDI{nCz>oplgtlS%XS~bWm<p!
zXY63j*oTDml543>hd>O1-E|O+$tr;1A8yS=WQSOoy_cTNIGXcp)0Pd~+IzjUeEpDD
zq+tIwGw9s4mb%$%P}*c*09F~!t)B-@cs*on)2-l8II=oh83D%ly-AYUeP!+!46lDC
z)cEn`)o-BRBaOTy=YtHQdVwfVPbN$MjuX%qQx5L8K<0w*^toDvQ!jKrhw7<bmQvbO
zF>m!LOngWs3V@tntQ*?(5)gji9EdhPCkdYeJW?5SGzKN>@Czj1yPgn4s&?n8?vF<$
z+9pvb0X0N#7|a%Fe(BXIs4=R9yZzSmU_ES#<ElZVQX?U_=D@}ZZS&TBLkve58N^tr
z*ia@W`Eh_^c!(DhFZtXDjlUq&HHEUaVJ|rDoNKSN_)BCxNSwNB9QYaS_>$dw+|G{&
zJSeeWgyg`ctBbK{*$K9Cy*=eBK^gnz^E}t|TN4=xQRU)q{9Xw4ESb9ddCezN{`C09
zKkjdOlTL4CQo{1a7y8c9&3!VIJ#5nvf7jWou;4!l^bk2ppntyaPK9!`K>$hGt)MM8
zqXy%9l#mxx|474mD93gqK~PF|D)Zfi*}*{`1NAeY2nvO1Qtfa03Lkh>$rH*rSZLs}
zX`R0Z&ON15yxea<&~6cteCe{ZPIfDp$*`PS`xzp4E`@OlnDeK?01Qo^a(%#?Z{5L*
z2H;G!Ue$(?T7uv;90V!vCr+DcW-t=@C&FBJZ)n96D<+iNT@8Tr+WLy~vTlkRqX5;t
z>G#OSKK}Gg+YqUhZ2ZWF^x{p~0cZjidP_@>mGfcqj8d-^Pe2Ox28iW}U;~!mmwiGI
z_UzylMu@^N%KIogYau(;Z`_`J7yy_E-kxdImmcRbnsnj);>>39ZPlk;-bZ{E&!r#X
z;4rp@`u-b>hV6$FrmF>qgp9y6JH1?I7ojW|hNoHLY?Tr}?mZbV`-X?dE%Ez}Ne0dG
zq9+^=YSf=I|1`Ya4nbROD`=;ru_4w4u{5?if3bG<hTtKYIm4`arM8vZ<n|GJwi<j@
zi}>Z7)1l?(a+b=T)!gM5@xh=+e406Rwd}xu?x^tOBiWcY02vVz#8)BK6gX3s+{k9q
zf|;rxTp~nev7kt)r?)S=(nY-x=m{NX7}zr?XfRs`u^T4><wZ<zP`j60@X*@}YZs2D
zOu9y2oVy2&%hF>nzf4rmqj^g(^o7HIpiFJ~>H(;qlL#N>s$LjG7Q<*xrAc0!2XI17
z0eNx-#=Uw3{E<&dx|i^ia?GR7EqpzD{4ua>^Qn@TCH{F!y!_$R`jw(`*nZ?lS$a(U
z5|qOnU{zd@afb1f_u7Isk4|p9K}l>F=95r+<mH8n-~zQ_2)dNBfDyfDYuWo{NbU^e
zWv|dXOs`;inx%&D1VyBqUpZaPXLxFzR_J?Uf-;8ggNWUUi_aahoZD4UX3G)s!R3TK
z*lhT+NMncFF%;^{!hy0n)-3)P2^<ydNxJX&e%RWv$0^}R9D=Q)edRb^=*(DZuwHh(
zt))7ZNQi48ozv{W+e1BqkG6xWJ>gcCyl|C$BcuG{O}5sBPpx+f&a!WxAY2mWTSU;{
z@wG+&>vJY(a^+xvq1EToWq?HAMUr&Dp5@&IWkE!#K^m4$67UUFeAv8zo^u{t+R}oC
zOqH!VT>B>sSLkGNE>dnVHUn=X)M4x%(X_u;=Dk*O>-OzWNhu4)#U{RlrDrzuHGbrL
z`vDRX*JyQ^id5d;A^7o~5=@~ptqbLG)kmlk^ebN~=VuZq8STEx6gCTJKN<@3^1MwP
zI50&33+PjldmgfZP1DU!zCevr?Se*gqBCo<pfTna+A5va2^%1=ngO<|_J-r`%>#Qu
z-Z+#D6_4GSAaX1V&Cm(sf!c*jEdFTf56G=KIOPZJIcEEc*`Yo|9n&b<d$Fv>|6rxu
z=WBpN$^qjqIcGR<NZ0133E#vxI>KI6OnP=w1I+(0{QkaN0G3dyoO$6bE$l65M1L_?
zve~N-a1>qGm^<dkqsNk=;+68mwE6HA)BJ+%cUR`%*m0|rDQS;){sO;?tf>!L<DNse
zg67_YKW7AYTB@SqKwKMZJgNz{V3FFG^3&_LX$>7pZ=#V<VDD5CGiXZ()8@F*sN|`V
z<Hl9HmO*4a-71nl52mU0AcdQ{pI$|AfJ?1Fx!&?DVJ9~n@C$|D^`r>6<5X*#M{Xa)
z|8zDmSL-t_eFaZX6gk6*mcrCR`&6&D9s9Yx$ltpEhJqmx&@9&}O6i}0nY#J!1sg-c
zAP%R2U=+yzZDz;Qy%yx#K+kYSWTp3gp&x9DR-C~?hbw>K;NCCB8P#qCR4sggAFpXi
zzpaPzty%z;p+6?GFceV|+FHEX_#08$4`2smu-)A{C`E4QI_d#2Z1V&Yq{8CorLJm2
z0np<E2!fvQ5~R@gAZJM+rKF^+`M4|<C~H0S>tdYLp+GU`!&uC|O<0Jaj!Etmig3FL
zD)evYae6s*`-V?NUr*HYrdv`vyDGPBAHc#w2t!u5Ux+RqZe1K@;$bs*5+Rl-aQLrl
z3tLjggHqK226h)#U*BTFo+5i86dmrw*tW!De-Cb~#ofA{w4TR}i9pGK^j#VEdDHQM
z^3o6NFcmTlnM}DZAVM-Sd+A9sD@b5!YJmK`uf(NZFe6VJ%WX@mO&MMUj`w{qn*to+
zKr@=3rAaz<9tK{O9@FLnh_;QXuc#kzrBif(y<ZJpxDo&JIMwJiAaSFrCgbJGJS<-w
zxb&Mxd>gMrPKjWg;0US2u?z%2JCC=XMV1f~sws2=xaaHDgpJ-;$Pb<#<UY@m)tHE?
z?{N}`fJi|~`hC#CN0o-TJ-D2Lp=#=nqvgu*ccC<@OCL?33*?Q@ewpe|*0K9G|4zJ#
z-HQ+tL@$pzYH|na6OqO`$8QZhkfhVPJpP`R5{vvPlB2q8>Fu~uYTCSf7IlkWzsakq
zQ)imfUDE>UPw<sFcOODk&cNL_xX?N9jj76Yzm{ii;b-t<A78(IXgCce?a>6Dp3R`T
z3y+zk?D(~mDM?q{y_$eGERUrye7Ko<<i7w=;jU9ni-p_^u#JXT;4-7h!XRj`8QI;#
z>+lAKpm0G@2m>^>yd#>|2kuG%hOjV3^E9Q_+-~&<c(&`{(VI?3&qBe~-cG^Bm@sG+
z*w_o^HG*glS9jS&56@RS;!AYzv*PpcQzQ5y1^w#_T7mm}-_e5K+Zu?~LbZ*&F19~D
z)~4?`T*GQ#z7S5vBz2<y>z#}tVr&6ridhTz(jx2Y-6!W%^-S0yqwo}hr)Tz&6&_wb
zsRM)XESWoy0ZCd?@E67nVD20%SV7S&yw`{HuSKXOOJ5v$lC1lMZm$tgjt`v2(rdMh
z!fV-^RGO~|>c4TYPDUf{_EVe940>|)s|8uCEDHw8%3Y_?J$fHQF*7ATe3~VH^KtTT
zXyP70j}r^^E03>*&h(|SY8-vS&bNZnd;$Yz-Qz-qV&Am5T5aR3Wn8(}2G8D$Dek&K
zoqVCq>G8oJ!A9kjkejMIrv>e<J0W_V%Kky@c~1@pXr@fFP9w1%B*|w954-|PTr}Fu
z%|oG2+iT9H9GqkWZxO7rd;=;}%K-kcRWGo8Lc!>y_u^OHB|n2Z$1b(g-v=}K=|VwN
zJQi4)M9_>;R!4uwu3?^@W8W>2$~XtfGAb69VepBl=iN-Ww0bhsEN?pYkF7c&qpTs}
ziKX#Zh=+lMjvq`rQDc=1Akl`vyJW-!<-0!M{ZM-$4pTf%7CBkGg~ri5sBw!!!ldB+
zVWDDT$|fA-hw(9QXfAD~Sr9~i5??NLRb}#Cx}veTUs@OWyrD}14(aV7)}DIcDJnCP
z<VZla6~YdQtl<zZg77M4t^;SwRVBxNG0I9Bo-is<nKUqPG<_gkU~qVw;WkU0TFLy2
zvGhw!U27VL2-^exyICB|PEnE~HYa6`5B!bBz_1Wqdxkqp1;U<?9cp4BI@Ao5|EfMW
z3QWL>D-7x=h~4vuA8?HVdq$YITPs$ZH4AxN1XS6*BgmW+foQH)*mbHhy4@0R!Vfdn
zROcOFl5ms*X9ywKj0i}!bc17Lz;lBOB2-9Z<e`+S)wz@z#^ewR+*ZWtX=pf4tuH-C
z&U99NM-Wytz87LoEJZIBF*z>wSg7xoI4F2oCxH2sAYiN|d!j-^k03KSzw~il1#>v*
z1c*fpftc{{&Y5B&K~_8^ZnHC21bk>`Va5<{%0)uMBK@3x(S=0C%E679%!tY~T9%Cy
z6L>;hN-h9D@NL#WxiK6SI1o~Y(EGBF{Ch8#ni9MA2%?!rE?J`-Q!S^0ZDo(){_fma
zKeDIA&B`y6j2gF(cig?KonHkV?|Zo8{npeOakgp-d_D4;!6?F5+Zzd(zy=;ASOvD9
zG%MDA>6a!<;LU;;WjDMK!GVD+aMT^@Tp>@Z1)YI`hp-euC758A9!+|QOheHV(d!AG
zFfflI(%?x(7`)Vqpr4)=gB4bne|YE$R7v>PnATtnq@UZ@_v9M()UZQ|Q@x`rCAlkC
zMnU8&a5gipI*$^%LHbBhHu;L*sJ&8m!CYjag%Q<)L)nl^%_s#?Vuz?vE>*m647vb|
zTBGqEv<r)c9#k<n$SJ<59uU^`2I}O4e-4`O#UUt!BY>Er*VRGmLeZP5!JejHlzoJY
zOBtH2!;q07Oz%^fd;aC(;B{p@8xRFn{+@qOZ`|TaPFLp0$8$VT)`xkIqk>n^=f~`O
zhmGe7S})(-y9HC)o-=)g7MKb9n06+jqqk|2=Sx^w%#*O-QF@w%znZ}Bi8r5z@b%l`
z_CJq^EU<O@b?XJ6idbfQ;bG$0{)eK5&fqgv^w2+h*31nZf_B6$9RDfo4xlM=b6<!&
zLr8KSZ?|m(3@cTn;ny*a4*l%5@VpR?Ved#i_}PFcxmr*)i=bY1PHneSkg*n!C7zxG
z323g0HssxCVj41>o_JJUs0@kgOL5BH<h0r90?N`Ayo|#pF(+M2p-G<X`8;xh8P-2i
z9fs8ZXxuA(rpgkog#3fj!B{SEcX;z=1f281Q?|1d*$JSr<r4z(-UL(%)Ruj(UcjKU
z{iNGP$1Xe}Vy!H_i(s>THA|6@b7Qe`LjZ;W#~p9{1%N>O%oTpU0tGaQ19ION{kCfZ
z&<I=x>z>R>^CckLi#FhUVX8#&@(mIF+~S9l6-s$i39|=P!e_p^)!fVPB4@O?u0uoD
z`#fDsFskjZH$a3d_sSq4-HEmZ{Vox0lb$5KQ<`!o#f4&A_Z04+!{{eG7PRhom~*r!
zUiRaoVeioA><;6D)CEdQ<^-nn@21FHF%`29fPu*tDEnR(QGQBJS?s@<yO+8fP5)4P
z=QqxJJ*8UW$RX&KQ;mM)$1?DvcB+Wg$AsHv9V=-ARY~6MR;2`OGM;NIF4K=pCIn0Z
zl#A^8QC$*Hi<8m_LY#N5J;R32itB8)O)IkaLpAzfMPGLoF*J~bYTzHB=V3SYg4v}W
z8s{jrSuLRFwu}WxYpxnyuLbInc?YU#GLu)Mo<7YabQ<d|%0Yr5o>}M>8Vu+=)LEW?
z-I!=<)#Fp%toPtPaR{?~=i+7_#I^~&{!&9biFGCncv`_L%{3Jo4WrVbw0eKWJ;78j
zZX%fPMq*J$@;2#gF0V@efNM>~*L#N99mORm84`wMs2(Q51|8fv6KD^8QC%qPOFWJy
z4i6RjuI$r+n)>Oy4`7LihnC_jy%!gvty<rv)xcODq_0;S&)zR+Se9BI1K3!GYjm5(
z0S#ek0()^|T0VI~N8-2c$Mji<j(Y*#^3g9+Cs}i?+zH-v=qzV)D6WzfdMzQ)2YKBC
zQrmx`<G#mN9Z)oGHLDoV(gT8mZ8?C#jl|-Sz&8l=h0s<)G|Jl@H>1Q2V<<>cf8;}f
zdV6e)PHa2i40+*wxN&^539~@Retz*aya9snINUZMy2xTl%^(fbSiycHK5;t^<$Rwd
z$a9=cBl(%1H8&a26g<GhZxwa;E{_uiQDDL1XNaxyJdgN&#rVO-)pWb0GLhepl<4GT
zMaoQ$Z5!HN!ji8F)>K8UW=>Z8Ze0mtYb|!SG7>x8|L<3T71eW=HA7a2$S<rj1ZI4i
z&=xIr|B-A^4-0u<!6g4-KrK`_`+@v6ab*M0&eLGeWuG;QctjIw3JST&X+jf?X)tmq
z42sq6q*6T%7c|JxH5$-OWC`O4TDI<_j4<OH!~<5oVtWHlQD{Y@)DYPa*N&`AzPk1`
z5r!?e(yuN|a)FaWuJQ#w?1{%wdxssyTRAkas)$hwO|EpUv;;!7+Oa|313HKITKh;l
zj|KL3LZihRkIDAH&1=k;uvbu8C8;ufLq+8wQdH)<_9)=`R_uWvUn~K7Emph?dhBX$
zRPN<J`2=#Y_$U(F@Q-~b11=LTqu~$W{_*MB0Whc<x>E%<#RAB;gI=O`r^h!GfPu_m
z1jelvn8Q?%JettzCqv&NhKEhjWk%jt*>3@0NP?N|&~k)X9ycq8t!VLP_nJC0zBlTE
zak4|@5}=gCMXUGx65M>^vBy0hGRE@o2uX#1fBb@M_7FL0N21N<oa?;Bm^XVpq4{_M
z+iTO)wyL!ee08H&KW>+zo8fB61_x^{yh_8Fq(oHd+m~}7BS34zwOlasTn>h-4-KZ%
zgeGGx&u%iDaKQ9dfY}S`6iv-T_wW*kOHANhPQR4JJ29OBx8zz+@_tQC&EbBxL27D;
zs%zt5?yV9f^6sdvKIMD=Ap?qI$3Lb*slJAZK39%qfSG>B)IVTIfnyg2h6^WEtAS>a
z1aPIEaInwsugs_jJUQ}0Rj1OsTxzar_dly&xn9m@U?>8Vb7$Vb{NP~lygEstjfM@B
zlZ!!fOs!V|g)M&kr|f!1mgHERanRO`C=65rE4O(7YsZ<oq2-7MC70A|z#o-^E6D}z
zqy02Adx*C0pRaEOf&62pt8R8};KQ;Eg|5iOQj*Il*-;t3ZQ*91GMRr*Z?AJ{ZN+`f
zJntBD?tq0?&0%WJL7&Tm77G`^Xi4D?-tnumW{1attjqq=3i;L$8{<*yW{%#9as1hE
zT=v|vY%sa;$k7rA>HYhaq3O<DgjDRM(AS5CHQznou(hV?S_!qwQUCiAZV)bAtJoOz
zWrFsF<zD3N*4}$Q$Eb4V*c?=~M(sBzkVYB;b3Mq0O#${tXd#+GJw2bnoSep-`bkjS
zejpSYZuI~aT@86T3&f6KF!xjh`u)o)q_9m`$pflLmN)fS^)(I38yWu7q+pNuN>QJ(
z6Sk7P6Lh#(vBxl(DAg+)AWO9d>%sm?pUMj>Q0%<)5e+AYsD>cyI`BC(Tf{shQAk*i
zDO>&)?Fa-PW!LE+pDg0S&>%jPWK7C<PId}d0RjP6u-d8t4e5gp_=%(tbf|R*PMj-%
zwh?BG1l5au7Mu|9oC|2$`Qe<F5M2Vy;1l=-&Nv=~-QPz|d^RR7Z|6<Eu08~HEiOiZ
z9r2|!k|!@cmmQJgi)e?!78#&3k2kO%9N(-B7Dr)@>e7X{!hLfAn*@$($sfuBu>2SW
zVup<$Pn?-6W$pTujyjhuUl=S~vw~-f>s;cq>DEJA$0t#!#pyN0J;~TeZ%!fGNAM65
zzBtZdG%-l<7r21_K4D@+>Jsi2*@G^&02Y#j8bHX~dNPwTnB&rj+S>Hcgf+MnZaq@q
zDWE~Qm{??5AFc}C%OCvv8S9a8zMk>k56<--xiSRybaB%uK~R>LVvphW#h4V($D)yr
zij1N#rV|cTJO+er=k4D)UWT4@2B9Dt7>KGka1&sMWpHZX{QUWI0)Cj(`qEdmUO#u{
zOLzwDH{GC1sP>f!sQl1Ntmna5k7A&@squ!LL2MdIQ_jFQ5ZRDHXP0y;B<#v-D`hbk
z#Tf`8hvUBdDwjdpLraT=`mc7*EQ#k_HIWF+F|<i?pC1kavd8_B5AkkM5>Ckmb6Mw@
zy|gan7-kh6jvEH^XLTJAHONBo@#qnlr_Jm+hUoJI6;%^iqr+d%Hl6zReWAuWv{`_;
zQkP?aTu9fm|KP$l{em8!{+D4OHvd;~P}>jwW^sNtNUML9v00-yv6JYi;JW&ppP$jA
z2&zsXXjG9&cx4HaHPMTe9==}Ck<(06*q1WTyL6MJN*E|I#a^q0ZLV(nLGFn7u*f4X
zC0aU?+0TvP6^LE&t(X?}ID)y?Z8VcQ;BC)MLGYk+RnoTa)Liv6kAQfm1f5a?8p>Z@
zI=bRkt#W{@n24{})F^E8z+J~P(IHlXy!jW5`bJ5F7mFyC0a;%nq5sV3^7%SG<iFKF
z!VQ0rZ9K;R%XUo%Z&8#l$y6UWV1vx)!OH~DG}b`7DTUCdkhA2wS>LN`2sN;2f_8tM
z_@Ji=9Q+6yal88b7xbQ;w{1Uup6@@m(oF5u$;vCwLcF#3r(fkCJp#*C)rEgx!(E}m
z$nAYs<B5)q`H*)bVedJfZqjtnvVJA>Q0r#B_ThlnfP<D?BvkM`%Z_K4%LESpd*L7M
zM{Str@6{#I!dDWs-|PSzp>Vw2d&Wp!#WR%|zH+#7$DXDKN~^M;t;&e#XCPY!Kniku
zuv_aE=w|jSx-AMk{G@42Bzje}sDi;-Lhp{{S|;)W>?uH<@g!jMh+O+3*PBRInPE(b
zV5(oU{8<gpaw#ccl7Y<cIaw}0AO+Sf4*JY)@>Z%O*jC}3p@1V3FCHk1`ls-Smk>F_
z&>h5Vh9#7B(<MQJvmpV*HPK-YWn=ulZyi3Mz|y|*SV#w+hf<z-V>+P|n*RCn5%`)~
zb-nIbyOo<OY~$f!1Z@C5dU3k~bB)D=FjK)8ZAC^JmO*a`nk&Dc%Tf<a!DgQ1X)3Bs
zMEF)!(+4V4T-cNLDJcaNan6Lwreu4RPB>436~b|AyHQ?8pJJ);Zzk+pF=^?N@tXls
z5r<BhL9zMDo;Z6Qxs%8F{>x)H#DQuT`A2cCXcK)%GmR<@;hATj5ar-_XXXY2Hb;Wa
zF$Doj<-W&KPyYyzghWhhs04)Hz*F2Ra0x~Opx)?DsM-?}`bbUFnTB!<N(&J<tXCAv
zTr9mj95Vl-`ILS=R3X~!k7wFtDd_}&vGo%9HEUtW>P5ETX0whN(ZCh!uJ7&C`OPNK
zeACL*(P*dS7O02O4=gZfqS&lsJRgShp<x;}P~3hp0|qL}^=+PfjX<(&<fwJ(_uS3P
z#g6Z;<+~UVI-op;=17v<y`Q^(06!t<Qw1j%m(9i{hC#MYAV#1O3>}Yz&_O&4m+k>U
z9PzDe1dlb9gC+S^Yi^t=kmTeN$|P_^C93I?Iv~)+@0{v6tnW#6jZwadil~wl=Wh>x
zC}WvD##oW}@-gdn1vWOe)~VYiGTHbEy@Fsk8v)XoZqDWD)s9Kn@;!EQx~t&HN+&?g
zZcM$o+d3(C(eJbi--AFcwo<VF{Y;>Zc*w%RA1CZ%8G=O0)m;Yare7S{@&XTZ$1*`F
zyMEwaR8Cc6u*SYCc@D;^wb5<m%r8oH1wQS>JoHl~Fs0r<8@w`xazo?XDo`|<K?czC
zxCzlV$_2(C_X&r-s3LR&IXF4rKWBp1NFBWx=`lde)cCLgJq)Xjen6YyfsIn>@R{y8
zE&(%<QvKg`^m5LYObQw-q_XUSz8neDv-jytuie~`7@nJtx`7oQ?_eRL3g-tv{cSsZ
zS9esSJWact5bWR-5zY<?rJ@aS=TE4gt<2i>^V-)()pk){)S0&FEwYb5bo3uSknCTb
zxyU$C2~-*@`>%Ooq16$X9wDjm43CVw74_v}$#ZlqFF^F>r{(MPmL*2#&TU@`fFDlg
zk~O8n)3UOvxArJ|@v>%&Q5fj+;pL;edG4GOye_^T@pA~6$R7QRgLd5r(=4HZ#U~st
z9*u3oXUrI)C_LS;v5@1V2wiu%{@z^viGkN(n%PFaZd_~gHYI1b#9LGH2>Vl+7H`j6
z?#2dg?dh#4!P+WxgPRGx`>z?^X&#()KYP;VRgSG7UB`&wUdsA}{3D;X@Ybpe*o7N{
zHSwQTrLx{<&BG(c3y+v1pHS4lHWQ0MvGff_BrykfT(<uQlS}LnaEM>D_5@W;*{%=R
zhTi3p#904ugx)~D-`bx9x2P7fw4_pavN4!MD2>thI_{mn-SbGx);x%ro{LmO)a_-^
zbx!cJhZIkh)WbGSA%-Z!=^63@3m#r<emhx|G*U$_Gf4!x4|(Nx#5!BgM<>r0gv9w`
z%n6dcZ|H<)OT<JaFojulrAC&0OiVoIa{kcPqx_Tf=b=j3UvUcUM&~W=rQQ|RJ&~I{
z+(?(SvnpHf_Yd3&ULG0S_C(lB{HQ#k2G}$BRD_M-CnSw6eCC(EVcZQmx#0TV*<TEq
zRSgV&b&hxrNSPoNK)6u4hd@H?f0FR)HusCblXGrs=r4F8NE%+1CA=O9If_=A=2<9J
z%k|vM`AVU^U2u4~6hk4Ace}ooa>A_eMkqkp;s8mYgvCyUNf)=0tuzWHF}T($5!v<l
zwimR_<+{4o+6|MD4JWt5w1V+kEUAd)FB#?zJEQ3z<PVfm%#)@L=x#&fn<<MQj{AG_
zR>YN^EEoL*^mU1UFsigMXREj%H}iZw5vJ5nK0Jriw?dV_6G|Ilh%67FQ2o(XF4||n
zc0&uzzMBVJ=U@1P@ly;?SCIYosy#HydPac*R@RDzR#KYY%XBmjP|Ja6A+7Ub$wyMq
zG0~(*6O?TcW9QZ{g995dZyfP01p5NBO^TGiESSoEh~?uvBQtGaK=SSAI;FELk&?;g
z@r;y<|FFcEH|RU0`Z4GVG`r9)n*|58k^hyQLBdgr>mv7SXzmbsQONT3m4tmRc`nC9
z;HU5V)z^Krs~<xFli-6uh&RSm1H6}E7-ralBkr^0p)uhGrD6xSuLk(1Qz73917jlU
zq0caA$Z%W{6cT8ANhhFj^;IRt#w?lL$3QuFbkEfW6^)|0;RYJkLC^DDS!JeY5Lamv
z!W9vZ+axGDQ0Wv&p1#v?&q-ln%{oj|!c3(oeLqv|yTwNz5vl;Y#L7Rn3Y&cWmy9ey
zti6sFJj7=IArvmjB6|1QmDABXzbpI%u9|R{CvNQedJc07K6eq{4gXJLVd4tAbUiAe
z<?iiwT={7w{jLVMt7-V4O$-LX@}SPnX^B)A0zfYT|0HbkyA?WS=GkkM2lnqjVpQSD
z)D@yZ=wqNtJ}lhrk7dY90fm$ri~qxK*l8S38PT8TOfT`^XtBmzbD!p)&yQ@x6ex@S
zleEBoHqq%vtNgLED368>x|NAp{4P0Mbx$tt@hRX}(!A{qEt7ceIY2|c^UI*0)^T~D
zk9&*Y)>YX2&s24=y?5rPL|qBi`49ke!%q@kNeTw;H~y4e(s$uhYl2mL#%AOih6*ki
zq7XjG&fy6m{k=RI=Za<2iuZ@WHa@Wc+G51KEU7c)glt*_W;+u!K-z=8OSsGXj6Gad
ztS^m%S_OM~%1khji4MsGTw&x@T;r{HoNK;RWx2%~K0y4tV$_c;w!Aqow!pn0wd_+t
z=zqy$LO7Hqf@pdRp23}32s!+4Rk@pGOwJ8e;>sv4(}OjvQc{<Zcnbz_z_+0)2<dT9
zb`Sfmm1AfCd?T6(vWGi@J{>GbTJDmJ<(k#gOpXGM+{OS6kQ<4PBXw4Gi=B7s?<ux=
z)X3Xk9Mw|KjwD_l%`H(IhYpqR;#PDTh&HjtF>4j5Pq>XIJ2T}QWM3}5ntCq4_Vc{e
z?1ZCsMdqDC<*nO#%kR&2QXc>R-Oht3t@nA-@(0sIr-6kk9c|}CgQy+BIuKgsif(_Z
zFgsZD48{VXG_Jot;K2jY;YWETlJX#f!5Bl14>ON@ca8BF{PFd}HyVNjG6E_D1&HQJ
z$Q}T24P`*@x`<%XA5Uo&1uu9;EXDhcg5{YS)Vh-kVvKxcp@(ZeyF&}6CI)`6!IwCZ
zmmlD(z4bB~#+YvoML3U7zi1HJ9d(2JL+W`c|A&(r7uncfli5daam<J>VFwp42LcO?
zPNYoiF-@3wbhyZWI2ONg9E<g1&;B^>;a)2iEx?4yI-)H4h<^inWXPU0->aTG@D0=}
zXb4Mt|1|U>Ytg*3%~uJ@O(b9f^Dm51s=hGvL^iWb7z+ZOHKgmvRBenrDhrcrRVPb?
zDpQ9j#O?a17R2NCp|E^DnVi6Ls~I>qU3LM8ZY)%J_)eD?LYg<I?U$C~yvFRtKek4m
zY9(i8*n$sXnAOZz*Bb1STwD<YBONE{KQ>AOPc{CdNQ2q{Wsa2wmVXW-?w11wVADR?
zYn^nW_`l#>Pf0fvgd4K>-<T;rcOzcv`LYv3GthMtEk7nWLhklK6}l9dJ@|<K3{a7I
z;Dg4Ua5u2ufovG(Tm~>=QM&j=Y1*8=V6j>IdEL1A;k%(ZcRIgswHN3No<nmkz$g+5
zC0f_RNLA}+#3C^g@H3`R-<4^-ey*Cr`=y_hrxD%``_oU=xPJz;MO;mRS6+G*q&nN`
z)ddz>{@b5st_U@lS9)*%+DgVY(+=$UgTRVBXc2Ri-B<|X(!d)Pn{h(NLqNg!In%H2
z<^~NYt8RK*7dKFhMe4)@-_h~sm_{u~a_-$w?Ap$(utWR7*L?9?2!>)HKarE56`~+g
zf+Zd|!BJR<uk#1abqS1%n3rBwQkj1xC{2yh$+TIm5H`YBIvF1xm2Ixq2@+@KB3X*J
zTqikicuDWv%rZmqZ6mg=z+;ZPnWB0hxwenf=F#j%NAnk*xhJtHf7LY>6Ly~w1-`(W
z9)VIaGKPqlUupxa_E2CmOrXN&i~(*E0Nlb`CwK4UOJvk@U}pmrS1mv~iJ+WC!wu2I
zRlBk4tu|9g*7j{pb=>T`C2-E9zC0Jkx7&I_`<}>%;Mjnsmy9Q|%B?={nTnmb@{gN>
zrXt~}$oH((BfM<<PSoeYEeu6yD%!1wR#fiU^z*YxCT*Jkr3*r22SNs!*v~(fU7LsW
zI69<P-_r?!Du99LscJ)T@Ff_tnfnpQ_AYKtpnL@HW#IMV0x$S!%c8zLFv`hYNZtb%
zeRIv1<S>NM^kB&yDrlewGy@LaT?WX~9(BWQyM=9{k>U`fPP$zS-bgToT7P;+T?EVA
zW)0JT%RXT38N6|3qW)no#exBgf0+dPNd4mEQsS%@v>P_I3WZ#M+`F-K+4!?Mm+3(I
zR~h>cFR&I9{T#DWK95aACW&7O#M`}UjFIe`{%uOeh(+-4Ius;kyy@O!+(%n6v|}=M
z{tS23e=Z9ru)o2z1ENL$+YT_MK@Zij2b61@L1y!u!$vDY;dK7V)%W*(ZJ=<<HfRWZ
z0whGp{H-X($EQv77RIyue-q!@REC2B@|l?J`#+MIL!5I!CQ*dxSOzx$hYA6uiK5Ey
z)y3i1J&xDGEp2K&n;!c3Kc{;^sOpn-Byo?U<>(s>#j++Cm!mdFa)hfaWV>C+#)wUl
z<qdiUe)9wdI|4Y>&E894C{{+las5Tvv2=|V5cCM2KsFT&KgKdNs2ZxZ^V)6pU|mWN
znASW!AzRY=>tjOe3o9Pko`h|=-Bv*5$Mv*mm{<PQKqtNxjD(@bdLL5bb-_8gTpGy=
zV!zwaBCiMfy^Gr{_)BR5tF0JFqceF2F)X2kh@KB+qh^8S$MfdKQ`^zZSDg^?L?LNu
z&V7XPXs`!CLS|CC6qyXIG&aE7>B2A4C9B2o!tA1=Z(cx$H-jEbr;CrbgwJwvnXqks
za|wopDDI=?QT$6=9c-8TOSm4>kTUjxww+?v3rV@Nc(3vsx4i(kMICFQEpcHmjq@F;
znRuRV56OOT03fK9TX8r<;55f7cyaY&eDAAiN8$DK=GFLhACJ?v0<|IRt^es38FS)}
zN60>}f3peC08;Yye3aGudzc<rqSnIii!|C;24uv=YcLD8wujuILBBx5ehQyeDBk;n
z1o*23y6xINT~u>PRj^!W3~H1$n@wk?o`XL$!E|>^ftE@a$N@1EHelz7EU-F%Ld{u=
z81o3D2hCSa9G?;mW!iKzYwQ0Z%=@;l+kGsrO#Et5V&Bn1^(0bITiOqDeGF>i!h8Sg
zRwuqHjN!tHnXHcsRWtnls+Lg!6ZdeL`wxTNV#kYYw6q2+HS0KU|4%V(i39B|$3&I>
z6+f#kLuT{DMrMp0e;uIOAxf9U9%r)QKp`WUdygMSK$=~wUA^JhU+i=axnK}cGPPd}
zXdu2-;(JnN=2QMTATAx8IJlG9&81aA;xh~(BLaq@T~A_kY0@h9a1qG8xuP!!jZKso
zE>C3S40mQ#8d3dzO)69a&TlheV$xZ+Fch&QnoM{bTPnEUmG&)wAA=5-fF>s(LiZGd
zsAt2Y$%Daez2Tx|28YRP(V6)5o9@#MKZ!$Ws`?c)V`6rt3=b<OCe4r7Xd&>H!`)GC
z<v-wU11CZY&7BfXZJ%>oY{YOQz=0&c;m|$>Bj2x>6`A$Ei9<Edz}R2J<ELf6!3t#L
zBPiiPW!bF~%mW$$M1OzQRJHWym6avX+EoGsMtCyyO`!<wYzDI7Qsx?P7VBa1##H{w
zf|ln@tE$y0&-_@nl6jgMMh+vg?5qqjm|}s6+icyO9+V2k8HH5I;w|`lZAk$AkB<ut
zq)bcCYsHlQq)n8oAps|4(-fHtySXA|le@W1MgGJLkB)2&4fp4Y|3}QQ#l;LUu9Y3g
z9ar)Ap_n0D|8O7nCv+bZXTLUQ6=E6Mr5L5IpRWrS$S76D$?7WCF1bnIz~e2krT^dW
z=x<J8fK`!a_De4p!-EG7utPt+^O$-C|0WOct0J`_A6Sp(XlU-k_ukTvFI|m@zk)W}
znQ*=2$q|G1b`eeyc69ftAdd?f%;roYu@DEE9qDDX-!py78KV#bBBZ80l(%C*QMz(|
zY^DaP!^4Mb_eGgAiT%4wtL8^LUTxddy`92r0yg=;wYBVwdS*m?hzru86E@`npCH0d
zq@|cb%ibd+5SfL8ClCjCPT(-}I(#q&3)}#c`DZFs-G6qylJo?)90A`GQG6*FPOzX6
zq0q&$i9?tloPe<jdyuD{4I?EbZM)c{9o1}D#v|mN`$@Y)xmrqNemTBJCL``XoPf8u
zH=}9*v}Vu?p~aS;+Xx=G;HvCc$0*w?q`@6c)4;0l{0jPZI>u=Kg-06)%Q+)&R$BP}
z66MQv)%GhpB`%o%&LzyID>W71snBc@uyWeMr*L2Ir-*9wo3)W39^Xz1&`(%4sO{n#
z4z9^nxj{_yw3xJNgd+fk0g^C6HEM>Jk=e7=0FRK0SObg55wS2#o{CZW&7Z%$fmIl}
z63ckLdMOHqq(JG(oD3I&DH)wZ6ueb8SL~q=VYSEY`qF2A<OPn{5<2N~s|am8s6;-T
zdly6D9cOWpst=ARNFM!%24)X2!ZH|JV|~`aI?%xN?VOBQy2knS?>ISBb=3jz$QWvj
z()CagCdzLhSZiT#ahY;m{yKb7`m9Kp_)ZWoPnWv=udT-#LsR0>8{=FvvVRAGcxVzH
zvJ{o~**8}KAo^^yM~1vIi$7G}wIDxruH^Xzj0ep;Q$=K7T~s0`4YPlIJtThs&L;a$
zCfWY&l@`vgf(Lg5b&COU&Whf`;fp)WsQW%6+Zp8P4Bdw#02?DM5-8gufX<^VKl<A4
z_?LGofL5?^USu8O!afY(bDKG|dd86y91{8WgA^HdZRxD>+?3ZmAToD~`ZHW~t@J8$
zKTo$B8>1<`x!>w-%Q+%%elowcp2)R*D9ZT_dpol=jzU~e%;`9H_TS&l0_2<bS6K8~
zE&laDu(Y5A`zKxE2Jv-4xFSxnMjgg`hlN4YGv1XGH)D>5Rl2V)k8ag|x!_Og{{0=T
zYz%g%pOj2>9&bWR_zo~<?s>?g_cD1bqDb=l=xzG>R)766gRGYoXWXBgeXP>!bg4jP
z{#JcE5NBhJtDsw01Fx?Yri%N<6+2uw$J3lIFK-&8Se~x<iS1~twmkW1J}^a!kI{Rx
zM_!odO{!H51kR0jK2OUgt_aZ4HbXW+YGR`}!w^Jg!MVeiK&PMTEeeQ--VT%PSo)#s
zBAf*QF*)Y&$vZp3ZxWnQGkO^D2}#BTH^4)p7WQoO)A`TCp_p!Lgq-ME!(Xw1fQ;ru
z2-NOoP{hCDgDGWa@FfmOWgMfm-a6*zdk;WH56Aorc8x2$27DB1r-vo){K~e(W{Ql&
zSnm%Bdr*GoHPLw!5b(#>id@s)`_1@E?K7JvRXqlK=hypA$R6uC&dX_kjK}IffLx?5
zy6la%@`L|#{HdW3)UgP^#yW(K`C`_?rZgC=VFs<ho|M>)Pe6%4oh$Wy^K8DD^eQOs
zlPsEJ|AOLJj8nl~j%+0PD1j$*p>gn9t@8yQ_z*<wpY4ndR~%&M)NlCun>>9a;&hHf
z{#%n8mMlnLPM>_?eTW9q7hkmyJ5p!5QUi*ZvI*Dq2J?oUjf%?hbNu%SZNa55ExSai
z;DlDGVsWP_*Bsi>o#N7<3hYOtu8L<XR{Jt<+rq|L;!C<t1wva^2^1{I8yL2V3(P#Q
zVq`0j>7PCV;f_j`Q}#4V2!uP4y%w0G)HTmw^dsWN^=FmC>}3(K@yC@@+@v6)GeYGQ
zkgzGu_Ph_+j+~M5U@k$35(G5M#&tubU-rUAn$2W&;ZwC&tT~<q?=0T%J{8iVNn8()
z1*)zm*Vbb7KfM4RfUsC<*v~({4tFu`J*n&tv;JVL!I#f$vD$Lvr@7Ffj@GQCxNA(|
zJ(BYM7=d?h7>n%a`=l2grL{s*msCzlU1%*0`yX_fDjs*F<DWXCB%(#Y0XQl!j35`u
zUx2*yCC?LBHi+dKENY<^(W@v2a<_wstpq;EgV{fXK;4(h{2WI`SqBy6N2B?I2fpqE
zAh)v+WVj(>@}6wGJuSq<@eYvDjUteD1xO!68kB(#vZOx*-1G>*d49)KiSI;`N)iy?
zYCtAz)<xt(0k5O3z)%dyL_Gdrngs{GH1d9(N{*r?z}MU=ngUA?#Gr@8@50c;qu$o5
zQxp<mhu<~SaIi@|csQ$jk@w)J|8@BiqBHt%DQ@YKWXkMsfVjtZ(4*IE%2mR7KE2uh
z$#Gs4nq+yZ#+d)3GN2;HEfTTo&wa2+b6P_{VQEb`OZo{qs&_7C>AW}->oJi=hMTs9
ze{RTVV<kfB``ZJqjF&##g3s+qdVy+~Lp-1f3F;}h?JI#eTTX^rIa-)IkT#Fc3_(#o
z@<@CD2jbt#!vNBkIw^*{SO0Z<l!jh*8HPZSZr07^VyF+M!5Uab+nbWZ!0>vLXI=}%
z6MES&EI+owMME<W2QKmOWSRR=E9Y+WMNP59>jTlt6a%x*o*zZB++1pd1lXokZ1+k)
z#&oMMlwQObGUV4P>~_Sz3;9mpX?o<uo-(Rycs(=H`Fe8VD76N-3H_A|fK@RHiAT!F
zkdc*w_7J$=gu<L!seY;ON0sX<8B!pp3cn}={A^-Ww-SWZP*D+|0dV4`gC^a6{FAU0
zV+5qWQvjKTLl#7YQ6Fc+|F&-oDFK*I4M;I2aP9niU>E5Yb<)K;`AB=Q9Y{uF{3M4P
ztePU`Vo&-Q7_MEnNq#?}_N?@WMcDFg>k^#4BT#X(l9c7-3jGK7H&q;1y0%6~x7fit
zcsx+PaUo+k1hDU<7h%Ka4?>Q>-Bfba1hS-~O$EUfn1~m>BQ3QV;}-L;J5kn%*tt#n
z#&7qe;=}ju;0VMNsCXo!Fpx*}b)vu=B=8E57eY4piRVI^W>O+Dj00*1nlxVM1t3c6
zt7|4HWdKxThEQ_mFd7icPSpS<)}9|zB4Px6K&c&yy%0YP+|3XmpmLorQ9iX#_%;Si
z`q7XPrBLoG_t#tY{`n=a8ze{_kLDP|tFuU=sG}2k{FN-Ux)Bk0KO^IGSc!Djuqmh`
zyF8&(QGg%ZA#-qk!u=RivUOmbQyx9SEz<h_rRyK@bjuf>er|NLzN7?QoOS>oTAqT{
zobYHFfpQM<H!6Nny2!6fSGE1fV(2XtMvb-($$vToRNS~j!tMPN4alg;#2^MYe3$Dx
zr;$1i(4K^WZI0DAZ_OJByJ#vfvSxy6;6**NEgy;-@BxZ2-4+FA(BaG&)3bd&U8M%k
zgDV3Nh<p`+Ja?>Cu7#Z7H=?jUUXroYlO8^1(Zs;qt`bH}kW;lq-l8196A{^7F_tRA
zrdVk{lzs!`XB{`O3Oqq+rVen1O9bqPsq5<_k}u5$5nbnO7vwJrP6CXnOtXLAT>2Jg
zE~8Dw47XMT`(JsB>XtknWaVlyT)Y%!*KvR;{Oo(zN^RbVx{*uXB7{ZAT=j-`qwq;T
ziO4H7XL2aPjVCDE#I5Gp1#K^xC3LbZjRk6{4~+ju1V?-Xce0Mit^j66Y$`wW%%3S3
zZc7g};})zWD4=x-6%t^NLrdLI4aQ@%fXlteoWF-@K`Gu?awr}v)j@Q$-W?MmCXm_>
zM{8eP8i^51xw=myVC@!bmQK-1N1(zTAAseJ|5m;@VDoz5Bgghi?ZOrifX$^nIRs(J
z)X7;N&|vF*GDrrkum;jxI7WZGS3xm+z5sJ6YW49GdCCXBsEB#C?w!K)9Wol&c4B_)
zbBk=n$L+=x;vt%3)7Mck1ML?nXY14w33^Z~NQ`oHdz=;aH;&f$@z|oViGWT_s3|a@
z(5Ec2Y~5M$KbRIvAi_R{1IfnuASMOUVNv3ZygcqR0Ub{%oUGACr*iFAGzvkkOzNlB
z3VwrigsW913zZ6TOicb{%1=!27;?A&$Jl#^bKUQM;1LlLl@*yGvXw0}Gn-_qP{`<G
zlU-5D%1mTt%O*RN8M3p>C|gNH_&s0h?%eM0@49}!KhC+%b?$SYtNZ<Vy<X4f^Rb?<
z`J*4<KmwgTO>@q(pREAwszPFqO-=B7CY(zgUB#AKNLZ>eB2)3dSlfx<I(V<l;U*Wt
zMvleh%4}VJuUVuOPlIk_Nqc!!>}w`BqnGE@<>w_@#inrGc986ahm0u{qEt{5WV*iB
zu=14Dsl|1)vIZSJIo2dXw%Uq!>a6*O_r!gdM~9A(xh|FrpyWAY>mtzAcS8~sYRH!e
zi>(J#cYdv^trJo{dtyp_$Unf|;eV%!%7k(?f(;EdC$9w6DI`b(TJrbjgi;B$6Xv>O
zaZ(77-aW29@D_wm%xr8eR*-AJJST6SQ~@VP`^JUhCKdXCt^{wn5Fb5g@6XM*cX8u9
zLVABt@ET}-jwd^NIRca_8--MJa`aztAc2Jm{Kx8DkAtxOSl<y97#b6E27GsXmffcz
zL8G)ynY;Ql6tYNw3)>Jn1~__Y-1d*{#t+urSEmG?mzd<33dHu*-tU}CzlK9XA9@fe
z)llJyC8sQ!O3WE?73i7fH&|IqS60|+3c67$H0>w(kB7<K=Jd;&%`DydRsAOE#AJ9*
zg50y8#!|rTb944fx3hL;X$F_-eWHx*$v0DL=`yP-_wOI**_SGZQw-+QqhDrIR-r#k
z#AY9~gHX6d_m};E@z_L@E1iWCl=i_Qha3v$6;NUyz~YB-@aBsTiQM=b@<7ykk5rFl
zW!Ijs9zC?iKynNFsMv~64p8%h_ushwuX60LCi*!$ed&@y!klbHNFphz#R6l8l?s0Y
z$aLYg_;L_C0Wk;$QN4izKNxJps9Do~qX{b&b>K#s74}M}5q5Yt@06~4^$K~ITE*De
z^~iVe#sa0d1?b_qfjI~`b2m$j%*uj;1<eByP_$W8(p*^F+Ld?*U3Ac){%}LyFpJbg
zwY@L=&%gPM!)P_xwanv)hMbk8I#6_`g4o@e6D|6HXiK|&pF#KNKnYl)qqvN$TI!J(
z*t8t3*8L|v^w=-Zd8uG)(GE!PSP}+-v>l&Llr;X`tUA!Aq{2B!61nsC?`pEUcTwI=
zDulxqLcY|zD{N3ltLCDBbJ6ShwLQJ@d|>La#_*l~$Di&Mwk&v^zjIlF48k`9_rmCW
ztM|4r(bMB-n-$pt(f8PZ?fwrz@4Pdy{gqFV^x|gaISXR=)ncyGR=p3BJUo^7R*47C
zu3gBf{MePE)heR68|@+8!xF}qjJ+#Tvu#8Fcj?hM4PI)55M3e$U@xx@`a4-%5p{{D
z?COb<>eA(~sS`3%!0or{TE!k{1tJraR|=B2RQT`1Ma$L?-eYBEZh_I@;=Aal>6J?r
zZ7DT)4)g&MyT7#JLTH3BH@BQzUCV9Hrp)N9twGL0II9{hO1zN0_;a}?mDCf7%TScm
zmLTYL!-#gd3gd;=>?UCB!a(!BI1bs1KW_<Eu0djeEpOOiEEp=GF*mh;DpPkChL=l1
zY$BCl6J=x5D@+}!@ie%hSjYs4=SVX|JJ$AvlCnbj5zRl31wJhnz_KW!>v`1lpLQKw
z0XuFSVE2=UG4)K?$>yu%C=s>Ok-i+4Po!3R-Fv%$r1>@7T=5<<70Tw@XVBTw6hv&6
zmBV^(i96ar&M}Me>Qz0l%=?8{XJBwS@T~Y;1&sn=h$Q$kp>JiwpIKr#)qj9^NZ~u1
z`Y-0~%S=nZ*y~XmXkp&SRv7$m$;jTDAlVGp<L%XE6`X@|^o_WA#?|cbAe~s7ks+I^
zD?|(F6YhLdq&jaEfznOFY3!dvZ-H7#jY;hwg5niu4sMQf?xqAc<?5Fd=IBe%!AAAj
z(aJKX(Y|QFV)l$-5M1;G-g{B;oF|Yrgw|1rZ7mdi8?t_>2s-_2t8Nu7DQY<T?2v*V
z1vFwaijn)##0*0u(&aDK0-d6qt{Q*3+6R}ts}YhMIl-RLJ0$MB5Yz86teR1Wm&Mg|
zQ7^L)ZKfQ3vAwmf3_--QxChidwO->NBBD&J)@$Lqe)24&D|CZYP2)MBWkYe%+M=!R
z`jUs>pT+k;FOV`C$PHK=u-SJ|3#4!1yw~4eDa#hKx`ZHEy&?+`|7;F$K8_hco8moD
zq}v$@vY3X#jGyr6JpEyp4_ZV`n7P_mp5JlL=AL~QG*?bg!vo%mSV6ml(U5a}wL!p@
zDnpl+2H17-&ZrO<M5OJ-x<l<VhhV?te$F6*7&IW+(B*)didlHHvmKP;uvY<IkBeI_
z8Joy+X9-ygprYj-{V|CAS>;SH9*HVVNx>-+BF5SI1-zn;s07ikFdl3uIO=xhw&0z(
zeWd%??>Ydl6oe}a*p8e5J^sKFUS@EH^{+WvB$3@7;#Qu-!Jrr}Xs&<Dm#|%)Cx3>&
z2X?Sr`t_ONr`IS|L(|S5RW}g?>}&%pgtxotQM4vW2rq&vya-EJy(Jc(-KYfYY8glt
z>9|>5N*HzjAO0L)d#^#$R@;OZY!eq$S*N+HR{WqzKl2XQGm3s}u;6D%ypv`bS^-<?
zD<XvnRSF#+95Q^y+>5xiuCSc5c!I<dTtK=59$=|e2H~a0ttUOd^1O-#03<~mNRAXm
zcl)g(De>kDAo1q(xvNsRHvdlsOWyo;GPWj|<eTzh2$B=A&kl02`@99eu+5Y1UptU<
zU?oL<iyx_vqd4@g#hGm{%`x1GybO^78q=3^u3UXLzhMsCBnHSz5f+01Pw|)RF@f$y
zz2G5?$M(J=dUchVkx}}>#>BzrI0P%sCf!e?n`k{Q5f*>#5M93FpEXGIekq%vHECxw
zOR!Ks3YB!!Zv*IGbo$jV?Olu7jZcIK7H(TCeUP`A<tLSq$Y#lDC1EpKQDQ`LII;{B
zw#O(aqA&Ayqrz4OeTHM)!$<fyxuE3)r0&pv9$-@@bx3~FSZCvG`x9UurBmxgE9#&k
z;#K(WhUb~vx&pC%Wq*1H65c(*LB7i#7>W~7mpjb%t-moI^JU6=Ji$!+vz7_pX(kb!
zu#?lPlzpccV<cg)V2Q!uc48m_>rc}0_Sedyl7YG@V}1j_U!<p<%!L;N+UkAJGwc5N
zK5q4SHMr|fck-upnqv02L}BVtkTr52M%#})Vz%LfQuSxh29Fh74~T$Qq@l{9Es?g>
z7U2FDm66Y$aa_H6)p^<u(((YA%0sp?uF?2NBsD;WE{Is1wb%u%Cd#j(y6`>0m>H?c
z#UN@&RZrXg9?DfrtBgyeHPd~~e2an~2c!O)8tu@5TmRT*V4)`al^8_SMX+hT<ub~p
z<b4TKSU4W7t0XvXKlOUKAEv@3Q+#<#fB4RO3f<SFm-7tSfS2ptQO@QLo<*q*P?VLx
z22h+xGdB)N0lG9ruMBk^!~pk##d`s5i&3(F<4v=0?`dp3nEVFh9tt--^+XTpO&vkm
zml5{}oGtO_hmF%b^fx<Z4^qInR|gx#reSoB0GL||M15#?Jnduau0$z)5^H7Vxs#O@
z{kHd~Z$dLI=m|qCnr35RB5ly_37{wc6DIRjwW$jlEMHSHA*Y=R$Yqkr8@(jp7zhMk
zYr1Ea@{XYf(tGUyBjCTgP(6xCJcLmv2BPT47>WB@?guyH`|pPBn(O}Q0RHbh+q^f=
zHoy1|Q6D1|XoC@uR*j3;D0n9WFZOqqsT5V;@SR=*PrjlP+{T802iod>M|?o7mnX8(
z8&xbS0kct(cmUq<(iZ3O+MhGjmLCe!jGit&9Q#?oY5mg}){}iGvwIHsTVHspCpdNt
zA&$i&`o)clxCI-KOt|nc29`!Z^IZIO$f^C!Ljontxaax@8(u5paY${W5_c~)8jd0(
zo08dfnQ5oe*uHHDpdQ*fgR<>vz63#n37|~Tz&ea>-|9|%fXr*%FI1>Cq*%XmKt?iV
zVUU*%PVdt<hHK{`1_<pBJapt_GcXYGb8x{8UieiA>Bu74XDEB<`*A&Xdr{Co<CMYg
zJCR@_2YJl@bd&I(aI$;|5^hmXR6#4GFA12NwZ4RMl;dvMWUy3IpE#D&ZYvSUy4raN
zLCw+$!yOVAC%ilRyWRbD+H^`Lvs9+)<t0#%uOS8bwYc_m`u|Nx)T7xiYuLf~xTf!^
zJyqkg`95p&V?tRJG*t@Mt|gacwdK?L37aX%-am)a8&V+>@_|cU)kK*oU_HF(?|oKS
z)=^UV*B=^6M(7^j5sKmGfM6EV4bb|aEcoR0t95AK2~2=CR(!j<Ym&d92b%`)t5+nO
zHvBnc4jMJ}$&7;mu+-?O{$9(l{k;bg&|at<Uz;eMM129`KYO!v)qpldNC(CAd4Nnc
zq=;!KrpqC3;V>bg!$@kb1Q@Ieue~9d7g*b+3c0YEQEWGEyi~nztg)DR;0WxoM1Vl+
z#B2X)C^-GUqPQU@&$sdITm8!=YJFvgkFzv`-m+v3dWtq^xEH=<&`o4+F$F6_p9y)J
zI@+Ox!*7UuX>M($)EKmDoAmi_0R7fe*x1yh3`wJlh19$+p>^pVQ=i^g&I391kU=_7
z{E2I9z;%||jdKx_bA-YxW4X=%;t(f1{(S1hhKDaVKd}q>$@C;+$7guUu-84D3)sB+
zg8lPS@{Umr*t(!~$dcqFXKMO3H{J-rkEfpG&hK`PdHM)zEP2rEwsGC^O>-H<O;fPL
z_QQ@+)_NG20kQlSHBujZx`w3@k>yC5%E69DiiLYIBstJ1>pNq$|A|b(N|*6hg1jG8
z)GYWEYgWacnGG3SSbmn%(*EnMhQTRZ;RP^v>}`|<pOTYF8?`jhWL3ahz`~PY$}?!)
zkR$E}4J`jjncja0ga}<RYL5w=^T2Yo_ME^?{c{nY_khm2NyhkSpBrCWFNC*Ou3n3p
ze(kzF|K;MWxU;{lPBhm^ZvKGY-dtGE3#)SPvmBqQd%xEegG^CNV)U$Xzge}t+%j=V
zZCUE~n;+e-JsVq0y0eW#>;3Di&mnG-sl@Rsug;>0iQMV6QQ`~Gf?BHOY(caewod9W
z2H(M6x$zJ&PR+Qhl3d5|M^lvv3MVT~&5HA%m#<Hlvmw0tFGyLhLBEPi7`qkoQ?9dp
zJ{G;mcgU_laroirX1A`Us2Bz&SGUTz%+g&r;>AI4Z&9b&CIV+txyTr>%xzds4>50%
zkWx07YX{Z*4sf$zZn;o79Tf|0&O?!bMC=o{SjiZ3Z53S3IW3PFp)?+i*N5k-q){%|
zczL`^mI~s;VXxK)8%-$TKOu8uG0HqaqxQDot-&ta0)l+xCmiL|<+mWjoPhxa7M9m7
z3F%@HvsWd90L|Hm*UC`m%6$TY{CKpU48?jy1yOPiXWZ6ND{qbEPe<i2>Wx?JTpzjo
zNBq*qIusUJyAl~+3ogyox@@j5bI&~s%2vdU&geQsW^qk4lLybKD&d0XkaIx>4_W}B
z0NDE|rtMKmsRc1c$n9Gy+JU9F<DJ$@-}sub1&}5jO`qjJ-`-RX9BD0|PTwvXLH)m@
z>s}eVTPX<&<&yC))r;u_QINA;f#@-a;!fF)dfB0oF+Og?>XcymsqsN+klzpEW%__P
zem|Nq=4OVQlge+xAk$=g5FZoxXfnhYwJcZJLfyqx4qY1K5GCHZ?z(HdtfM%mZGT52
z;p5{Z=}{bfOxNT76~BY0rV_c5b<|olDe#ZYm~Pl>CIw*6F%Djs6lgY)=XNU(SdM%1
zMpjYjn@zarh05i7`FnX?e;;TPDQpYz+muo=lDedUSA4;mtN@Qd_0legaLvFFVZ40f
z#i?J1F_ggBSN#~Wb{@@CPxAZGm<U?0%4t{$>b>LUw)#$Rkvw}mXMN$r<O+)ynMk7T
ziXGII2=-|``#EUVAmfv^AUg$kFg>pOMZ@&hSw{S5dXGQXe%?Fi%13oR4>3xQXoS1-
z6ww>z_@(Q$Z7eCjd&wIOA-^;U7|^<<X<%mpN6><?HLMy%ra}XYSPcpnuOA<7=wEqY
z9=kXX+2F->$A``5w(lRIE+&-a=hj^f9<rjIN=(3a1|>l=?14xycaf$k49$m2fPK2o
zrSX82CK31+>OM>2AEj0K>yFvGJ@r>i!3oI+z4)lR5ZH!5gdSo$?V-gZMLc%1fD?NL
z1gCblaD-!4UH6v09Kfa}ppkI6m_S)8we#vuLxVVWk>1j?_k5nFN7y0;zCJO;)*?*9
zG4P=eFb$EMm>XbfPBJL7P3D?mYG<&YlvVU4+whuXqY}20Hzjn^)K6z&PBU=m<YITx
z#0mN<BtbXZHxT5s4;lhb;)n!E-X-9G#6?KETXtF=cc14(I|T-|VlDAZ{Asx0_JRaG
zQaO6zpq9>=`xQTjM;%d1P^$rjYRt9=vI-RnuDw}^`YDnkZ3IS%DR7iKU(6`|+#1<t
z8@qT(2v-;xv%$2?38CA?i^)Cfa>4Tu4`N<__pq}Kie}5BxG<z+!JndQKy!~N!;1R6
z)oB<i!(%8vf7b51_!L4AN-U>$t97=J*bOsa<YKOOL(d)p)sasd1+5e+c<6jQ3R^Ss
z;WxeAds#WO0U<%~>>vY=o<Ld6u}`LH6GyP|ztGpY$evPXJLXGre8Pdg@zA50z4rbu
zHATu88WqKY@;4w9MP>_nT@j$`Dl?{cRN33zAeUb=S@y7Z5JB^fM~GoBZ8BwOJ<SP>
zLxHQCi@ge-^%dw$ec{`ppB2aIE*l<tNa=54OV|iAalae0aNUknVr|}F%~sOxRk=T)
za_=oZ$&~4%BGQsCWK)SlsxkhgObVyQ2r>xEwjK@Bd!H()Egu=ih<&(lfkkPCb)4nZ
zWRXXq@bKS1@9$z`2E276gFoo4K61k(PK>VDz$qQU^PqYkoZoIQ_J1iw4B1VUFlsmD
zoo%%r7U=9$DX|^Zhut!^x^JPSN`;Vm=eA$5Krw4Z3D}HP=|Musr{Y0-0t}^yV+KmP
zw>05@$%8Sx0kNvkz!Y2iiQKQ*-SJ@4&IldFRNwe+6_jFs7cH!bZ~!dH&{_H3Q=tk4
z2M}w10~y1s5#8KdpjgZ>B^$PcEy>7aJ$$i$J$NX~j|S~SfLx?-Xi}Oha;J|l5R)G^
z2CWh|Rt?h}v+^tayTJ#ZA57xET#T0Z{Vh!-urHL%s|MgcjU?hZsIG$*PT7XFUZZE9
zg6drfQuC{Z<U1d)o8btHh27q+Htyl1k|O4Zr!OcPL)wMYzDZ3DO;h6DSG>x=vs&c&
z?YZ4i>al2N7qaLJxpo3o#_Z?heCYA6KYN^Uv6CLJ&#Rv15#`0X6}gG6$LVHMRj0-l
zKCPC<KaGg6eW(1hN<wu1c?vsY+J8JCOo;qDm9N5ab+slo&D1~?Z$FvW)YuqU@m>VH
zp0vx?-yZQif>9TAa)FmCPkC+z<MiZJdJm+|o<)7&=nUkwhJg=Yh;#J@D(1<y>Sggr
z>6i&&TZ@d9>mQFzB}$VbSOcYE&p*$mVrtScYIGg8DKocJ1%WX$B*WVRt-&$>T|ZIR
z;MqMUGvX`uz1(ZBQZL-#GnsHh`8`lMQ;Zg(P_!6$R<D&Uq(vKtLyTdcZLIlQJ%i)N
zkE0aUsCPLp?+RUEbA|IJLa=g5L*DSA9K9?K^gm&~$+7$Tp99#$xUdD8F;2jy_1IEk
z)QyB*w=h^NEVBe~MSHfNNbS~&M7PT?xzQoXP=DLv6OVj!>T+reNT*CsF^>JNi-mQM
zsb~3a&RKjYT=s)SrfXjbF!GVu_hCiUCLy#swFMgIzd#w=l#21!NKq7D!s+hmN$VUu
z@@=3%c}ZZ+!}HFih}6>qQFk|P*0v=d`1b*lqz{LI;S*2zTb~+6k{0Y>TT_?muh2KO
z!rt@GR!>5&e6|j40IK2Q=m}d<9Vp1Kf7QIQdD_Eias#H#g;FF&gY9nhwz28_fNnhj
zHAw3{!rcT4twoKihf1|lafA_uB6gT4zl0ZKsHLTq29cdZxQ|M0@~cooL<BZ_ZOVYk
zUAg>)$@y$!o7?2U6SL)~nVBCD*yhKJ9l=ln1__C3!%JD77MYz~@=sDqMTS&4nHpcm
z614$E-i5t%VYlz61jOLr+hJY6tM|Awi&SlUu=C!(kvzbLryvw{@(_?);=xHk!NoCG
z6AU0hY`6@x5`1qQTE9!3zx2W7<UPr9cuoWb{AlW@@l)vb{%<hf7IfUM{W>J<{k0<L
z<f~7~W<zx+HP3nHB&})2?qK5xf3drzeMAmNIHQJS#<Vw3GvuuAvOq%kz|5k3iDF&9
zc>igQ3IR2OD{d$D&+f7hee~35<Q__F0}~Pk**UENbY5Sgk^#m41`w1IfD9wH)<2Hi
zIfD5D3&`4^OGr$Nu6{rC6{<oD$RA2>tA9K)H=d@PaIgH>)Wn^^N3;Wgw<^P9G6k%#
zDzy%TX}f)Tz?usD3)(@MKK!#%I;2#HKz@bAROVjFt?`=CJnMu*!k1vLB4K_0s2h0U
zk=Ew-+hk=FJNvkzycjx<*eE`v-p4UUeVkyLERM7a^l|FcyhDFQ{lC9Bi7j^J(QPJA
z*1>=mc!eS6O7Q%cjd;9;fAUR$oY9y4PVJzyCu+p^tFs&4jk?NTcGB44XQLrsfr(BF
z?Ws))zLUW>J`o(o@O+(c{ctYDk4C!)-(dw~KPZ-nJ!dKA`PMeHHrUOXp-dvOsR`OM
zjE!Po$d7c&AOG7Rik?YEVC-0lnJ_$G&S8_XquK@uYEK}DMzpHH5LXa|UmAO^=>W2Q
zn2;F?{@lPzZlDw#>W=I+W&koii>d?z8cG_?i0kn5BoLG$R&`^hoZLhUiva(dhUv9~
zI;ucU{J|Wwb73eouWWZ0w}Lacp0u8!ISD2_7(yTB;}_j_>nZ^Cq8}GDPKigcj!4pf
z!M;O`<B5OsZ#((>$V%QNTxi9|xN0`h2Nc4KJ;zXg37<X@w9DStnVO|)dDG4I+8Fx7
zRkfJfJ1c7p4c6Wh9m(A<ZKxXN)1<diciY+jCHyfFM%|OC{q+~p7Z-x+T&UGrtXG|{
zSU7ol<|rmR+9ae}pE~hZXZ8EF$wmkAezxE?nEyUw!<xHePBQDTqr?o-x59yrL|z{R
z%0uc!{4%AVHJr>0kYzz><4hF4?2V`0EDeW`CTa_1G{XZvZyzs6co<`!NA~EpPOdQE
z_UOVx9QtlE+T<h1b6o|kl~{zycmgA8qI;;2Gl_P}9#GX*u*+ZcWU)vv{<Fx9kJ|##
zv5@#bPI|Ca@hbk^RH%v!!ne+y++P(H_LOTP%cZd34DmkB%d#+#rzwOnqV`2&S>`rl
zGs6k_#{%Hd`aL$})3MI{ZdxTL<n$}V1h>^b6o0PzSWPuHec4jGI`QM0X3_<8(I-zn
z(tJz|pY?P*10~a`4^CO>-_~d=r686xRaEwWFYV}jGWPzLmIipiPdqfavnZ(4lyv3%
z@3R3J{>r>V)&Or@jF(w)%7OqF@C^-pTml1U1SJ0iZdP-zMli_}bDE~Wn3udW4FMS(
z-dYx&37o8|&+>=w(c_5{S5l=k_2qqgMCBv~acCM4qQMGk!gp0eA|@bs(gW$)RC{bs
z1xJ@1_WLOHqnMjLFCLTks{^XGn5<<m33{p)c+O={@xws_bs=_C?dOhGm+#}a9i`^N
zK-s-O{fDq^@qfPbX;@A|d{b%)IztDl%efskBjCfp>~62OLDUx~us95_i&rNkzV$Xf
zt|%!nY<b4#Bza^M?N}!vnSbSQgZ%YoZah8S4W@d`-iZz+7xLVBD#mN;|7@GJugfcg
z+93T%4%ZvPw^aJvk+V!tBroK>K69Q@ia6`}U16N-#Y|-#eblNg89zmduv~wiYBL2R
z?LGU;-vV$q<p9O!fXR$*@V4<(`am~`*5X}?6L^>BfwGWcA<Zx+VPiX33R1yk5*`(0
zp=$_@n|X661ERbD%*!7YX?s*bX$!bjl-}a`D^kSJ1fC#M_}KCt62x1<OMbI#mMKJx
z5zt>WTYVMesnZM&o3zU8c|Q-ZZ*ffx;1Pwe7eGF)88Q`^LHO=m&J{|yZ5dQMsabIC
zB5JgaT!?AK&cGB|Y%p`v0!D<r_*9*|9f`TDTy~N)&`vs=XU2j!wi)i^X|@OBZ;B@N
zrZGl%RD5_e@v#{=VH#67O859b)0na3Hu-Zv%cHs6*6XCmF6}@ruyt*<ynNgimY*oG
zaj4gE!C8T*Y2}%4wu4#oXo<Mb0q717{w}1HT=`xDdQ>9UmERr@dX!sn>A|ya9sGT#
z<sAjCpM6p41DyzoeRm=mWe#+SSL<o~1Qtkj3X~2l0HU)XIx`w_y@o!sD>-6nD4fJ@
ziLU3)EcMpEDm465a)<42*JyP2BHP5TvlGyFfY07+W#aWNdBoQ|AlOi%QKh+WjK*5V
z=B;!{EI<Iy6udKKM$e`_^ZjU0TF<+NQmIoSj-kK@);WjutiL9}K+_)YJZ5o^U>P~f
zQcY^T!k=2vX+MY8Y>rJ=PNk;!jgF8X>!N^isJ~Tq(W*aSb&>GSDS+i#Z1&jRv|)UB
z<<$GLTmo|R90=%YyPBr;?*_w1@(TVPIQD2f``H(b;?p+1PNu=^=oOjrxdYyfOyv*w
zl(2<le9*SF{A1I^u)$-j^gqd05t6N0{(S4olU%m|@G*HANaR|cV;(g?LT2xr6!mg+
zcBmv95mf`DA3BfvnjR9)usA+5(KJzzEsvco_$*vw)y1_qx0c$;_URd;r(b({y-$7E
z=e}h$*ein`+F!)sF7I6c6omjrqM@(a8R}AKRnpRh&PC^MrZb>52m?*ju+2~89YPMx
zviA3v5}^W9Yun#V?|}=>3jZn_oq-bY1T7e{jV+2r94Wvzi<c!JtF;g$>XqhERNe;<
za!gYS=S7K#yBVQHH;^0G_JUO{A;J%r&_XZoGq&B(r{(e(w&suU#q{l4;n^x$fo-c#
zU^zwvN1GnHz#C^LmTK9`8+%Ny{*bTx1r;;K`_aw+gnB%{MuIoEx(`yu+pE{q;m@d+
zz%f_m%&#g+Y#8cgXV;x;IepxrCE`Z*@1yNgF78d2Gd^?}Gx23lPlRvnTU_kuc6Eka
zuWa?%$L%{wT5hk(E;B$ut*<)_r(3AaH2TB>oNjM8-JU1z+HKFZnJpJkn~fJJSGe%q
zxE0avT~K<uQGVW-;!ozzKM&AHj=DELKZ;d+2L*gPJ387BDV6HJmP4VFNq<0IaAd1z
z7g82=oR>x{bsG^o2VkFb8_m0oV^b?&YKP#owZ(_onrRPpDJC<2WH}G3LNp;}W94cK
z>}*~9#IX$R?HOQx!cj(eE<a=Q>*m`f^Uv`30l|)jmR!F~dv`P+pm8>>7mu7`HSEM}
zlmVJJW8h>6J&Xb19HIBzL1o<yLa=|nI}>09-rol)d{;;SD-e4dX1xEteq`bC-nj$z
zEafQM$TJ$!XelOS!VO6{y}T<T_<b@&q7~LS6yCC{l*En6rgtb(83w>Ep`W+;WB()i
zfEGZ0Yz8gWV*d1m(i4)fu&;Jutl9wNJSp3i5Rio=;d`xmNP;_`6g0y`vk#OsG`%l{
zz>d165V~l9RTcF0pQ81(4^OZqmjEe*B7!UxJPdUTZ*<ar0M77CEx=tjDIR1HDp&#b
z_5S^c>r88~6aL+}x)0u((+_S_PksMseczc=KwcnLB(dT&d9YxyF+1Tic>R2fh@2HI
z@G<I#pxA6F;CF87+6ghI%q;VyI>g7ZC`DT+-tWE0y#FZ>A_O?%s!&`Hl4sWq(BnCI
z8IQK-bqoxvmW?{c{Q^;3Y0xh(6M*G77zn_^_5`dgl^k2J4=faq?v6SPjMvIWA;E$X
z9-<Nh3sHxOIxLjVvdFl2trDp^;bB^c5{=UktbmbAv~8|z%w73D5mmVPD|6opjUC8l
zDkbE|&-rr>zPI%Z0D<waV(hiECWS=SB+shjj|DxABu3ysy5HtNCIpfll&QtF7zP))
zsSO{ql_GAHxtEb~op&`gFF<2=T(ckb1^<8GehB%$0{!8Qfj1ChihL@}0s{TEugqtF
zXDNh`VmDYJP?BiXh8Z3}n;#)-=xXJeFUyBvb0SIC0i0hzN7~)G3ZSFv<?JAAw6_Ln
zf1R0XE4mV){oymMJK#3qi}#DR7oMCJFg0_}F&=UXmjLS{{Bxuk-CBX7pagJqrxVD=
zF89)<0)lXvkkMWzwBk%M#3<-Lhh=!DAzbiw$*X~R_<XG39_!ul$<j5Sx6<6Mfj?DA
zdPp&VLm={;vvMe#*T!#vX_$4cv9{u;2CjUD#P_*OR^;J^GCUYr01X_X3@vVyecHXD
z-TlP<S_`NMm@g#6WuKxdAo8_8F<7<!j3{9<JvEJbxU&1=B>%CVEX7Md&g>OVK)^^y
z<g`v)i=xN72O~Ti@q>IzS$Fq}?<b3lVdqMY$8*eEIW*z4kjbBUx-_W$&#SU`>aIcu
zdpzLkQ8hf6gp+Zh9ap~zEe#pjrBFAk@tSH_k2*T(0E&^9(U6U%=OL{D!GtrPitFyf
zW{ApZ=)uXIh(|t*2g6GdRG}(qR21rlp_mGN`ZUXEAkphaaXead@iH${-BUvfyYz`W
z$=!-EDe)M_)$h>aDFWDCbocs|QaEZ-eddkjSpnu5bq@hbwo;3(SnUwy9Ch5t3Jwun
zDwny{B#SGQK_Afm<fO!OobY!kBa@^&y%w?A4^j3or+C419_=#b#>$_DN)qgo7hF7<
z#Bug_5ul$~SguMtGh9m;WsZm^*TXBqB*v49D3s-Y&A+cbsrvdvminPK@KEz>U`-}m
znxpLku^Gx5jbt7)h?mzqdO>&n(^K=I(p^8X)ee@?r&<hb(U;~-jLN>TLCRF)tgus%
zO1_u%(P8=k;P0EShlpEz1;_x5`Wfk>8=^qsc3Nh<jyCaP1HJQF#Y42!7J_FojH*7a
zbx?^mzIkKmJ!0%OO{SrSwQ@s~!ad`7_si-ZPhp+fSbDN{hQE^WN#HL#lMaJ*(Z%(4
zfK@0#_ZNt==ocW{Sb+s!ZBQqR-CTBI0H8sozw+G`$QHPIdgZ-ffLZ_Nts~1CMI!gi
z?YA%p1KNSkYoazuPl-n>)Sr)`{WXWY+?TTJYnxB_*x&2|3ZdY**A_6l6|(&oI4vig
z_;g25G6&7#jwCCr%~ImR8K4Iqva5`ZyA03)*aot|vtm-IH||_rL1G5evp$JG!OcOS
z;!8JSUr#8+8jm__EpEd)p5l32zL%~*D3}BB;rw|d1k5~0P};USxdiQl_}puvERJCS
zXQc~auP@4U&4}C83B6l!NZKd|?LQt?DOqQqlD@}ime6~iZ0;iw`y=VdOl&;ux;lh)
z1l%?o8{@On4)rIUgxqUmRLjz=^#kRBY)b<<WH^XJ<}psZlh90{Tu02eDaCorn%t{w
z`^W812##d{TQKuRv*wN)lm+>yY4=iQLw_286<MHi9$VW@Yk!hGd2Y3|9mO;*0kio<
zHE<38ZY_*HA0^&?|C8?XXQ;!!=nMXlHX!qJRdgAM+1p>o&4R%t*-Og=xWQ3d(DOu-
zGRX~c63(=va<?OLNLafahWY@YFNk@a<bBXIPY8z$pjv&qmCieNG6~swFP_{{g(6DK
z-K##0xEb^?hh(c$*aN$xxiy)BefU>=FD_Z^Zn>M*i_Mg*?%>`Bza%fg#L&z2*4|Mg
zpz<3mJ5fV%Tn+CmI-aq?l+ycFN_=q&+6b5O_>_Cl4LIM&v=sl71>nm2AjL)xv_c;l
z5-(Uk86Re;wcQ<FAwJ?>oOAA_LF0;hW|st(AI+~-{+==57u5ys3dhXiAAVPwQEOfO
zHYa+!iF<y`_gR5Pc(|bf=FP?b+x~<>@I9cwrfPKXMqzM)4pX;6`!isxYdH=Fuo&pV
zCnqNt+vAX(SXm#}>r~Ek_zDA3hh~&}vlRR~vf_>mtH1~Ha$ZT2^u@+NLDUfJ$c}3_
zQc1uO28ED3T3Q$I26lcM86?2lRh<=Jc_yH}2R`m;US18<;J8IP8H?l8Yt;6R%dQ!u
zFHX%Ja)S=_*1X2@8m%r_Mq=;-I3=&teF<9^p{aq2N&cZ$!kWHXB&v6fnw^MgHPw$!
zzsxq?tw6}5{6T`vR~JJ_m_9SNQ5N63>)l(zvcw{fb%y<ltkLmLH@}tU&h6Tvuw~Ff
z-e9u@-V#*kyMLN1*R#-E*}WO0^<6<k%&)f*w0~vad(Zku`b64uc^%iUusCrsv9)aB
z9n5mjXMd)Tn`rO?CLch#1EI(?6`sj>2~<nu^Hx&)z~AuD>yVE}%tH4n*7diKzNVbl
zbJ6i`?tMcuraCuE0s2MfpEyKzZQ8@1A-)93)7-u!1yl`<8io2-T#>_-MPrO{q^RwP
zP=Ff$*DMkNxdXnT$A!FkbvA>?=~+7}YVtX|P6f3jsIaWPWD%rjqxzD}rX~+3#If;w
z`Az;q!e_ASX^1Y6)aO+6&8xL$M8x0!p5@RZJKNp9MK?-hNfWuhk7U^62pAL;pCFPh
z1osz|FtSXCcF;#cMJEfHcPdV*xG?cfaNMz22|_VXcZ#h(YH-RU&&u~{7p(UH=-*H~
zu?|psBeIvdQ0SibUy!E;*X;baPeI%w`qPfF1Z?U9tVGlTmx$>u+#2m@rpMFCOqc`{
z1~frOYE<WhG@61^EMc@q(8Ucs9ngac0|ttoh&RLbciSKHJD}09@bXcI78;S{dR4ED
zk9dopdfGj&Aw+6L4lvO@PV*xh4BxK7XRH4%rQNGf(ng68y{13tmqy3_>&iXtExt^W
z0BA!3nI#P)la}yoKp4yY2KUbM+^r7fUB$SA#(8OV57UlK#ox%%vv31U^VA=*#*4@q
zMlmE2K&!XtIB!?Y16vED#9Xkhh+Y;KwktS=<8#)$3o1yiLqq@VOG=2lVN4~2*x(Jn
z*OC~~SpoTBGbkz-ZTb5lV~LU=Bcj;%iD28~b6e?KUq<^H>u>W-f0#Q)aXigMCxVV1
zepqZAOTTOFC#Mo`;yduX#yaMwqR_TM-S4sB{=xfK;ZklV6*V(kO9`{CVeyia2G)M)
zoo$aGXWKz8CPpoK5-DOOTS%4ePLV&VSLvLq779YFaNsHxORPG$djYhLu7VbfHj)l)
z?UPLc{T>q5R(guTt07voIhQ?D+b)<nOcur4Q)Xv^d~Gvt`hUVis;fQdzliJF9~nEA
zk$xV`m&zvvZ8_j&B(j1Z?`gv15V$}z23TGBGkp$_ec^RpxRPEeTp5)Ne8e|U77LH4
zy)3=!5$a#B&A<f<l3G6v6`Cu9F5E3@ngewiRiqQ5$%6W~f+SrDp!-OeRC>9;`%tG)
zKRSN-^8t<?$eUwF&IR-S`2moJSRN%PTlj|qHy6lwX|N;0-Dxpm{bE2n)6uVLyLffg
z-@OlrKH2bpZgC0^e5%i#&!gcIw1AEMS*h$PG3Yq@xTek+RTP6x>7f$0CdJc?C(F=;
zIg~HcdH-U7_Xbnm@xC6KIc30ZX>tn<qR#4aS@vZ`qTCdM6K4{Q)C*t?M_M`%^>DWx
z$6s*nh-rxzEz2Mz5K}S1IDhqJ2i~^=EnKn4t50!rdffnNiDE-hkAchmg+uP`E&GCd
zyKN<S)SxJ_YGp7Nbs;q{Ay34`QLXdAP5H*y{`!&p<5$DQ?%Rq~k%c{>a?Zbd5GqZI
z|J?pCur@<&_@pKQwSf;8Mn9&q5lj%s|2qg0I~vgX*M2v(@zk`m7SLc8tNTNaVdbdG
zJ@Sn7!{UH7bc2AXC5$Nm_T=I>gfhq=*DZ0Gj<@nj;L+`>zTQXei}r4>@CG>MEQb;6
zXcbrSw(iL+>^*-^D_H>U!hpE!0}tV#D8S!)DzA(~`{*_G#mfUEKo7ks9Ng5;P-oEi
zY+?+mnZ+Tj5@~VISxAE}v*m|bgbMP{oyLAVX4kbr3Agjt$Xr(dkuvz~>6!Amd}pQM
z=n(HJq?)vW*5{dk1~kBUzUU$9y*+8GiA8H=$e{%h#_f>OG+jdHUi@bKw#Kf>FVHS?
zjE3I%`OR|RdB(i*hzk6+1wVE<hv~zCziNYKcG_OoWY`@R9_1zgl-Za!0^E_+(Q^d`
z%|vkhCei>Rug1CVfK#syBn=OZcEx_)>8`KjjL%K)&X!967WcZ*x23k~wzczY$C6I*
z0Nu+^gli*NcLS?v=dE0*f2R!lu|Sw&zxrnL_(Gtsu1CFJb`XQMX~Q#N2QyG(_6PB4
zm%;8GPC#wGW&{98841aCtswY_x1}Q^OA~LI1dKGQyj&<?7<Q2wR(`n#Z!PhF0lB+_
z)T3;AJY~0F<5f`lh*!(ts%8Y~^xoq_8*hQiHX1fGYVQJVrF>-}EKg&UotR!PYYFjA
z;H@g#9*zu)S_?X9%uVLb4ao4ig~e#Y0*tPYA*+H`e8yT=UmvEZN@HRgn4%{CPEi!O
z9!C3zsJ&|kJX|+rmo5%T;3X`l)m^l$faJj10e@J^y>TR|qa+)T2N_62$Mfn5Aw2Lb
z3<oF3^wrap$k1}C&b17OmAR2!XeXy&)E#mnfCCi)@A3oplK5V?+kLc7G9JJ}TwHrU
zJPW$LLeMFEc;%3u`r?JvQ5W^kBjAAGTH5jY(4W@PW(NtYy=z2Sm&_CMOw@hJ*>mcg
zH3gfqYmBPhbqHPTKby~{52+bZs6+Jp@F(nV#xHN{*o@_0^`FTOj)j%)N`PILgp!h-
zOZF4&Z4j@D)!ujuHl*TtA#JLJyL1`f4vvz0H(fl%O2KX1>!;dV5dTx?ws|tPenh&+
zkG_M75t-R7&r7qqPOPc@%$&dfkb-+Di@hIpD71&mx2UbDFHdUt-hv~=zzX%Q`3TjK
zJvQ&}D8?vl3KEBJjmOolY@IxHDt9zx_FY3$uIP;_jrC_xytex<!c}|1U9(XyLf0<n
zjwh_m6;FTsc^-K?m!=>#nrV@kYh|*FP!`$}1?XSv-ot{Y^NUC0(phKw9{dD?)Dkog
zEkw}MuWd*Ztl@R(Z1`0)n}OXd*^hZdWv=l(wBseLH!}!DpQ&VBSSu!VdBQCPJ6h0Q
zm!UUorPBZeD0w756#4lw;abwpq%W9$7<DI}w+*B%ZydoawCl)_7RryGUKZ0*X^mRh
z`6Mhutz@n*6h@#-(A!&Lt-HAFnAn;^(}^q1&@8ofp9inA>0Ef(FXwGVH*xcH1{|ZO
z%u1hMEIr?}P`IA4KRpni39BeUh=}Q0#eay3M{sdvEHZAM!LEP_H<6kXq457Ms2dFV
zu%7AFTE9U_sG~CP+|^Ix-n-T0w`}!hX02iUmgCaMhl-z&)xZk71M@rZR^L7M=xp<0
zi*vG;j*L5QCM^Mt+QAp5IDG$f6f_AtfG9ie!1_AQv~dmo-Y0B#6j?RZNFO|W*u5jn
zF1T8VkOQDbpSu-63vDE`Ev5aDWb6_ODTnz-baeVP9$f@_hGyN{2<-f6<br9~pnHoN
zMV$+dA;E&!M#%T)#fo|33NEjV5_QSBNzQ8FU5<WMfYXXIBoI0+4mLilBgXX2452BL
zMtVum(z#!Z@prjSZE296mZto&V1BypYg>hjIH0T?3VEEulN$zUq5}J1R>?*{5*x2J
z9>F=a|1@4gs;A>b*ea6r&4_SyJCL!@)(Ie8W*w9`vtJuuo#)#k*iuF*vS`hg-kpO7
zK33><vQsoHKxs);e3G;J`c9GCvM`m<wA$!X_34=rU}trl&OY}em4s)rA;cHv4JD|a
z5X%A6^G_(DG<mBQNwreMu!JTIUJ{`kKCqi&pz|y{Oi%3c19KI6q3MRB5{?xEe`H9i
z1C{#kVUe81mgrNIs}HXUGAPMs=eAWH#<<SJ-k1VuK<QE&HG&t9`<h#aw5F>4a4)o!
zEVI0l>r>TRA}!gLnqNl%$kLTg(<pK!coQG+1enrt&g5+kudz<AK~H|JP`OBUcBHqG
zM^NUZ+<zcUN__j<UX(<PMt`sLUZ?Z_t@$&Y0HOjlY(D{-TXa83N5_qD9q5PjfnT&J
z>v)eywyDx#+b8hg>FZ{z<LTU|V(|m{IzcfO)-F1f!z!~>PrQxzKaVAWyk%ztBw245
z5|x;Tsy)gYoO}&kAutocq40EyaFhZ@iXvhI=Ed}2r-Bu-0X?%7@%V8|L$Iadsn`*W
zE6g`$@bccjYf^5-Jr7OHIb7qoGNYPd*P)fS5`8lkmOc69h~T{e1h4DN2?B$A`T(Uc
zw=<jw=D0o_6BWf({WdT9Ak4YOw>t*U=!L?Z`}Y5B&i$X->3^6M{3rZUxYhpR*u5c=
zD~LVXq3qhv4I`=@Nu0pP5tL#O<^T&*!4dWZNP>s%XUZ#F8l07sm7hStQ2<FWEL^H{
zj)dJ%hJd(RmT_F+kNffLiImoEZ|-JR*;Le5HJ!fS%IVu4cCS)8d6-7vDU}gsV34YO
z-JwDH-1A2)q!8A>XP+;GDO?$>U-3IwuoXKLVD#pSYFazkH9-%&uHzL)ZS9C#poN1m
z_ltdXIw2TrY|+*u8rvOs+nWKWqRA`b_<kBJB`x9l29d0Y+MWoiOFKI--Ex?EdL{F0
z&BD^^h4=7VG<xYw_ow1KZ!?wut51;OZI+@UoiY6$_Q}Ar!NNAo*(EYme|l}(?!8@q
ze1stR!uQ-T<E6edy<`BfrlyT=?N=+%Ph<kI>%soUc|QWbLAY{?_g7O4E#ri$PbZd(
zbk2S<6n>F9@PU=^1i-jhc}od``%%&a(q(I`cP@?BnjpZyoX*K^3H(8hwcY~CZ&b-|
z0enN5b3Ih11n`{OIKo2jZ}4LK!X{HBciiBuLC<IwJReLR>q}09$WGr3mtpjUkI=}6
zhI{ObHzBo{xU2>~BFa7^9&jsKK>{-<%pxOcmvi;MfrUrN@P^(Ou4bw2=kwqOi-2+W
zit|S@G-E27c7_Fg`#wy~yyjk^wzXBE6?TW0&r|(Ki}{3BXW3#c{aJ5eu7h-J^VUU7
z*Wi_}zY!yI9?+8mSBYmNi0LSQ6d6{y%dkhQhUUF{l4|j)4o?KSYUP)_@U}i^tMH$C
z^f(+Brnq!p?`X&66Z_uQ1K2E(CHFw)3GaKFKLFV8mw}!UIRHo-F0qo@!|dy$OnaGe
zf5D28H3j`0nnM9E+mVoU0a36jojUEKrDIbhL{4+E^!eBp+m9d9<4L>~hLI=F?}T;_
zTH*|g6T$)Z+j_70xe^4DhQYMrbN2qQ1*|1QBLzLHAHiLE)^qdw?Q&(v5jCt?*t9t3
z0@S|vS_G_f|8Vsu2O{>W$gX@{<&Dk18B$jHv)XgJ^{2V)`a`eiNkuH>eu%4T`=Ka(
zr<lXNB#*UHc0m9~_-<!c04CJ`ByK%TyC;Xf9}P!tFH_gYI49QW$GD=LI$t)2O7z0T
z`p;<$h@AOr`Q7^*pvQ!={^<Mdq}}}ok_UYy^bZShkU(Hsi`@h@?!RFW9|x>8^;Muu
ze!_=3Ek0U?H+wltTM;Q5-@kv~ahA{Z-P_U6W-G-KPT3{;u_Js-Em;+jYOnF>>976i
zmA(+khBjEgtY!h|(dENK_&5gI_3A(}8QOkcY=u-AW!qQN<mr`}5-->pTE_o@J488W
z0PO-J;b+I{O8Nkh$r^l{Ir2*@Bwip8mWz@6z)s<}jo7w(d!$HCt=b?mDqxu_$)JRS
z+a$Vx(52;QT<4YfjW5$8n)EKj^I~=jxcHd-b^Pg9ATm+^0Z`#Oeml*SsikEi_Nngf
z?zDlJHO$1_`2**at9Pnq{v5d*&)mFDtfosW?3a?ofEJRK_}$#a%KdY;J^6&_HK7nL
zPZ`+Rc>zz%;o8}KhY~y^_&Ek;stAPF1OKdHwR_*>eJ4l`90P$jlAtziQ5Jl>KYGKw
zxtOy;N2mG{E!CD8*>7PAee>&;>!uLX)eS4A1J=NutgjvDTdPIyM}sohVJAi;|BapM
z%znK}{o<QRX=K2bcYa_o=G+AfaVq>})pb1}yE7b_ywi~u52o{YatGjQkzTuJ4WMwa
zq;Wv3gX;PXT=ZV%)=g&%rj9p0zCXePp{BP!$g+s(=&(n2*i;J9FT0%Dl$UEC<!b5}
z^cFtZR}|ix8^cZSboV*~I;t7>W#FhDfurhnNrVuSUtHSy?r?XBTX^r8d4r8XL8_0^
zc4l{psCb=hqE3gLS2>r=KDWQ7GwS;Utunc9%!=(hk*2KBG>H;hkOV5gCuYP9&Ys)%
z*Gb5+VR(qJtNAGhZdeAC7(xT19ooeRjB?LrI_(4m#tXBV)KsBVr*7CcrS&6@9#4yy
z@Ue8=Su|QYYq`Ij(ovH(n^XsE=NW{3hRf5sHD3P!qq_naK+wJsWaAH6Z>#zUlu|0_
z(qk#8FVs5CUq(5}J?@|@Q)+-r|1J$vAmR068v1(Aow9DxcTB1hHruQ_87_L`N6uMK
z!@^%pDSV4PwWrtQ`XzpfFL5(3?xv@PFbC|~qK`-!0LqV(r0@}43@gw_9E_F>05QPq
zd5da>kJ6y`JU#tpD4d^bsM`!=+3=|R?>sspD}hXWg>u2Z`+NYK3$FYZ>;N5n*+3u=
zNcp-P_YxNV{>_2(zUZ#a+tl1i`nvLAt+J!lp{MMNfb{v+nJ`|@>@Rx?9EfK?n!nc}
zDg3B@hwZE10d44=1kRgyK-hqij_xHcFA*h=q%t$6G`?sWxbYJucKPXm87YFqE?&T&
z9xwe|4*c=sacefW?^dmS<cf!nbtj&{&0M=v$#!>=VLsu9GMCXSu1Si;uw{W~g(g?|
zU34G&g^&Ea9LHQ$LI!u^GE?z*08mXQ55xCz5-X*Jx>DobFe#_=O1R9?I^lF5zt^MN
zf}TZ7sZ|B0>6P!W3yn)MwVpKkL|kO}&o|0LlDICXY`y*Wo<1=E8m2Y%v`aYB%E&Bq
zl3Rc8e?dZpJtv~FW_Ek+?pGi=AGi;d>#;i7Y270rLl>_*|Ef%pz!T@+P&$NKI*E3+
zc2RBO3ionE%a6JH!SSQRImb1H`&;?$MWgRagi%mYi9}Te)JCVp#pjRr7@g;bcq6ni
z$_>z`G)P&>6DToAFQwzXc6YSUw0&W)KnC`fE*D$*(LC$Uhe#_T63R0B=sREM4<EED
zeEI3?R(R_5>*IDWMm?sIAH|dz5kIfGW@7*16~b0MZsuN6jx9mPhcNu@0efInn51L;
z-GY^2a^m20q4)waRW9+SoM?YmE(7QHi|2I>_vG8p3f@=rNx_T1ufFbr%d+ufE$7sJ
z#j}r(2KF8Kb4pL$NqAnu`I|XKt-t37CyB*I)U~d$+eOuQ%npxuQEP_BLs)9b2@^TL
z^>Isz-`D)b3HYre8l+Q5R+N4#(kJ?OhU{nnJ6CQ&{uBT@T9H~LNX+Y9f_NqQ%nIaa
zDMQ0Jyp0TF%T=qv!;9ZX=wa}|z<O_x*xuGZ0^(WAE0Fbd2;Wepnn6u%<h1vxN;65D
z-OFK%Ea=oLl<+M*@84NA-d$e`G6`;j!KSrj-ZEC#JXUv&!y_f3(adGng{iHBG#`9G
zD^yd6I!=*^@fQx(0R1jU?n`%N+u0!w#jNO(-S%CJQAM}wuT|SDDr%hJXvRO?pruc>
zp8QYQ#+TWf#$~MjgObG{t@8^y#*=|QOHj7Is9wVT{q<c>+rwsMw}pZxgg4OI*v&ky
zaGYg?8y5kfUF@sa;y+QBJ?LkCnv}?ZzX1Wcdzgls!o}|(D0v5g*Eyg-R3X$`VeU$T
z0TYp;Y#PaT+s44{rvN+Vmoe1hx-%_L<lzJ|qpqgS3Lj(N(f|R|JdzXYZ=ZB+sSeWX
z=Dj*9hi~TajnyoA*568@<}M8_naA|d5x1>L@o#H%tl~psh3`{$`ckDe{mi3usR7Wx
z#UuU8rX1SUs|@W78+hq3pGMK%^d+Y9CFtXqzJu!YXQ<OWlXjtFT6yJX6cu6QzMis{
z_wd_)p1&OvRHCWq|MeoGw{a48Blu>O@Oeg40jM7B0*uAKtw5T%n?DxM0wZ=bZixaS
zMk)noeer1V<1W1a6TJTY*v9oHJ-3}vPIofylGe{*f?L|}e@`={ztKoK`E+CW<FX(q
zY^EdrEYO4~0?&Zl<NFX48NF)v64{Hw-u0W{T9}6Z8lyi3U1S<4*i{F%iWpf~<Pb6p
z<<~Zdq4_l@DK+R_v9QPUB-mc2Q?6<|4lkj{LT){e<3r)7S*1|h77&~*fv<+A>uxXG
zZFrU%QCVu#-i+eYqn8eq*`UGwNHdiv2jlewZMH7~CjYxMu81~!Bt2oR_$|HEdF!%A
zD}kyi$DDHg>;LoZ7WQ7hn#lJ3)E_=21C(B-h|Ggn4umVhrb}Owoc4}g7(?PxEt+rZ
z3yYA%5sWa3Fu6cLp5ve02a>RPX41~><7lq$nbxaRFrt`UWx&RQl>v6QNU5emBO*?3
ztuL>8_^c6Q@@*!N#0aU3!FHqo`>t<Rw$5$=XWhK=u^~QB*Q62hhpEzGLNeh#Z`EsG
zF_uAwBTUJxA`U%^>O}k-53`>kO(QnsD{FtGP^xX)bL+>Ghn%PNpZ3@o=H7IY*KnIj
z;>#0wFzEEVu?BkOBpc}GT&fO<fJiN2>S0hg(uDF03^iM2qIJ0cE3bjBd)@SgFV^3p
zx%U_G{m?k=qZ)}Ha|x!qCr9w#?PbpgD(s=tH4rPBzvuxM6X0mJZg{8i7nx4|UX)cw
zhgI!QV|(_wuH1U&g4*#96c9y#>>j0oM3^%W7X%1z_q)(nCN^fUIPyH$x5FVm%+e$g
z=0Q`vg_ZLnBKiP%u|lzND)$1E6<TMCfcf5fL}0<I_$*J*3Rj9)PEqS9WV0MEcNSOK
ztSwS>ru)I6-elyKkXJj)6866Ota9$7-J50cd1)_NBA7kQ-*7kCP<U+DXwKreZOvL4
zFH;;A1@PuV3e90DVyj|&2mB7+;kkRy?-2JKRt?(CEiqU<J==Tw>o{)J{smwpJg^gi
zmH5c4*nemi{6QpO4fdx3O9WKZKpJO}J^QUGqc7Y><uJyJQ=r;nv5LH)0zP{7E-745
zo<GO3t`@u2X4l~UyOGUhzXP?|p4P^0Op{|gV7KW~L#5%KO=tr+=+nTHLM(p{wt{dO
zf8S*pF~t$SB7)qKC=>`-{=vfLkJ$vCl&~11!3hw>SH-tJbbk~nv7hVB>;@_?eg3h&
zRH*-N1#kjBjj)qm!n1r0E@R#w0TorR3fsV-lPbAHqtAadzKVls#x}Y(*!WxO2d=UQ
zK@Y;yQfw()b_+fghVmFG!o4KA1<s=kQZPSkObTfG(||BxPeVmTZ6kK_{Qeo0lxt$C
zto-};@AwqSTl?y(<P<(237EG}6BsAZS-Zi<@f%5~qx3}zVW0^gWwzs#oH(N{iKuC~
zmON6{H~%d3njwK2N;4^WmSjc#pAU88xhD(EWEc?(_GT$ukqv%a{${BXQBKaumK8{_
z2U0)YIR!Po5F=~94HgI#F9%;~<sY%ltb593C@K->*5Khv$F>%T@0UIyHhMvQSf`q~
z?9r^j#r|7j%&38D(>+MR!u$E2zGkwiVdB%10@HA&1Z3F4PQZf&9Le>NQFZVY^@}Nr
zoL4+Y9yIi?yN52?IduKZ{*}76_tzg$01NtcrZ>>%16ba8h6B53(WM1@-v;CaPJ_oj
zxY)Ms;88JpI9So3WtB4=pXK@8RK&t?*vwHchMfc%z2k%R0?k)JM2up}g~T=c|GqP$
z58+i(0JLNp=wM<&9jRkrZnZDx>LS$n_FU<J<2fA|jbTpubThIv2kkVQY9<Ys2bCv>
ziRW5mgOeQB>E_o8HgiD2tKq)zs?d$f5uTf!!5YrH_260I*2){7$Qx`oU84=sG}ip0
z^o(~+5`jiQz^soON_K!H702_xooROPh3nl0iv`5TU%!h2HANUu#jifFmuKt5_=^<e
zdu};5oJ5((KViu)GdLOV_gUV5D9zi<z`&)C?5a>R-$^P^@B6O4fe1-y2&Bdrd9)AN
zkCB9KgM||3DMV)=l<?aDbx>=A>avj#?Q@2R(ttq6X3+a0&JNm*V`1CrJ-<)u6&$b8
zA7xb=W0%`*)tSs|H=ljEo<*~QVFS(5&fe<}ib`;#AJPyE7mh<xnOWf3i~6{tVXvFs
z({q-LFj_fP#D_@{Pd0@_gBMo#7QzKD-cuD4JpcpjS&)n0F1LekJ_XzrT5S>lw^iYv
z+d77CSR2ODsXBzr_LJ{yUyPPspA7%D;hA;`R>PlPfIyq0H&?Kwh+m&n3<KU(4-#D{
zst(DjzYC+BA$cf8Jo=PXF`eZBsiRr9&c925B;X|-9m7}jV-pf=-P%xM<-fXL$ryz5
zwd2JsF--VElS@lqaQOHCLWDpTAe?(T;?2%LL?zRbxBccnB~KIlDqMVYy0%QEEN>t;
zR84LzE7joV3HT*GeZSxi!oR#G0?=eZc%Bl3&<XkHLT&}x&qMAVVuB)uP*x3xQAO|c
z1-V1QIeJB?nV8BKN1-okb2=I?gI_?sXaUInRBm_F5xNV8a*rrKI#!dO#mDeS$BF=8
zPvJXxBd_`Rq+qi|wB1HwW#*4p)tVRFa`8;?`Hm7u5yxJXc1ZxoZ4*~_J&iQy-dWNN
zw|lZNc=P&EBCG$r_xS{1nuzoh-2Yt~(RcZD6zc#9&*H@yqCZ$u7^7rp!|r3p1Ti<+
z(d)e+1XG}L-`va9@bS?B{^BgI9SOc3jOP8crQkYKU;~}O!g%fYtW*j%$;op!CmcOz
z$k-c`T$ac9U|malc40d;)fA7*lHn{A#VCOcT!5$}pdk}6Oefz<6OCkN&dkgtE`#nr
z4E*m3FsRYw30`A>5_v`oV&XU0S5~9=`R<f@J+Q7RZ5!Sum}z04tI=*69<BHCKp18q
zcmTaPnAqAGdi`i<Yq=Fg1XR!MBqqx3J0=IP_Izvl)Xj(#`FFd}0Au(~%JXClruxu{
zUJ^Zb4rpZH0HnbqmrwTvt#$^G7Co|FxDLO+q8Wb1meb_qWEv<HuTFn$>*o&_Q?A`=
zuzeZx^}twyn2%r%@kx>EA_&7E@0TJ*JLy4VDk4q^cn|}yVlHuJ*#sM~c4Z2797RIZ
zrx0hS^F3B6VhG;2_iMd&H!^_MT>;*<8<my1%MjAY1<?vM4W2-nv%o9nSl&^H0_DEJ
zTdHqSRMiiAHPI3PHpohha+2eKqJnRZP#kX@3$IeOoz=yw-Vb@+)C9@*M~-~`pz@x)
z=$tezb7oztqzd>5B844haI%D2ze?8`)kRZWEXbt52GleR=y{I<`J1*UfgU=1`+a!$
zow&v(xd0Y{-2L=f{a#Daa5via-**qVUuLizg`VU(8RWx-rlWeLq~G7ql|wRA=5iLq
zBe~7zmdP67%`MgH6fV6xhJG4f`{U~C{SdsX4nB|LXd`fcHSD-)hV<OCb7TGRQ3CEn
zuX%MXFGlkD(YQm!`i4hvil@Y<a@STxP3@@CR)a|MRd}0{ZDSnIL(YNfOy_fMgr*4&
z13#3pID*SKNIn5A0Av@T$mii)=#%+C?4$9TaO@Jevm%I3i3AoM{DKxoK*&!y$2VZ7
z!a<?jFf9<tt`iPXM25X8p*5-jN9iCoOXuTBw3KHo+p97T#xK@JcSPv?%abjgANNUj
z+fH0V&nl{IUYF`)=_+_1XuBqG{h-giJ920BU7_80&BS?nY3tyK>lY&ap5sSdOyRcs
z`P~_y6CTK{;=kYfSby7LdJF=Pg|Dlqk=nBW%+600_51HLeFQFrj`+SK6<}>j$e6eP
zz516?OL7!1YYA%<{0}ldAiPk7oEotEh#1nbgBue+GAubBUx;E)Ykz9WerZG)p`r*U
z{n>1mqrq?pIuNAcKnCQT;RK=E(umXLw`dRT?bV*gsEw5qV#;B;m073lZm)xm3xryw
z!K=^>s_V*9v%FP+`t?bqA;Zk*$1*sYLq(UX_*a+oW?-wy)t}HN6mgUpCDwCkdqU}d
zhVmtyBYW2Ih-pGWFRT>Bp~r}B3J?Vi6^l>QE`ZSYQbC8P2SA;B|KL8cII8iL?4ECJ
zLRTqX#%1ammrOtCPn%AvRB|m*E4RA@{of;q8F@{m3_ro-QdYl)_*uo#hd&K%>P7zZ
zJ{&~;_b#72Ed<5x)s~3@ANPre1ap@Mo<cV!4=0ipZqRpEZvjLnz+S86z$8A+;^SD`
z(EoIGMuq<#>sDmp1szbDTaQ8he%OD`y0laI#z`?}&yfsrVW@dQ{|{mR0Z(<`|AFH<
zoQQ}DDV40UMM9_}%FNy}N@Qg35l$+Vva&@KIkvL0LK;?)>|K!@tIQC7ulM1)zW3Gd
z_kaBF$K$@Q`&yjy`Mlq+`5Zp`%CdQ&LPoN+0EE7g{Xw;AIz!;<NP177o1I0K^754{
zVai*BeTNNWF&P<|W2%>c-7o*xhsYz4R=Z~aP)SNau|>+LMrT{$rH+6NO7dia<pEfJ
zFbiG`2LsouJuNTTfD41vexM8bw4S*PaVdoQJW`0L+*n_Ia)bNg1tP^F_@w+;9M-l3
z9T7dmj)C^jQ#Y*yIZ1*0d^Q(PW+FT2R2?IRO@TD$2tX=&s}aQ?eq|E3K<1589+Su5
z?>iYEmZfvJ+YJv$mHlVmVzp&=b1t{dgQXAFe#S>wdY@M=N!|W+2=#!&u=aWM{C*T$
za7@bvr@r>*8$pwDxnfPD5rQ8g(;7c~tPcCF!C&#;SVQG?WO$mr)I3RI{t3-BB<W51
z`#&IkY_X$=UT7Kv9hpZ{9O?#N%ffs5B76@t(Kg^c{cH)V+Imm=@Sa*I4xGGp3@{42
z_w4!BLwBhtx;#u-#`-0=3mG0F9ymXc#s|Lv$3&)3-Ep<&#|HBW$H9xB?Kgg{_*!p~
z5-A$PG~Z(=Cu7%sPb2d{{Xf6QY`mA(rpN>rhQ1Q@KMHY_>;Lg2n}Jj_jP_dslR>OH
zFL5BhKMHVVvV-k_FK!fVk$$@7s#5^)Om{L<SYoxvGK3O~Oqc*OlfDm{%nu&Vo0fRJ
zz93Q_+`QO6tgR)IL*?RnYY!X9mt2>@z)LuyW|eQLiPG$_h-QeiW23;^ua@tI6Ow1>
z$&BG)pALfi;Pft%EqM5)uaRButaqBLHs+x8!ey!d+y%C95>m7Nv=-C9S0OC0JgyMp
zr^h)Y;s6)(a_M~*ls;KeJfeH5c$ogDI`l;wf4WI_T9A{V#Q&);?6Ro8Yu+Ao`4HHR
zS3i$c@?b^=Y?Gd2<ZkbbfVpAsg4sZR7RCX=#)y~3N<Jg*El>|iLULvx@CqWv#L36u
z;VHLH5rQ8iS^}kwpF>uP{-+74le!Lbgcm@sR04BuE&%~y^#<Jl)%G3V%mUVQ+m$XW
zJcLJD97Y|~4N|R72iA<9(UpMK4%#YI3XTETGa4%SyVNzUzWasE1CeEdz1aUe&IdpQ
zbXa-(!luQ0PjO@6GkJvp&ZZ&4(NCH^(tSrTB|snd+a3=O^gb6v8Rw-EJ;oocTpX>b
zvmeT;2BN3_8ua*q($QVH!xeQ7TOlS8R|AiUN`eQ=9vFBR&C0&krrJ>L<Al&I+p2>Y
zu+ZG?fl1w?nu6|3q%|o&5dH@kyCD@DF|I~bd?=IB(`zh!1Ppz9ZyQpMK>AH=!9kR}
zl&gFez?+hzbAeds{N@frr+1(D#@b9kI+UiJ=Rn5?wgH{X=G8~)k92e3)~J390mHHh
zL_^l{-vsYw2PRJ$g5V;BnqSN-AUze4shxp#<V9!Bu2V=hKa$k8d${Y{cemK?Zi9Gf
z?-yY1ST}xhT5qzqSViSdAE4SQE?hUw?=AC~GRlMkx=#cWCn&A~?Og;fI+m~W7&v#6
zXEM3%;#a?`9@dd-+}kqlUq{t*1C4a8A!dul0IL1{K$nptjA~AaU*K%5)yFOVxAv_B
zf2q52Lgh8j{&}i_d5EFksPYB<BC=N;3D<DebQoDgU!?k>vsLImfP{H+U*&>w3MA!T
z87tGNH}BA0r`kcFd;$hq7wnv2hz%eM9KhZjrfMNV9s^oZhWmP`)O71KOR->qw<`cI
z5$F%$0gV$tl^0o6jeN8Z4Q9w2@G~-vFC2LG7JuDWA^OiO@f#z#@v-zHfq_W~+8Ywb
zMk%(RcWn%8?ut~FpC9027KN%f5^`KUz<{m|C`XT}hX?T@u)r)#)SCbkk~_LGdmhU{
zNVx|+VDH-T`E>r`7T_USB9I*tYk~XZC&1n056ltdWRcrKysfMbpc-k-%TNysC)TW(
zT($ghUx0<|td(8-#NtjHGqr?Ho=KmlR7NUjSaS0JlBHuVZ)++*gEDn`S<l>C?mN8r
zU+hFP3FKl`^khr!(TgJ`&s{^^Um4aY(%!Sy%oIL_XZjKpAj;0&#|l6!UV{Tps6H;n
zX6}K8!Kcp_?IM-n{pb!XF?5y&0QQB!y^0^9N9aQ_kZq$7Sl5IP1bJAcy@XZ7L0^Cm
zI*v~<3JMBM0B+`jt%6~RHqD{y^P0;o0qYhfr7ok&HiMOA?x*7%{p*2e6U_!Z8!~r}
zVKC~2$4-F9eh<N^0dCO<EX!fdl%N$kghC(hDa_c*Z_ql8>xPrto@=ZRq`<4CmB2GW
z80bcU(*Db({XMEa?@V5y#RsA-0+;^)e9ZfW7Yg}zxdqC7r*1z-mu13RbQy3v0N9Rh
zPJH=DpVA+MUy&kvTb3>#lJb4?=FQ9gqjSn)R!BwzxENNZTzOgvFRnEsJW!1}K1!F;
zaypiBXI@7&vG*iD_w_!n#igd|gtHcoH(1YFJd(D7fRlU->wf@H`Hr?%?oylrV@8vI
zE&z%f^fsh0Zu1QnKng>zRd6PC@EK^QY8hjaGlQ=af<1(9CBrR96}Kt^#+x4+=*)Ob
z-UZ<sfVB&9D4ij-P-mef4AwX(=~`qgV`4f`!^+HjHFWzYShwQX0x<}+Pfoc$^XeF5
z%5}g1^QZ3uF_m|r2klIN=}K5tI(PvIX<tqDephbW9CsY)z)koG4}nsV(^a?@jo%o9
zdd`9Q0ZhSH4j8||N*aWoU`+mrXv0u)jgl%2s6uTvzT`N}^Fb%t50pToZFtMW3i-U<
z<^Y?Ug-Tr!LSxa<Q-q(CS2JZp8w*UH^%i^2SqYqwU5k(c9l}NRwFqr8242?Ji;q_<
ze%K4&u)jW7Sq<FcKfts@uzwr-Q_zGE0<!G0XnB9Up3R54N(;p3QQ#W+crfnh@-$GA
zv(;rJ<*wIi9<qHDhn`L$HDG6&3gV4N!;%cE330_AI+{izo~!E}x0~DP$Sg~D`{9Hh
zw7KP<%K0U<Zm=5dTvN(3QFparXymgSN<c9!N;#c;ZfGI(nrz>H@&z78!ZD_-C`fL;
zR|>gdR6_&LqgVGjDAgjRj>(wgd-g-LyZKJy6YPVi5N?$Gn<5X@%@;5TNtN~~0+C%(
zo`yRXthv+S@ags67KrL|qmRlK-Do#@Ye>(2Ko{wZ><B9FsA4I3Xp~=4s0wu0C?Mh+
zTTPjYZY&OE>kA>L)~S*moy`dsGA!TpPnGKqrP%s=mw|X#IhnRFhJ{q%Tn=KD;NAQt
z!e@seyte^CWO?;$7DLCfiNe{7MSiorQ~t4g1hLT60ja4s3=7Gd<}uwZMRoX)5%}4H
zc%u}#Le5Mb-r72uZc?8PACz<$;daC$WtvlVLLm8s{wfs~(&O1nQPPeU-$Nujl2I29
zkZUhfDq}R=J%R)==vj^>+zkvx-RT{bPn%EEiw&USl#18BYNa6GQ4l7K?JGd0f>5F}
zn|jH@D%9S71y)Aq2CiC2f!MY#lrPszPS|zwBgzec{Xf8@9BqXmmr^L)`Vx1^sBm28
ze0Uv9vD#8>@c<5^-mKtp;pFePN>zrV;mOH@A$o^n4EYvSjHqM?*JWMoc^(OIe|^`a
z84N&%wA~3!Iz$dxx_4m*;eQejXma4UrX>GY3*W3jLME$JpU<Gjf;WftB4DW2h@ADZ
zXN-}g(i|G|`&pa~p%*rSy6oxOOj4}fM&=7oYf>g|&U6^)=cT@vtCqM5U;x8ru7(Gx
z3>=V=uGn9Moe5InjRIwH!&vAs8Xf#W@T@K0?97BupEz=pfY@fpEauQ}@M?v~hwd|c
z0M+YojfK$}n0j_UQS~$V=X1)Wg$%r>keXX6U8A|XJR4h#H%j=a<GUCJ_>=*JCHJh#
z=VTNcrHuOn0f2>AR0$M==vf3+i$0W6>4uRy{L%Eu2h;TbH?#%9s|2-8f<-LPmJZDT
zq>$F6?pA6TuSF#`!hr{bufk5!=#(4sehozsZV%}I0nCLh_|BpAKsV6KyFCl043ug7
z)aYY2b2<SyN5BhdP;^=$RX`-i2dJyw7ibI!Z2X*we}o|s)ng^6aw4xHC!~4r$j`NR
z+9EE4bas81wP%-#F|%4-)Of*|SvaldH<#w-qKcKI1`x+akLlZWv3331+M6MLa?M9y
zED=~DDS>mpX+8@d46N@z{b~#ohMyn6Zxq$gt$$XTKPJj?&u4SrntK~p{k-VE3{>F=
zbRiP%+<6AIVl4*bexeb`3pyIK?A+evkW55&dLV?+E_hO#2}YEr;%YJ~-=O_4k|bhI
zE?jv6`P$1Qv{VW@Z8?N(g6^@E$ghzS_;S_voBcuCRfq-1LURNeYDjrHOnD)btn_1H
zOTXMM&Vvfll!khAW1@Ke;+0I0i}vad#yMKc8zWLZ5p%))4?M{J>~B0`V!BAy2H}E2
z-~U;z+7)rM{z$6=?I;vQ@!0;g58n+<tUzJ$HqIQW=-uZ?-#qcd*^s4PH9miQ4vL8o
z3NT!T>Hs7+qd;V<A<qsNf6NO@qR{5HOY*7^f_i^|+~tQCRhDB=Oet3Tukja>tk1;k
zVo+7^stKxs3m~a=zJsRzryvxVuJ<#g{%m1Ez64Buk|V_$axAP1*hOQe9v-k6-`=^u
zJ8Qp=0Gp8Kw<f$6sLd9<*ArIZs6QgH60v*4X`}Tp;4y#rPp|s8wP#1E-*!886htal
zjVM<CyVYgjLp&!2DK<A&JQX7-Au6e<#lY0i7PKC3gL#%R5`dBsP`E{Qvo6iSX9jj=
z9@^zZ^}D;duE0K`m-ahSIDu9S;Z$-aI!+QBWZ~kbc>FS0oXiU#mjY#G7-T9y8^w*J
zE(s#`88G)LTb-^Lg%x`^FuIgf?_aGYWL@lVSjbxU=yz!nn1xsn!X+Vliw6{gaY#`L
z+Twz2{aZT;A;G%WfofG+5%{shetIzb6v`h!hAYc4tuoU2q3~S^dw{SGWsGXJ(8-f0
z=eX&{)($dv#K3+_FVA$X+6n1q+pU*GXFBUlHA?fS7tbVl4#oos*B^#Oz0S`Oy%wfZ
zCPTXQ3a|t!-pHSy?9qmet;K+Pm^j`xgwM6ZfpODAL&`IcY3JqfkW*GL#Sntw<tM=`
zo%e3vl}rekq0!Ktpm{_wjW{S83*P}NiF!cur-ATs%v(V-5SIK6jng?3Dj$$#&!r`9
z2q9AssHDqt!lEI(;Dn(*-G_o$4J!g?TCiG^W<j_S1fVaoI7~?1GeVC<uEDH`nlL8}
zYECggtL_#PGJAF3t|1tV94b&Z;6A{YU&#(j^%w4gv*&L`=o2(*r~lO5Vc@v(`{$RK
z3pZAj!{s!VGg<M0TMC01k;=U!0to#$py~bvRNj<6_Z;(F#If7{$WR#04FVVLSi;kt
zs4t>JeUZtG(bE?LoH9mHkqC?lG)>S4_@6*|GnXXdCLjS@Z#+K+Ve>!|_(&60zvKXu
zfRPT(Q75-Ol~W$6KL<b2ezENVI)PxA^z5CSi)C##;wK#k`C-B4TyBcJ1f89n+T+tx
zZU_1pQOSyYnGw3b?=9Vsl<g7>pNc~Ibqa2Pu(x=2MIx}H1z^IuSDl=Re{>D`>H*Qd
zB4t<8sjK<O@d_j-P+3)YA*bo!LQA5vA+V_lL#fx3bn0(WzpIzde6GX1?_=EYIzd*B
zIgc3cg{aIfQ{M}*DS^1sxADsxe_A%Izi^Xp-$~Cn`_)!U_86+i!34`tysyGDAE+;e
z{(pWcOo%8rf(Q%`3>#C6Y&#l1*;ZJB%<5}4()=Q6su{5}p-wNPiO`sEtXj}R%3D*<
zDp6IW3PWvqZCc6)vTVoTFi_k|Qr;{#3g9q+i0*?hD7t=Ay@yVoUIfD{Z+MIQDxIkQ
zK@I#5Safv0lAw0!2J1o4+fWBjFN*=Ss(4VemtO3)@MyXrR_Ig5!y*<C?9ef!TNTLZ
z`6adD{|YETK+~yjRy%+8cDw&L3`)kqtTNZ%wnR1;uD<A+P`aoZ<-LDzS_~XZPOpPm
zT=lY)I=k*c>L5>}?@M!fwa@$cT&`B8&e=P~kNy(&{Z4`@y7GM54_kW`MA9%4t_9h%
z0lj@#e};dw08$;w2VXirMP?UMC02tqa5{KF)uPL+#icgR{tS$)xPes~2_m;|J)SSF
z!;}}5^&pO@9OFLCgp3B!ZqXaIl_m~E!T>S}gv;q;GGpR7zy&?r9^KLk9sz+1P-ErZ
zVv#6kYZ+K4-lzhJRb85s8n-`kE3C~~T|{_N<U3dRbwE&G{Ev`PQN2S2EGQcediEay
zh4<hA7f(^*1|J9=gjIZ~G|E>nW{nR3sTrf+x2cDJx`C8nIMRNBkC%Z>C0H2BC^bD)
zI=BZ($$Hd2%bq$|+@Tv@2OLfzUa#|O0itvB?+!TD!zA&s_whNnM}anIX+(l;Q6Q1a
zluz2nH}6*hBqZQ(z*!j1XIlmO^f4+z_*Hj@w*$+8fc!!@%XF__Uwg~;FmBI%9n-OV
zv`;BXy%*vf!B?hPIM*~l>wNlTdoGV>$FXA#EL~qjAqD!Q(h1)F+Pg=avt)Lh`Er<o
zWp74+67j)&0deZ4zixWthiRb^nrt<xANPHBe})<F!JJ?AnrZ#*(}|(4d;xe+e`)3E
ztK!?H<HRT5phc$4`4ZvMJalGWL8<DHX#gebL5c;wvDhP1O$hJ)w1fY@(-iujXAyFw
z>IWtf<W;C>kXHhe;Kd)#yI}FhM%-@W`Pk$T^S+z32M6d)ky>4eePUlHfh`2WY>$1$
zYg#0oi4xC%OJ6)F6inToZDvmIahQWSM?F%<tEiu@dg_r~LdalR7J=ua;=eL}6ft){
z4|ME#m>}*bIOI3?;IqB7$tS?n#Eu!Vq6i`BG#rTT@aOMWdeqjqPg^zbs~G)fgeXwh
zOrYPnFay0JU5n(giXT5}*8H>3X<!CC2E;A<@$VZiJP(^|?X@9)Z)RwY$2QnpYM_dY
z;Fr~TBCn~$?nQO%j!SUhBMPPS%{SQZ-Jm!l93GYSWVcQvCs{;uBqxVtfbn&4SJ6Tl
z=Zvf6zUyn!o}W$NKMUE<y66@>YN`R45DpX(W$0wiVUwt=wB8Sc=v5lb4ot;AXJ?n<
z=!iq|LEC9TeNi;CaRML_t$1}U#<K_^VqOqfz45W!`}2AWJ1zG!FA7zm8t<BU6L~Bt
z<XjWPu~8;^I#DDjumw}G71F>Y=T>0(;r!%XgkLN+Ny+{zHUd9gjhc3JErP3FEaxa2
zy$l%WSa5{P&4=z-B%uq;FX2OC^P3J7dNhU|2sM7mIGDYzQ?a|P+(CAR<?@veXV!o1
zf)Soex6uK<pk6TdmStcbj@$L1e;UvFwQYc1#%OOe<pCyY+B>;heb_%8nHF@ZwU@H)
zMMASxvgbF@z0go*_oR6Zq*p~4eDF`CinH|>9w+wuLI0=%PNYo7E`gAl5z_*@QR><0
zrFSH!gP=N@3?RnU1F0hohO)@`9VtE?tAM(bq=p-BB02zU^)wg>b)1T(x320dvO#8v
z;e=rZC<kI;3futeXZNN@$1QGyMVpdmk}|8f!z1a1cI|Q!Dg^2IAxt@J4aG8Gs&2CY
z4qob6_EDMC9;q0`9~bZ?EymU02dv^-I|CZwM!$&J7hK0ay7%dVlD9l|dhrp6=cmDf
z%P2sq6Rw&fh>|@b4h652x5pi~0C&F)Kdy0D`5@;#B$=a{a6)M{A!@}|r=WHx>uSfU
zSe76@F8iC}!6C?^dv(L|tJjStEUJ8n>0;NqHFW5AG3HzX{H+>COC=JHff<z#w^Zx`
zF^jqW5(uxlj3s$C0wFo6@01nQ*WvJCfUPwP)`&gn@x8Nh#JqCb7}>W6JzE~bufQ>Q
zSS8pUPX%6-^jxzL&rB$QJj07B6D0HN%fzHy0qSUi$d?9MfSWJMj`aO0SdXiem{ON(
z)Vp;uCsN_I^z7rS7j948yuonCdFxc&KOA?0q$|kgg_j9#X(I3dDDrtot>d)_f$8jM
zv-OTwQRi&!E{d_VJ3JELO4JUtI9=pl=GfZevQREledD3f)ykb0$}YM2e!7<WrGW8&
zYZmyFd4Yz9S&4cv^57Q}9k_oa|CQ0<$rIqA<_(@1K?BVvmKV0)tk6@V9R!QF^27(!
z!zmaF5QX_9mW|d|PzP`kxx~6M!RTogmb=Im(&S;%9AJs|y6@7%dO9V6s`nFGAI}S}
z&8yj^)U3aNm_M;ySwoh(oyIL$8^-2E=DU)|ZF`LSbXtRuqy&v$O6dS_I^}5#U}-j^
zovT-S7^;X!UK6h0c4oqBPa2KEd4s_G%p?OwCQTq~c=2K7$80wGtb=*TNUo@6-{87O
z_BV5&mPUq$vq2<q88~ra?tI7F$7R%aQbKhtXX12*5x>;teIUn$8F5&38nxbr4ypov
z7pu;hM-tAj1x&C#-Jp#J&N3D0QcUwiB1`=j2zo|N5eKC_Bu=`n?|xIOR13_116)nt
z5Q{=b5bf*H?ocy$kfCrkCgvH%NKI%#1~^+z_T)!gW5lzvXMT7EjVeC|u5AQL!M-O`
zZZuGy%Dk$}yUhnUTjT?r?8`Ch9;&7Z4j0*iE3&^wj3Iwh_M`^2%X&3AOWj~iT^P(s
zOBxVUoiqQGRXsA0@6i-(Cw{i4U!*Tesgd{Fx?zvVf>_p*ownl2F)kl?YmOf_YUu_a
z#H87>E9URpA_Y`dTFtvR5v>0=?EyT;VOF19Vc2xkoWua9JqCgUb|4wxuWF`VUv8z8
ziZ*TcA%A_)y$Q4{n6|OC#=eCK88@J`oUi4rW!ZyIApiL1)mO2|4bFQsTED+U09JNr
zM=`&C@^{t}s!0%D4DNkXuJrrY&`u)3PUK6g5IN@w{a7YOhY-zi{(d;3=*?xugIshy
zbi(JbSwF>0062w&<cRy`9s%Oi*ruHjarH$Dj;?kwgjYHj&w3k{SV(Irdkk#Iiag4=
zz{Z|X{`=H3P^nDEI@dpvEOemB*2!cC#7DbH#;ZpQP-lhsENeNgz+^HV0BM<epbI0*
ztKS*GovE+ig`?vQtVgoCPzqhp6_jPwiU^0;%gx-nI<u{dcTat-Zjd*{BoE|}<Td4N
zQn=3HfE+S0A8lx-3@+!3I~UFZP><?jg$K!zg6Q1}05{8}=>a3$Jhb}O75Xg+I!?2o
zzG>Z(2l6<Xs~7oqyDt>)!btvZyL1rK2`d+wli=L??RRPnso~IMdg(S=UO6`^xa=|Y
z{8$&KNJUsnxZ3wB1LL;z3N8_;6XT})+8x?LQ;lx#aB2=dZwD#SnP+T0Ss<j5+>?=-
z_U7M-<qt3#YEy^qJtQNe;1?2h{jQ3HwOXe%%s#V7IGMNh#aohBk$uT>0O`qxAdI2e
zu2O<mMdJ@7iwim%{V%eIv9^I_)#I?ge}WPB@{a~lb2+U7^IDxm-)v7PK1LDb14NJ0
zd*6v20XrrTkNldD_5nmEKF@-DM@9N$fIPsMmD|>-a$>%oBka!pn}_S&kgpdMmk1{z
zP(e*E>^LZ0dZKtZ_d2}vr^7|tU49HzS0gvD(62>9u#{;GBqQ%{V}EY0V^SF}ZrzH5
z*uf4?l%|Q|$p3Ai!ZT~Z_5h=>JeK*_NfYpuhT#ob4?7s@{G?=bl6P#&&gKXY@>wnf
zd~vTzYH}Z3bnTm#aIbG{>$;=&gNt}4@HyrY<^6?px2@IzysK*;(@(5z{Sj1En6My0
z!v(vKuH@Whn%fTB4?iWB?AI0x`%|I9B6=eF6q~v6W`qy2-zB+!lO`gFzaC%`{z-d1
z9R9ddKO*MX)wYx@EzSTYmG%yTvIt?|m|>$>_O|za)-o0uIJ#xmcTKH%mTA`*hJ*rx
zx!>yjPd|#LIpFkuTEeY848^4U8o)xN!{e6~*SId50m}(l$Ez{4^V@G`m&+i3Q0Uhw
zFMdGId7%B(Kg-twDnMoF%SP5<eFz%(Ljp}m;tj+c9c*l6^9KRk^1|xF{s!=5$@=57
z1NeBckn+oNwXSEbw7UZ+D+1Q{{bEXqApKhfY;rE$^3t0~Y4a8MMwWIEX;e^GVVq#p
zYWG-qSBY8sNn2$*wqS&>4?85YuH7B5VYV;>-4CbZV5P(Od7*u}%rdSEjQsA`#z0m3
zYN1XI10$i%H33pTAPC-yN_MzFpCpf!aZW-CTL^4v7wKx6pT`>>bn4Y10KsUI^GhGS
zA>nne0LnByhT;96cxBGXFbJy!&ILM`L~SBNacmf?og?95c6ym66(#LZGt*!aVGM_q
zaM(T^QQu2}?2&xK#Hn>J#lrK(8DOoN<BlLq4e+tg(KJh6enX|Cs7N(&x>>7xX(F_-
zNz`MXUXt&0J>Q1^?~`?#LfpqmciPEq6BIoxH(U|5K_9RhV&XF1NPD<!vW!?JSlTe4
z%dx7J(8CG2z2s2p>GL#jo4K#W;^;i}>bc0JVq$XQKL>BG{rkO>(V#N1hL(*3vXkC0
zjLwa=1p>&Fvi~-837X_D!OWiV=Siz3f;ihn+SVaxKd->LXqZT>YR%nVkd2?jtD&jG
zC=4k)nofJqUIDCOB6?TZrTgorCHI}~xGlH?S%>aCc&eT2T#@;q<l3X~YmOqujPWat
zSL~%AM_dN1>;6W~cZjnHuc$12qF5-(-+6CBCyoxyM|1#dq4{X86i$Z-R2)<3-dCU0
zik|N{*CO3&p`3K4eR&v0Ob)r~Ftr1$dqw)V!2DF7K4R&V*ur#2#AQUW^u+rcrR7=X
z-0-{dI1=Y#+33`$tssBhz+B`oD)>wg!CYb*mZrDTxVPDRu_=;Ot@_I{F_cUqVJD1P
ziEW3n$d!8f5*HpSQYyr2P#_}$<ihLv846-$&{$4t@<`D9_TCiSsgug1O5UUd8qPdt
zs6VoO*ZBm}xfwPaZtOT9;=kx%SU?ooh__F8S~qYjO`+-XNqZCX50^E3&DBO;grTc@
z-DUwtH~!5bBAI7w)q{4o9J%>lR9bur3`mECf7~trT`vPIw(i(SGoaP7oUjh(>LU`0
zO3A*|QpjsqJ~$*})L!G-H4fo(7x9*uZY*FyUosu3yG`rryW`_?5!QuHKDp~3W?-`t
zI$&MxL&mLR9t?~1_X3Zrv5cH>wc7Eom!|rTp1Zf#x$k1W1AaTB6W@<&8XMXXC&1Qb
zf+HBh-sF2}3Vm9d88QQPugN7l!E!7`gtd-In{D2T2G+qT2}3)Q!mJ~oWZW)!9>q`7
z)6wVYDIG=9!0TF<;^p}wWpzz;<i##+4mIi~ijMvcu3$-(oJX`<nzx2Py$zD@fVnu=
znau+A_6{$wx+F{NknxsMJ5TBIaw)V@<oB1pNkxT3df^9~kH7XrFcl+d>mmHG=1E6@
z4W1Q<l$V~_o+}cB)nIt6PjuLi(;rGiw-+boDiFdxVOZx<cZeUJFT7acFIx2AX{9dy
z;xCc;2obGkiY3RoZVvnx82FsaypuBcc?V4q6TX}2M}JwNAIL-cGobGL%%7h@&>nqs
zG!?ztdVv0ra!nv&=Rqe^H>l$n)LN%GgTS|t_q02fp%HZfSmvqEFrDE+Ew`)Thx-Sm
z^uQ~5IOe<K7pP+ajKB>I+&yDGcUI_6S_K4Hr(6Yup{UU@OcZyFfnExES(=1+BnrXY
z6Y+az0o11gD}?^eTduWt%F{Y7Vg;9Bm)oDfVhygNLw?#cH@}qsLXJMD4cUx_AEOV6
z!XEb_T0^A0Kz7Ck3W0X=q;j?97}-bi^9@>po9_U>LtZpIXk{i47!gg-bMXfpJ}GSP
zLYlD~D6z$0B>gHD36(27bamw!n#Gl~*REnD_ib0LrAmWY>!Aa9Rtf3NLEcYxr<H!)
z_=P&M_>DP+yKe*OBnvIW4LNLmKHYtAWqa}a^JhLcZa}TF(ZU`lnZ4bGV0<KY=MMv;
z7glPARb3nYqN>twM~Q8x+~;G!1LbFT0%VBy+CrCQcfqA{@=o31+u)GU2w)tp)29d7
zb7xI2K<Sc7;X=^P2>k=BfkDr`U-vIOSyU0Kin5(>YF|ZaFE5=gmA-&cfe`33M*^(p
z)Mtc$Kx(mq*(?hfU---YZ5TsEWo6{O_tG*+^lB*Jv)*mJCsue*-iS>NGyevd@tK6E
zi#>)U;=MASVF|J7_L|fVPq;7+5c?qRM8icD?mp6?sSHC2gLCi#vgh|M$k#}~3wYO~
zC9?GbsO0%Tib-7@7N#~JxXRYS2^1%k*%_FrJlQB2ul#{p5{au$v4rfeHjlZBqq7>{
zLB!*JfbHx1^K8Mbt5E`^S8!Q=cxH|_rA{_Jc6aNSucH&k+lU({Ts!aVj}$e27iU{&
zXTj1womK0{PX#%qvq_I;B>eO=(|AW#T))|w<3-yG|LPoZ38WbHf%Ykw8x=gN0W?&R
z>;wwyNKL$b+aH8v1*%ROY%gQ%!b+(#+<%ro?MK3TFu`UGxwNbsum=P<xPa!SsB0=g
zybcRNB~{hyLc)@gJpwgB)I_I0znx1@{lJ<zH4R%Hfve*JnAy@|=q*(j&iY97j+H-j
z!n$Z!9+GsCNAy=DUNO;i{mkJUAgafxW9x)3Q(uK=erO6u2VRR|1PlG73hAQ@BIB?g
zkbY>0AlLjyprhw1zpKn{PJpgW&YNdTU7{NEiE1R}#<&{LMWN>yDfPcXT7lzgtjQK~
zoS?YXs`PS30TR;Kg?88H@Y0)Lp7N;TFv!zEJ^8+jby5||WlN9?@3kTtdBF<ia{FAg
z`2@Td^SAY3fy)4c4FliU?^1dM2J~J7m9XotoOECP{uSPA;%p%JMiCbbG3+7H(j}ot
z;eI|-Q11hAt6krB!?Y;(OwQi=R9v1HOK<UgX!@9pEPDRx=h*l$R@a^O#k|v`f}d`F
z7GFeX>UjdS>MyP|?0)}bu@CfrHJ9|{l5D&31VA3;^MEw#0t{>lb+U9Ca{U(4&ny%_
zg@xADa_W90U^w<7VR<-cR~qR2I@sX(QN05c#vd^YAk56rq1VNOLYB}YJS*$ybWb6W
zzb6Eip|3*Kb?AsrHSuj7^m{62&z{Y4|DsH#u=Q%7&g{4$27&Wv$MI*|&!DYXd$l$&
z-EcLwD2eC3CkSD_Fhdge*H$Lk2H@EC;jJlL_DQ;=?L2&1I?gr{W;3Yws-r-HfyhnJ
zcp0FxoylBf+h^3EtV96;WIDe0x1e<qng|FgxLP$NJQZ2o(4vf_(@)y3-@+GkYJe&4
z>oK6sx^2ChQ%UExOx#~ld^CN@ZT81agXV|RdgP4I&)>Z7sV96&_<5&h1h3u7+OMUs
zz4!nC^0}GN>v7H5U7u?DwqGPgjz_8f(dnYwFaEwk-5|U)EPRbN4KaCZu3&5qR!&=o
zuDpicvU<j#-5*4Q3=tcCpzU=reW;I+X{7<kReje>2ZC4QV3qGc+jRAYKYY(n<6bz?
zBQRO<RGBMv`ub)EC>E+`T|;X9Ftn~n4}N+*54)eCzE<-Uwcl~*?lgt=I`ox%h?rbl
z1#{*`Sa0^X-8WEFK@sXUjt;f{fWb33<U9FXr;@l&+l-b^D@`%Wof8EgD<&wNo2%po
z=<nR$Hioa4C}iI3gDQ8(p9+^7D0fG#o(oM8G1mbq01`lI2N~l<YuIf0doqGN6|aOB
zsAeJs)m}$H=bpnd?7HFuV0w8Ob!Szy6=PEEd(Qe{=xz2!WS7<Si5%(+wj&Sm2UQOI
zBhY}VFZ|O@bq72>_M=%Px0#?6XJPE&NLS`5mew!HS>7W<CXF|bU0M48&G6dkG;A#-
z7|OBH`hjc1EiP;CF1&v-?*IAy0DD*|FSq@~(S$q)oy`jHpNYsFU6g0;cdVxj!2Jmz
z-5;td5V=K)_~`xs=sO<9V#Y=<nO<woAPhR9@0Rj*fc_;cg}=vig8ZpBSd9*}A+9<=
ztK8uQ6(ESF=F{88C7|Wl2(N0Ot^F*7Gb2I-T%FZ2=Wn-3<TK!BmH}vr!*KNiWF1)T
zR;06((JP~O6oaE{?C2AH0Oz3*+A=Pfz?`4eJ_*;HUXLt~p$<7U8E8RY!&UJ>ne+;}
z71<*3tnEW<s44*<8D_UMVCnwAz7U8Kr@^tw1=kneP>CTPx}0lNjDYUnb2-Ieml?NL
z1D^wI3|6f^QOm|O80k@XXjB8-?lY}HI{+W!Vb$-7>`o&=4qP(!siYiRur;yJk>WKt
zulQbrDY^QF7BvcQxSceQs@*PF2}$vrrBk>|B<}97Rnw066K~m;A~o5bmeTum3R{pu
z?cC50DliD-szO~z6{mtGn(M-@Y{8haua*W~FI<eZI$QU@|GVqC4Mj(7>X8TO8t6Jv
zY+k3=h=&e@oX~SqrO^Do02z~Nvn8j#V>Q?+pCPaqAfZwWZdjYiA(?4o|Eg`Y$gV6S
z*{4y!4AzOGK+JyPPoAX1lEvSpg!M20uDf02$^5QYkjC!0&{?+8j5xupy$Le09M8wG
zlQ6?()YWf6^AjYb(sD&EHYiiY=);G-^<TiKv{z*Y$~m>sz6|xlAY|&7<+d9|u0m?3
z2iBd*sisu&(2n9pa(WOj6Yh==CSvDE%Jzb+wdF)0(QqQ`BtTWu*T7KpZD;nzVg+C$
zN5L3Pm2q#^<qN6ha8fz&D9%Fibt88TZfl0>Qw2yi$YtqCvH)SY34BJMA`luRW5%ly
zfH`Vg@08s0M554o!EAP=Iq7ysip>K*krWG?t!}ltWzuo*{zuA-JN_|+4#dS#2oeJ6
znsznCLRo?JrEpV>RlBkt;?8MtJ1TzPqkZj%f|3q>yYM}3*VHy)2*E4SE={6SDoSU+
z&!>vr|9b$nbI<UoU!65_1E6nZRGb)gS4af|$F^6;MhrauWYv)S1g#vLkX!4>F6FfY
zF%WbEtKY5zOR%@eIhh5h&E6~^6=FOK<KQ%yD8i893v9r`VP280+@&G`SYd)tV-nZ_
z*t^4gVFaoFsc6&qOsS}_1tS1E5Td`8Jf>#ky)%dlC{;Zo$jQxZix*@~%&wl1@|X+<
zxX7tU_QeC%==wdw(NQk}Xsi83G)%!yMlV2yONSSq*`|bwgyv&VRIvEoI(3BJ5DL18
z1uI9=yQK1>mKO=HgPkZhY(y4vZuHH<<t+|;0fE|6<!Y@dW?aj(UB8w!86v;0*!@l@
z2`sVg+s5Dk-=(I11#x36e^BPPlb8!wG|jfD2;Itlc_O;dT34AZ*l;_P&GC?Q$nmUg
z?Mm0~2Oi!UFg*^n5ZgRVwbU0Y{Y<{+N${gY_VH1vKQ7pLBZt#6@&&fZJmk+#+tPF5
zZFkrxB)$jLxHQ<7;OJPsIz*V}hN$%UU`=cr_&0yf=C^@Q7mrd{tLpi7EW9_W44e^V
zz7){~d~!9(zk~7oxMG`sN7v1E8v8HF^Tgc6)b&t84ON)TVqBL=R_qiavOv-1(zJ(}
zIRks(7>tUIS3p`*fJ-?l9E3eyrpNN>s{sXs2ej9nRJx4!L1th4`h21sO@(3qln$eV
zyoTON)77@!^_PJ9ghU%l*O2>fJIxGHMg2>%PGURFgS^ks!W3J)5xejK))+=QHDSm?
z%^e9PF-g=71?=&P`DQQK3N2LsH9Tun@QRN<EElyuadL9jLmew`2rFaAd(1zVme`EU
zc8FFSSdgnxgBjptdtwp#>bRPFm<Z|-jch#u&059l7tj_SW*1Gc#_+SWNP0kF8nEU*
z4ui@EM)KP5Ppe^gC4Qr7@q$)YK5^am<+|DjfcP&Rs&)-#cyuIOj^cR_BW63S`%oLT
zljwAB6VQgky2X`hVcjG+Gg?fF`4u;!^2@~YqlaEBsH?nXX-{U7DJgz}h746!a7MAN
zS0gW(G%1e1Hk7m5Y5ZSs6qP)140q9Sw0U_TIE=h3TW(O7Np{Yw&_$3!F6XwH<Fy7e
zv^|`F0nh9l=gKjwHubfsfT3t)R8(5*xq0GLg7elCri*jz5AbR@koL1{&v9@7Amv4i
zPgDA%`hDJN2!{ffa(w$S4R6Hg1vE)G^rxR)wUI$JP(e=XmMgc~nHgauVa}`j_1m|%
zt_2g29n+v!DtC9{G==Fo7<dq{IlqC*DFBm3_nfvhU7`LV@H}@|)sj7}0|X<b4izl!
z=gt*j>sB_jeUL5;iEvlSE@L5=JXQnio(7=AJ#2h|@&FoLCPe#w&va6R8>_toqHor8
ziU^hbIf%YG-90>;2VD)Xh;@krBTAsCca)j6u(+FLL)C&X%-(sId){Gq_vI1iAZm2K
zj6MPSGdaA&EMVv{hwMK`5ULYCsSW;P#|*4VMj#c*ifPLPX!QXGO^`5kzp>J(!)*3l
zPSSZsp%1p-A6`8bbu!0~8pQuM{OD_AgT`F1n0eQgh2rSGAQN}ecEODTX+_NioMwN#
z;iaLLRBVb1KIu-QIOM_QZe+E=xxKirI?N`uV%z)53WXGv3zT`P-kOrF{qIKYG$cz2
z!%{oq+YXfO*|Y*h!HV=#>8V1t35^BciZ}~@iVwuLo$?xRR{*Y-7U+o$2+_g|R2}r4
zkpW~$C=-2+yVQbKJ<R(o8zQ;T=ng5Hml$Dby#NqmQg~`?fjD(>d5C3!u^PG}J6S3u
zz6M`88paS>$l9hQr*{@u!17i5GQ=~d94n?y_EZN1WOZcO^WM4D`?wb@80V4GmVNji
zsD<;fK0o^vX_NJ9X&_vVl_?u{l8y@DEY03yYyoQvv3=`E92)6cv}fz9(?q7Mm;;r=
z$SD{;qn$VGkOjC@)@Quqs0%&;!3MDc`e<(QTEB(GQaUU#Twupgx<)W(;I+OvmWJ5@
zMmC>C4GRZfUH8p$<Cuulb1mHN@)CwKCt#fbJ!^DhA?R*AdG+HhVz|c{Li;YdMCUHY
zCITUW{adr?<}>}tO%Dw@gn$@`Hj1zZt<;r6>N&*qecF>d039aV?p#m$2^Vl!M*Wsl
z)`!&4a(ely5tr|BybPkAOj*ykk>E=+tpT2qh{B2DUGlke_TIJm3L_s{JXuNt7Vstu
zd1s(l*az|ssN53HYvEzpe68a6CjI=TuIfKrk2jn&)c#z5Yo8R)%)7T*cW^R$<Lu-3
ziG!i##|bj5_o6Owk%ca|`tbGQSw-Fqu&WpG2<A>O?aU>+b_><-q#}N^l841K)Wi}L
zQm7SU=M3+xok;Egd+Dqep|Q+AKsrZQwYwo9pD{FqZS+8#o8?zd_M25=5qG$l<rbZy
z?>KOu=Pqvn+Wo!l(wEn9!Z6;f=}LP={TpN$R~p%&p<!Z7s}&8$kDcem-J;g{V&4Yu
zn_N;F7@Q)`U*!i(vV{6F$B7*q6s9xa&`Wrw^N^a5veM+*7%!T^<a7kksh}OhhJrA>
zCvU6IE8d_*+XnGgXClfQ`f%-Z@If*1&HUtgixeOXAwbA6295yR2<08G$VPC6;{x1-
zO0*gYHNpf%^@p}S?d13$JDd*P0rbfxUmLeWo;|z{HcnTdCOKUT6-yTO#zOqv=#sq}
zOBYcF6m1O_IH2|vV*We^)_v_53#sX<fS<LLZQ1(yfyCS!Qj`5<+P)=Ek{@P{iNYqW
zGr0oZ^0_}IG?YaTg{|1b<@O!tp~fHE@mnhYw@-}2mFOKK=4pbKn~hZphtnvLcA1+?
ztFJk|lj=t}*J-uyl^RWM%u1(<E~&%Zxm$^+bp#DQ-8D{-OLBhy$XQn&6QOmh@V1_J
z$$u>~=<k3v@?qBV*q|hKc-ai}s1B>*MYbLyZa*~2+{PmZiGiQD)9&ZYubjQ-;9upF
zEwq@KLJK2~&jas_FT?g70cRt@In~jy$ZbI@iMmD@ru0BcOK7Pa)Z>PHiUdT7$%Xk!
z`PX5}TmXh~Q`j0DaS*zl+~lm#4E4uMj73M<l@?NoDpfXsj&}k2MY@r03XV+D@3+&|
zg%qW9(FkuWhoq8;Ng%Kt0^#j&XojB6a>E9bUN{Igbyh#H#6X-IR$`ZY*W;?q$7SvY
zrW~cF*VFsE^xu0<b{_$vu9CZ1(BR#Jcr*Xyb1(=wC`J1$9g1icvW4bz*m6Cxd;mP!
z_N9p0KI$@lJR_20_7Cjh(kSpSsO}6s52J7yKDLRzXM0DHcd_CKO~9Ip)w&ZdH0jl4
zVrRh0MKPW@5d(=%ZkTl%g?%=KyB3PK#=5-Bc5#@pQak%>K2`V8-!<_Q?wTeLUBCJs
zy1J*NcIJUSlqJa7{_e_t&12eds4Zb!qXpZwfe%$3zQ{QhI<g+%<WGAf--YwLhtZ2&
zQ1GK`G*}(LW)DuieDU!(rK*|?Q@`q|lD<C<-*c}+x4&4Co4=En>ekBRYG!AjrD*}g
zY%e@EIuFJr60zj*U@(H9Pu^3-U6oeg*D{n2_R5)^FbFb%9_b-KGxq7m79}Lgga^IQ
zCes7309I_`Wbw5gEl>)WiFx<`=MwEZ2)aCmu#A1KJ_+ol^MhXBXdinU*f^58o9ST7
zElJ1pPH2sIbsB}s@7Farx?*jNq_jL+@Q-;x-v^AS(5>JatzGzFFiz3CMLJ}`<~^fu
zGbmaSTCPr~9Q~2mWR>$odc)c<Q~!7=&xF;LbeIdkMM~5}4W{W`H@4uN&ZUYoC*P*)
z4%|r9jr)JS4?(mJ&<svR0WnY(1GhaAjpd{HHfVBkw-3Sav(IMm8`!%cRdXX;AoN*v
zj^40ai2$6E3cR{>@Xw1pZgD=p@ArN;(5~@DPVNY}gxnX6&%^wG1SurIB7vB<6bqon
zbKcOV6*KodeE9IXmHsNR`}xz7GiLTK0&nX54MFhKZD!B}_PS%`Dr1$@Pf|_q>*0@H
z%z$idy|xVIJ@q|>S55TGK4fpB%ld;*96FCz5}_{5m}-d-0(lt#9qjdnSaeV+)Bgbh
zn$(VzAm~my4ZF}}0!B=uoyLyt&%6_t!PTee+u(v2;<Gk@JV0iJ62ZL+)Cfu*gM2_^
zZa<)BnxG3BXfq&nK<v-~8T#DtLGZYb<e%V(Z6=XpWUvQ^hExFz0lqOB4+Qgp%`XxF
zvpb~MZB{end^v#32yG?g^@k^-aCDA)>W`Es)eJHyKHXZmJvi9(ee-ACR2lAFIPk)2
z;Zsx~9py;9k}KBjN1KFFx!BysZljm{&02Ipy(nsPXrcJ0q#yzU71OFBImRqJy2tFS
zqb>j1yOC*=MhHAk1PXY?=HWFEjz>`czug<E1e4jZmcxrOnX_Fvh8$>x0-2&7DmhNT
z@}N$?0XP}RREwHvr!+1*A4eD32);VHtr*6FVKAoc4P1HiM&90x_vP7_&tT@+00W|;
z>x7cbYD*Y_eO}M#)qNs=H{8S0tr`(wOB{y;1J;)_P(AE2_E7+KKd}I=wX^@JXXF0p
z8h}*#)E35|eCE>6yJ#e(Ya9Z|2|xwUmXtY^6Q9DEKnWsou?uF0e+o4<T0P{c>asB~
z7Rb>GYsd;<l5|=bMl)|n5?*OMd-bfSU6<BRV?)~?AV_-_W^RBj1HEuI3Zhdi#9O!#
z3J9Vf6J$0!h?Jw|SJ;Bn(m3ICg@N_n_>_?f8Y32^q?(Eruya!&1!WjG+`C)SrikEB
zH89*A1tvlSl+7m4Pd!^x$2O=bDi)}5U~)1);aQE^KMHO55p)8Jj~oEAT(GAh4UW>q
zd$w;@GUV7ivuYld&O^~GLtKg6DI_41&Qc%kwPMK-D~$@t)mug^M*-%PVE>tnZ}i=V
z))HE)?24?Bi1l63QT*|3l!Ywbo4crgYhG0#)lvp?r#(@I<AhVCq1au2<Qd+2OOp2_
z!Uu34R1&3E41yu&iQ(s4Ex{#vVd&w1e0pjLJAl)6aHB5>;Bs12bBc-ssS~m>*bLYT
zc21=^!wHhXzdz)widY<__p)7e;q?@y&;eAU%O8P|3=54`<!(VK&p+c=)|Z?6<)XD=
zM&P_}!4-)RRPO)WLjHKU?G4#H@p0#%0sQSTpt(F*LGJS7O*%o3&+MaD{OKYeoeBB}
zNumUq{lrPp1a#cT4;-vNBDZ3k=9&1ljrg&qx}Dib<!{Irn-*zBkhxfR7&QX;8o)8q
z$L%TGEMN=D5#VJ;mKcWXg*2Kr5Kfv^QP3-f1ZqN|c%_~Vv@2o(0_2rD2-#O2>tcS-
z9)Cpcc0H^MFCV~n!v2>B3!o;R!`J5{hRVyfdlK|=6PWr_5mAJCS~$QGCKvhlX+~O!
zA2GA@<P<JHeT5^)@F%twh;6NzTgnqSPKvw?zwg(_9p7JB@oaGUX)upD)qF@><%Lyf
zZhj4X^W6#(tHKXS&F>{RpJ%;A8W1=}=ZKxcG1?0;e?{|E65{FlTN@U1)i00%z^ZVF
z@b8?&>A}97eaj35k6|gay1*kq%0LA%=xg_uItsxKYX0sga484hWOcvqm(r`S5KB2h
zggzR2ioS}}B0AVD@ub{n1}z5uv~pvCmoSz&OSC$9wnk@jH%F;`kXvW#c+X(OwJ@jt
z6qrJ1&(@$I<kMo^n$WeP6W+*GctEFIGQ{$Kk9aP8i2ZsE92-7_v?~pJjb+`=PGJ8k
z<#I@?>rSn-5;!@5Ygnog^NrKbZfzc;V^GaX($V7r=eGUBmHfmg%o`3vu|I4uAD-`h
z38QMI)U&rnzP_!mW}YtA%DM9r2cr{|*Z^!qs{JtxFP*-CkJe(}lVn2ylkyYc7yQ0!
z)*f_#b`I^uP%8$#^61iFTnnb|rORMr-R!h}MV-MDU(MWUem3<p8f`KWTrf8R+#v}~
zku#urY4a2A6H0;S@$!q8GA{T^&xZbpL;r$}@x|%<$JT!27m6rhp-&JcD9gj3O3s4L
z6EgR>W5j8D6~(4g*t6O6e~&uiOwp+0!Y&1b^2YDm!xk=g-(BJZX{K_=5T1rb$%Ksr
zMh#LG2|a?^VyJ}~@T?BiYaTtI-_o*J`Rm<**=Es}-_3Q(zBCZ_8ld0gfk{<?r<@H4
z0UH=0v<mW$Gz2(xci(X^YPNxJ+TBE(UQ@<L*YvXI9oEI#q%=^v13LQDdD=cUdcxK>
zZ#~3$7*!QW^Z%vy+_OJLziu}FK54kfT4wGwmC#v#l>Ai8Nxw5zy=0Km+7GmI?M8Q;
zN1UZVOG~ZcQattO2UOQbN8a6Wi6;zo@tc8jv%PQTL80z@`x7)X$4ycLpwweEc~j2l
zO$=P}=N&^uOk^lk`|zf1MtOQkG7gMnCHTqCwri_<+u!urRjC>;f;^P+y$vn}-~r&)
z05HqRAq|wUW0)W$216~=kV^}+V1|WK2Pk-6z{nCJ<O7gz(}5c2TTIK)Yh=<6Ea3GQ
z=dQ>ABKCE&8ojX2CbHGsx>Q?Y^EteuAsq1L12qyGl}O8_OW!c3qWGiO_N?e1e<jP^
zOCHpFp`AsCiyb|M<H3C4buzkh^=A(Aa>al9b~c;p?*m`})X-gY0_2M4NC$uc9RRlz
zA4uE8JKpF3tkZAXM)pOzMs~B+g}&i6#81%wMl4E5bW6R$3%YYKwD2)$xz|k(P=g?F
zcauE1UxkkWR}+cxX~{x?f#{eqRy>JxBA__$Enl7+Yl8o-mGr3MS-<B<@PE5HvuHe~
zAzGAaj9V$;V<CRO7Q78ZM&*?pD=n^RA^h|>po$=dZ|RnkQ?v)D=KNVpQ(Zl2=y7m9
zpmxtSM%19snPqV}^He_<f%aX`C&A+s+=PR>YuBCu$oKYqY5l{c*l?y|8j4a2m(9*+
z^J_t2WdJB*coa0sZ&5x>q$UV44}ig#LoPsEll)-a=wQRY)$a!uV*$J4v7zZp@%l)O
zHyg|eKt~!Kp&ytpbR(?*=tovPE!4&KeR?q2m=N8?Hzm*t?r??$=J=`uN)yob&SD^{
z6ac)et|t`sA59>;CI~mg`Pujf0vjDbl^&K3`5Qa*zx)z&ZjBa%P6|;&jO<W6`<;XP
zR;&e@4gD40we(nYPxrO+9lNgmZYRj320Cwr(QhkT3?BMB2{kR%k6SUPCMD<x1L^iq
z>O_3S?s?Lgr#KTQI~BonHQ*CzHA@M@;#8->W|15_7M_TGQe&O1QGe~u3t~|t+^Z4~
zBlvBYPUe@ujvxpz069|~it}gJYx6-X4wQ2kC&WXU{&`y3Fb6t{6{UKU0|i4I81ayd
zQ_OecohV%l00r6N>xJaDpmS<~*~bNToY<2iKfx7J3HH%Xp4%;=zJ-Ec1K=De;)o)5
zdNhM@6<kd;M2Pp})V$U~7HGzSRB5!Hsj?_U^7z9v;GRRuZmKk;%{vtS%)!!P>)H$a
zoQQd7gbSW(7<;Yt>8ez0%=wPlTj~6Kdtaw`@Xhab-<g5(WkM>9x5h#%XIm&$l)wLe
z+3L~!_i`-;6fJ|~nr>)0Zw(AL15C|pw4r}FM>vJKv>P_`+=}<M`e1py6YL`p;s#cX
zr|r=E!28QB=FMCQZFVm0<q}Um6gazONqXfVT6<NyGtYz%AXBb_lp?NQUSpB)@)h{=
zkCSL4fQ%gg%=^gZQ*wOOpYTfYt6`Tqtc~+iq`d=n{D0yq>B%wKrAd)CDvM1eYc@1L
zeh!=5o%NdddYo<XF}Dao#;QEw_PRc5|Kw7Ci2@l%*kZ+vNBq*grt5HbiVw#BRWeS(
zN|Q2pk>MHFW}5ov&ATIvo0^%XHpP8-3g%J#qiB8tZF?L}f?U!MFhJm%80Izb3u?b~
z#Rzs65Kr9<q9ACZ2z}%~L6o^ZMj-K#)KayxB1}D!`r%^AmQFwmqE+SNE(vx)u6;;F
z9wsm&7eg1D^_%D#R$U*!1~OpnMq7pep7qH<%jD4D*46Mm?1}0Slnli2#eKRcm%doa
zG4@HJ(N(z`tN)llZ|%`EW}#6rS`ogTf(Pnn&hCIhO6a+G%&b^Ci|_9hilguS_;=M<
z!-T3vYnqGuH^TubCcys~sKN~hao^W*u4WsglI<y4oSVU1mg**7xEyF;j@01EcOO1H
z<30lwR1{!_AevII(jb`;fci33o82#|rVxOz%tsXk6FQIS{_z|l;K(fjI`DDS+)6Ls
zRKT6s9s&4$>%>vuK7tAg2d}|R7H*7WTq-kx9Ag7YuLv%mmTT6tC0L+B{bzB^CO!dx
zPfIcZl)4+nd)k^}pW0%JA4H3a5(XNp6b9rN0JL!qG>-r(G#H`3B4eN>^8Wp;5{|#>
zD*6Jz9ztZ;9(Q}|<!^m?rrpploJ~5X%OI=>dw1D_y_=vlTp;EP^@y<3FwT3SL;w2p
zLSDsmBrH*uNo53h<`|Z{Pgu>IZv&CMf(N*{?;lX+__(;2M@wD@cN?lnsn0j!5nKT1
zm7Qe`pl~UOIOwjD38Dz`u;=^q@f_+YB6HNBy~@5#c*JkoD0lcl1YyV}ypxlAH;;fT
z)wS;U-L03LUtH@w=!FYzxvd{7-BYD-PfVfdwKmQ6;^6lYS~(nuD`K_`@UP?7PmTBG
z-8_-c*h`z;YWx2}mjeuB)Q>dx{!gvJ&Q~LL`ASvK0gTmNxC`%0Q`Jb9<vAxkls&|m
zvX#r;cXtm`T04N@BG5+ZRf>u_<dAS4kWIJ%yYthajvbB-h-4r(8sHhH^~yk<rU{8c
zf!LD6IG!KYfAR?FbF~w6C6G1Iqoz-fmS;19L@o`Q)@8Nk$v&BP(C#lFxyw_H)ExgO
z`*AaM7fsw(HHqXu;F0{xxaqizL#$Eh*lc189Scl@f=$zCYXJa9`E;SW@)bssbF_&c
zO-sLGgQ_(bs%3;BSO4$LHy0-}{a9)o)<%+J{<q>{bAyUPuGB(>Vv-%0A%ZeWQlx?`
zhZj8p<iZGpg%wV(*@mN|puss89~`~N<OKlHGyuylI&kI!RM~h3j9|{woOK(nAuILJ
z1tVF++(Y{p<MQfTi01%&`$s)6M&SS@Cue{&eGfUjtC4bUX}=y8V3_AsYN;eZ>^^KN
zCXwf1VNCPZafL?r&#P6(Ww!=@!|FO;TyVe%A@-yLTyJmg3F@_Yl4z3_0;?mM^Yn#-
z7|9b^G5X2=DQQ<i!)TtIc(EDO{&|*EZ*HKg@a~3;J*ke`%m{E6nz>>4)8lh<g^k`D
z!axMX>CK1slKLh`cpY+;f%SLZ+wu>Or99OT{Q@>)4&(f9LUADcaRt`6N3PeN_K32E
zjyq0s7|}+2r^^&jiJ&LrF3<<y5Ka8OEdBiH5;ey!P;?tf0w>Wk&!M*{lOKr7Wvm}q
zV9F%zr2j<vQ@PiP#QIkZ1yx&+;l<!Xw3{Cn2z9x>8!?$sUrsH4&OG@FX0ZUKyO!l9
z#Vt%22(401^aI4uYTK_@A4S9E{BFCoEv@fv%hB_nIq-a|ar-A3vOOcAA)iUzdiURQ
zOB2Gvapj+rOk{gtaUiUDRf7@zJAkTD1WW|`xqJ1nOW-sqcQajU*Kb;#AOGeq!#62}
zd_9@~)3ucsO7`wMc`Xi#?RR#4jQ~qUOiU6rff#<KdewIxRvLSB-vLH`pmCqHS3%``
zn~H<YC!^ng-Uo>|ubb9+Y?woz{T@n%I3X-Y$`-hd_3_qO9@aJQdsQeRz?D)*JfkK#
zx3@7-iF;Sx_lqTSJA-<s`YVbqs0Uh@s=p0LKp%Qpq^ZEcGTi{rzb?*+^lbh;5l|U6
zqrUo5vj^2CL+#%~@dPjtm-qha5yNmZKv;+jP1#TS7mxFW10)u438Vp|d3p_WX*mHP
zvxx2lV|^r72#6t@!2ryD=zN<9gvjC=7-XY+G7I}iZ2L!$K(QXHME2|;#(~_BJS&i%
zDMH2#xJRfxTBws<{y9-~dulQD_T9E@UFXe0f$jfff~X7(%zbAKQGU48(W%YzDO3LY
zD-B}^|AsHm;dPT_r|^;yO{Y<(v5*ct->0_TqvMpy@7^pQKf&{NdT^rp>GDeM1~6{^
z{pRTZLBGBIuYOw*fG6-qm8EP}nE%lN0H?u`rq~uc4n)U?8gZ#SnQ@U>Bj_>G!dVrM
zVgl*HDJ-DHj{=P;7Uqv>Wx2fdbug$+2llg#ARj9hT02yxL;tdM_qE+@x}~;WlEFFi
z_jp$*2-TLDV){a?3YClH@{R`e%007?3!z59UuOZz4R-Znz9SdxzVFK2_V?GjBT3r+
zm4&9%{^zv^?tlvDgzQlbEX5Ok=oA_c)=ZOXr^}{nP?aFk9thd~wCM(9z%p2Np)&9~
zwhiFcNZSCeL3e3H5r|J+Rvw0IkLVbh0cV6v%J*vZb+{Mw0=;|jh`r;R^M(ErY=@ze
zzVoD5<gl%QqE9iel>tdgJaYCIF)CkwHSP(c8XiY?`qVAmtxuRD1OV|1O0gf80{VYA
z+T`kNFE-AuCfLJB>${w2zvsw}y#nD}MGyXJ0dQ>Gq&v+m6?FcuI}P~9<D_pWjEC{i
z9yg3G>1E%s!52}u7?fO#2mkA#>>=2FgT}QI#I&zsyZ9yrTEV{-Mu7J=8zcDfNGb>o
z#t=amRJo+LorNVa+9V=t2JNyB7kzN2SGa$AP#7;SdNM}>eGgl~vJC7psx5LbeUhTy
z)=9e~KnA-dy@1a~458kkNZRhdS(#M1M%c*sZr11j{!J%Fs9O+;@v=4`AFeG*Z7NX<
zhv5i#R!Glm=rZ<%Gn@HouG*HNF=uYz%8a!8{U{-`!4ms4G_VQW4PdumKwY0*9fA%S
zDb_x@!9wEJ?cR-e`3PNa1MQ_Wl);grSJgh(0v6TH<m@eaDx=M=h2Bm7#XlzYZ|-e_
zRFs}!lOY#qF?gg_=)WOm8|EEpcK1B@*%trq-|s2|7i~dDQI4$y2G(!T!yr9}xg9t6
z*I$Cs6&$$3`LyK#u#7sU$xg_R@)fwx7_*l|aRWQ>NcW63_&PWV3TX1UVYa?YLGyO(
zuK<-@^=XcaEG3WW;4K^y2x0mELZL$T|H)pH;vpo=*{27eyTnAL?%xl#`8rg&NfT#_
zC#U{%d2p&!FmBv4pvuAU6ER7T?t+m5dO^Y$*<b|%Go**0FnVug;UsiG>S%FdFE)Oh
ztpL;)%V0cXk&=Fb0siTNS2Y{g9|H~w?+iC?P+Rekoo4fjM$tC1J-`P;I3iWQRt>R#
z&PyIV<fjuYgB(Io@p-#0qNU6KL~{B>(quTXN;dmHO%VPEAQktS1T2OIvcN;$EV|J{
z<EOa^Z8cymP+(rZVh11#WI%HDkY2Aolq%b2q0{Xt#glHio;k1?08z(;b|$$yVK|vR
zop$^<@8;b9&L{FntF0>+7y!q5Rh&Jz3I-fg9gmCtO1w3dq*lcAxGjl7`EMrb)`wPP
zbx}Y-j!gq<5WI54mL{78#n~Mr1psD8ZgySNAW5SFYf%AUF3%?fGcGb2z~llk7Y84t
zxkCddih<qsL$IT00=72SQME^{y(S=3h#J#0nD%OwxmY$C`z@}KEqm3TuIv5uEBW=T
zLCBNa(OWi=Xpb@c<KOzIk_svdnw<?LjKguI_FZklf8EoENZ=v0A&x-z-<Pom+U$4o
z;-3{M{HUD{{UrHOp(C=Bnd)S?9FXX*zz*s3gcxcwU`M+dUQsHAMybe=91XeOMByO4
zoB$a)SS{uCO~Vqd2?#7{fGsSa$c>cvR&izd$`}8-tiJKV*KKR3|9?<<;I|54cG@ms
z<X+fA0JZNc-W2!$H=rOSJL(QaxU)!_y5<acn3XiZBOHZN4K2!$;nq;>750u33oqu;
zk`cC}ks#=b2$z7kleb-(JW7k1H`)KQ8(J6sqfkwbmfxr5p!a|`dFg06Zu6nt*bU#k
z4pL|u4699-oX6X^cwm`07cy!p`r>s723euNR<f_;I{Ji{RZ`$%hyeEe@q>?0Rv@oR
zD5WMuAIIzO{(rc8?|7{L_is3&jHZkVr6^^l5{fz{Aqiz<t5jxIl<m||iBLvF+V;xI
z){=%KD<chA8D(bQ&ll>G&-J~2kKg^fugCSr{m=J%e#H5Hzs9ki$8kK57e$lh><oLS
zmC?b8fZt=04y$>}c8b-siG4Kj8(O+!mg)8xCxb~LlU4kOXLU^0j-ziLRUX%{qE`^s
z2rDwdI~Rk)OTU@EA-zZYAME4KY&eCnnD_TZv<E-;8lr36*4xi0js(LIplzG0Ldq8>
zd%>&_cxLk6BtN8VpmL$|iNF0-6H0&k_DrU2LHLuGc>G)*Jt%WUNQEjIHC!aFVPZV=
zNozyWVAYRfZ^BC86y_xaql~Tu+zR|C&jGpMl9C|7km}rb2gSzfOs?me797gys*g?{
zz2l_*yWn>vr;XcvZY5D#sa^ZS-KHxCJ@3&WNQ&vYoqA{y!sXek<^NI`lcKlC6J=y`
zJ$JEB9+5Hw$@CIeEHfLI?qhez_4+LvOl9-t)nwTo$t)*BPLLUPR6bJVO4}DPG<<eB
zfV~vl_2u7{+ZYbYB9LZtzPNQV?x@k>38gF`eYUwj9y4#3LI!CKN@%7W#ii3dH~I7L
z9);8%phd;iHPBeSNqB=qWVu%*!4yj)Wxi*eKwB3nYD44bV@ziueXq5L3887!0?>q<
zjQ1NB1&Z>KUSk$@^R(J5uTN+Ftymn{jMZ`sT!I~)mV62wTr0PA=53fo`<Vb`Pkv+s
zmQO1G!?K=~1u12(0<3v~bz7INfc4JC0(X;=6yrI#(352<ME5RwMN6Gyrx+p<fpAEH
zbjJsM2Q+#W3v~iY;=<hYCR`iIetr~4WlZH9o?iB3@|VFZkmD}k6%o4WJ{-gTg7nst
zh8}dR*(GWhEeC7nH+{#5A#iEjL)&*LAKPUY)a~I$zTK-AD0G0mTOxjW1SeZCB0=*h
zXKM9i_t;#&eZSRHp%1J@UZKCG9*v(<oulyle-RmN08!{luCU7&+b3_tK?+vQ-pa?U
z6@dTm*c^)<Su+v|skovpSr~Ss<~FApNwRx-o})+iXc-1SW){UetGI(vz6mjZm7cur
zJcek|H(<Eso1?pD3HD8k4he96T$=Y`oE#kdB=v}l`te?*$&bhAWhUAWp<3k{nBumE
z7#v5wScFVQfzX?k09Eiu|MRPa%#ZD1!sfkbC_SBJd&ntm8P2+@s6%q?bVwv-`|BN2
z)|1wkK^~HSD>}tD2zoA|-lbxa-Ssa$w`Wq%t&wY=<i-c_F)*k*D?7B9X{Bls6wJ`7
z_iG89D-Z)LPA(vM#6d9vZNaW*Qlh9_V0u~yETFeG6q-tzB-<dXhP}y;^J!u7-_Mb3
zc@gj=-*@st+tUL1C@ru6ZV=qm)TA+oPys`8o3+792_X~3K}&t$+W|%p<@h7vB2_&!
zQrG@^DfM~`u+IlR2Qo$VbGjyLBLrFgcP&l~JW-JEQrkrqh;>q`zNSrgRsFsMJi3!E
zf$+kCNodR1fFw;byI<?uW8sfZMI&{OF?!;+7sp<^=$Vc~J_t8kc=C|P=x7<6)Q}e=
z^n<^KbH>1+y$6AQ=iFUBv@A;rA3XQAEm659B+nYZS`B5~!iv7J)`0iq>4kf4%vr#-
zj<C~j8X`rtXtmgeBc9|v1GXBW)Seqtdh=N5c7t4_`yyf@{!e&S0`FR&L@4xDQp_1Q
zde^bcZ1a97<U8f3q0!DoOxCxN+s8&U+Bw79exF&P(H^WlI<4ydI76SneZO!)UUjJW
zly}t1yy%5&<GG47a6PQDa<{}pw<>bV&ThMX7c*o6`M2@ezhX)&6fvTvps0~@Kzv<1
zijHs3N!s`<(sQ_Hl*IgEVqyc!Reg{=s-h+2h}QW-&p$jzT&IlE)reQ-yA!mBxZdMR
zkfI?!`Anm?BNJdiyarG#8<RI^adX$abzjVIb2^3S>b6&D>J@p6H^FJDi<dL=GlIAl
z>@u0W&wu^T%+5d<46>N}+1n2L_PYig=*&WS3K=$Zp_0l?_9K85q8nJ-!AxTh?_<O9
z0=3D)8EVbq9~>`Mfj!3#AXNz{VYFHEMwS_FdidCpR>rx{ABmqb$Ijcbc&);Mi7{)^
z%oec-^dTwHVsjJUlLCGaxRu@|qz@oOiFlQ`3I578_$`uGSDgMtL#mDlvWI&3&HW!F
zXg_063&D2JKJZvSS#^)gVz~}JH>ab3N)O=c?yKHG(|hn+VKbanAqV)Rio!HiMYasL
z_UT@2g|u7Fvsa?p;hl)_aVXzI2pxFmm3A%KaErurj!c?UzrLQvgUSVnb}c&U2Fx(o
zaGRIfFl%b}OE4OcPS&Q3x)Tqrc_%%r_(FIiO_(6$h$5|g*B;a^lZjptW~9GE&mCPt
z_rQN&+Ml?Iut1Spd=>R$e2wlS2?WkPz}^{R*0`2a>s+GB)uw>8^7905dM&0<o=&p<
z6uz6PYs1Yvy%<J_V`-q3V4;|GUE;mzr1>9XOy<U)!Q6Ot?n%*Q1}Rk$ZG=gY0B)mK
z#O@&IQWe;s`xmeHE>1p&<?5Z}^_{#P$)HE0EFcVzDCcmt`i#EGy9ZMa+hOu5$yXTB
zY;sfinE#SE@weC5{P-j*{Npir?;<ow2t|^If!p1jg*$;-2cr48WtdPLL3UeVe{>KR
z*E>`jl=9I^Nv_Ad5bB2MB9F&qk63>v5(ZeYG?|egEhkW-4W4)c>cqe9EzbM#WGef!
zB`+iuN(O{;!nmmcs768~-Ia8KmCd!008T5E&5*Pu_MyP2)20D^L)vR#7ZHoEZui-R
zLOVMH?u;SSq?2atZ^%goBip9={h^Rf>e&QW1zE%c*MP5>j_m)&A)uiMMB>xIgt7D6
z{LdV@iQ2FChO;#MVnugef4pqoClHIV^ve@6UQ?r>m#BPMr^)ncOF*=8+64F6hux38
z_N~ID1YI!vnt6MFf-nkx4gJ=O0#kmCZlN}hH;jA<++(*457JWM+-FM4AB`sG9fGK}
zvyMV035$tFjX#ema7a9Wo~>PDV?vl8#Qx&N;tJVs%cSf~Vn_X~aMMHqmN*FQfj}LF
zuLm<<y4NsG^}dHoVY0wU3*BAWo0T?Bb{Ww89I-T#@$q|(I!e>%htoWmi;~a6)(gK=
z(b#wycHq#{yX*da2Cf9;TF9jZSMw=dMYZku1>6GYeya46lyuZ5p?>+@i_uBnz)e{f
zVA*CQ{v!{CqPm`QDi^W&e?pN-g_m&jQ8w4ls3Q#_GZ?@Ns}Z>6S#j|ztIUSA!1=QV
zR_ma8jXUl}LA#&}nM*{5W27SB%}nZwnMDWv6h!Jo1+?N@>(|ySV@sCpJc2niJ@6bZ
zG39=iikUSVbu8D&q_I0!whNHDxYsOYsZ1<l4(%G(rR?tu8RhFU@A+J_0nqDy5cR7M
z23AgJY&c^>XD8|UE1BeF0JA;FWA*5)6)zS^H~N-XTXkQBGr=m>i1vCp%RFOXu03$*
z;ZEwApwi()v#qTy2?skH34sj?a{{Rv{7urIV|{|gxzE;~0<SfDKT9iUt<pSv1ZFAW
z55J+<bN+{$^ATW!AqrtN<Iw@C=OdI_h30rfPY#zx>5eOU+|MXi@*J?f%G3p79d-4=
ztkA=%pAcrVIf(j@QoZDdg(x{7$~4uYS|sxr_U-`0E?Q=j0OqUOH~j7n7TI<TPOu=-
zPYy2Z66O`_mH}4gRo3v>p_8S0%=I`J#}vZZ0=zkq0gR1!U1}vd6_F>8-BOr64Fm9K
zz_@q|6gz5>7}or)y7j-VN8Dr#Tgepz+~%xERQ{qWWGysA^&bTvjH1w3+2k?JGwxTi
zR?FIRK=ebb=Yomh>VjSL89&5hJ}pFkHvnCUQV%;pBCx`#Y4H%6nj_4dH^Kdx<d=hL
z>eCw8I|AM=(r|a8Xp8F!|5U;s6@tEY!dCiOG_GzKE4;|G7X#fEWS-2U?|=M}|EBA0
z`r<4Q$~B}CnoVZ9&%v!nW$9VSK<TEZ7nc~?W$|k#=xQQnS_9|G&gj;|yVczS$<N`8
zJJ?O|D^<$n>HjL4yYj+u3S*vc1;>4OTt^h6JZ-Aw-#&?-SOA8&q;3th0B{_Y35Q;4
ze~2jU1#I1|YPZRXn){@naI=A|Jn9tSB}Bh_IGv^6Ok<74+*y?b<1zHqo*R*^U5P60
z;?)IWYkz=&hp;Fdj<4JdZ|tx-g?_&l^=f})gvkqkSq)>};=5og1Z^5Y423Ub@!lNP
zEP9r6vh7P4Yq}F$9p6Z9-bGuc;li5iP_JoqT6h9M{AA|J142H2_TWr5W8gWbC*xY_
zggdNoSaS-YUGIUSp2ls}K6)O6Z+ET<>6CZ8>CxGdC;J)17v)~u(Vlau`_Kd&1Q&b{
zVLm|del2rt`*&F?Awrm}^KelT_8*lvQST4R03Bw5sKNa=3N}_K+k;`ef_7fC+ekQ>
z^F*-1Iv>aJ3{anQu%*f7nY3)BXL<1%CX+ONqHi|$`+b@>K*bYCLr8(>i|_q8Nyf8K
z3LAhPuLJNbIMiggu2&RX#U+tqf6`;I%b-vC^>8;ODKUR|AN?wI+2a+q-=y-8DIs*a
zZ%u;9_5NS`z+&X-NXP|y&FZj2ZcNo}z9b$JC~CA_U6w<0!xKuP%aaEEe(^6T4e^O>
z1ZFQke?W`c6pQ*!fmS2Q`=Dpl8|ZEnRk*UEa{NJ}jAotY50(CBU+><%o7^eL)qigu
zUA)jIVgEBsvqr#(`^!=nB{8IbzK3K$SV_4gN&H6o-q9pR#LtuIbRw<5h|8Ay<8@gm
zB?$}-jh5M=PJO3Yc(lLcmBaqGsTCdvK)F;eE@j?JXLn(_%(mS;mQxi0-pLMP^uL4o
z2IBf=uDB;N4rpzRF=Yi5$=|k?`h!$MA3Ck8upZf1M=F_jgm`{-;xXV&yXFTDoa$oA
zDqLq$!O0p4f-ET<L|MAoyW8gor#Q25)SSAxn~uH?Rp<P4YmWY}unKJ-_!Mhla{GgX
zqZ|;t1VQ)zdYyyOAu&1+?|1Q>zRBsV(<X|o*sBhh<ff?9aWG88U1?EvOUxNXQ(_ZB
z*i$NY;nm+y85Pa?9TVs+^P^{z-Z|C{#+PiP@cnM=iPWv;H+NV3k;{bc3hi7%+H<`0
zhV99a-b|FAt)gePWiJGMBhR8$G#(QXBw@}_dA0#JKMkSFF6GFvFii@*`Y4)`nlK8a
zEb)$8<pG1Z%0vS_l5{0SO(t(68?*;BzH2Nx4}kMx_5yV(ONE0eKzzb_Qrmwv`-ELY
z#V2vL)9RHf)<B!Ww`5s?_y2aH)tQJbmNh)MoNTxYa84{_SLX)Zs*TO3P3l@D{iATU
z?gmC{Cnc~vHqttizWw9kSsL^X{aeUzjF%yz-GG3%aD<}wUb?SLZj1GrHzXpjr>~H4
z>W}Y0l|`4}&V$Yfj?#HrS6Q0BSi4fRhX1D9%>hTJxcVJo908@ifw>855z;>9nXa~$
zYuP^;k;B2PIe^f2phH{tn*osQocR_rE(Y!i!vhlUpMU6k4eepgtZGJfufZxzw{5A-
zG`wQJMZi<HZ9dar{oeVBYG?8_DQ8fUUQ%u3kud+ukJl|$I&My?^eSpa%h5l=i@u@q
zIi9DRSE3!_GM7~OrVmy+^9N41$2_Pn=VC@fMKC;Y!b?ZU;BFg+>ZO|pX;mW{l1!EI
zsBGlw*UTcbB(FD1y$6kmx@ersK7b^_Xf@IcLra$1d?oI=i2nyi>>}ablEI(HIv*8G
zGvKDtNMeS`HQ47%wJop&-{7J2y-wu5ihh_((WLhNoJ`|cv2Ve#4*J>i!>NRRHZ3VO
zmMNNMySdsM)V4>s#pf?&vsAmH;cKh3Ay&kBg-A~2u{S+L+c>*rp6WIvI6Yw!h*Wv!
zC+3j5k;tIdA{Ydwyj|RjH)z2jOv%4?VyX#=aM@t0goPs3n4-FU=<emEVmnz_SZD?0
z+?M@<sjGf7LKvRm*~BIss6NxKYhdtx&PJ7~S`v0eE#WKK`75`cj0SDrj!#zln_@lL
zW8GhhLUAR=g(}RV&Z6t1m5Ls*5jgFWq%@XuV6X@N1cm;WnulC&L-=Kx#k#<2Quu88
zTGfTKKA;CU?YPvfGz0oIWXu@&9r>@lng0PaA?K6#ucKWIz>5TByZSbeKx#7XGF}58
zK`$mV+Cmb#<;g9@W&e)Keqj#g8Mz5@e&ml28W;n;t_}Hq#;O>IbXt<`oSX=qp6~OS
zP9=mdXV?zR){3Wq*ML5!m7uAZuHeVlVALP9k1g#M@d3VTZtx~du!xN7%X@*75@ZM&
z;mni8j}ZirrpeoeUiJ~-6Hd$S>uRVI{~(o=zdUwMwg+07N;el@P6{d2{4uw`JY~86
zm#r}ztz@7Ba6>zlU=$X-LSgGSOaix2K4tKO?M0E^h(_{-x7$Zvf3KF)oQQU*5y(38
zeKR^~ggwt&gdH(VYV7x-i5_jYcwip!y#i>f4#*HskK`l_E7CKmojFrct@lKH8Q2UL
zR&Nb&Yor<)$c%I$Wu~ik70j|(TjavEWpx<7_+L6l2g{q{nOPw2Coq@u?5h;+0w`i`
zV1`v~g68IzJ%{xkB#LX=?0EbAz1qORBtseXK??14_ldDnB)M@yi%laLW=A&)6QZ`g
zM8EZ&a8=Qt-$INg2S!*2=^@~#P8(S$K-ncZ-i$nlUWMBY_&<+2z3Gw65{`mvfuh}V
zxV=f7kvlvLQK}(N|Muu585%a%40Ob#RNIi7!jQ#)*k_laTfb(fl+Ff^g_RyjaT8S{
zdKvM`X76rB^u5_mI<V%&DGrWY+VjALd78!Lv|Hl$WX58L%scVMC}Ui7nf-T*2G4!i
zrWFSoRE`|L{gImYZ}kJ-8POJn=hyR~t<9-8V`Zd~_b2_h+EZ`;c8Dvh`*?43QWYm_
zS2W0(64H1Uw8up_AHa#4fZM}4ZWsNwUvS;!^K9;eTC43rwozmKR6Vbb^^(rGTX05B
z>V>S5o73rYL!gG5q(raNS`l+K+I`?J+F4K$N>qBDVl&z4W|a)}AJemqMWc+q|L8lQ
z(pv*>6rg^H3}S?v`1N(%u?in&LgR)OshoQ*BJ);AA2bA|UnUCdeO@8kc+%+QyYHvo
z8qe~33=-SK`OEEu;u-<=n?@#@1%yfvo#HnT+DUGU!N@*(5i(=sD3VuxYb8AxzkIq2
zD49-(58&Imtv<1<jr|eSpD|uvg>eazA91&y=r)p$pdv5&v0zwOIgEvor&rV{T5#t2
zZ)SqGDiy|=WJYq%oWbj2FM`^?|7Z`(z+cIpP;#U@tX+s281RWl&x#hvip^{z<Qn{u
zrytjums<%%jps*yeWI6a$SDzVpXj!1Jof{Nw&Z1ebkU|pD8ZNyx`S>+xWKs-eNK!p
z{g=~B-|b{R`?Y)P!BYxz`1cJTvqq81ne(0BQjqoJ{!^?n8li_0F5AhO9Yni??uBt$
z`+ZUDkt|dyeSX>j!#DGSOd>vQsL4t$D1^r%?<TvkM#)X(_@_c@(TQVWqt0g@Ykt#+
zLSt58HFM^YwR`=Mg?eD_Q+o{1n_acRse8IDilpRsKr<iM?6$x(Beqf5yy#}Bh$)KN
z`)+OT9U6GANUs)Hw~OaJs<8-RP7JM3G<6t$guUPwJhnE9T~oEX?dHVj142)msFdA{
zFXPaZc*|j!OK&IXG7!%S%6xPxf6|I%lD;*kZ6thOr1W|Ln=>ui5xMhAaUrUWXC;mt
zIAbAQ7e8n5{G}TT#r1YI`^Nf2dYk)y80U`O+E!Wl?o|7;x6oMTqGw}s6PL(x>5D_j
z(+CnygL@xuCo(MMMN)aG_MY_fg$RRI17ck0a@37p{ba*+SVIBNCoo|K5{#zfq(zGm
z-&h#3MJ0mBE){9gB&YJ!)x_%N`^S52C56`>G+<m7Sk&gfO3w8}tKh8oxAS$(oM_@2
zMKfLJmVpiKLbNASvJ47-OF~3%Y9OX!eq4q~+_T-Ly<)wT$9V35<qM+?-JVRFCEqI;
zbq8JF0es5%*d>KRwdqDq>adEO-oDSjjfST@JzUjyes^kD5eC$~#pi$j3$DYY#i5&Q
zcaatMdJPS!M1?c13=I8DV$QJFK7>zRA4`QeTjw)r8;$GaI3{2)!29Fkn*OucmKNFX
zKO+flrZ3lTm_LF~dYhIlYhSyC8(dsFmGWi|VO>6wbPvjRA9Ge0JwJD=%&$*pdZ0tI
zk7^r94FhkeFR!Q$0DSW;_&z)04t`XvE5x46JBvb1Ly~&Qc&W!T_nQGQhRodd)AbnP
zpcj^O4!`*dC|N@)0U{tME%@uc;fBTY3ODrv54{V$eWE1xgn6Og@==zOlIZMKr_i5`
z)YPiDxb|%h`e2_#SEBameRQ@zSP!AyKc2{bkh4$rYD}kYp*f<zzH*dSK+OnoS9(&v
zb6mpk;5(S27>$@f_R$T1Y*RfZ+&h(9k*5FYtg@^%Jpm1iNPt!33w0!%$S*f?=@fTr
zzJ7fCX5PrGt8wuh#S?jF4SDJOSr~3zxW%i<K<gE7TX8Xs@*=A8LXMr*Ap*kXENTG@
z4vQG9Pw`rmSgt&eX6H^(77Fc7MbD;!<L;*h3m$H>I@R*}#`U4sy-gpax}DzYZST(E
zcPYI-xJ~|H2MQr=--uVUHZ(dZ^vr**Wc;-0c==!x=M_OG5~Zblxg@9+sJVhxShP$E
zMTEVjoT&3MKT%{Z7K)HN`O{eR;h}pHR|*gN2nTQD5Y2AA{pS68nN`R{F;e+X%dzk!
zjDeP`;L^kjs~z{%-S4k>ambtP$d!Z;Yt{vwd}aQ$I(fxw{pIBL@%yf&c+R4sr88e;
ztnMOwtukN@js14zJGX@C7#8tN{S!Wvy{6T1v#ELAJiObpaAHX)64*j`(VxgcY3$ty
zl23V-*T&$(Nct5Te^e-AU9N@axt*Q+UbCD@^x+F!XfImeI<!!a+wR3GF%Ssn-iD*s
zB#H3}z2$+vz$8zNWRwCIYxZJ1`AUB^8Af~&Sip8sg&vwd*o7vyB`s+1<!T)3I@R0b
zHuTEK4}jQO(S(ZU__Y-Xg6$2152FsmgC~E?lW*y6)%*{AKb|bS;v<qBJ@%viZCq0R
zR#=+IkYLL`lX~D3KYzW{&M!~6tf{55CK2wZ7maB4(Zu~bb*<ViDpV&{^bGRE<Q&vP
z2gA}3eLJP<=5ss?8Wj1k5yrHMi#aU|H-}J|XiI0Wj}k~-Mdv>7@Mh3F$7H0w6<8Ln
z(vjBc2%@u8Ok@Uez{_8tve=fIhcs?V&r88mmK-hz1pMW|m(^#o%9^Fm&R4wZMde7D
z+`?y#;{%y~BXdh+y!m(xZdLnAd|c|%nXY}yz{#vcf}8(l%O0=oYpeT9aE*1uoO+t-
z(_Qu#!`|^eIwsf`CTV*ZmBM$+M9k<IPA+bBZ4?+9%l&+wGrQ?U<6c^oaR5)Hs%e!h
zw8h+~^lvy+=?K5mj^M+MS?xdC5Kd=o;bE0rG$(*g!{T}Mw^RA76ZYNTV;rm+L5tJ~
zgiPFseu*Mos`C{Uj9tA-ZVax4+OERe1FJU#hHDkxu&%vTDK`1j-?!#PC`D@}*Rjh=
z_6&;@m7;pR0|jHg+^{db2wy%sF#h_R1D5Ki*HF0v_a!7hW_o1(`FnlQp4n12Ad0r2
zlyTg?x#fD6qZ=F3ejALHd=A>LdU1=APCm8bKxL5+!f#)R7TiFmxgYYp&kzpRN!afg
z1jv{L;pr*gW7Ws!|D^4~%@s<Ed`gq}Y@VX*zFI1#X~aYb^vPYh()<x8_up`=HN)=i
zR((E(iXGpkC*GI~bTZQ}@GNSrpDkA4T3mM`$LhnvS!_a4b1#~u-{W6#E=(`3@yy+>
zomRpX_xV&7c^}?b&bY{wkEZ&9?`GZFJv`hrE!30xL1Au?#LwveV%~BRRVsdbl8;g6
zYGmV4{gjX4sNMWXItRr>qd=Z>-`q0;H#%dB8Ov-w91VHKQSj?o?lHdVugD^Gf5l#1
zY8rO#del-{chhooq;bNJPq6{XmPRWhm&YLE_=5Hr(d7)cWsG|40f9fV_`(<U1m)CQ
zyDy<`M?_1{Dm`~ICmjPTd#mf)%~GOP86S?F+mvw+T%W=mpTrsCzf|hdmP{y#=Jo9t
zIcR6muWyaz9HpX$r5Fy}HU>YLftp`!svfl|4STKKhjK!`+#dlCTymqI;PWoMpoJ`r
zd+!hLjH;sj);0TN-ibub5wOn&r8!UO59;|P$<cYA&+k1mvd*}4aB<n^8H$AxwUZjZ
zT{1p-D0MMU869m&=F2sW;|IIapRwybg_h*Be<QB_DE|6KBUAO$sjaumDGQX6sh9z`
zCFx2K8hJi>G27A1+Pc#9wOH{}9Dak;`ayTx@XI+}*@B{D(461dEmw)kQGJ_cuPgDs
zl;a*Mz3dex?xubPPQ3dhGmGtGM*gSEj^B6}@Sr|C5{wJ)uDG+pEi!nt+9kK3X6JaM
z-w2|@!%yZ*TqbNSR!>mi^~H1cf(v9?*yW}tK4y94rpl@^*#ik$1#^}1&ehhY%XHS2
ze7R^6aFqjsvRnt%z`R!?GnPdAq3qCsE2#V+Eg^AxjW8u(RH)Axu<T)Q`iiE-ndsUH
z*F$sH7X&Vwus*H|5))aboH>^9cOwU+m6D$~Ip@|`ptZ=aV7xy`rfWoIuzm>!y1W(M
zp!#xOsUVmv3}dSkpAGXwP<WZr(28|D!r=?zXQ2=0n({x|7ISNFyb|$dz0D`;+~@Z~
z3~O&hycdMfvM#PMk4yLhx%gfaE`5AYq=fial<Xek(sVNyr^bIgvgF0B%44_3S*A<^
z8}q>mL$RRCDyxl4B}ypmbEee6wBVZ<^29A%uxa4~DjkD{iZRmKV&)aml5KadGKuJ_
z(n@eSRH)km?r3J$)JXWUl}!Uj%RXBw9uq$wS-<Ldea-d4s7vFBVawf1ANd=*hH5R6
zzI{{i66r{LzmMDOVi9ofcPSCMkD3CNzxc~7C{Y$ne|MF#ApAtO*(j!vi)!j84X}+2
zJotEd1YDnX%tjv}`hiwT>{VQjHHU1`i11)w+y2gszUOx~?bMo00vDk!EdBgK$@d>>
zU1;faRdHb7eIKn@(kk#tx=iz_qaXw6h)uyrL$;3g+dd-80js!+ER`*qr@6O7iO0KX
z>ks_2-J!E&EL2n5Kpgc>x@nigt@EWE8yu={k0^I4?!D?F14PJ8WY!^;&5wO%QD4&4
z&XyYO2`RXYuHvLt*^=&`;AZ{`cS|I!*sqyY(Th$76UZL@L?KN(8+Mq9`37180_fIR
zd#tHWQkDL>i?8TPSZDvM#=NBx4THPQ8)~hKWm879&LvUz$7vMN>F(u;8?2$TGflY|
zoiinNly9``>DnjO9Jgl(?R_KW=PWyE{eRo~4;(Mr&phdPWwwfjwvs&*bDQ{lo)LRl
zzuL>>=L8^VjylVSDhJ$V#i_F=i{c1~3(z#*zWNih(>7&Gq0|?ve|CFJY!fAh=?Id*
z-HM!BPnv@DEdR(hCwrpbXy$VHhiIAK;l~J*dCOP;hY4uvUm098@nAd9y%d-%$88yz
zU&q15F{yzkdUs45=nFqoC#YwUkZv~=$irop8~Y*nqjTC|+(#-`2B>#Wquq#C%G$v#
z!tUkzpN$h^jRse(mhe!b4$<lJZh5U{wvJjscVq7UJDN7<>DTVNvD6~*(Xk@-MA+w_
zkC{bT>$N+d!u0SJv7N|nyT6@2<X5M)M>vOZiiJ^Zf9G>FowOa;mXq;n3sq~8BCQnH
z=}*+32&#*{m_-hzy#mSnao6v9jRM(?e$8xPtdaQwbldXdf_CKJ=`Jz2;>ra`H#)&V
z+GZz|(~<h9MOiQ*ZNGx<rre=t)<&1ZC&CPQyvGW}*X8HjL(|L#?Y>nx(>|JG%jMw3
zQ)1`94pmw@J8C$LIkZ=tUmJ(?moiO@CgwJy)uwd|4XrRw**fN$tJb0f7VX{L1fwF~
z0eIfYEXa~?rQ}WUvKUqsjCES4l&lRudDpFFTjNEbig(B%JzsYGDI@=lKZi7*lilOj
zla)P}MxIOLFD}Vz`<Nd%K5zNv=ue;BxCfi$E@G<jA$_<ikF33f4=U*$?y-t3$y60i
zE8F=jwNEN<xm8=7LLPRvSDdjbx$S<qV|X1=>-BDf4FRvkM6rm?w>>SM?&cD=TB~LN
z^V#k!kRV2f$rfiY@5F;oc5ZflGvc+=u%oLPPlpb~1yUBk`n#f*d{y<w{Ir&+GYekQ
zs|jg6pS@6e5yZ|Zr2$4+QJsge=?9t-pQ3AOPgN@9tQZROz@!TN0WyOvnm@iabFIl+
zO+n#4Kn79NH{>XKZA~kf0}j{{d+%dR&FG#UX&qLIvZbGNe!QPY9MIi2B{X-1;4{B0
zy!7^lr_{|ytKRYaeH@ES_Y=!zx&K7NYIs>)XH1AI5xFaM-6)>&N?9;NJ&wME_^y1^
z#ajyk&wYPz!){N@-PCS0LpGpFh?42ro!h%lDCH>&cN@HZ8-3@td)##;;ec$VMPT`I
z)An=9pwh^+=%rXO*Af^-#~1<Uyrd*5bFJXM?V*y69W|0IH|!6?E$A$9+OgynT<w;7
z_fp=4Wb{?ufw8)%7n1>GCg|tYo1=F*tv6;8Q9pR-99wqhRVZ_6Jag9W*TW2EE2(|-
zH|Ca0XvuA%*TNr!Zgns(dcMYE5HwQX)Vl`nd@7Pr6he7S%mf$<-A?8um7NjTn##`f
zsV3hZK*Z0Nf=OxC!%ChMUwR^72f<^<D<#V2vZf`DFRBGuN1vksAMesOjy6ls|1d32
zsT8KAd&&GdOk5C&-@N*<!6qrK)Q0grwE8#Uv56gidI9LDk&E3E<iR07dgH^p%{wRF
zTF<}KkfAS<-ov)nlX5N3TWwvaiEusKk#nhAf4d{BkfBxHDtAVB;EvF|z4ey_A>fIc
zOZ#oRtA5g4Z<$uFc>J5`I_iF($AWE&3GB%$ZJ#b~@LD(eY3_~-n9!UUc+9bh%%VI3
z-=L#r-$&Qa%tkNV$`b9rsCo>)mjr{`Hhy{49oOn%ok}E@V|u~Hr2M>rTEP=ppU>x}
znOpTa)aPXI4*8EKx%bCOHu=kaffxFw&fq(}M4p+lk&Jo2i+^FTg+)rsGES>Bb-qU9
za95Y4NuNTA_!Pdu87ztlXhGrDzIV*XIJ3dg2ct9ZnaUA)ea@z>zqt4HCe?S+ZVl=p
z{&MNGgDN+-%dn%w;5KvN)o0d3ZjExjo~EU9Uhhx0yOV2Hx|*26O1G0YpwZ4bDp}Z3
zo5EWQ2GWznokSh3!0A6n57xLWeM%R+!xnbMiSI@e?ZLKZ=5M)x7ngx&86FA~mD6c`
z?d+?r(_$y`cH~hO%yo59>lhXXkH^vNrpJ#kne?v+r!4T5x|tR>6-Pu<DJcu7{Tc0;
zKv#?r%MEo;xD|LD5QsV)X8S7&C0FtWQF6;y5S<Z6c_jb&+pqz|2y?Lf<QB?oPLg64
z&9xy2RKRF_U(mIHYn4MmbdIMn-)$TJv-E01&XR|S225y1fxRDoC(qG&L%(nAt;cw;
zB+^j510*sKgV^MznS6zm`WQHVSR?8N<8lQ7=lzb%J5AggpLZJ-=N%L$(1$D3W?RNK
zTTUXaeVdm!mj~xNSF{iE{1m9JLg4xoW`GNjTkCU|GTFJ$kW0zFrwm<A38vTQJ?3;W
z&6i>Yvt<4?KDdMNhVAoZw&hCvL6$YadMB1C85bh&Utp5|v5QESPIR!4?)UE&$f%_>
zIM%TVH{TCGafvVa9B#u0)=(+?V?vqrb^&y*Kg@-xDqBAAA($*k*Lu(&j6IH^b570Y
zaoQayJ$LxyU_J-YkWc6h*k9awU+AJ7B|EL!=0vE4sZ6Uj(Xy*PrF(SI0=8xkI57%u
zBt6K_X8ijwqJrpBhd(6;J*OBUU=XI-`8<AQR$)#vf)Qfe+d7ivGMh)ga1>lX;Y)#@
z^R2Tc)DooXNI>(lE)N_19G*^O$Ut?$xOgDFn1K-(#gD)dUSV)G$7%=h1Q1Ra7q9fV
zk(qRm#~VrFtyMb5BXV;~cC&oCrDlBg#4?xAvT9pV&4$!hBH_%*?;`n!3fwTTh{V~M
z{FNQY?-9e3;J%P0YftSkf-i+0VI$1!JtXm+^F-FF0r-_C+E}0${5280>2NRWT&=bZ
z)~8k{ZoM8Y=v{Zmr&sCzga!(-I%t>YikNhFYE(v8k{x1JiXFzedfbf5?MQvKno(LC
z5x>;l#GG7|pWUb2@{Imq8Kb4C_X!ECD-n*hF}2SGi2Tgw^_MovMu6&E{}Ev>f0e%d
z8+}SflIu6YM>oOQ?<ka9NGl@}r!`mKaik)vpDJ-6uFEq*>ykm{!7bBjZLG%=jr0}g
z;C5yw$6cEeAyYaUpIXjC*V0nLXq#Qd#e?K$_;qxbn19*0+j6>))t;xU@p-IVsRy}P
z>H0$!@O)OC#K?Tq9<OlyQFJMvrtj>MeHR{<ZWT=0R^t6FkbYHMeCi&<VSs!qz}|2P
zrRKt-ZeM*R{P<oxaqWE*67tdM2y9FWiP>P#-b+ah=Htkk!kMo|%{Jg@<y-m51!^)U
z=zaCsd_R|%;U%vCi8mc3b4={VUMXAFokniT9wmD%V7Qg2I(WoaztGRlW8q2R2Q`kr
zPchJ4tG23%)8gt-ReyTfPG)%)Q@)e|Vq19`u8r)tWw=trif)?~UzreqnKv5)hu?8h
z4|`N|2wK1e_l4~j%|}3849HANNbgx&I#ZLsejI;w$2mhVqo3lV;SIcLNXKAu!1%q2
zgfEG`KUm(-y0_N9eqlJUl@IjP7;#DwEW!)V!ruxbt~{`*yP>H*8>Ep6J(i|-e79U`
zx~`#R7?J_HL*Zx8efUr#&*L3GFVk^z3`MIug|VsZ?vj0()8;52k<`aia%(mRdrCDP
z{CY2ahB+$BnuOi5ZWE7#T231z!X&D-N0=jdqT~1uA3*Y{dItin*Kvbww1`5Z)Qsd%
zF_4h2a7$YqEqi=#f-1i)IFgUR9JVjlKR8*JiNkv~U!uNDSVk!e_`qTG7+gkKHAlUX
zkiT5uQ0NiAi`b6LI$KObP~clRC;sH((j}_W0V6*~q57Ai)u2PmFi@>6%LCJaQoaT2
zHPasDf-X#f?q?H9k`Ws!8`mxvu58VX-omIBKxfxFOvVk`uiVqIyzevW3;gU$r5;!e
zC_WSpaz1HMF|h_<TNJ+)x7gub`mR5pHH762x5pJ@Z7f8UXSz)as5lW&&c&Up)Le*w
zhR2EN(e4=iyzncW@NV@x?D{rEOSik89<BBm(~i$>kCoY)Vc)EayhD5RoiC95ZheR1
zku@Ru8%ZeWXJw6345rwv6wtA0uXLu>HgM1+kZeM=tqkB$3I`Ru6Kgp9N5Hs#1*B#|
zgVv~mFtF=C(F<kpM&?di!@Eiu<Zs}o`ch=3P%&@k*K-`E<Ky`ROqH8=b+?u}6zA&u
zx2I0~x4RCI7M`KdJChvzysV5?Mu*if`cArjA+o=(4_c*Ht{zg;udW8pU`JgPXwAQe
zb{k)#ejm!no7v-x96ua=L|JHXMTtFyUzRRd*E}XM$F8pAr(k^YP7zN^^}G_V&;zP6
z;#&H)&*~EiRKjugbC^Y93B6xySb<xP29Gx?RmDl{H8DO|Q1|fE=-%N|ZTCWU_ssXk
z;Dv5Q)sK9#OCp`yW8g)M`|H(jI`sH#{W#x`ojnQn&lv{!jUb0|1eDWV>sB)^2yo4*
zkSG{>6N->(Yxz~<Wd$P*-FA(mwf(a%ZR<=sK<F;Go$4C-sC0qW*V>2oJpYA;vhQJB
zW*PVPUpe7BTyR!*d_~{g1dyX&63(buy~4Tm)(I6+agLm$!NlTkWy;ze&zm31XtE2Q
z4)5$+%N&*{dVS)Ct4eT%+$g7@&Fdb4fGM*ef_Z;dSN3-;J4M4toPG_ufWBT5OnecS
z<8}lj@{cSA2ds9W{5!Z1FU5gdnv`0(E<p(?`EMa_8{)!Y<o72j(-XIE7q!Vr)kV#_
z={QT%XQZ``U&#rS=GYTiV)9wMIU~k)<d?(&i<h#EI_n)*WLuhl6AYb2y)X%s6m<Yq
zoz%*#i{WcBTyDD0wxK25(bVDcl&+o%=IA!A^{<@kz3fOi7n~Ull_-9n$IP0#ujo6J
z#+0JY+BNoN6w)&9V!3<Ef<jH~RQ2NGD_XE8Z&HUPd5nF(t%HGDi%Jd;5rPcU$kJ^u
zg(C~F_%8}OhSgi-vh&DAIYL|)kb!^O{wu_l=ZDe6ur~5mh$Swskjd6_CUVSA!xosG
zouB{XoM@Am`8+Pk4WRJ*8VZ${3+5ezhu>-?K}Jw>EZi!)Q<-)LAM>T`{6K3TH*j_4
ztZnhycDFKkk4(0B`vi;mkW1P7jI~tNK<npZB6E)T1Y+yU1ZGnXAtx!sh;X9Z)RdVz
z%pke2qSzqe15R>qjVwJWZd)NCqGRd;E%g*mKzPo+xHM04hNzPL0)J?IO}u!u{-)a5
z?Ch_30%lRO8OeCE0J<NfW~seu4qLSX&yciLj|ax%ko^?-0oBk#e@c;7`r1g{IMA6(
zPZn&8@7c5ewN$pS{}b-@bL=Fu@Zl7louVxrn+oQb+W)qhWi~)bB<lUmY*y009@<X3
z^~Q{oW6)Ul>B$h=&XV0&lmKJjN%hm!<`$1UWnxgB{ULzPn4~s%hmucQ8!`*0sp;F5
ztO=+e6dnvi)a^Q$)94FuGEToLcdTUMIC_n3yEqwdLfZV?zJY~f*;Q*D84Nu83gXc=
z$wUcjzwN`VB#dKPq1%>)5Pl?p&Z@OjFlL9D!mVn8|AA=IEz`gV&eqpA5;-1m2g0mk
zD(>hVnGn4dTrfA!%;In!mVR<%;G>CHH2~=&FjE3m;%&I*(K(2_vlmk5kWamo25-vj
zjeP$49PoROKKlN~&L^Kq+=bXffNSe<3XgW?<lL(Sc^*(7F7#`;WDZ|F%NNy%EFM$W
zRmI=?M~Moc2d%B|>CvCXy|mJNW}QiVSx%2WHFjbaxzupgdfVbNGHcFezWmJvuuwqT
zEC+F##-8o8dBC7#GA2Ub545{waw$!Jr>P+vobXeY{qf13QbslM0>72%)X79=khnlA
z=jKaZC~z%0=ZEWZ?dxzRdG>i%whx<5#%!>6iS~9M20=bzh)A>GVx^W|g5}47c}j&B
z+y*}IgRLfe;BMd!lxq%U!HGOlXO-!TDt-y)<sb=?ZmU*x@40#!6tc=ynwBk4EK3Q3
zas;S^Na#KOuiFup)~~+hMbD<RmzIvNk;MTr!cnqV-BPw}4yWZJ%=%F*`dV(TtSZeU
z9#4+O%7TTf2{d~hRigkk#oTc>9HvAio&lorQqgoSb(j&N^49Rp2R7p5^Z8`&zOsUw
z#1!gsfRB5#;7Q*GL&VW5f{qPU?#3*mHh8LaV@LnnCcOr<5!HGX^ULxZ0-o_2;hVNk
zM8R#L@)_3ws$xe&=^L53-<6X)&1>%>QP@)SGvZB1%(1KRAz=!qz~5Jt55Kiq8gW@f
zJ|B+*f|#(e98vw>!qLx85Jcyk=$rVd8KX1?-V#XfdA!cUt+9&ipNvkvljp9x_tEt$
zXX6D7hrhzD6S(l&WYV+swvXdoHWz!05GwA<>0S|0aD-RqtR427V;i5SjLrsNdk>mu
zH|4@Xa2m-zfpiz&0MItc=FlUDKRUH^X6JM`mOn@JnJDT|mLU1Ir)U&&?@o$*dh$+T
zB{yu_+iyzjC5<>z?xCv`IuidDJH|+$Ewi+eSit8PAE+l_aiWn)Fn$rLLEc<<?+tGh
z`tTMx00@q~mC|qzI$qeVlvxT_F*@7gdMlJi+9i=N5loO7ewNH4S;i;C%JCmn0ag7f
zfWBfBs<%|a#mTf9gCk;Z{}s6A8;fxq2%AK`jhtcFe5flE63z$jmeXUXKsX<@z02H+
zi;DV`bs`0$J@x575tM7FKT(A}B{2HS<^)Z+*V~`MquHM<<0F)+2*-Uc{h-qCGlG{-
zPi=bijMF+dfH^pWBqp`IL^a+w2kJd|YQ54mE=;~T$U#w$$2)?~?!~4Q*{IrSxG`Np
z*J;WJrMzdrL2s6e;bmhA?VZwge+%94I%2=vI)a}dEc7>{qLo4NBf<eic4yIyn!^Ef
ze61ydL&sj5q$slR9D;?tVj|{oFA2^jDSWdl;tD=^_~Jphe1bBYMTrJE{N?FI0w+I3
zDI;ge1i!Lyx<*$cJeOst7B79(Sm(<ZC6}sr?B%ufj~0Gg)94k}=X5Vf>k<J^8ipgA
zR9#uVn*1z5G*}lGd-bPNQ-Fe<meK@L%~$o4FVM`M^Avf*WlAeUbnYK-)TjuzRc(F&
zCA$n=cCLxS1kpe7nx(O+5q_8Rs8u}XH`tZhQR&%V1gR@6@~2CwwQBQKXeo;S)-7vP
zI}A@yzDM-u_g6PM>Q8l(Ew~R|y4fK}B}xo2{qiF>(o)B1OZtBRd$g1NwTJ$bvhG%}
zz7(U3Vg-O~t&;QAP^8Gsa3QvJewV@5xTpTT3!6Oi!!v&bio!-e-uGa~Gc@n6W-N`!
z4Q#Q5xi^kadA?jT{@%E}f$OVlzZxd)bG-OmW-zBXQ{wvv`^4cmQyS`X7HH8ez+5a=
zm<dlEo0=R{sKteZ*u9Y4>r3bOPCFvgp$PjTBDw5Q)$+KIe*%4zeYly;=Xaf9sL;pF
zY;F>;w4w6Z)S)_gB>)HIl1M(;m41?6E(AK6v%Oel_6qTbPsSP@uEG8!kGxj*ZU1~#
zg&YrKXl|`}+%x!(y|LEM3ykRPUT_}fq+7m!I#KbW?e4qj+b2vXWS#`vE|t%?^tjpP
zv~zzpg?tFpAvvb-5BiP&d_Ruhe+lGcoZxlNUV-3(cs!;LCfz6CDn=(gQtxO{x!^D*
z)*vnBlkp9?LcwLNw_2%=e-;k}`4hy|7n9t@lg<NymI~~P%q4^3ezokFScYKevUt@&
z-e%m<NG`*xxn^_iBrQXVekQyLj!`nw*cqicjjU)z1Po6jE7#cI6bc2*Q(24L*$2;d
zbkhF4-CC2mGv@n;{`%n})5C~=RHU=AcSr3+{KIt1gjbDX>?53UVeX34QH7Bgz(V{K
z)x6AI7OL@34Y17Hf(@P92F!ED>Mb9>ISe|GhVnK4tSkCKAvr)8IwyrCL=G|bWseY~
zmfI{SSI41J^DAmVUmR8Ok^(>>o&r0mX=nPTNB?B+wX9xi>w8*_hjI|v{)(k6h@lY-
zrG!sKw#LrC4a_Cz4nw-#cJHA&RJNsUg{Sn_I+raWSM9Qx*^I^O?!;A7p7(Gc%ewWS
zq}2k8_tCDQC$$yN0u)y0Y*0lD?N>x;f{*;k$py|yM5do0%dD0bb}JMIgrO?w2bKR(
zfoi}MtBrO<p8j&tFZQi;@9RlKHbc<O9{HGJlm>hZa6?LF652CKe(3jF+!=6}Z<MAS
zkNlL`TlGv<d-K$Lm(>uLaNw7j+grGVk>OHae=mrRp>H=iIG6Vv#sBOwLHZhzfEsx7
z_8ncAEW~HoZuOPzYXCJae9Eq8sh&Ix-~1WM=Xi&7Iy7HV7C`g4)SrB>@+^#nclYHB
zh=7w@$M>8fK=`q@k=z7wy662u;m>A9MHSk;66PRSRdGI$a)}J2DE&J3G?J843KC<_
zVA1MqU#Kf^v%&j0QS6x4sDJuBvu$<b#C6?GyE`)MI<ne04hR|n+1OP=@w@mIr33Ac
z`TXS!YBcvuJ!i9*r+laHhCQ_hRcP$<!@^zybNPGm&*)tx3$lD=X53g3u_tW#GOQd$
z9LrNn*3${FWkFC5BSkF|>Q?AUBL?PNuaVj%7Rz3Ip2k}5<+Y#Kf%oiEtnIxg>K0R%
zzlS9-imfD*>&miCYl)bhhLt`XAkQAtq~`RG<g>DpJkj2p+w+bzqGbC>kfv5!O;<@c
z+8o760@hEC-QeWA`o{0@4XhP(Wg7L%*k63^Vx29;^^yI>p}AD9cB!ZBCupXf9s0F5
zMW!!u9;se9b;8eFA4|6_-uG%I1e!7lfu^vg6O{GO`roYjv^+)Xxw<GrpF9p-)Y;EU
zgNj)K&k#qp)bC6LJbSmgp$4J(m)N*OfzzWvBDpb5T%*XX^J4nErXMI|ct)^NwD6lJ
z<=4F>N(Rgl<HTC#bP5Lf3q_9*9GB#P6*5TTDjy)dgQbhypM!~dK+9>FEGrEMiQBsY
zuv!X<0W1fb-XPmzFZ&8pE6({n3AFISGqR85L|Vi9CJ3*F!jvkGF8w7it#VCw64%>C
z%^~J4TyG+le^yF9sG&dr6&$*A*>?STI{Cnc<WYR+p53%6f=HsL$;BO6Ll@{d2QI0u
zNHGqa6L_LKAC<$oXZbcq%MO47VrW_M^5%B`_1U(Fo`bc%I`SZ7LMdXhVhIj<XC=qz
zBjQY|NHnktCT1Nj8jDy$ssopbn`@MI16gj*6$FX(K#kdhh@NuGe<ZQ>Vj+-yPHlZ#
z`QlY2jSo>3p;hdt+IUj9`{@$Z%Sc}2J$QCEaKdZCX;qhd>&=*J5v<O?PHFpbZy7rL
z|2eq-^B!qu;UWYtJr=>IwjURR?@ENqy2-<aa=Kdm@KJM^U_K&>VxpVT8RcV=Q}Z1j
z%0)mcYj<B>x)?m7LczkA3BNMjDnsYpSc#wRT)WUMR=UI$h6fxG*r{uJ>zth|P;MlE
z;Kllf3_@vKK^sFcj93selX`qSiidWfi#6<2JaIjzUruK6QW@uyT)<wqy@jAjf5NeT
zwqIGeGN9_J1ftJ<T(ElF-ed$Vn#$V!I-T=}bl^X_?8kjIvO*f3<HfB=x%e*PMi9qC
z*SM!`^|bEXh707<(j5I9KnUllj<lKK^m1jQIysfqE&dj+GA>lwg&*PS_o5iVpBld1
zQnQ=&7pnS>AZ;UD+s9{9B$5Bq_aM@j_7j1fB-?^<w=B@}d-Y+!egb*tklFr$<hP$r
zjCW52k8LT%ZYkC}B3t)#q5F?BO;++rsDtAN!7M8Q!n+z_LU|dFsdoj@IoALodUf<6
ziDIVqh@wvAlhm#LWUObIB0oCx*251sHmBA0nInnsZEFsBJ9O+D?*eY@(U@ZJR)SWl
z;($>DT`aGErHdu4l*Vz_pq*3lvH2GCsfuKDw#NR#8=@sg_<uiQ>RHiYlc8+F8KxO}
zmU;;}Bdxt_c&Yc<Y*v_}+Ad`^E%3sNLoGLVppik7PoXjJD*e%tOga+v9c7MHz}yOT
z0>O;}8WXO|&o?1fp1}R0@{)fPR@5{4z=DRWh4Ymt-la+0jy0>?fBG0bg;}YVdeku@
zw?K*y(bW3@Yp{Te0SO`}zf9cG?y^hP$-=DswC)QH+7~;w7-mm?DRtv?9C4lcFnnq~
zrUp2Dy;B`}=`GE!aele(sLS|AQBV@-NfxPUK|2EXKbt*^x^6_}bIIK1JVdG|Xa+BZ
zxw!TxCKn3C#j_|yQOyei8y}d+e!v-NtH0|9<t%{inxi2G`}NW?fRMCs3GDR0_~+Yz
zx5B81kn|CWp$!w|*>(8#VP}4osJgS1jMyciY4<riK~9n#Qm8t!VCy>3WVvF$t+`SK
zfQY&e%xkhS!DY9`)I;w8&7*>hQ#s6p&CLBlQ;%s4z@f@~34Fvz_j+jU(p?6BFLD<3
z{G@YzzVN-`KO%<|=5^?UwVY}VtIC++*BM4h-Th2YLOw;W`5L|AsQxKNKaM?fiGXmX
zAzYpTP^btFpzR|I5uKpbi4U$Mze@@x#lW(hT6Fs*``X!mZWDE4iRAbAhhLD*S=X8Z
zF!5<7@CpK`9VW>e&>c9wI@J<!s(F03ir3DQ?glnR^tQ;9p4t~%H`43z9wss%^FW*t
zk9UPB+ukU-DZ|jZbT#1fsnrd^Cu(u2F?WY?94bDz@NfC0@NkmkMd*C)gul0fVgV4)
zGT~>H9_=}ZHB*eD1w9qO(i&1`2%E$)T+zViDaXS4g`^^Iq!|iFE76TnUWTOCQ`E$U
z+LK}*mx_VX>n(5y>K{LXhLj>XqtcvQ{y%r9>^K!|OG18N;mi8<#phLUdc}U_XDG0`
z<1sNVhLL+@d>=8e0;iEu{dyR;n|`ozt5`qEUG>)FKYCZ5-W5E|kalf4XS4sk+VQF7
zQ6ouQyhPY3TozTY=hZ9b{(Yp2Oz9@eK1e3&&&$9Md`Om^vAA+2tqh0Fiq=QhdBYTV
z7)^e7?-*T$_|@o{(RfeDX*fp$UHV3}l4;sKXup0+3a)_9SgE%E8mdfk7bgiizs}8#
zAadd*I1msZd%`pLXx74@ip=)dxaUxNE9Y|AP)Idh5ZCI3|J(*c=26rD`;kN#L3j;A
z-noA|&v|00$HYh@FXNuzbJb1-*SS1SH+5EoZA`_r2NBo)Ud;{ET~i0Y0m%8(Ny;NM
z6Q$*SWCl<8KDRFSOv$|r;V3%vJan`79|6P%Vd4|5Ij_Udc#YL&#k!Is$C(-VWVu0o
z@6R0h177@70t!rk_je58$QrxG_HfsS)*}TdVh}^(BR`gblu-W!F7zDX@^WUOWJ+Q_
zI*~N*bwslFAR)IY(0Xko^c!C{C`LN<J$(?tNCpF$Ie6^)(ZeW0AyLdTD3J2^U%vfy
zPxz@#)&n3aXr@i=&xj)aIO=P-z9$v9Tl+*0+qQiC<vRNFotdU+W?tc)frH}tDWCUu
zhP$)y_<K!V4gbf+f`L;@HC;eFCD^Rnc8~}=L8;$6Hf=(TCf!9*H#?y}$3i~zkr=zX
zo=ZOJIz%GOY;*sNU^ijak1HOdyy&lBBUPVAQ2xz!%K6c^U?0Q>8yo>rUb+$^znjWs
zek*<fZQFExMVns^N}$MrQRk4lG3s2^_bzKt^K$8p>>sN?`5~%!GRyeYPx3DXjQ1KA
z!jZi_%fwUm0~yu%JI-$X>q5_>2D62N|3n%^TDHjEwv5Y+bUgeC&!w0YAMYjM=fR^v
znqN>N;>_jN_bvq~TZfKUFT?HK$6QC-Qp{q+TUuqbeXS%83QT3TcTHwTjUE3F$B=(7
z)^r0EXpc0=lp2UD!OQq18Mo?q{#;rq6ZZ?7{V6)rAwWl2$@F5=Vxia*w)P~{S=;A3
z6dS&sai&ojWbY2@(Wg(>=uMyWYP7pb^H9)>#PzrK!!_w=TlC^q{<t~6+WhHfEx!{t
zZ%Tzkt)~|<w|;r3GKK|Nd`2#kAN*_2fd&y<dC5S7b;kWcI%0Xs&UGIcHm4xgTCy-P
zapud&G2C!_()+We_`$uZsv`rs;u%9P_s0Y!8Pn3Ga&Fe)nfvg+ThQ;hu;sWVWu|01
zu-f<@RB+W?^5O!XzI)EjO*7s)>L|T?Eli0hNw{P{tyE($%!i?SU(**)If~|+#?8^W
zLL_Lp8u|(5aBWuUx{7o-|Ms~vT!mS1fyny5EO_xxrh)vO<`TSt_#XPjrUzj~&vS2`
z|Brm*i>YfDW6PE1?3d8hoyM8k%wl?{Pij3xR=KUF`Z9KWlV9xinOnzB|58Ts_2KaJ
zf4m9uqV2pJ5UUJp#A~5*X02M6VL&hC&H0`3f8RBpvfH#f^2Q(idbOncZq>-w`~qFe
zyz#)PeP2D*N%7y_c*-NB3M~?wXWj>HdZAB=BsE&07*(IQx`n!Y{#o;X?|!K{Mkb))
zNqF1%){k7%?wS2-(Y)!awSQ4^v<$t5{4w3a8*lr5m~zdyTbl#Oqg5(24W`d%>g%c7
znGO~<C{VnYn9icw4PFA^X}m~0fb7km=P%c#@OYnJvE7&7A9xRm3HGW@|1)JVcHO5?
z@q{k)q?xx?@t;2;A9#+?ep|tHEdEj0Kk+O%X9w0~&m6yBv1`?Mr=DrPc~R`l*9vBn
z?9FcZlF#3$lrT{|VVquLp0{Qi-k9>)j{SDGqb@+O<&F6y5^&c@c>2$q<0bsXsCFn~
zuhM@r^8s72ZUb`Lqp>kA@};xImorq(Je5~WH{RjGCy@((3l~0^Rw}JywESl+!_*t3
zFY`#B{)_qRP$)2aLD<?srNSHCOKzT-p<FyE&&Wn->HA-v@or^(wCpw#V~f7?P+z<2
zMctb5hBUMZt8JdAzkP0W(^(?D3so-56VO+M5*cZ&ysmUi9~0a4@%Ib&K;OI_Oh)@$
zvup2_loh{K@8oe%w$PEwzgsYUCf1e8qn5NV<k+q5T1u2A-OC0+yA`HE*Q0LEct<r+
zY6Z&Me<-JX%bwj{DbvqlimHdnMuIi>Z|?gIuaFG6gt?!Y3GP9Ff@<B2^+yGe7qGmW
zZz}lry<)g4XiRe|JAV&9k{pwFFBRa_UwRT(<&BWdF!o?fH7{J3Lv{3B$QO(64A_cf
zFUGWK=l%YF?;_0n-l;WGc0&=Z>b-1d#y;c47DZ<lnpI14E|iWyJ@?gGS}9A=xm6ac
zQ0cZZPaS@QLJ2wyruLT~La^=7={ln9CvnfemHiA+n*4qIWEoI>@7J4l-CaLpvnf29
zG;r3wX-ZPVrZ1WJGWVeAQ}!j7jMp!0%b+|inNQ@A8Jj%Q=Kp?>43e<kuU$SaQTTmT
zrv8QcoK>>F{rHj5v!@PNjFvnpV3n(0;jcTyw3foNda~JI!?xdF|6jU^Gxu$J;Tr|~
zD;Hireogt|@AsJAsY?2NdCvP0fepqQo^YY^FEq-7V6prU{jmSIwUvm_ykEz?lQ{aX
z74KdI>#=yZ5+~y|<Ti;K+jAv6Z~DcUNiy{Dna0}&5-oZsKRHaty#K{PMguF~VPYt;
zi0xm-Z8`Jkpyt&tMIN~!x(N?D@|-R6t62Ug*5W@soaYX$&X~W`?nj$_X0Bet;L|m<
z_)UZ#=JB35fwr#elm(hACFogNg{FXulPlJc?AD|hM8NUyKKt2983`m|vDaH?rhaE$
z$*Fr~qR11sa^u@Y0r&s4s~YsEy9m<$OlQ7&>NyoAzchaXB)6fNeI?Cn>p7qn0H3b;
zPeM3z9ew)bLETQOTr)db=$0S&)FjCNtZ{9Ai0)U_v7hZfE5$<;6I|(+@GJ_Lgl0k`
zrr(^p8u2U#HK0BX0{MPHg5L~ZVCM6tm$Qs*QXpU6)0w&YsCHUlJ`oyJI&+cf^|=Z}
z`%is`DJuk@^XT2Qhq`!BI$zLCGdJ^_rsXply*w*7CdK#L&3q~9a}He3pT%kmrv<-x
z?X+n9*Y@o{?dX4{)|9lRmw0$I8%@SR?+1qLEVG|~`M)H8k|_{&Ibu$U)MNV5=EoXk
z&*w}_<MPw{&9+Xh**_5%r)^k3#s&L2@O+ogOf~W!PLSG+Z}EN|u4m9a^LGm;L~Xt#
zL4)rduAaE5_U#mobD4%M9n<tS)6Y?6ncip~7_kOvQ*qqY%H8@NzlG&L9wU(>sxxeX
zTywvp`i_}eFgl2Wk@Tg%N<xl&UFB09yRCgyS>Uw8ro2xfV!BmZfEZ1dHBMph8LaQW
ztKj&{=bE{Q-wTdfjKFhM@9yOHyJ!9`Vvr>w9&1;)4;WRhi#1+$yrT}D%PU$b)~5lA
zQG2GJ9W>{l*wkt$y^PRzu18`O?jiOs-##$IVfg*HnI<Jxm?(AasnVRke?^B)a=ho_
z&leB7A5`UT8sui-bQ@y+zo78omBwBb@Ta_)PR@p@(z8vk!*?}I<(<pd9JQ2DxEXxc
z?mzG6j0Fb4pO9+~N;edlxwmD%Xp6_}4oeT$kHk1M#gCnTdOpr$)&C^a{ZBW^L}OC1
zt_wN#FMSZp%+RFfu%DhK`q30Vq1GD{wR7s#qST?lt*-@?mzpz%_ErXJ%*em~=e0&{
zq{h3y$W5(!^{<lh<wn@Z=PTu8c%~FwX!!Jd{-0H&#p8z$OAm~WOe&t6T|4-;{3_?D
zl^Elr@X${=l&j7aY4J@-5|ReU(iTg$%YTCg_@-s|-?^WwChzAR>zDsBoW49ZuSLff
zU)ixkCjEY7-o)~sL8ASt|9`R^lEW<%7x>qzswVKZjJ7OK88;vMpTtW4FKr3v(lz-t
zd8lF2F0uGcCV^?Jbh$hLfZfa1cgCY0-bh&K+f1)vrn;OyE&&9W-me8M^7LoQNz_lc
z7^b6Q6|<+**qKMa-IaAj$`>y!Q4vH5MrT?3(*O9qe~W{^IuQctJNjY&5`tI}n~I~@
zL!(f`e!Dtm=|NBN!+Dw`nUl?KOw;<4=}f{j>>njX1klg6!sjLQ=gWkqzcJ#!b>m-4
z(E#uc%fGwv+P^&;S>vXA6Bi!b2QB93+}YI^9JCk1Q*HLzvL$gEwa&8NGo7)fUPGv-
zy6oW3>q|DYY?9zxJ~LVS*Yo~t`DQ4P9VQ=}I<o&MIGEIYTj~6APwteK`F6&y)4TxB
ztiAn`x6EKVKd@bE|L?7nf}f*uRtrzmvwGSe^1oT&p9e0A&Ybj(tvUMt<s0iV-sOJr
zd31L5Wd|*tc&!HYh16iy={%#|*7oTd4Yi9f|3T|Y-tUF*Z}PpL|Iau7OSJzi*B8ob
z;dKeGh%GZU1=SohU&pY<{xjyRn#q@Y7Xn2|PNgiMx8-Ar4xHAk_h_`{O)DN(qIeoZ
z+ttnB2Bck?jt~B;6;sV8-GIuj2R*;{YwEQqTJ(OV%uO3sp?~BKJdk){e>p90qnxm5
zAl@{P@jeJ0$l@sdIMt;57?y{^sFHGVP4oM&o;dlcrrRg=`Ow||3p0P06`9*x$vzsi
zzI;;(eRH``Iq)s=ps4#;-v7hfcgJJ>wtwe}GO`jP8ChkAh>T0JiG=Ktky*%=xEhoa
z%1UJ`dqp9nX%<PDMWJLwBAe$pFPFaYz3=b+{C@ZIdj9D}Tt1)kJdg93@8f;6MVZ(#
z(SY=bkjDS3E5G*7p~?wJkM8-T`@eS^ebL;4S>ES&aG~!7wXWDq#}lWQ#^)I|{y%8d
z<$ZJCAktwosO<iQ5#H~UYj}D*<^Pih`;QUiDjN`RB|sI*4SL`Va%D(pGagg}4rC|l
zM9MZ80Y1e(U<J)H_FY>RhExZ;&i7<1vB@KBYs4tY4^@XV&8^g-yZHu?b1+I#+qk!j
zhpsuMc-7u<jf88Lg#+S@!-e+Sz^0@t{W}{_4P^-HE6rEhRe9j5VX5G%t&LoAAr+cz
zmF+tyyhfu&F+ko9-HCXGr0zvXQ`&8}4Onv{aRL-qLFC6ZLvbuZ&T!L!aM36M)A^p5
zphD?1g;p!7Es)LSfQHux*WeX$#oqf7U`m9`9!(MDZ*S<Uw$}7x`AH#PNuHV$o-3sD
z{Vb$wvzurZ*QSY!QAlcHI7{4u5a<x+x3Kx|Bm7lLKIj8KPj%bwgfe(Njax1T6MjIr
zhry7k5^vNwx^GS8hFFvxGF6*4DbSIo39LM}eq$2!+x~G<_P)4hLT9CUDi^FAsTg;b
z8*E`Jw*PmK6Obd)t2P*Va`5puqMZNyJM7K6kSt_=ur2uY8dZihBBCg5*jk58`b1Ly
zK^Ss%GaT>N91l%xX|)xjEd34hS#QVOiot>s>)OyOX$Dp2$RPFc<aYr>{5EVjrKAn>
z?Y)p~5|v<Im!7>zvwQl<f~L4G=O4?1Y_1q&D^J+%7QcXXToJ=GWmJ0j^Oqv%;6aDj
z!DR+ws$Xm_lG9(r<?>lq_c%)!c7IAMy4@I}>f-l4!vsVtK?~{O#m+XpcCB&F-POyP
zfDq~(^()UUo)~HNUi09Li`B`n@}{bZvDU2enSzqO27aOI|1O<H?7kbrc@ujPh4!S}
zJJxI>sJpU|V8M&YYqgz@K{e)k(IHV9dml(uUV$|E*-M^l!UtkeA=lcA+AmJ$BNXp-
zTBn*Je#lh*a7gl{B%}_+XP&p5S$S`ES=joP`}*!qq@?}L%zZhrYxc2O42sX<BSymg
zgjjK9lW~q@rH6A+I3u11HJZL_cVgBUSnO$1mTn0Ad~q!C)*@3WwuHzi=FW9#0@pzl
zoKW)Zh(%&`6kl}Kn6KGKdoA=_=T~o<o&Nm)R*P!S<BK&|3er{+z&m~9Ogj7Lk7?{Z
zam5>@uCf+-opHFRWPHbHnJf)bfCw*>COtw<xSq^LaOIR)WnG%ifdCq9CvjP;J8~cZ
zOSXD{yLe5(8BSliP+!?to@^Tw+;^X}CX~XPw_|`p{0i#ulY@#SB3b}Q%>Metu(#V~
z->Recv!}<M7xv*x2psLn^z@$m1n8zsD36X$9!Jg3p9Z2(Pd69<Cp5xGNKG*-pm*HP
z?b29;-jQ{u%i5y2AO0OHWm+}<IDDJ|qwAgaAu;SSna#A~mzTi=)vR;58W<<HLjywC
zYl=Pu{6d`y+#wAs+E$dqwV6cwWq+HTdaj@~l?AWCj#AeZjWL1}?0+_({Rf4?Fj(B`
zI=rSDT~A(mSU`hoBknG2oBbTedj<TFHMe3=zt0Guq<sSrbB8sIZuE7KzzZ1kvqKxf
z#Gw5SV6|DH$*6y+O^An+NRS`?M4@dU3#l#r!xz^?tk@fnQAxJCF}{{Xi1{#x&NIrJ
zo9HsiL1f>l(GzNZtvayN%*V6wc1)LU|M&^*mlXMdQO2{eOKsAHLlYGD@S5w=iVkmF
zMJqzd3C9vz4W=s@PUvs#41(Md@8JIKBb8orLR;YCG&Wn>ljEdO9}by0p+Og`PYi%b
z^3;Q9{YkGen~O=@<E}#vwE;P7yCM|ts2EMXTD`yXjc?^w1V5<_l*eTJYaFpVd=W>w
z=JsNf9s%@=+3Qs1wl&uz<6_Y8bK~9F=dllfhNDPU&Y0kkLqgh*Jp@7DpwWXCCSo}t
z40mCpXhYQZ2)qjl`6JMn{RP7wHwa}9Er~ZCgo~1q06X#SzMifV987b3UtZgq*U^D$
zDK(|W!_KvLT~z~4#<-}SyLmsr0>Z=$j0pzVgR$0~Jp@pla_<3asCrJS_%W6AVA6Im
zJI(pKE7c)(Uf1djUhDcDZ0UF1I%tQV^EC+G{$HwF*r^At?1O8BSCp$)davU7<UPbg
zTmT{G^?66Ih5EGU-G21d)WFEqpL1nEq3{BD3hu<(tu$LKe4W*P-YO}7nDX;imAoDN
z!jE+DHwt8?aaGS_pOL{fxhmd$!UW2Y!+K+i)SCT@AR}UHY<eyNz4(6|pC~81l9*yL
z2k{UOypp8fYTjQT)fLDug(78!D@R&&nxlCGCIF>1SK8lBk8NA^vjKw(x*1_`@4|t$
zf+xP_(ar1tL$GgWKNG(i<KqA7T{_S&9*S9t#C`&R!L=6MOt7d-TVVnCeFFJycESM#
z#xC<SKl{wtEWTG;u#}zLV(yjf-Z8?9O`;}vU~HS-C9gaadm3K0xw|jh`x|@_QB}yi
zVr2-cPWl1K#^!<_TqSXTx|#rrmR<{rCaBIKK``!>D%eHz*C8Rf*yIGBpaBrUpOtd|
zVCVp2%d++w)U)dl*sOL-947!<{@mujUO^AFhEQgAYrR;jXD(v$==d*O55v+??57t0
z(L2P9J}(|t4R}~K;i{SAsP`3mYMaFu`Pw28+6?s}oqD|JB>-eQi2LgWg6Vo2?%ODJ
z23VEK%D0X^_XRxq>+I}%JAiRL98p;AeDGVe-gkvLK79Ca)D>S*pELu#Rkec;hpi-7
zSHGq$>*Cxs3o5>fl);0>4THP?{C`m%kSDzC4Y7kfZT8T|i#U#)82Ut0_oL2HXw<JI
zakH1VK}}bupwrF`=mTGF)ev}-&@C=65B*T6OZqq{@E@oX0k!7MNwk@seu3a%s3-6A
zGNM}=1O_-R>~p{iG6VwWrn_e)pbjI6!|ekRwbNdo@!syoh$a!Cl^5rxoO(WtegzQM
zXT-b-OnF~^RYC*t4`|Q*GGGG8KF&Sa!~C<i1f;y~zmga(&$fQlIXb$#eFwTL8IP*Z
zewC#auAFjn9Gw}c^|Vg6Wl;Z@LE)c!@%zvnA!H2+-ZxaW4*hTi6sHio1X*zs*boY;
zav|jQV84*Sw3Wzq;5o2{UNWVwxfjA`_aapDIR&Cn)~Fvlr+WeeN8t1k%Vv?FnRVDc
z1k?`&`ZifFfLT{8pgs*DnnUKt?}b4F@7}}EV^#{pR?_E$JUY!LM~$6w=}}{^Gk~;6
zej;kDy=P{|&&~6xbwiWJ>42el$=qLFxivu4MNHbNZP|00Ej)HvinC;s>DgAly@OXM
z?*7gDUiF24uNl5*p1{b(^-LVy@aIJ%o**{m{o>Zn4!ZyKMnS0&X?*7q69YmQ9tYOh
ztAHB3@668#P;oSM;5c;h3*C|&X8I1c9f4>O2~looCbVH%a4GCSO|cc$>cD3Kou<bT
zwIcM>?jDOvaPEd?ze3<a@;YEMIrzp6I6A#F^2M!yg{XlfcJQ|Rt6yGuHTYt8Mt-$6
z*u@I4ivgPzMu}1Dpuh9--g~3(YO18xqQU=MAJUBEcx`g-ZUdn>NZ5xxq@{;7iS#3h
znKi+)e6mUNG^)EKeIr2~q1&OM;Kd?(zdjJoaQEeV$4o7EiXH~KFo{n4)ES`u%F^}E
z#8Syl3ARGJ-=-I+{Sf+Qo4W20r_Q{8SQQ=n{ze#}0$&2CKGs_`e)Ls#EuQ7DRiAwn
zOZ8aVvC{Cz6^i#I;e5Q!Bbb3D*8F4pNkByheEG)qFGUo;F036|jffbQ`Onqh7?7Sm
zzA^6+l$Z8XZ^K5eB<F^>psSPZ5cEBz-n&wPXzLV%w`3m8o;M$b9vLPeT{u-vUyOjg
z!176t8U~^bjQ3nWdkvapP{Z#@-vhi+<oyKz`9!>DZzE-AO?GEm13Www_0j{!*=>qu
z5!?DG)YIGtn{o{XPvvt4&fgMyy`S^&{!bTvz3Og4?u8Pg*)}w~$h`~rLQK1%Vb0eV
zF;BFEI3o^U1e?TdW<~TMja{*J{y+JO@X1O~`k6YZ@EZP8O{%$+kEK!LeGB;Zn{q5M
zE-QBB{I@?X*7}`GQbu;oF!g)(s&Bkv5X<58M=-~4U@6G=Qy2Fp4U3{r+86+>CF><1
zkw~c-eFU5=eL$CbHaxuHDG}1|h#2Gtpyj{rI{{QRS5O<od)TImNl;cV5mPMCKiHZ3
zeGSvau>iL~VkzKh-Enk49Lx&j@5hBy4}I-B@HlXO^r1D}4b)TNefS!1sfe&b$4W(U
z-ftVgeG2Y?8!#YzO^@!(u=Up{c_xRnoa!;lM94tp-`!sODHOio2kZHrov(6WHQY*!
z->^>n|J)Eqyo*{g^q4@cO86N&3bLv~1Tne0!N!NC<~1Qh8ljH^zBxhY<Bw&P90Lww
zVW#$^8j|<pK?FqqjbnXqKNt~HnEIxYQlPiwBzry%+wf)QEoYFLacD|<fsoOt2_8_m
zg<XXgw<7X3)0=muft{TlSkB)|Wp>Rwf+K9}|Hu=IK`+#*31%+-NcJTN_E$)z9ToG|
zMf<Bx%}N+I3r8G&xsYa+3&1|EN8JQ1;rFlU@DsINF1c<%jT?dEw()i@#xrRcYg>W|
z1F{JJ$<jX5-G+!oW##fdZiY@U*%%|cn?SgwY<`pqT7(c$>_iK>OSkJY${)aLA_7yx
zZahR>C>2-fPjU?iBjQSE5+B(40~)+v0&m6b*!RAv%s?s~d+gi%#|IBvCI(PQhZ~|;
zm>H?*D{+`zW;3`S!yj%Wyh*6Ua^Dp_zyg^Oq937+wyTiJ->dlJ3#>5`>o|kJ=s$xE
zB1}4oDR0cL!dS>wfmlD?J<N%=R6MjAOcf)LSo?+Fny-;x+KY(RSlNz^ZA6mS4nzx&
z_VvL&TtS3kh-&NTVJU9Z-%w%pj;cRmn;gu~iW7IxHJ1h6fJ^A=3+Mh+t^~3Pmq9K6
z0}nr}DCyklk0<0A^0nu=O4ReE?G_EsW|HWxbOx@2&ixU?{uWRiGd{!(E6-;}vi|$@
z22mVNaY+j!h{W^U`Md_acp9Ne7Bwvz!aEVk;R~1BFcS1L!22%&&2z<P>!RyF0QG%s
z3@8zNU(EtZDhq1tF*nW(1GbA5BP`%R?A{LO=RP3YJ8CXVEjy{>T>M7a*Wr%j5L0ck
z<mhO(7aq9+7+r9g-<!A)w-9=hTDH5`u1S)uv7e+gyP+x&5=;`$w-)v3$PP={RGY;f
z;RZO<tJD!+kcWwFAj6YMfQG&Q`D`N48@U15)v??18cZqx*@pQVjZQ6)YM?{iXJ`fD
z04e|hMZNL}gnS_DBRH9#Gw&q~0f(RsHxTu8-fEM=XZUrF;wl5@3@esi^Jc#DeOEYK
zGoc~$Z{SvcZu;u(w*vfzn$oE?U>!t}5iXP&)_JSF0xX1q#lsAxi*St;BS4Yo2F)Qc
zE?=QPl@`pwsYZ|&+n;uOp8;R;XP8m?+Nl%BL1|9hc2h=+soLa@dNMx)g1=FU=gE%i
zfxsc@mdSVE*E?vYjUOEyjFUufrgIwi>4Yr9sft^GjLQx;k&oE{ZvU&z1{YD}jWRyP
zt>UdlYeZ*t6?kF7`2ny!*-BQcmC#7N^1lWQU^?ql^T&vn5@E2=&6QhL#47?+ET}E;
zwnvNFUPy$Wq}2S2dH5x4eVa>QDi<Q~L%gS)`DKXbodpQ$eI3#z{B{79r8hKL{J1B-
zxCadw%nD7JcLb?o7!~Y1g70YuenNoU@fr}}S_$gAHN+fI2d#PQ-`4}aSV}|=Ju};y
zHD-i^i#k`D9gU3v?0=CsbLWa@FhRnjoxlPUT&@5Go(%8_Gox;EaKMGp<wvy66JnCR
z?;)DChURWx!N<Tuhghj=5gRhlD>9mS9AN?NkRF<AZ;=b~pH)fkY0QJ3t*X?kf{vnU
zUX9O|7wFHr=L$!H{^eDlS#EzsH@-V`@<-p8J6AXR2`mi*4Q@b>R;O2ifq2$`Ls3ER
zlcB6jKJ4zhW4mj)Loq^m8A&WDm|gb-&eZ}#3aR|Hpxxg6H1ylZIO-DP_qqAC+_;(0
zIH~6yIPdFmH#;jkTkf65y+=n+t(|J}$B+}}t<6*E$yy&yNUj@%L4@h%-4s+18nhXP
zNcy-oe-m99kP%PnG1RICLqhEAUjYlT?O=0qaB;Um4oE7vx@q2HAFtoMc{8>1!UbZB
zpsS(u{GoICFgB@!ZURo4llSm#5oc=^_RR;noCaC=@y~of-Yt7F`;52rKD+tb2l|;a
z4e#nS4&6IW40d*7&FWkZcKkS~zyFiK*vt4+Hjs3~+8=EvF4Y!yEjSm%i(cQo{hRp8
z>l>58w4GyiuWvipYdk)?Xam$^_}2$*2NS8p?VGWpqM}>$V}#6%(Ns8aGZLeOEj~uX
zmb4UFs~D9y>c7t#6MK(i89AAsd^!jJR_%O1Z)8%|!GzgM+dRe8#U`XA(!lgofQMKQ
zKhJyZoBg?vpob(L0S<$sN0O)4Y(I7ZUjJ&MtiLjr5r#CCUnL&J;%C!T`(wb$z`T^;
z?b071B4#0?+d_gd$RTz|ldm=JNWarjjNGMU0}*3OvBObCzg~42m?8e*W3}A<YiKJi
z6F0!S-?!5yeXfpeJK+pt{oY+V3iCDpe67FjIT5D3N_>kwPxiWBlOS_Vw5Jn)q=2=C
z*JP%q(o$S`%>fLI(~35)^kee784<zGtDO)9Gt#!VoX<tPS1tFn*%ZAVp8R0*CPgV&
zyu=-pVtp{>jxFC{KtM6jS$~YMd$2VZOO-{*=Xz-Wl@Fto_q3uE(4W2(R&8eeu>J>|
zuJzGb7_LKEseE0XG1O>L55`CBDwnh7jM%8c`yZAgK{hzo(tP*({T+YY1h^u&4MgAR
zHewBxs?TF?Z<g6V=mGOk(ACi*mN?nHJJ;ml>o@T*3cePozS>y6;m8sLe%*wiT>15#
z11D<9A;%Aa?Db0jY36z%^9sXquTe);F_X(MDgo|G8qBi{9)bZaya&8SMAAr13LjNI
z=~R$L?}{BE`ChTD_7gFV<v8^#R+>x6L3vH>FjD%N!ac{e@3?+K5%B_EgnE>#IM>Qq
zJ8yMBO2GROG9oj$0)x_`#S|-_ou){;cTc0l@m)W#i3GovaOu?3iwJ@F#){?3_1`}{
zO4rXeaEJLi%QijJmcVf$ZXdyU>1lV_=>g~udwnD%@F^?;jZAf4TJT-AfzDlsVx<-1
zm+L_w(_nthlA3OXkDJ9Ux~*Hce6!>Cdf1jji62}_+RjNcjHNn7RjJGuC1j(0qMZW6
zN)K0N2aZL?58LAM&r6+0E46=2?0*O^WWWE>0afMB0s{P%=-TS`T)J0{hOON^kx=*M
z0t`v72eyJ}PBZQt;}Uhh$w3bJ#mO5EXA5BbkmWOIfOBZS3zLA0yhb~drj;QS&+A70
z-tqB{?T2$OKy@M}<_m>#<jtE7?Orek#~q@S9W9LppaYC{sNzMwK&AKf-o%E>Y{Ji-
zPn@EBWKeIxAcNIy^wNAm>Z7`H)5e<>RF4s6-~N92m3zJV?Pva^VU1@qSq0u}wxaqG
zEYZ(<<eL=62qd$JYlg4|I3H3N|5*zW1P2W2Y%H9ID^~}cIOrIvDqyT(EyUja;Q2*d
zy?T{p;QrQgkV%(tiJu(nk<T{#bpokC=+M+nQ;i%chAi9sFIgCXGE!VEoD6gQCe+l_
zQidih>ZAF2VW?pYgP`%uTrNfpa#}ALyOUwaQHN6Bn~x)Re&T1SR~z6GVlh&;xgyTz
zei_+<v!@uGyBdb2dc*24w4xVQm-#ny#C|C#Pnhz&c&yEub#K6Jr*KW&Zf%*RQcVO4
zXAw~qr(HSXSPFDRqJc{bNtFU74{~5yv;tI)=!&lfDx+hiIRw+1xc2RP_qcS-|3xT$
z0FEBgQJf)EJFj~Vzn}eeGd0QL;C>v7-}HAups4!QH8h<MS9>xyEj@i!ZmfJ*=%(Z;
zb1$4vv84akGkD!opbr`S2k!U9h>!0|8Ww{|h0kqDGq<gZ+}e{|T_Wx_(w&VRy1J|X
zL-_p>nqIj#RA@}|M_Ab;E1Pi$Uhf7m5}#e1ItDqxrCt06IR&j|Q>PnYR>%Mv|8)ks
zb33{=QS>K~S3u+K1f*kZV=1RKV1AntiNUzVQ=nu=gMM?3OVB*HrCaf51A(dO%;?Xi
z;zyNB^Ir}ibrTeTkyQ>9x_7pm=KjvtY-e+$_@lSqpUd1%M1062;H5JA=4*%7p2OeP
z_&@d=n8`D+n~^zfyd^1|^IkvC6wT%4gcrX7y_Xkv)+*>PLB4e6ocx9RsTdOS+L<4H
z=`i9|_r9IhGoDYx?m9cJGViOOwy?9iYc*dTEf+9X3*0UnZ@8Q~b!}4wUiZ<Q#y!3^
zXIdps_XtZxp=`=;tFg6~<IBc6DjC6*1M`_HNc*|Yh^_n9iyRq;!yA9WJkmdZH2!GA
z^ySAU<CH}C<gTpaL$z-S<Ekg=bZ}JQPYQ_HHAX;L@uoxpj5n?dq6RwO>$>Ucv-ObO
z7A%4ZyCV~5BDoB7o_7q}S-`AU=n}7oNWZZtAn-^Ts9jO3r^F5)E6Q%m1sJ2bzc5Mh
zXSBAaW)$3HRnXX)52-axAD^wv|FD^wbxY$!E}AnPBVe=(2AfK}dibdFZWufMez4nN
zz_K$5(sNqgS}`S-T^!Zd)`{S{Z!3gA`6%XxQ!J{ixu~DHs{@W#^mhnj=15&xn@<SX
zlM!}i0cL%A=-?>OSYnuEf1|ZkFVnQFXyHeAU=%d@nv*jK*eB@@gtJI}rYO<@qkyJ{
zamYW6236Al312}NcYXsVtX%0dAI&zpZ&#(g?4nN-5*4-W$h?NS`gPc|*%KjCW+h@$
zQbxbNf8cj;aL~Q0$u#>mafu7YvQLtPlj>gv_uw7t1YWVGa6*D)aKA1E=Et@byYksr
zuVDRL@pZDTjdur*p2wb8tDM8Y32U_%u_wp4ZS9qz;q~#lvQ{L5s!w4)s4S@*PSm<`
zv*ziEc-5O*$@XR^hSFlO2R9AqKB@j(2&Ev_p%J>jU7$brzWN4c;dm8q$eqJ$ZMXtS
zsrD9i#Z8z5+FqI7o5^=m{}f0N`FL`gT7|c`WFt~Tnw8}WY}uu>b`|RuOH5OH1?k$K
zExZqu)*Lg<CDWk!=e48lTZtaRIlDHfwotG8@I7*m9kXn+pPY>w;~x*+<k2lZba2I0
zr+I_7;`^ol!@~>Iux+uy5}t`4);XNSq@d_GJmtf%ZiBTxQMlKzZmq^-3j;ToYs2Mh
zgd(1Gao(!6eL>cr^e*Y`_BJ>_WM}dYv<}j&xwC6MB2avJNEx@jyW+*M=0_(PK^Kp5
zYA(}|3IYvoF)jC>$Afn1Qabp}$1^oPfDf~xO4`MiM(}2D?Iy0V0&k|$p%A!MXlU5L
z$?WTUZBE;)1~K*FR0WfD53AS>@r3X-=>RXn-R*tA8Md@;zt(+gj~oX@mnZhy*K9zq
zJ3P!xuG)hdCqlfR`U36{?Ybi+KEAuz^ZAL1x5A9p4?<p~=H1YJLH;u5ET8fQqBp<7
zhiPhYssAj!8;EWZ&%8KuE2)EW?Xzj%**isb#OZL`L1SLx5Yk^O-Y_sbxOyk@7@&`D
zyY4mxtOb!-Vj*p<LXUND%-R!LCk&*enn;d>tnVIJ^OGQJxU(LOI&XA{kMqDu+7+X?
zV9l>!4KamO{o}7QhSO{c&smjRJP7Y`li+W6_!QBv{q=R9-hhe0d!c2V&TAwH{suiP
zH}$D1H|c(3IM>vW@R#eBTZ&C;7;owyDZ8?Fxb9JE{*A3lFsoL~l25t!`pWaGKVAOY
z^VYn4kQNf);>`IGn9@oR8}<v;F!r?{sJS@QBhfeo4VXQO!&cLS&sctZ`!`~YFQ~5w
zUO5Rk($x>)zhZB~m<5|bHri<m0ENo70salgG9ja*7lakU;A1I8q4UZN|6xPkfTm5i
z5}#K!`|DLBDHedCFk0yfgOWK;<$-f&aV9v8%IYx!-y&Io@)o?h!7p#pPmx~Wg;SE@
zRbPMFG|-eP^;|v1kG4R2#d|Io+it1-*AxELSFDUEnNbsR@%!(c?*uY(j)i3QIK>%*
zs8*Ap`9qa8A@4aXLm*f<5Y+)Qqt{-fCnVC;UAyF$<5?@PtfxQjc&iu6xF@2<xSfTF
z*qv@g$GY#>H?i)@uK)H@3evHu?Mhh(fA4wic~BuapJ*GO#51N(ZR?dg(&Uz`Puji*
zcnl7^tamqpu!nJn6(80Lf3;?;wW@2stD6IqA@21z&S58s@lIOEVA$S$>-TM)P)1~<
z_c-k$ev|aVo&u*fAt~Ppo4E8!lM#Ts#(;<D&lr?|->iN}VuYCbYKoD4v8+Fsavt5I
zX3IBv#K*bdgp>>P*4sX89Jm;ybfL#d?FCjeqWziXX7M!w_IvHCTG1PCYR&2=dCW1I
zGL+g0b$35W9U%v?9M9U&_O@0zVc5y$q;q-*_*#BfH|Ap56TmtCoY_>h-mcJG_GAx9
zx4xzR<|#>%G6V|Xwjm?1m&GG%y_~OmjcN;|JC<Df4`3q6MSh5%k>a@(d`d|TUcK50
zW2()`=9ZTq*;QzJ2JG+Ee0hHr>}E)BNMFUctgeHw%JHJ=QN;(fWv!EOv;sT5_e3ai
zfYfe@3#BdPTG<m<pIz&YWZk!xH_!>W49PumE89S)ol7ZlLo3(!2s7G}P}q`+F!8pw
zirGTXGbkk#e!yP|%ZpwnY?BK<DEaG+##2DPD)tXecDSru-t%48#OKyopLI`9%0*8v
zAxX8mw$<`;=G3R-MVX#r56~a&7OI{SALDGC+BX*5YMBoQ^2_hOB(VpW=cK-jW}bBi
zElk!eVRL06KA!cyknoN+z0hrZH2>R>LWt4wvxCjXMw_>_-R(#?T(e;(2CoXVwSjMU
zMHUoWXJ4Z^O|uaw+SjRyUF)u21-t9;+r;`=3Z}ioL-T#N)IL35SaHiBfMeOX?#i$K
z*2g@^qH%hGqt|f*PlB>sg{DMU_ccKze-FuAWy3yRat{bg;NJJb0KBF4Io$)8!!+Qr
zTsV>vw(c*TWMGdte@i8{%2x@gu)x}iVjo+Y!T*$<LY`~7m5f&VLx18ppZJG`Hbo8}
z7uu0TxLyg09(EY0k51RTDSGl-xpPmIS;}Izv#EvIX`j+^PEh!MKHpd{R%EWMq*pG3
zWm|jR&tY;C)!x+f<HXJcUFLj%xSe*r5?u4Kpf-GhY052m`3-G>=nrvee-8^mO$pDQ
zxE4-yWwPL%zJYe{PYz)%DyML1U=%Vne4QxIHucLbAWV7>y|NMP)b8`TJ26zqx2sl2
zty}Yi4Vb*Gr30DT-1`JM?E;MDl5cKPuVwq+3lF?x>%cbOA=VpOKcCmUgl7fyk>r?a
zgRxz5Jsm*V6MFp@7$Ez{QvQdqk|@6#MmoCwGR0Rf?^wt*EmdXToF8^z(KB|%LigtN
zUH!OjSJ!=u)S-rASe<?RdEy*R8Bd*upIh{JZoRyC=MK^Z#-xU230rHQbVj%ZlzdTg
zn1_2^+!Qa%k(3QK$xJ19oo`C&qr?vRW;ve+H@O%Z$LJ5wUB;RcEI=s>)yO~RAF8h)
zhhg%wVe{8P3d5T|#%1odFdjy8atG}_=7*<xgdkae<imj*TB-GFwrHoUiSavS8DLme
zWh`lZr5z<;{H=cl1$_ri?7_y-xss;%?Wgp5;jW!u7Ia`tov&)^aei_C6?d|w+%gyB
zBe78+I*?8X*-b6}?mt*68oWdIq*Fs0V*q0B_mTXVq3um)W(K;8=yo3?^O^NC&d}jd
z)6$}Jk{N!V%m2u0lswznk|Qr4w=1UxO#5@6@D1}S>`TaE$ZslMul2hyJ2!)~6T$in
zOGPJ4;*PxRQ*5@Aa_<SgPL4{#+<G&EG~plGWkw5`jx98pH#FlD&)+`1NYOC+^9^->
z(#wl=ci-%n&fJJ@={3++<9YY|r@x~VcpakDdeaM9<5E}{Skr*xS1mIch&IC=b;Vm)
z{NI=1a{b;drFQsm3`|wW$HpY&+ajQTQV%6UPD*C+@d#owe84t-4(WN3`pGw=t9m|M
zW^Uc7tiAW_Nnv<BPn5!OjRAD451lqAlMT-;<&>8a=+MPTN)v*;Wwq4QB2$#Es6$jv
z_NBK`@s7bfB=`qIo*W~V;R7!vb))vfl?`gv>xH$89)C`HQ^vQ*#>OV@K@p(U(3tl<
z<`tz5a+g<b_<pV4FAB>sSH9ic>-4Ms@~d=9+!;cEd<b8C=p9^_%w4zP*lHdkGYjU<
z)&Q?QcOssIV96iiEzg21sUqtOcY@yP6(GvH_sL8fc$w4ExTO4l2|QuB?Rn{~F&%j0
zS~{u+tF1vCWF&e$7*`H5jD(6+X7?Rw&+Cxzz34oUrJtP!NgIxaBZQOYCH2@ob4so0
zk{Nr5{*8DI+yb$?>(fH)jyn)yp`_pb>9~}EmF2Cw`M<xV0_~cf{A{~J-d&7&g3oWc
zwHf+DZqG>(hZdlsOsRHtLmNSs-b31z6PEXV-n+Mi)h%(bg!bNJ(tgmJ5pOSSBZ%Uq
z@M=I1?)mw(iQ4HaRD=tQojXKxinguznpkgeod@L(7E|Mz36ecnm`ePQN`3?82$ZhU
zt@NZ-xD$}|1qqpkd<zV(zYaAS@f7P|ODfvJLS!m>x36sZQC(fGJ$v@}o05Hr(0anS
znH)p~uiFoSS;j^wWKt2X&aSO9FfJq=lIzzK5)!QEAvZny{n?EhHwHr3&W;b{n^zil
zK0K9}DnLdMnj=Yz)vvBc(msmGPw=ShZ(0s%9}S9!q{>QilIs6HIf?fqX6Ur9FbEFK
zlN>uIRl2}N@&P5IchU6vF)vNE%14|};fYfu-pq2#o&zc-49~EfG?zdX<4Py)pC)Yh
zj8-oIS7IPT#U!I1wVOjDMj#UECNnMi5g}8<Z4JLhy9U7eZJo;LdexBsc!>`s!-|W=
z_ZStW>E#%{dG!H89k^%rNII!Yx7Bipc@byxT7WJJ$5Iv2-L)gW0DcMaKg$%Uj~Vhy
z=4gArFNytV?R<We=T`ZA&&EppYUq3Y97p|_2f<y_x!}nQ$&^97NAlbN&cZn$b0^^}
zp3>RBNxA6!v7$rA2POiD+BdT+#ve04xwSCneOu`pIh>9>5rQ%DktOwbM<wVAF=>Bq
z%9Awy1XQW%ViOe>OxQ^;jjtxI(=va28x)2QR3n4T$_#@p-+RxSAllLTVwruA86=1%
zc|+UL;kLI}N4)MH7J_~M&Ut8_=29i>d;5<2n+f|)DyWldyidw!JSn*sa;2!M5iFjc
zgZQ^wiflC$y#d!^XH@2%&JjkB|A;qfx47l~=k9l>;_KQps?G$uZf2;&VviE?3U4<C
z+x@d0Durk>Hu(FEyDC3ukP}L$4K^q|H97#9?m2Ss|3>_n_otXzY*aFA1Hb0b9hHy=
z9G2l=y8Ul=P!KK|F{4f(=l;EWkF>P36j5+<aj7~wI{qqetAa)>C(grxwAkul%xFS4
z%h7WsNAqHJrkD4+Y>4LUCBo?}zie%)_+bdOKWfik|C|lzRYR)L&s}8Dqm?Aihj|u>
zgz}<m+CkWh>OvIBJWkj|CEeDBrG4*74*_ReWU?746b5~(<(z(9Aeitkn3{05SkRt|
zca62x)b^(NI$Xq3nKQf0q3ZP|lmOa5^p#M{c`g57kmw&(NsuBqCk}%&a`Blcp!q5B
zBdYFhJX^0gHHct=TF|h=$Gm&EXV3A1x)UO-RyRF+I?OLf5f!xEsw5l_M8AOBF~~ZW
zf3h}1KRbqtmv=`?Me#vMjD<-AAUk-Rt@hKiOv!4<@9bq*3CxenjO&H}c1-?|#=rg#
z8-ont{mvR*j4v%5@JA$=;=(HIUFd+ngHni}9(L%Btj53I2k^~WT@Kk(W58i#E$F!|
zyAsV=>h}V0WE=fRl6o|s?gM9|>|0J29|a<w!FXzJz{b>_?h+Fd8vqbw><wORbsUBy
z?ab_iZjnu$Oxq4a1!Lp*+xy)Pwi;&c&C*FeXi2Lve3Gi)bz#P>$7dVAL8G_k*QUgt
z>tTG(ZKYWc5~MxV;W6Cc_TK<hMT3j&MmS$~7M#Uviv^3qMvbqf8a+a0RcAEr8aK=>
z%^&OW9S`JQI`-?~u8waVUCS{?mWrO|d_J=)2z!;_H*Tlg|88X^|8kSicG2ucmB?zZ
zqH9<#2-U-0<Auw8vJfTtdRf5)yeUiA#7etz6V%X6P(J^F9kQFRH~w%tk-wtME^M8F
zL6kPod0PY~%7P{&XCj*aSPTq2QsdEz=K^$Wow`@+SD=j)2$=sx|4yYO152?(!g)NP
zX)bY&2^D~hvO5x?R8WaZmi`|`peztt0qD~Sz#cce&ovqN4D3TSO9rK`pPsgqdpeZ?
z9(UjrplrE#j1#-F&PfOu6&-q?Zy|h3!sR+tc7^;WayqpDcaZKkHB_|>hDR99`08gU
zz>>tn+}p&oyLab|eaCwhEV=@nipCU4<a9hqgzD8@{GU<2eEcBc_H6-ex!JS~OJcrp
z9J#WHO)ZTB^t@Tp-yK%D#uAB!$+VtORt}EZedOswwgI~b8H*WbS!MlNV*{XLpaE#K
zKT{+mzw8AlUKB7R+kMSGn5BK^fOibcjnhX369X5o_?>;_kgT!T;r+dPPrWwDms;c2
znSIXf&C_FJu<lsqeU3J^KYnY6OL(wA^^@QK@!JtuV^aP~9=;qvNBR+OLKfI4@CMPH
z77SwLumii;ioT!)eGwa(w3GC?Hkgowkq)tyYyX5;Zh7JKw>POFBR!Ah4QbWX)h9^!
zMQ*jTd2nipzRxprVvb<H{Uvh)8Ps44lt`*(04MyNf{HfhYM>uLt_Gk!856(HX^waE
z;u3sa570Lo$XIqLFrH)h)sSKFxC#J#9~iMug{9_)0d|mnEZ@9i*x+$S&^7>9$6JZf
zA5uWIPQ49bvF3ht@_PoSK~Cn*3Cu7KzR$fSU|d`)zh{2F+;cb$Rj7xDuBbaf{xM_J
zAlKNo_NPThp>;@W_OT}Ux^da}eOc!&L*t3q1R{Y|>7`awJZrK?Cn@Oa&I1Mj;dYME
zLIw-n5+5Bx6~Fh1YV1FU37?@v09AbZf@2TKAiq?OjL6V~K#s}HpaMy?K?Qo`-@^^9
zTpb0tIy;qd66`b<=;}<J7+ZNBF*#HLJ0+84xw*MT_-o3^9mipMuNqis_x^9U3vBAP
z0un*vxt6B3b}T3trU&wYOBGwTi(tWcTd7i?NI!IWtEm4k=Wib?{qtT(I=8dKM42mq
zz7Z~qX%PGg4rC<4=Bd!qn|=ap?yE@FlLy=K%mSWpFEl{bUxDP*cKNdvkh!gU><~QW
zC@?+R#rrP%m}FWu{ehQL;QhP9fYAnM#E7(BF5|vH&Ud}*K|euz2hFc2n$6QYohC^M
zuE!r?)E`bhvdtFEj`d3SS$L|Sza7+;u`SAcwnW=(ij*8V7_rnqSxOd${}AN`67f%e
zaeM>_g<8<e0?B#bmgX-_PlIAHMLiam;*X@O(@S{_Mp$(Ll0p6FmshqBLk4>+NAFS!
zojHer5N3vVN7Z8hWww-^J^^T!(07GDdL)MZq3mh{uCQ)~PI3-c&sUi5$Nuo6UpB|9
z`cU%Y(8FT$j7J(k+DGqQ%YPgkd^0B_hzp`Mk^c65c>pSmf?#|5uRA3v=gwp-!<2*r
zM@Gp^!7E6^;xM?n973KNKz0WgWfE@QWMZU;1R^0*u;%jqWx|jw9zkA;ISa)cO8^o7
zW2}_ZdjVf)5lQK}wEvx9sjF3wa8mGn|EXi$*RNlHx@U<0na$gR%-xsZqooHy{6;oP
zWo*1SoXx*fR(h8XvPO2hzmZ~S!Ny&RyGFSp+?UF$0lJ?C*uCSP*V_Q$lz#F<`H@rF
zIedC_pYLfW3W3)!J$g{<dg1}^JS%T9GP0I@3wgzOV}}prC8#?j*kh`38Qf-xw9IHg
z3^f2ACmlR!-Be}DS-?^&8NEu7HPMSWU#-OSy}eLzVxWOBZop)$y)YF(I(!GT7g;<3
z7pC5nAe#<Rbk$<!21=Np8T;y%WCN&-et<-D4BW3m7ZuH0C>fg9bMXg?^L`Xm<nr?J
zHWb-5Y_YA66&4Ph@9Np&;K&VN5d=Luq9C-uxDz8a^H3$AN%W}8hpK=_XFxYs?LrE?
zLHKdEPWLr8S+9{ey*yK?n#W~OUI}Hsmz`H$-*$I$(xc;n%Pl}LXqyL2>CZyTL%QXY
zbMBv>k|^FXTFoMj5wg)AWqAOz^Y0H5VuDwANBJi1Lzx~p%7wRl>TQn3E~O@$z^8&P
zLnfi!OnYNlS=FY{0#;!NYWujG&eArkK;~1<T0vvwQneR5+$eT@S1<Y;uuOYkzWBX?
zQIC#ZKyy!#D2$&9kH<bf(ZssU^>%D%sT{@`xPN|5rg#>@eA_1g2!bvKrk48#aCf*+
zi;8fmqCx!J8QA-NXoUHpUB7g;Ct$h`1}J}_xZRw6OqfZ+IS!Th8(&i|q;$5XC{c7k
z_{Nc~c7taEKq@s645i?pC#p8Y2vXMZ1-$_H5T!#|?jQpw6-^(Q54feKl)N|T+_Bbt
zi=JlsM=$kB8U|Z4Z&>w6e(<=FV_w<G=<zh-M^`9w8??u$f~~e~PExQ_T0vTY?w}^@
zF00J;Raol3*<$P&<cw}-qx>}_0oj9>UZ3~e(E9L8I2d0}{Q}y>+FzDZ@9P}%?x%h-
zD8n8iJ%0!*44T+4Y;!LDIL9AUU+30H16L%57UsI{0vmR};*QXce(s28KpbKVZ4$wG
zt$Ai9Fxs30k~Sh~oAYE~My06R=R=A`<c$S^%V*!JE$|Hn#rO8)nH|V9F3GZbWXX>Z
zRdA)4ieR7k0U;@#Mv@0JGZk-&+8&xx3)9+v#jH;nGAytMSCiEs*pCG-Vtb3>s*DjI
zB-Fi=m9L!vWw|y}wf5T-C`V{K+wB?dC?f<~wjHl!YnCJ51VEYL0unL{?Tyr2cqQ8<
z2;ta=&m+`Ci%>Eil;kPKYjD7zK@`2!^VCVto8k`Fp`WvP$gVRgTs-1@`Mtvc)&N8x
z(_9MXIN*cp0Ke%zH|cN$w*xR)*0plzE<qd|!SGET@H^Xt%}vkF4lpXxXN>|hODt_R
zzZ7aGRcB^xpw9|LcFRJW7EgXH1zd>*;A8CU8=$&rTOC9cEN-y6P@E1Sxsegc$@Y&X
z2IgI$vXhAVhg9Zp3XvbJ3779_N=(hji<d*RkjDt2bJLXgIQBD=g7Q14x{&#*il+G}
z+_8kdgW32eR_uuY&#%O(?Lqe}wfurk=K~if-vee6z-#^t+|YCo{@|0lA4BqPM~k~j
z+315Vo7sth;9bb=gqyX!WuizI5U(sQoh$B#uzkr%7_H$mZs`l{Q0IT`6M6R94OMRk
zz`>IgSJ<hD5<5}+$6iOgCSTPc5d=U$L{~=SKPUWxu-j1De)|g{-RKYq4jp@tv*$O%
z#%u=Qo;C_s*^)6oW*-r^0LpDvAA~@!j9#~sCmrOk_Qm!TCZkmF2Fypg+S=L(a(cd6
za8#duKXd$e3d;d6P4!rzoxdc?>94}l!4F7R31<!d-VJjfMNfT{Q8Yfu59s3?K&tRW
zryuluEJ6Xp>RUkcjUJx7lN<`8BVd?29x@tg0{N~aQj378Xf7_U8E@gXF|TXGFf_pe
z#`QS47vBSAH37XU>fm9Y?vMN(t;7*T-!4^Y)j#{!J}(4^fQ(7E+ET6_&AJUrT$Old
z>`9;v!uHB=^bQO=Bp<v8K*4=EZf5~1mu6Ua+$y?6cM!m36W}aei7%zAOW|Q8l&_iU
zzBu<Cf$fr&9{1QmBoB(>?DjJ&KpN~uqWfwiW&4~~5-UmFq4tT44(uU&tzXuH#oDr%
z4Nig#%qK&LbG@L85D77d!pt927&hUeO5r^KiQ?-B`_$BYpWz)VePeat514~}_>OcP
zD4>N$tP_v!^u@=-@W7byTqyl~C^j4IdKkO3IDduwCGF5xySL=3tJ%~h(hGA&91w&G
zF#r&Z;=qA*Sf91(fo2?EuddwBFex>M&sc&FAr<)XBbLgMnCuJ9VqY9Hcm+5gkQ+*n
zn=!X~mu2G9wOS!EVz3vl^i&nAO(4%&vG3y~1G<aVOc;QyMFF7lg0pkBL2f^ckGc+C
z>P3O<(>T?QQZQvj7Nn&j!+R7Z)<DHw$eD(rT%splv-E7w=%pNb4C>@!_hEYIV{zqq
zRO1O5wHCm#nr=K}kZqs^W2`y8eR%XRicg=<r>|~RB{mTZLxL>VxBaV^=C{!NDIq~0
z@MgCQw^()*%<d;G8U4qq@v!+N=OOzF0NP!!3M>VYu!S}SJzvzY@QBl1td+!P6n!Cf
zlg5^9Kg*#eYA%l%l*eqH5T>XZ&L<H?*>vQe8L@M6DnefgAAs8=E05{pcA#zoK%`y=
zoOR7nCFa^>fP;gB5|c5#49!W`$>?~9bG<u?Y^$);;F#V}4K#p$ULEtxHKlowAB9+z
z8+s>5KNe_6fyTT6ZF!nNUGgb3(7((pMw6iZ3qZ@?_rgdSGAon+MnL>)%Scm(@*tdJ
zv?ja89|gtoCNSWOcH!FWoA+UePtPpw@81C%*z2*9>-A*&3dryVxb2>&ct+nIf|x~X
z(>XhK8Wp$zZqHA&tvuxdO*?o38DwA3pr!%vMI!n1vsA%grbDv51?Fthm>l3e(1N7`
z2yQf}N|ln-p%a|+Hp!qp?@m?0v<T!wzLnv>4A6o+`fxY>+*tsfr@>U237Ea2;pgW!
zzRA2k6Jo@CQ-F@17F6NEkUm%Gg)v;QATTdZe5LAex=2oc;dGqhrYm;R$}2IOYa_=Y
z#ea_3ZsWm^_a3#Jl_jo_`O#W$L4!xm2JYqSO!fNG$m9ziqS9h|scH;Zq<RQo(`@Rl
zM>@Y=Ir1Z0$0w#*)h*L@yVc|{gAaf8eKh^)#KC&2^rNJPfWNW4eB~+OC}=jKD^Pr@
z#~ek=_HSU|8s71OkU|~iw5J)DINrTg0V9%I9(kXtYiw+cOpJ)=rN}!{cwFhoax|a5
zdePa}@nR>gK@#a{0VR%Q3rjHlqndXGKQ;J=e9l0pwe9!rvI!W4y2<WDz6H$int&vX
zI`~od-_D%dfbk<;_@*6r9fIR;Z?<chM8?JK1{Z|R^R=zQ2HYU2&Eg1Sj>P>$j`PuB
zcG5VOluWfhkUjrWocaL|t?|q>UXuZ1g$(R$pu%lSDl4|{+mJp2B-T{xY}-*!SHR-f
zEgv#9+B<zYkHtXxVh}foPfY3VRgBhN8e~pkM2rCJgJ`imjFq1lx!=Uxl2TZ7`s-l_
za09PdzJzE$1|esGCx>^SAy)E(J}3v<cUk(u6+X~O4<o9SNYY487?pcudVKHh&?z=~
zc(Ta;%`Qdb_w|IVfKhf;6G5+TaY!~8VQK5`MC}X~7I4dUKk^=v5QNN-1tOr({0T}G
zHv@z1Ib{5&YA|{to9T%!1M+VIBnH)k`U9LZiT;BFQh;|u4f^(6z>*RPYlE_}xe*`F
ze6W_O!{F@YRwmh$XxP}G`@|JzWCcx23o<O4y`Mp{Q}Sv`KmcG9QMwkui%@{w$<(<5
zOb+rU-bL9j;yI+p1g$?rStptjCSaT@6YN#Wuhd^(A8*^Ei%)#<sk)uisK7`G7#3io
zLY7-nRj6;u2UZ2@Hjn)v@3;I~x@ok+LK!kv3qKQ=qf}UZ4?%po;C=`JkO3|)npOxh
z^Mer-@r-rSM=>-cT?+H#g2hH|S}eQ9$_1-s=8*g6`$co95NRCiZkMd-OOO}7=`$3m
zc;_U_Lhgm)Mg5fc-Obyho%Vq58T0I5Dh+_`e+K1tk8C0AtR`A5A?tA^`DJH`Qv-x$
zT&+#}gTtO0mO+QLE%=VXs9N>64x-YAmti~7LFWX2h8bTs;8$8B7@P+nokCB4rX%p%
z+N_bgCh>QOMl#%rirS^}3_8=794m*kd)+7c)%KV?aJCd3hU};c)YuH49aFfbFB0;=
zrOUW0q6M0(Xh^4io&X0oxx$Q}*u6j_e(_J6vnGUJu@mMfj!$6Qz9D5-H8^Mq$^+Px
zt;l}%L1<Ez;J1(%#QUfzAO(*NwgjAsWv}Hb==3u?4LBFz0cgO&PZt}KU_@@YCZ;A`
zya17^I#hI=+K)qCt);~2m|}c~Iu}e4bne&%;j^S3WYlC)8jM5mA^kh^cPE#XwBF15
zO=g%vNBdHpR8=7Op)FbZ{J_(4m80nzmdfi|cJgVf9K9?(HC<iZB8{xXbd(R;=h#|%
zxDTox&n*C0^1cRZ#ug)NAr!duGcos6BLmnE0?cGT_CH9+U%22>Q?SPXo0#P|JdHeW
z6XMbYciEBwtTnGORHHNAxi)Vn{=5r7X+k5v!4C^aIwIoZ`5ccg!VK(1M>^y`+MpCq
zaYY`q^|7naTr4~NwI0zDIYAmLKL#=yP`iJ5f(kWpnR<WlRCp*0`60br<987j09BXr
z9Et*;6!om#mi0G)vR@>pKe!;GuD#VP-;BukHXMRKPf0{kG%-Sou94joe|Hnug?ezv
zC|~GXZt-L(f;$c51n!rd1@uHhAkU12enUDV66cFSH7n4F&Eafu9{(J^RLkKj;Jnod
z-2xyaeZn|t&m2%uLKKWb9*#d9B$1u=0XuD~y)%&JEl~pmFvdLA8-b_<3O$~R>LDv3
zlk0748csd*%hOdZ02VzJXd@8d6H(|*NpM4rd8k5BE%0^<rgKA8qhlA!7(xPwTZM_f
z;L|I~9`Uq?E?ov-wfLW$P4I(mjsch`zKuSP-IWZ+Qws`eZKb2*78840fKrnKV6i3B
z1|Fo$!T3B{#YhxKKz5wVbEtK1T6Q$FEI=S(=1yEw%V!TAp(VnQRs=+1RS<o;dh~Vh
zg^f>^ehSVx8GkmbuMY~yl2<Q7)&L__0GkS=S?-t4$b@d&^O@p{E2Juv&B;u5^&Y`l
z%G24o5igm>MzaoUgBdoC`yW6$EI;p|(@u3`If7uJy`ISV3!!Eca=d;auK_N(4JZ$*
zec{!rR(p>ON`l`jwT%WF^m9Ya>cxI;!Y+(sp<<I0K>kf);F6SzB;X5pV6te`mshv0
z#OI7X&baJ>q8(wH^W@D1)-_ZX58|8U&o1~Zr&NtWkaGis=d0=%)FH0!kUZS_ye(${
z8ev2oTX)~w=VV}@d#y<a$I`m#?~PfP$PXo9)H1%3@Rce8GG^h*TvawZBj4_Wy(qs^
zca=%(5J{?z=l7n+7Ql6~8=AHwCnqPXKX(j!k+>IFn?9cXvLo|Is<Ky)9!)qj*QkQ(
z=d!3e)A5aoiK*z!th+DJe;|-?$G4#??#G_&GX(z_m`J7iU1VDKxqwqzWhv>KeL0tQ
z97=4cd9gRQFXH1rSDBvU^O$a2R&7GyKds_v0B`Z;ljbpyT3jiD{V+>Tig+(Tf?g5a
z{^SC6x>^na9`Z$=>VRD#7!0&3^lm4FGZ3IaR_46pe1-2%l)S1d*#Yj-bubnhOAE6B
zPiXa$f?x_IM~wtjo1&q5ut#MMngi2d-O%h8&zTPmCsoztP?m}WOh_F#3;p2oO4ZNu
zUZAT@hTdkBP+xvk^q5h=uz6aRfA;`zu??KA6&mk~1%fy=WHP}HQ#RI{L)VV@XzZNS
z-f>}-F8W%9eM0a1;&0y`IlxoIadtlCoM~&8e#2KU!2EPT<1vejk4Qn>n4xi{Uy1N`
z<!hQ-A^DvLLSB0y5aqw1_6PbnyW*HKC8^ILLLs7-NLBKm)#_(qd?BKt86!#R<NJ9J
zj-}g`_${igW2upJPi@Eb=AsB|hx-I<=W#sDP!ctNJlowuD8k<sjA(;S+-aDU7!44g
z0ja?sK<Z-yG49<YAbv?fLy+(KmT_lo*vdz9VR`Fe_RRw+7;@7JpErR=-;RaLOkB$N
zbyE;Jeqy0NG~Sj(>nM>hYW)Jboyb&`aJ$(1ViD&%9RzBo@HN$9juxezyhA=#5Yz!{
zSwGF4y+@v~K<$QG(B#3@!xvl61vL<KXDpkK2dXNs3ugcL7R!co=Md*1joSjf1DEfC
zXiY?FbfI0VM84KLCocB~K!+=(rF!`qgmZ7)(in(u3xM9NOI5-D8o;?ius|xY`*MH3
z%`B9%fX(UDHPX8PML;D;4QTQ@cXAmz*J|aLXPc5{mzW?0NJD=pnb5noEI`H!%d>3F
zI8wDNDketRwK)$`*DBD0&J=gMQVs;_$P-qC6L@<rpzL!fe?Pb-(V%VA;mrG(YZXfN
z#`Mhr#Y_Wy>im(oyGZmHn-RCx-B~WE=`N5!_cvnGNrO`Ik5eAojK{&q)d3q{s?OUn
zeW)jE8swR(iG=j1%>-k0R*(=JgAtLqRkg)#$&-{)&${0gTXa%X`}TegxbeD>`Oefy
zJ_pV3=B*nK(JyL;D9P@ZJu_2~5z;vH7&<1^Ax{bTpStP9@*l^JI2rwmQ#gRUUVD|E
zj*xj0h^ahg#y?ptPS{plI|t!l9hl0eodHl<vHkS)qE&P{zI6vXNdm_bR%Pem1Y^=Y
zWHU#-LD2?2?A?Jerw8!pyU2SzD<Q}rrz>O_GtDc@7bf3ocZ30X&aJ0s^KMtZNf#dS
z4n5Hr_ld%>0HwVJAnoma?VSbl7&v+!gUt$Cw{F#JnB~o4--0SMTDK%-c~1l$led6U
zk2=iEWCtIK@!ewxU2DN*IAaS&O{oMrDjhqz`hVF%a?2g%!=IertaT-G!#0Ek0{8OP
zmlND)7k<=Kd7T1KF2G+Cqi-)}o@Jc>=sy+>g0FsgEBBdiTJhqM&_-?->-+gKZ$Hd9
zZZMx-P8rf45H>B1xU&H5oOLjEGNO|8Tdd)=Vcx_upAXm><QRUMmmRZo(IaAM-4Da4
zpHDv;dQu+=2%H~`1zT&api2;?7H1@-$pyDdxu*%wXm86B)k{7n1-W!&L^ml)1J!~h
zp(E~U-}Aq6W<i&O{6ro(zT>lqGMSo%kTE`LPF78}*ZJ+;Fhull^3kY-MfHWA)472f
zP-p};FT7R+9?D;`{rX23p;kZW?|Kg-Ml6I{YauhGh3@u{KU97bbRxX~48e3H>M2yn
z>lS8z2JrBy=Y@XE?t~nUAj*3NBtH4&vd>J`a3|ChGB!WuV^?x?%n~NOSibkzD4UE=
z!8^V^e@Apk02rmE*>029S6xWz4;?7blzHdvY|jK%n`Sql)^!dgdgk4myLz$9_-Q~&
zmvYn+yQLX~69i;CGAx4>89hgk%PUM8G*{UQMIV`_17KAnQN{3Zg3O+tz~xF|AB9Za
z^fc#=qAGcI<TGbNa`5^5k-(`1=x&Sw8#22zXaYWVy8A$VOOBCP&%MAGe3#BUhR<zu
z8w!7tDQ9eUjYVo#U0q#~_n5hb>xrrP2FIeJxqvS^@^5FyG!~D@?sJ>HB3wnIe@_(J
zwu8r}<MazK%Sww>C3EM!hP;FIT|;!tf!&}hTEM7fT4iow_{z&^!?0|}`{bIrXWs2i
zJ!?_v@vWtbos+Nm5_v`PS(RH5A55ZXZL&ty?4e%o61=M1T=Y>+j1Vn^CvkI9BCHfR
zme0wr|BBc`NHaptzP+a%TWDQV4^_WijwR$nB9_$~EGFMVee{rCNT{F5y~8`-y66`K
z!<ge!9llw27U!p@N!)baSmX)}3BStBxKGcgN9r#&MPr~ew9mx2*q&CrGv8=d#{mFw
z<-#p_Mp9F*6Ht=c*RYH7Z%SJn&qJboNc_bUfF`{29*^_M8f$MB1?A4J0~O1pOLD`e
z+EYvv4@)l0A9@(Jdr_gf96(pYcFuq4F?^;kDIay}iAnHTMv{MN21lTMB*>W7wUpqL
zjBmX1<#%qZaH}C0T*L@cHT&gO#>hdX(mRk>DRydm?H;f^9k|R3=sw90-=mM2M*^6q
zC-$M&sBw=sy{|-z5{aG8lqn?at@M*Ini*KSCk*ta3M59CUba*P_*--}4(YzZ>0dBm
z_?vlZHbFt3U-ld6d!!;i^m-zXf6lD3R%}3Gi3K+vw`(rvwCuYR0^RSoey95&kpi%R
zVF>+mZWWS%gyR>v@=s6|a)Up#XIN4>R%t-W<oH86N9F2d0xVdOdUt%-s*-=MizP8d
zRCL_3va(r^)Kpa?W`BNd*y-JEmHE*X13b~P#tFV(m^7v~@tXposiVN^e8y(d<QPyw
zy5LHcCA~-p_+`W-16(i*u40fp;il&|h=L+k{K=MF6G@|D`$z{?0_^vXmrM#rVgD3u
z3in|^9%oji);vENP_xqBhl3z@qtc|&SHG+#!_r)CC`7v)`bLd;4}=wllGYUKwl07K
zya7MwR<>;)D*e3&&+%|0EGF<3aKxRzOsV#ffC8%sFm3T){V01#HO!qKfD~(eBx~*k
z&_NBr1i=ltN9srQ;7X&;Wj@P;;<13s(QQ4eKYqD82E3Ysc`)TYNr_qVqyS1~p8pbZ
zYcIl5z~`DjBaIL1Cwaw-(B6M|8<6XoMAe35R_WrW-|%uweDlAO<AdR0bDVdPKMEON
z0S~JiTlwOifjgTH3YDjfi=$U-#hyR8Rl`~NyX{Gn1ZMD>ed{dAc1Q?3UU>XPKO~0e
zJ<sQ!WDRGP<?DKQDt6}quUg2VaY3;5*yq~V(A^^>Y8lY1eof~7g>I<HJXl!x*<^QX
zK+5Im1u`p=5}h|R(<2>p9)Y0CbRnUiH2cYBVYVqS1}?}g20^p(V&I{;utnOceDB|7
zz!Kku=AvYIB99))fMUacD+53+2mF!sU@O_d9#d4})GkAbWbT|43gqz{BTLvA|C(L0
ze{75%d(hm}a076Vo{G=k3Pn|6<d`DR7CEDkv+;Kbw%mcqvmTsn&7b$To+|&w1DFM^
z_l|jqWhCykE{gOv#~~|;idjdVFuYEX<p-}Xu>XFO+zt4>I-o5)-DRp5LHaHc^k%>D
zNhl7B0y}AvEsz{=b1GjYjc4OiEcpb5rC2}`yiGyavkVIs>(vUzU-Gnw>USBBSp+@K
zLAq>eCpWTm62&MVM!F54TDw!4q}z@LynN~+_(|@K3#WP@O@R<NuH{T7e_TJLC<ghV
zwXfOw21FszC_6lGnKyy7GE$9<zVQ>}TB87o7uSFNLeN73Xy??<kd-P9swk84HG@$w
zJue(3Xo80bxO5hi8LJp5Lz)Ykkdy#>!3SOWMs-qh`UT(uNcXE!YZidtN^%?VpV~mm
zRyXC)<?5j`!w9_rH31_)OP`2Oj0VKsK?F{EpY`o4kfw4%GD+ms(aR`w1~}ZNSJcr!
zEvESmq_z3@D#+blj%kesMB45Dy@pu;cuWf+7!Uc_oOTGjfHO~6KqnaAFuhJQevM#D
z?pQ}8{0?vxNC@dQ>iGPVX~*(R_aI8TW>Gt)hFIZPpo@GZ&=C6`utmNgU+2D@ur(y>
z*U2&5Yf()`AGfP@VsW}h0|NJca7DHli+4+W1Tw{mvB2d6_Oj^0g4Aw))g+h-+f(G$
zr~Gyo<h$wURd=;ghJ1SoalopcC!BrIC?@&)C;cEz2xs^pYZRbz?ee|3Ym|ZrEslY9
zk4>S2_g_!xB^J7xXh5rH*QFLV;v58WJ^2#yc5D9J9Iy;<Otuqma@R2diYa}dai;OJ
z#;YO}A;&xA?XBY`BRWO8(t-JL|I#4$s-cR=#sYu`!Auky086HZcMIenUWIt>a_2{9
z1hr$4J7WkirN*NBZ=X{3yCRSzw7sN3tl7~OJXhp5<yZ#B`8>l>7Y#KvG7y}-2<<fU
zDdfJv=`&PJqKClyj^&iqdkS)z20+>Ck&#x$;w^q!{G0@?n7=L+@o`k<c^2n=d#BGq
z$TQASiJ`l-!#EiBj~L+Hl-u5%xHFt7XZ`6x3}EOV(5B0i8ZtuO^}|)c^{O7#CI<`m
z0vXGLcv0I>#Bl9F!ow}}XhhUPBm@{GJGo-Hh`S*MM`eQ2<@VH&)t$xWVfa)P;0U<|
z1z#0+0}C_{lwibOQSiD!G8ge-Z4#7*T4)&na?XFrz6u69<JN14OEjQ!#XXl_<mh6w
zMv0cYySt^quQ|wgD!ohWLZnSVu%W9{1NhKdz|7t;p}*^k+&T!XyT!G+zju~2xF?=B
zP~Nn9JipKRxhZl=_Kxj80B{hrC3gR<x}g(*u-6?WOo7X~IXW(CHy8=wI7O%iYPqW&
ztcC(CV(U8pDExrmZ!E)wZ3K3ro?=tT*w{j0$CqkbGvInVO5IYw$)UP0B*J1G%ngob
zV5y+ecO8K-K&m23E~g_l-CE9YnXMkT0HS*cr4N&4jhOfbUjVoaNy}N#e(s|sZC_NI
z1Qn*#NHRoB^wOS#y?-ij<?o%#&3b50;@#f}gaC|FuZPI`zjD+_WhKCb=$Jr+eI&te
zD6^UVW|YnomL+JESA)v1Cfbxr!lw9#OZg#@#9zb<+|7`~_l0z;Rs-U-EOH-EM>SF~
zClP=xh&#FUu*f;^Wq>?;XBqI5YYtQMx&f~-Q0Kf#Qi|4wEGyDC6@ZLx0dlgOcPxn;
zH*JXkTlH0yI>vQo+zLUGZ0zjf8GM1uydTVi$g6*GcP3F|YI1^calHfHw|V~G`RE73
zwcrS|f}!+X3ygLLkJ;&_fJLdYqylCaaimea3FZ9nG9L$j(=yOJz(RhQi<|pJaYgyJ
z9g-(+J0wIu2fkB{v5%Enl6&+rA{HQ9eAn0Lax&bSQGMA+A(YU_F)U0P1vXbas*TL3
z(oqGZr+=j5{$1N_NZUasJW5A$v(XOiTM}Ax(_@Jxjx&vk@{cpH)Zpier`?Al*y`;i
zeC0=R!B2=rASv}@Wr`>1f;Q48tYrV=Saj8gY(54xl;j$o+psAykiKaE`bouO_PPg2
zP8$N}4-_$~@%n_9pijmOQ&m>q2-}Y7IXj=L2QjMs>_rMjbx5lqr?n6w`)@$g*!&ri
z1UDd;x%pMNC*<1uAe&KF6)>;aC4y}1+k4>-=9Zr;r$6#Tu;yp!{%a20{#}yYu8P0>
zPFg^wh6hzi93r8Is}9^xEmRX-eeYEVH~{ak<Zl-|eI2hs1pjLda*}<s{*Af598ifA
zW|?Z20WSeA9zqS`H-ek65ySgaC092hh#QB>EGL0YKdB3N!8D=cZil#hH^4-_RMACX
zkpDmIy?Hp*-P=AKQK&>{GPE;=%qc?yMP?a7#xfR)OclzoqcliDQYmxMAW2A)A`z8P
z36)Bc2ALWN@42?ko$j9ddEejncRcrTyvOq&(YE*ctaYvHI<NCQFBJXgV~M*I^TXT@
z34T+a`y%&J5OrHYT2#0{Mc)Q2r3}iHTX7^UJY$$BAh~quAjm1*#2t9&@-J1G2QvR#
zvyUOI6*v|`M=s4I8nI&Wrc{4+=u0w-C*Ju_b6voDMpqBbS)@s3)f76=?p@MH>FapB
zh@6sd^n3TSa(VlxF%!kk(heMeTCg>%npW=&MXz56H^*-C&03UjLXt*1!k|eO>}AqC
z#<O<_2q?7iy2Im*gtbKSWq^rEbVdme7;KB~Kh!lH;8SohdUEd7740K}-jJz?n_q~Z
ze`-W>^~>eTQA2Z9=&yRTY5}dZHU7^^TSQEAq+)YHFq3k-t{n%Ir4o4%2uZ90O-g^_
zb^zBQ_FCO#T!tDq9S$z`02bHr^wK=vg-hXq>+#$QK8`MplNM6>Oqq#_V;LUfnQfV6
z!Gj+*)PmQM^g|*QT$G$`AI$6I0S}QH`DEJ>LYy9-69o_m%4q9it9RTq2+HI?KK9W#
z_D|4=sDvw{dS^+R1{Oc?7GM4MY)XqJy_h+q#Tz(gC~}+INHJJKVnfj}Ak&#}53AXz
z1Pl0M8-m0kCD~RpwCsCJXiCtdg$+4ipk|^x$=+gl*<p|G#|F7S=FyjEl{!zV^BUp>
zQJnf8?_Bg`liVIuEu)~&AOtXEi+eqUML0a;=?Qf9#Qkk!sp}(dKj5d%{k`uWU8wc8
z1)<y<v>P3L{i%f)>k!E!Tfcru#s$$0cv}aOD*b_FuuGdG@3hbLbfMRZfL@mxcLPPi
zzvq0pMB19XL`|XW5$)dk{vX_1Y92n~jMg_mMtRD_L%>#;(n(|=JoCXHW>gYLWeX5#
zp5%Tu+7C4B3$qV&0c$c+T7_D~=#S7iob%8a7cXVum)01=d&GeJxvo^bd5)9L99(Ds
zq!gcCOqg9^5BFhtAl6BPh}TE)#Vv)KUT?~C+?$rv6(I&Fe7>cq&#Ni;`|e~3Lu}M5
z&7NyGpJ^3Nq5_i<-<9KjbG3GDCxDOz>PI)u>hjk%hVErw8?p)vdmBRPY?<`I_l-}A
z8Wjf8w|t8Qx_n^*g+<8!cg_Y*yLH8y_8V^{kHp5dPH8hj&603DjnYEzEZj<-%zb~X
z;7*w@yRmxq961M#nsWpf2H|bmiNs?OV8OXQK6jK`!c@#@@tUcv=&3UhD4!%DUoQb_
z?Zy@UuA=Gsj!QP%9-ArGUYS9vD?s)4t=|6Kj*3qdA6GqGuGYo?T-S5&OKjX`{-=5U
ze%K&hJqD-UiZE8LRh>0yVQg(+u<f;ZcPW00<_yM0dUDbzAg?e=Qh|WA1Zb|<;pJZ~
zYWRuXs}`tk(vR)VN^()5Ij+xg(usEUjJ}9oTr7ZxgKRFZZeqijV+_LQ#Mr?>7=3f|
zv{OeZ9)DFF2{eDNcL!>{Pr)7GT%F78rrTv}Qzvi6Lkf8S9IcO^ACPzCp*rjGz?3?z
zsINj%J3y}TS4^AqYlOyS9Y~sL0g@?z13taSNP@UciGfRc*BI!6p>2#YH$uCW;~Sz{
zm(e$@|J@l<e?3o8S#W&Ox7fM+$v3pWsxIcgn5*zIYwFoa-Yk7pHfVYs1;qQj1Q|=9
z@!(=>%}74+j=iV;Uwxn%9hT|A@98|l!I<wOhQ+DS3iDo;`h4I+=4#mk;UK3wG?;R0
zlw1WQYO^%gcWmuxdB~SEpuel8?^M9Oabx)i3%{&p=@M~tL$)+zuSDC<Ia+T9;{)Q<
z4BOcRpQ(>bMi+?E!3f9AC~J>vA}%*b(;j_rru+E|OEal_YZ6yTb)5N%6zYs=$E>cI
zk`_^?@}bc)o-VbUW@CsXX;PKwsLuMYC&x!2dvkKlBaO+lJ8|WId?!9)Z%z8D1WpUQ
z7rfg~os7+ZtD;PNDNt%x+GnGC0Va6;tkE6&Y@&?dGQ=&+DiCyePA5sNa4EK(vj#zW
zm&&#dVN!smOIYP*5N#@9R#eR!iEsCEx1f?}#o(I7e<}+Sh}up49ZRjrDRTHa+exi7
zLLpa6g0w)5Gw{i^(;bUudTeI>w%>!-xIZ*x8@A??`wfiab4d8Vkw|mW0vXRvekIQ&
zM0*k6iTb)kEr7b<XO}S-Uo_wRk-Sgo^_S!$qt~+9N2dDwPk;dE@2JUhxBda^e}kTn
z`o5S;204t@5QZEuy(ML$?}sP%k8Q~Juzhx8XTIw*Z~%1rzkb?$LhJ8AR7s1}e-Aj*
z(p5leG=2HV{{MUZ*TPMO0+=`>6F*iW6((UWtpCTutYX6U_eS8;L1PW*<$?ZGpN_-z
zmr0`GPp}1-TgcS*^~*Hs0c$Jz+PJw`RcWrybGM*Jgw6&)k;XnPGo#c$wCH%LwQMH%
zF6yUfN_E<@%w<JqXf^CX9U4=>y{-geng)WynIGFlRFtDik=lrei|YjSzhAI6i?E?E
zZiaYMM@^bGH$G0>B0EW@VQ)Z4)!z%B2SD1>ldqV$-jF62sxx}%q5IFGhfYvTU$FVk
zeW=*yxg0&YgctIkxlXOvojZ?O1aPV1`uxRXd6WtpYzQo^^0tT%#Y_Ug=jD&;{LPDI
zwE4B6&zttC0OOCGUKII{h(4&=ByR0Lv7!m8EXyTZ8q{Pb2ks4v3lksN4q?h6+yvaB
zoxElVIx$kM=JT`>hCRQ+%gCt<CVLyLfV|+PGju6uObLE~FkA&<Ik7=L<Lq(m?M%z!
zUDy89$$-@exCAPCN1k*LL=8?o1W}txL)5$h9o=_L(PN|X`Oabeb~SWB<J%$e<*Q20
zp+|MzA$H7AHDP|X?|T#D;4k-op&d%Ue0tG(0r<>|K3+$@uyR-_DVc(fZ3xJsI1m%O
z)wTnTT+KXazp_M!ju6B|3O0_p%QMkjprNyR3qcYg{^~+8KvT_y2Z4^jBIaIvt|J3Q
zKFp1e&(wyYEE@1S_=5&LpR|23<&c<IK)UrV$rLQ=qeQSkio+^G8LgN3MB9?1Cyka0
zhxCwQuYH$gKvF1*y=}9ZdzOIr+q}mV>#wGTE|yB<z2U~PK12eqfGPlt{VcUXpv;1{
z2RpH`nyH<b@bGKxSu#H)gdfeUY<1<lfeh>XtdGst7UeBX-p-UlH<#1!PZ~26;mFo&
zL9H5tL1EU|xW^oiNywr`<R#2$<?iYk0bf6gOIWC7rz4$C5xg`M{Ppuhs{?`?etz$a
zN^KQew|_U^k))8aJGLAXb^Z3~8P#KmriQSX1WO9zJwdkD=aQIs%x$9C4xSgRVB8%o
zFg<Frwq|r|_1mp|!JF293oCx^Gw`*(de;Eo1K)LK>&$}P+KZnDe%Nu4XW`6?pocHO
zf&$#XhB0tXFN*^?PZ8s3i6qT5A{6XKJ60E)TF5zO4B#0%609VTDP?j@yM6-3bf$IG
zA2+ONg+)89w=IF*wy8#<JLzq&&glyqD0AX9j-1T!4Oe0F$;)GAhEXh7N^SUm9A=?J
z@emdfn_2J&*mB=|k=q6*AR0UC{&%?K(jRKe`4Vs<0jo&lpsAhl>1+EVb#w)|7r#tE
z_lV84Av;-Fe2eY#eSuwfW<`dohnaoZxtq<2oH4j$p9Y9>*J3MpN@*!^xf*aoV&B+O
zLM17jb!1hgCRD|5(P!{AiIpF>iO5chnm}|I`V&+N<0sIP^Qe*Jd=d5$#T1R-jD0Jb
zNaOdQ(+Auhl}l3irZav<f3QH&ar)TZ#);4n#geKq4~*o6<ZGUDLr-)llY?)#-sh#~
zF(<py$9l0P>!=%o2hG4gbZ^>-y0{epX|9yxgCcO7oxgs3)|Qstx6Hr0BF1oqIB{`?
zDbn7zTDlFmGKd2=QK^uAr_D8Q5sM4RRw4B(YuTQA)QP1vJ7_R~6l1yS5ht=fo>1H>
zusCt*=o=jkPaN8`SRxuHc;m4dV57pw2o_qtN(24E2oBBjXwyl8Lm%0){M8D%XPU8Z
zR+hjWh2h)Ak7EsQVd4Ze!#=iMh>^u<55CS>_4uU<Yir9T#f$#9l?m+Zx2U*&EP{WR
z$o6=yl4*M#3RcooL<_$uPHvNcYEUza0<nkm@h<Db)6|3SPa4cSO_EnJ6F;Zcs$+(c
z{7Xdvk|_+~Vpbz8U$E*AS%m9;1Xp?&X7U*KOq^2*-8jD<B<dRc2kX+oo5%tGrZP_I
zWSy<p7{)uc+fUGHXrCp8$0)Q%pY%WU=#hL^z<9u;X(MQ=tC`n&Udbq#a_$9k*K5zD
z+H!~&1At_wg5VsC)mZEny9|?z=Ll>~DnSM$MZ?Ep4FCQ_8OiM+${-b<L`6hgJUdLx
z7XBjKv!e;wY#KT)Zc1Xq?`NQ8cUSb$&hkKql*5!kElLm0uKZT*X`?42BogPf#>=#w
zi;?B~d_+gp*eiL{7?qta%3R$<Kf6(j-xypR{SRt!Y_!Do$pzOYO<ZVBz|8ZU1)YO1
zq<~Nnl(&DhNZVpq2gq6c$A=BWC+tsEf;5CfnNt21y6iauS0u7QYrv+BOU<EsNR4e&
za1zRB0y>BGM|#g``5lWKEx+4@H=ve*{LY9oFprksRX*YR%oXFVp)A1EjLY70Yq`MK
zn-nBItel$rSAel*PQG4fI(RWgjeRJf?t~)!N>hBJCpG#XLM|~9F*b&VG~_+YFc$iO
zV#B>kqpBG@7fNScg5<lbmOpiRfuR7ou<J9|E+{RG2KiH^M)l{b={ZP!rXIFjUSxFl
z{@C^(+j*6=^0?Kc%>~}(Wp#d)o0_G$3>Pg)Aud@-2T^g@Y6=$3voS0+(C*TN*l8<Y
zCNouICS15GQng-=k}j$3cvky7)Y%zXg!9Q!4Bo}s8ui^V{(#oV;Y)j+f5eeqJW<m)
z3NT^lEkaF1faIm({&U24!nkMAJ*6F=6EZev&RrgmOL(BEuOoY=%#;)ZVyVz?juWUg
zAhMPcj+XU?!X2HbUwIv#yI7Zu7qnMq8~a3~>!#*Fle%8!et$8G9I9<jWqwbZAo<YZ
z-vt7%6N>Q2dnO}--krFnUz`ju6M~&wblPbn5be}gH)vA&00X4Gi1#U@z3Bc0#F&PV
zRFORBfb7K8Ibm5x*M6=A<Z^>8!hEPAE;#l2L|l7@BUA}KEiKgrmLZN}Lb>TmAKL6y
zBm)L}*c&4<XTLcrvth-<j_IbpeH~`Xz+)f9og2nPyrT)qK9h0|8D5&G%+$CHTtiGK
zDfKjU<gDtx6g;Sh^xgj1RbT)l{_35GIYeT3{XB6=2xhv%*?ivKd+Vfl54&>4lr%TW
z<ZT{i!@jol*8;j*D<ccRFxS%OuGOccCK^0`XhL^bY8->HmvLCYM{3#7_cz4IF1r!3
zvlTxuUX>U%*!z2MnW|0Z*=2`|n{c6xrz<^|?3nzDsN0a4y(Jt!jCj!n<oevyewqG7
zL|6c+LR)SkE15gUR^@qI*1J`Ej*4$a*pS_6)!-n@0?%Xx*V;1YLq`sRWo@al<Fh14
zm2u!?@7se@xv9D2e9|n*(Z@Q5KStTm?n`OsNScH0p<}zsKf#mJQAn3;g+dl&-Oj?d
z2X6H#G>36-g`mB+{D>8YwiD{D{mTM+Y2_*9I5V<eAu!#Fx(2RD>{GWBUze3Y=1$53
ztO2%bK~c6>eJuPFoReynnk4pB%)(^6X#|DA&6;@whTTPzHZZ6&m!cz94@!T*m`1yw
z(mjCqB%o@X=ESUe$>y5F&*G1O0^EOQU20^Pl$bze<NE@lCb=CLE3;J1N+$I!y@q`r
zg>bx&t&p=YW5Prb>=y-v{~(f3T)$y6y>c%Gg$Do;lt;_gGH=Uq&RD^0M<79I5YBg_
z)i-<CRCb~aqcR`vq!mBEs1somQR>!r&ObTpZC2z(w98W>@4xSK*(4fOI{fqDa0MA!
zg=a<}-h=+Q>(e1IE+0OE7t}!lRQ=o;Yh<E9F`Ebvy`qkv7t6~lA;ems8~X^}k7Qs{
z&8L&Ag2(FnV~j*hLT3Ij%p@*NgcOBt<Mj)7EW04ks6Z=Q<7#HgwR74z9f<+$nF2pE
z#Y(Bcpe{uLiy)wU#J;=-R@pLrH-`sqYVMSfxD@auLm*_Q{{@NLEs!`()$3C;Tm9N>
zI)`JR-R2z%XV;vH-VpR-HMl36msOY<`<X56U?|A>JA;JUN;8u6cJna}BBH#lUhh81
zb=6<vfo^L=%6sS=i^kSOxY6)>%VS!QL)*^=UaAvRMp08!E97XgMfhY$|MF+J`|p9Z
zkt$$?8}ktB*N*84BHc4wSKZVdz$6AICSOAJ&|torZC&44OhzBor#&o58S-lioxVbB
zc<9^Bi8`FBkLTG80PT!WojL1rh(St!svkc544<TFCtqtm!H60$%+HVwx~ZI+)Akb*
zzXNRg*}yrGk!)M6^A#^7yW26>z3GX`8Bp^zDJ*&pvD5mrzLUy+_RXdI>H9=12OkyP
z1YOGn9;bQ%X@|7mWHD}rjC+BJmyp5WgYQ7^G?ojoP7~vFJ<F%;y{}(=XXxbVk+H<$
zNBA=0cQ2m}oD-F$d+#l!Cym$T&irn4{QJ$ng;YW=O|J5so?<8q!;QSe%FF7MlInzU
zmv8{Rs+;|4SHqzXnbRx88=1cC7gzbhw}ep?2U2vZ`N&ODqt=35=?!1>N^r8|KOSOR
zMbgMOhxa;LSky5d5$%<MsS(TeBt`)x+xmHO*0}J{=U)xYuGsWtUqUie1KOO7gLk<D
zE+(4M1$RL5vjsrl%*5k4!-$e=(s%bMAVv?S*+%Z0atQ~BMqmow_uzfXklJasM%loA
z-zw=fu;Nz@UJrf;g=@i@+>ale)CW1Q?|2#b^h^H2d9KZ$7?nZmH=es>i<s)r_f*@f
z_Wm8i73dz8*6LtOwGzMM({J=Blb7G!v)-THr}R@Sum25E9xMOo*H;M5h1z|ofhZ$8
zQj87`3HCejesd@Lo}kG`3EN31VfuBi<$Vmwt53G9kXt?Zbg5<ujLA#FZb@L7@q%+%
z3a%-xd+y#`a9Pr}&{V$g%w|s)&g&emtAYsu4(2nQRZSu<0JWnBzC$R4dg6QM#>%4$
zk9};ssz(Ze>nGXHc)mNe;~uz3j@A!!+SI2n7e+OHRjLqto6L2WPG_{Rt(TmG7Cu!R
zL$N3z9`0|nfoVm=P)5kiM{&VF-^{B8ao^?H4BAqG4`Q@Fc(89@UuFwu$4l_VMNI6b
zE#9c?-MX(p1}wR#1d{6c*r0RpuApXNCzE`?|4+pJ7mM3cL_ruSNf8_QY=652Mi=ln
z&9rl4<o-ZT7M6k^Z=axdr2~o|<Mr<2cX0jRcfjN%<Z{cfgdOwG+%>}z&W`Qa4Tabw
zaf_7{XB1LtA8UMCMwXB*VVS=PcyN*)3;aRz(t-P4)+<iu=5%eYe)VrA)QOOxJ2NRX
z+zh@+_K+qeq{jPE-br+aC0us1J;?@DhuFcww@T&0cXY3v3E}^J8}M^c?Zg#<)jGHO
zfM{uh>_Fl-Xkn~`WeT^Sd+i@k$A`XUT-fnWmPa-j>`H1{_CpeOaDGc`0S7J2e;ITg
zKBp!UrWururLr(rd4{cSVd=9<6N;z-E@%Ei0$`&_Iv)dqPZFsjUUO3}5s=>lv^C4B
z`TL4<6mDqjr&O(FJyO2h3R?Fzs4S)o1V2KeJO^keQUAHT0g#e|I_da|zyQw(D5b<b
zLIX5D%92NBM0;cfETfNX#F4bKJpRPkjqrPb82Ldq+%waY-OpBPlq91*;9qqoc{Z2f
zn~*gzSbMbyAoz}9J@$%OXID_py1<g@OnZq>14}&V$jDlZG>j+@7=h8opiNo$`3?kK
zX~1f-zSY@><MG@dbVHF|ePVG{e8oKNbyVroySjD!;kCnv$lKFe>8Rh{i9$_e1G5i-
z)yR2>>jmgY2EK{FWiGC_q#Pz6X=t%}W2B>6bl36aJGZCGz{MdqZSoS+8Wx_k7C;y5
z!py2>U&8MkoHaTB{z8fG>d;qdFpJ5`5GV~%`yNz+Th_#pn@>;XhA@PLIU>)C>}OTX
zZ|usn<4oMuy&t;dG8;2b+d~gQ)p20j!6G_TEDt0>C}gj)RJMR?b1G((0abW4t6^Q>
z3uGlkyS@MGl?!u)cp3!Z;Y3VJd$Qo2rN)LzV&%h?q9p9Bx>Urq@2iha5_A5oMn;tJ
zs4kB4I0gUtag34{nE24fsjad9CcZc1BA$!8|LYT#WpxGJb;LiSMK|O_=0fQWQ+$^9
z9bK9}_st5q^t~rTS5s2pQ@T%*PP^lnMTJmS8;vlV{N=C%Y^of{TG;DF9?Qi0FR}GB
zIl{q1pA;<#Qpw|JU~Sg5jJy&Jx1gesh76Kj*mA{LgOj3Y(qB!>JcN`*2Z>N=t=VnT
zOM-($e(PIINkz+nBGEv7p0ZmKEnMUJ+2PVeNvME^jDjV`n7K0v$!^M^a7la#wJn*>
z<GYf-2@j<D`rHrOseJ4GUNMsm?wp9MqZ0VB2+FIt-kL)g%dJWLM|85F_JOI$a)<0a
z+HVBVNX7v^G=Zn>9zwliX<b6O3KoO|Gdb1OR$YR$vVE~YZ~0F8JvngsjOKr{(1K*4
zWgRx@Xq2R5p>22nut?Z{c7;P7H(obN2ers?rs~_zubtjF-{Zi#`+*Jzf*F){F$IBq
zSMrozj|MU!Srh4a(fo;Is&mjGXnLM2HV=tNM?9Np664H_66Qy71n>t0I4ADn*B|dx
zhsDVh?%Umrtb_ss3duciYbNTHpJf~J#JqKj+WPWGsaoj($cbZ--V2hlNQi~S5Ga*!
z#r85=7%?ntdziR26Ph<JGJeW<P(g-x+@KYEJVr1p80bez|IGeliP*VHkPXLTdMHWH
zG7X`DVVEgzMX(bI&r&e1q*j0He7Th`T%_w8@)InxI*3}Okv}jRIEG>3q%iZUC0GtS
z+P;=yBq_<PfVZdUklpjyYk>;TFU_%%VDvJe{UDEKYw`*6K9q>tje3R>x8rM9z8cpV
zK*skO15k_bT=B8IuYPhXxBKpaGH!l9c3y@U083nx&X2rwnWY_s%_cD3eR|_N#)Oc;
zm7`$>?J+Ujdl)f#pkcc)GpiAmbJZ~<04V%ySH+a(HC^4jK*EUXZ~W~gdZ6*jl@=R-
zlM!zq`}dolW*r4UyC+>s&*f$Dja`uc#C-(M`W`ALNmq;!4<*c==rc<oqtZewT4NEA
z7KU1^65ctImX9F8>jZ(o>fX7|*I=(mxo+nr<KzHQmzX#qsMi*FIqIEg+^6svzp8(F
znI(RCW2SnaJ5pA1)#$14;`4>fAUVp*iT_HTx9^G?Xg2Vy(S<7Lb-@Ct%VzS_B*Bq6
z^4p`#q?4?J$%E$WJHE?oDBe=pcxBUJV*eC9(>0YSPjTVtk%`3QQjJr%DhKWZ;cbcB
zz@4ut%b@2li5F2f7MOUGi3I4!-jJ38db5@dM3Lh?Jos&L{^NTj&32n{p^g}YAXl;0
z>vEy;L5e62upu|r?3S#l(bTiF!MI2z<6c86o9z3k@i87sbHo6;hQd&U>sjy<4_82B
z8dLo^&EuJvIJCvlr)XDhrlm9OM|aQxogI$)2LHTWSsG~JH?Ph~u*C>KD0}j2qlo=y
zz0}U?Ae>M%5f$Wa|H$Tjx^`aF+FaJpTx)$>l(#VH12og0ePM(;l0;QYW9{4ioZu%@
z=PFcW?R(<yA=;~@5`~LyjcLoIp4TuH0j*k+RJ{J{N==UMl|HhYJEMf1*hjk<H@Wkq
zrNS-K-Wl{Zkq!&SDpDhsgfLJI-t@p4^C-@(b?wI`_;_R*w+8cNSsBcw1;y^ZPTmJM
z$(ujB_4JOz@M<NJ4||=pydLF)(7$`*_9atoZ+958qNf!LWw#z2)Pz$L-G$86&~6ky
zMXNbXM)x8$B8dH3K;O9?hpuJpS7VpX78`5u{Q1nQZ6pHA_~VWa8<-f8?oHEf|5jPH
zcCb85fEaN;f^&jq%>H1Nr*+>Cugv+12?<X8mXT8!-eLI?tc@)O7j7Q)0wCc5cj8SZ
z(h0^bt+V(3Vhn~Y9{g%eHql<^ZBi{8R|Tgz8r_<N`rDz>dv1k2T4tIPCiFZ&6%~h8
z%kh|qKJdCL;R$N56O7P#d$*JYn?{2X*LOfLLKsk*+j`Dx%aaS1vY$Y;n>uAi7kr=y
z-zn5t-pY%s;uSp9qp_(snWlgE3EQwr#M{!ZK1)A?<r$0Pd=JdV+Z}AeM3hQ~xRPm#
z+|o8mQPr?2PfMb$rtj!#{?Ar$8WdDq<&pdgWs1J6D%<}7(H@LJa#o|4?(i@O;I~$6
zUei9^{mvydcF?|KAOf{Ik*Mr*hWLTSDJ<ZHhJt7W^^AFZlaT@3%7wpt1+s?SO>S>N
zvWSLIg1!e`YLciIY_8R26~=cH#g;!h5~W`cIXBOOfg^H=qo|MfJl?cM4BUDmZ6ofO
zUxrny@7}=;t^{RZsLw)G492`==-zlBeB^53IJ|YD`xq1Jk7qV5jnKo^ddn)gJS0V$
za^@x;3|d*FD?u1(IX1Xj;<)ZMquz~w-yk;;VB(?)g1l620w*wrR?nxsy6}Y(R7e^I
z>xV%+AxL!!;{;-zh(3w*;>}O=8{Ww8SQf`R*fu!G)@p>PO^T&!uB_9F?J9GhiJ>P=
zOmR&du!m}aDB4-Yh#Is)nK#|z{84)CBZ-T_gHah&SlcCAUu;=->cg@=pElexq~@?6
zC|CJRsJ=LiRWFB!c3cCO+TuoA7#Q=j-<oo;qeY}GMGIyx{*_=B64J?d85_m79vel%
zvmEh*@&4Szj`f3=?+RQ=Q+`%)OeQg5>5c~`zOY>IZgS6Ou7jtD5Gtf8SOX%8LW#Rc
z=Sq_&91B3V%JJ)R(A&17*(-b*0JAs+McxiSc%shpI#wMyfCA|)84gV3xyXuy@{(r|
zE;PBRp_J2-!stmBVKOnimv8OAa<0@$c2KL=w0~O|r4E|zYGzGY8R9uGm3K*XLhP%4
zlDPSGy|#Z#RJK@;DQ8r>%D9MC7%n?q`f%eoh3nH4G3XZ-YS=9~ri$=foP)NG$o=z&
zsv~zM&4bzcF(TeVP2`Mqq&iid%}J28-z{=z8g_#RZ8sGEYrEn2shZAplyGd=aAxE7
z+Vq3O9sB4r(2dB3jtc}+M-T_;O4^P7j`Q*;KlB>Oa1i7IxoOYb3s_ra(EuVW-}FM1
z)s<ikI_dPd30-+3(0CYKYJX$~>Buj<-`9%hZHF`c#}_x6qRYvHCoPR$z77BKgD(Yq
zkY<>;VrC+l3JiQqBT&WGl>eM`PYUDGpWykl2_MHqYg(ETw{V@bGWrHKKKB>L^}6tO
z7WG8-etcFJ{XAKHGSQ8Q7`Ova&;o^hlg=gcizhO4^5+Z|xjd_UrMdC$a;rsn{4of;
zVa^NlLDvaSkHIxsDUL4O^1DaX*aIS`+~{6|4#IJG@^#ocO1lY0g-($?@Xvvh5;cdk
z1~<0FH_W1jp!AtIqd!o-%k-vPGU4y5DIm#t<n%i6L7&2qW7LJ^z_%>5;@+$(xrVI#
zoq6nNi=vZRerfVcgax3#>BK<MenLkO2@6y74#$20)UHG33Ml)OM?Z%im#EAnU13yk
zN@4SpY9w$gezbBKKin#BMXWRkR`nK*-nbMb%cJc>@$*YeL&CMGoVVEu^(#IWG%a2S
zzLHRhW>Q4I{|=l#6X@X7W+b9qYo);;A!EqZ|0E1Df#K!oO29QYmW)V39FWokgW(wL
z;Pl`oA@%;q5@?o{VSTqw(h|iv_u|V}0d2(fE?k$T4Pv!3v?6i6;F2F?>yE+&tyk#p
zh-Y7$>$;ZQILgl~4$rK>3r$I3q`>mDI_IHOezPDdpUY|fmybO2%Ks3W(+u0_bi$K~
zcymQK2Tge$=qx8M504w&D0G_^kBCgx@jN$rHTze=Y)v*u<MtE!fFpp^&o|{eAlIe9
z-mX9#{mY1@fu_>D?Qa4)CVC%Oatoy~`?ByuA!K|kYp5ChWxm)_u5k6^+Dx(=>VJCk
z9-7ts8%<%h4W@!|y&A*&_wVKHN+y6`Y|~Io=K7$6w>SNK7^jeC1(5OE-6Zpy7uaiT
zQK$<G@nnKAtVvMflv4#xW(Ez{E!5oc*@N{xDWt%Ln-h23EOC2~h=?AEuzArRTrDBE
z=I$W+AH)qo$fz8;STa}UfW-!i@zthmdv)ZuZj{Q|vv?{9$<;&)l9zzU#s5m3OKG_t
zQpY;@NTqEPC7b|OXk$^}tHu6d4R**OH5LFhvv?Zj11KRZJxe$R=$}xiWa;H7TnqZS
z6YR5zUsjEzqxGlY=i8p{XWPh8h%jUZq+Oa!tT=MjxoV~mODTHOtC=JB<ux#SH=w8^
znF-jW{CU?^iUDd`1537JGLxNUw33U=sUWt{M8^h<=Fk=%zgJ*&%FUXVwUqxGk+7ge
zwR<45g@d@OF6}H@3p`_{{<^zGcDERzrM(UUx>9_WxmCW06u`l&X_v7rE2U`h>)5}w
zElb{q4Y~05z~rYRlr8oPBDU6t$Rno}x2`u{Id|1Wwy@7>yZJKgCU4gRK=T&lwMeTp
zT<YFkuUC_4rS<Y{Q!&o{Tr>P{$dD!Z&KY^@36B!t+v(wH+n~X|b;b-7LU+jaHeA*f
zrXMdmU(!}YiT@jt?78lH%uUj?IfXJbNpysf8^XIVpgK0GjlEu=EY%{%&R$0Bmi%Ea
zko2x$1`tKB<aICT4#O}0`ys1#G;Vg&nwb3|>JWpdgdI?qgV#92DN)jBn+@tmIm^3J
z4;#ZkwKb1K3k>xiQQu{A%3}VTtbO)qiB~*D=~T_Ui;h8)y|oPuXYqvCqjvDdMs#_e
zWAG3{v?7oPNXxS!ZtZb`WFhO_Q2hP%!Kqazreq{`K^SO2?sDQ!bL&C`^b)%r??xU4
z#JHZ!Vr-U)J4(!3kmCq_?(l6sJWo|x!bSbULAMQkq2wg$(M}@rq<;Z%`s*a_poFWb
zBDoi`E%YTuIg_IK0%132O&|)W(Lz(Z&qcA#?m}Pw9<~Gjr?s#tf)CMwT$e|y0BpWh
zf%%sP|I-aa-LnWQBz1}U6kD`l<b#KH5fQQal?ZpmH$$y4hI3bn3z~C_$ZdL%P)m$H
zG0P$~mI6!YXiM5cNdH?4Q9^i(NbIov@7OiDF%H3zAhc=wLXyD~?%6FsZ`?;n12_v7
zhjv!EM~U<i!uwWqj&2EZiRwo!IPE7`;GRzaAY#?e8E+2>_IBF9o0QYnrwS~yz*XpM
zy~N~9#~Zl%i=x*TOU!sB1QA{28CL*A(I##Pu_UPP0^{#iw}jYjh~7Ax^Q=KuR&3mf
z!-m>IYWwO#MFeCyq2THe5Bce-yu0tWH^U)xXR1tx1=3BPt1=QHp=<$i$V-huU!6d$
z=cY9Y;S%;QGznSCR<|JQA>kZ_A4b%gR<LNdoDm|E8o~FUO`V9=&jL63n{{siJ$OSI
zD-^Ge_F)<zx-~nl#x@oCUm+@XQf@$$#?!aR97Nkg!DMa$noo~RRy>);Q)7<;A3VzK
z&(J5cDezobFSjnkXeGL)v$uadSJZ7>4qNP41n%<s+_{O!T6oPnzT=E;WjZJZt{vtE
zt*V!c6JJ|y_WW|h_5R2U*7<8dciNOHgNYl8<@!I)1tZNl27*LE$K*X=JpKUVG3j0W
zzZAIWU_sOuT1v?BFQtU^BuLwjR;`8oJJs4KBrjY<tDIK-JC)ODb~$4`iDc~a|D8xS
z+Ld6Hp(Vvp|4#NnZ}U;Z2<&xQ9m9P8ji^IMZ`E#3A??cfAYzZSD@P<qMZAt~W1`MK
zcaclXi4rC6lqeTNIB?!=EIi>37~nNg#ekHaa%3za=Y#O)1*iK%5>gC26XR1RKC~CI
zBevtW?Rdr)ACy<LJcr|d!v(;d{|y&ojqv{s7i0DK|AvdvI|OCX|IfpPN7h+TfMTGF
zyVEe^`ZXRU(2+krzy9=*2VvG#Cn(n$Te4v=u;5FCQBV<f{IxW+uPDc9*_X?I7)}+r
z3>RP^CM*X$Ef(F8Cyrg!Y&ojv_O~R4KnN0iBSj=18J6e5xW>2F4D~;Cy=&(*LLcwD
zz;C@l;TT%-Vep?NA1xb+TQFZrfN6vpyTD2l+SJcb@=m5dPu@S9IPnSD9QX_w!=JFI
z|GU?2v;H#h`qfPjAK<LGC8T|hs0?>;c!HCYyi|cR6(83jOz+Y2|Dv$|znY(hoZLUU
zU6d~zcQ`(_U5ppwS}xO=j~GI!VrsTAymEodG{KdyT153T+3?;nU$xnB?h}ImPH@Wa
z7I~}^BuSbKqx}(x`Tra_jBNvR^4yOJkCx+h3&5YRlq<6_jB(@}RAKwfH5h${pQ#r<
zx=HZXBoeD%H}f~Ly|%X96=3oPk$TB%x~{UU$9iScA3K6$8&8}jr-+@)#J;2o>hu%N
zxy9zyWotKFbEUb}ae=?786CxHW60&*@cN1CXRbGzhNUbnLYOsmGC4`;w-*-x`Q_Hc
zGRchHYV4LFx<{?(NQ_mWLHpEyq`#=i6Gf5i@%ze4hVec+61SP>m6g?9s89DG&&3hT
zF^5-V6>HYn{j$3Ky+;N#0?T)wrf`sPZ(Y%6ZO1PN^#UO~nP+S)1~1m>MbzA0y1``c
z(2JU-<9~O2iZy{MOzceUXK?z6d4Y!G%Gb4g%ZdGG1sqI>jS)vEAqCecH1GM4xi>V$
z>llU%`*~tA_6k5kUsz9U9DOcSVQ%95LuPiOV&@^&M$Oxs+@mq+Q+SyPJ5c2rIyAIm
z=6{~!i`_p6T~7}~%1*%hy`Uq;IPSZ4TJ)|`|2FRC1^prhOUH(T5#<l9)PeB;@bp;d
zg)Isj()DJ6Fj+`znyHm+MIH8C>hp!=jLJ$rDW<RKW(oTTPNlpX{qLjY(%3$t{orpJ
zNUvvb5-Ma$r4jNjLz{DW<r`D}r*%rV7Pwq#ULe|D^2nVqT$i=X7d0ci2fR$kc1AFg
zqG4rA7#OW+W)j}kdoUmgci?)ls8u-4cME(B$>$#MI!0IMZuHCsJ4wbF4^-uRJXY&4
zRpUI!Zf?H<sDf-0r$33W{`?VG!e}C#3#r27S#udWs?~YU5+wpKUpzAtU9~~Mx9DD*
zbi~JW*}+Xz93X&Ux%!*<95rw>B;jwxG(#UlcXv;l{e09bWE~^1`vM)x@~e)XUJSv4
zfkH^W-%`Nk$y-6a?D_7zu4CJah~*qKifYeJ8tD6fPQ!+OKZkCQ0}AO~A?*V6V+Y2{
zxBWVbX?z@vSww>=>Nq;7U7JEBVs$npjzXAv5hq}SjZJ$(n$*}M_DpFW=Dt&XW6n-W
z678=1M%%NbmG*xY=MWy|BhCTz7>RSVD5k9f#zBmSwlamuhKSD$%Gv4rEJ}?LNd7T!
znQqm-0s2!#9RJ}g5a+w3nY938Lk>845gUGj))e~cps~@~zx3FFjh~D6#A;G<aJ>j#
zFI$TCUzY$w<3qO*UdjCPTBrZp$r3c{+c^}bA_S7qSqHi<h2ll#O$->VjWb0cx^O$B
zsC7_H7$N8g#;A=1WU3Qq0cKS@!8Xx8+y^ieM3{aSDy$Efn<sf|U%mXDOdUlZFJ-qf
zQyiTrGx6k{pfe{d_fojkz%a($z3PVz?(P+#bqSqxK33-j=DDNY#6QO`Z8Oq|I@96j
z+5CbnW5Z7+Ohm~)YI3rZkvNtBVH|iNJ#DPrnJIzye=c&_-9shA2wmS*r#zvDE?`o$
zDbY2_p!NjP1;F~np?yF0(LVw7d}L-~obfe6M+Q5KbgR=&96!bE&r-mN_2QKpz6X+u
z>Sn70SVtmR=1A<Ecya8H|1pZBCDBMKo?=Q)<31oJxqarkI(w1|Rm}z)x9n(QIbFn8
zGMNhevjPmwQG*5?ub$+`ty^!qQznZ<>pH$#O$j$$q&GCYm@jz)V!{`_M%$&hD&MVs
zUq&BqGo$mm#$oyk{(7bU@*ZZ;%mi8Q>Ew_5J(L>g_r3~`ny5ypBT`TUZRXKWrl!kQ
zHlt$x(WD}_)+pVLd<z-ZV+z%+g*sb5%#RAqKV5$&ab#hVUhddKV|<PN&%cKy;Y_kH
zF|j90$3qRd`18gC^HV}uoX~EjN-j7vc9l`Yk>%N`!hS6K+x%J73jl7_%{$81qetM7
z@A#0&LWzUQj5h=7GUGne*ei@KMYI}$FLczZ=KmJzG407=6+7=K&62g)|EIl8h|C?Q
z!&2}!&Eb`_X%hD>TF_NHDx?*ujm|6?Zyk->Ct^qKg);D*TS1LXDPG%g9mN04_?_`v
zAZpAt&M2JZt%SMQUHnS^-D!jtQ2vbqy`5a}^n}lTnAuBQK+?{h<$bxdZ+IUw^#bUj
z>gJUjv)1Fve{lFl->i`j**pBXM*WwkIeJ6?^--z9G%ZH-7w_5Qo`TwK3@wWHRM!mW
z{9=QK?YlJ75jU}r1&nAu93c%F&{;Pzd9j#Up_@Sr9G$A;<%nOQn)j+ekLCj67AY<i
zAJXB6wj-HCTZ{hwy1QPWE0Zw-WXcg~mI-CwhKH;SSem}sr@<QMrO_FW$!u7%TQ(yQ
zl!sZ8T<)t4lyG%iAY;8<)R|RTB+)%M1}@25$K-xn(%v_3INOof5yQsh1#M>PbTXtv
zb*SsGy|a)A%A$49;t6%EHQ2o_rgNb0!veHuB2Am4&+J}8XcaI~=OrpVs90;-)su|i
z;F*rXs{AGxg!dc|FBk1-NRafHs9Z4TAO@-4gF~tt-=&!1_$uf;)rnvM1Kz00%3Gdq
z=H0f-&BnBZ@oGkAGv1x#r*oZ!-t4yZ4_n<{Lf2%O(;RQPUBNvgWEv4o0#C1imSOjB
zr~c}QPA8bvL=-OvH6$A1V4v;v+xkwKzdR0OgE`K`!j~{3$n3uIypCZq&6|*1AyvQA
z{d_HzkH|jXH|1Yy_O6mLZlX-2THwm!IB4~xQ=JliLGjF<@AOQjYjAxHVZ@Kd^1pwj
z#zxBm_UsK&AHBp!c2MpNoVoL_k`|#Ifm`0D!IgVtA6Zk82<U;#<BZ<2an(_2+s1f4
zepeH}`cn#(euSulMn&9<jQwxi90GASNWI_A>vp@~goelZ&o{oUhxjo_w$eSD;p!U@
zBAgqt?Yr}D4&G-O;<zA*VdZ=@2`c>C%Vr$u&)ls;`lJp&YRbSu)&;pks*+#DHwru`
z%i3z*r8_>yB15@{DNJ*xO4>a*oVOlK0{@=-imMGe9=dpK{&1#<trtVrwmQ}C$@l>m
zy~9Ki#MyJ@-bT>TFmvqbK6e!zQwDC+nu$HH%d3kv*aeayHjruS$;PX#&wYq~k}(Jc
z^Dz!;zqxR{KxVMSC&S==nwNlP)Lt|%nX9y{Do-uVFB<rKy}L$z`s#b4)2Q*}d2m_B
zb!t$;5v{07NckfUsj+(vIn*rs=jYJ@qr_U-ZJHrM<DW+_c^;R#k`!6W)bTu))w_;3
zj6vS*R@+0SGfrmb+qb@5RT-8!$B;|tkv^H^`Lzs>-w{IlHXVboU^wfS)~1S(Mr3`O
zR}bmbR=`tox6TpK6Zx)nnh;)dNjk2-=GT>?u%*F>aNK>o_kul4r~sD}UQ+pB)FX%~
zBhhF?X?#2!Lo7|Jf(M@lKbyIYw%j$z%?`FFmbgoR)MMN5+0r$Y(FaY?1s4p?vSJFH
zBIDNhs_Sx#I**+`Ee(m4JYzJ(Y0nbgx-3VWfeE>%kdh9lyD6W!GJ|3DqUwydWWQ28
z#k@N)f{as)mvfYt&|JY7i4|_ySNUxfuDVkRlLF~g&#ocC8ovK#foT$tZCYVMV}3WZ
zK>2zh;{2{bSAGV5=$YDO*?LFjB~pjEzGLPf8TQzJGi>S;;>L;ab~GaE{T$ObJGaB(
zk`VIZwj&qv`0@InjHK(MC9oJ5M(+_6^el@5l2*Cea1ffKe^B93vvgGG3H>3sj!gch
zc!|XNlp+3Qenx@@kvDzcM-wf1(ZS1~cTgthzf8_%A3kkVyvyHl7pTcx;!WB<SmU1<
z0w4%Yh)sRoqI=)p%o`Si!hD;><qNq#;D%3B-qSYMx!w+8nAi%v#ue@zjx!sK1E+&!
zwg%1Rt&BQy!7F-e)FEc7&lu3^*np8Qtx%B|f4f}u{oBo9scCZ+vKOur`1=Z1w7=_u
zjJUr!b@(M8W^>$Lllu2~^6RCZDMpN9O!=lnqGkQdWmK~gW(EA3)D;wt4Wy{;JG*V<
z$JLC4D+NN+!6u#cm`Oc@XZHLfaqx<?JRLmbuXhY8H&-SdF$nfWme-bpSt3M~UruD$
zD1^TYuotJF(K_KmbU|Puro;KfeuxYo2H{i}?ao61W?%M!wp@t~>5LVn4gU_j3GQjO
zZO;L0Zc{bX0kbAuPYQZ*9X`bAkarNhs}4w;tC?LIK&6u0P;5iY<PrX+#7)1D;g~n&
zOiQ`><M(HNVs}O)O#<m*itmEgXXO*+opmi<5O0>PMO~T}0p^Vv#gX$=*MkD0m<sM&
z{ylIyBm>1PbJCXK7JyhHx%f)QKtf#H#!4SNGB1oYs6A_#SaRm=6_}jBqE9FjmO~jK
zQ!vdY3f?Vq9IwFOQO^?pZz)l`fEg9`HZx3naEeq$<VEbzlt%M+d9bW9CYp0)OmA~B
zc5TYL2vXfJi{1hTh7qK$&Cnl3@EiSwUGy~Z^|w<=&c8mcarg`Q0hJrSnr?q!+QjgK
zzdmmQmIS6OQD3RWvF{}Ew1P)<vLk2FepC~<H+@sAtYi7q_s!RQH1FizTNhKl(Ci%d
zOof%q%iA%bd~cu`m6e!*iO+KF4ochK2IgOe&NqpNMMU{eG<J@wt}8Fk4r;HNe_Y8c
zrS)Or22}Rk+IhSB`72%#9R!?pOIX;68MTldu^N#Kl;4`+=mN>@mZrS>30w5P4nw*x
z>Do3;G0#mk|LuME)z<UCK!}(vcRTtNxS7Vc3?NO}qO*=S=vhWyF?a&5wQ2jUrb%OJ
z!?9hkOT6_d!o2webm8}uJQkm6gE@;IR1~*&Co<j3uoLdMK8%6llcRShN=zsPKQpwc
z?Wp$PqIXX-jMa%SO!0+Zs<I_m-;(znP;<of;#w!_1>y>iYLCN{5cKjLhGH?1OZ7qS
z3{`q97xa?Vitf@e?lk@Nc(mncby0(?XTi9G8L|XYTDL{qzDyWZ=P9s{clorD;W8Ye
zvM{M|#k;Tw&}1c96waD*szc-3^DPV*2$0`<9<r)#%=)sdU8Tq_uuO~$^+qOYonwA!
zy;S~-fb$|!wHiH;=a54obE7xfc^G;J3zh93U37e|zXDT+m=4@pP|rk$5xp?pp*q;R
z_B{%&^ev!{5cOJ|LeE}a>#K=P-EDb-7s)uAd{EJ}q-_hGDv5teX}O$Hi45=K^~)HQ
zi@+ozgDOg1&4ntVjZE&=2Ejw>eKPMXX&c$%4;{<63p0jeI)*p4&v4bB=xoXnONb}P
zLMH0OJ%sClHXOL5T-{1tWsha^>!_6oW39RNYMuvz;3C_u8%K93wv=x+9pyZYTgtz-
zyBcQo$V@BUsXI3Jo(}5F!(#U*>+`a-O949U>`LB2dxcDqq#^G?*q3d3I(a!=7x-cD
z^^MZeFy<?SF;v6^R~#V$a@0}nl(G(F0iu}amIv9THrIW~_u60Hyot$+4@~Y7W(;i7
zv(v6U|FSeHv}q`_8%|G)MVqlc<v9aHEDLLpxZD!5#)q4_soMnu&rt*<SW57slFu*!
z3ld_IzKASe#Cy3R<LgiMR?pscWAP2eMMT8lU4Y}1qz^~a=$Y>7%%6=`nyd<SY=&GE
zY^9&s&jk-<N?>T5kYf?#yIgsPm6K8Er!Z#6A0v!SOg^z&<G|1G6a}fLL0c6&F6%wV
zHDBXiAg)hkU3x3){)^?MFEmySS4jO_=1AJa_R(66qecr?$e(c0+st_$&)t}WSsq?C
z{6W_Re%+%`ZN~MmDps6piK`8>n;+_tl&=LS?8}AyeW=7;k=!pn(U%2WQ~Yz)*kOhk
z2X>QXhY^(_PTE!iJ{Z|)*Pm&Jsj+wXL2oWFbE+dI1dzL7^)eg9ymO5CZ-tihCze5;
zKy+!6tLq<LC9LL*&hV!Pt$808<8lG=B#a2Mt|9r7?R9hTz<A;Wq>_>FcR$;dXqkg?
zH4+%ydzn0BM=KcWAjv(%EoG%}VVBEo__Ma+x2_6*8}U4I(s7*lKO4V6%r$`@Ie5p~
z7VC4v5WcH$#qt}Rh&0z)RtJ|YDfs8)`>r%7TX%~~kmfa-o2x*CVj??<f<`ndTpCHH
z)|A#S;QUsCs64}J5AjGO*)_7&=~X__QDRg+VxOa5tvIiaXkoCI+R|DIv2|mbx^L}W
zWlp4TZqZ+Xm`W-WT~$%fgQal_mzFd)W^5mX;V5GguJ*g$72Ai8NirLZNeyEKQG`n?
z!@<a@D(-&t>1T4M=wy78cVs>{(zh#Mp@vuk4=&*8IlGL?_tWdl(eHFjziVJkjr>0+
zl?P~5)SbQ5((ze)E+?{fTKk~P09ku;yX$OP@TTr#B-X6}bTq3E#-g3)G6&}6X&1t@
z2Vg&w4sldm@wr69?VT^AP>@_nI}ZVjpE@bA^#YJKN9gp#l?`txIPMYzMZ`6bF3DW!
zBj=)VkLW_HOX{EYO=3RGDnJ&zUWXCq8A5dX=td?P67r?{)f7ic`1Q|)df1mv?>9l4
zy4rw48sNRi^0;X-TTHKm<JnFoyv|d7;awq-U5cSNt~yny$D_rKQ#35(c+}n$e#p9#
zcKye&^Rq5hUQmLU3~e~xL-Jz;<wXlFM}}-B3o6><JQ*p*ZG9mxZsd@qHEK>l+kjLC
z>D0lK>Yr-Ud$dl;+sTtAjLSzTYY<|~Sx;}|tkDKop&7c8jA6yPem|@R;~5YqS5(%d
zLN3$|12tV#6DWf=Q$aP>fy(wR{GH;YvG7k=PT1-E3`_-Dodf$oy>9qxrhzX<f@RbG
zwqMkIjAmhh@REDs(%E%)-Z7-7;cvr5%a%r!!Z<*1?HPT!;)N1^q5V4`VXg7dY<`0D
z6UGOvw<gb<`r!j`iRnnY+{q1V>73fww=_x}XT24w`E0)KvH0zrn=E>~tujUn#xySC
z=N>GExc*AowLz@yk8g)isp=x%kgVKJwUlIjrdFZ}bb5)BSG8c2I$xJ-<1D(4@Yt)f
zoa2^iEMSgh3Bsi}*|{&ZcSewA=`Xl|+Ab#F!MC;FU1}s4GsV#WTzvObDlOlru_m%X
zQ%Werqw^iq7<dZoVVp>6HF9yhnEWPb$Ke(I8M{zi#zmtV;^NULDS|-9+LVL2X8<BV
z?>SV~12d)Er03ckc24;LpTabVw}^v--SQ*mY}{fLL_sosvN2M<h*7!vq@p32{(G-v
z%gpx{De3RID_$W{Fu4)1jmRF-uY=7W^KivLb#&Y#P4lLVuQ>Te5Q0A|+;;wL4a!Jy
z(i9iDUrET+q)Tq%ON7K2&!T4~IefF4XH-hEW(LNiD>XoHuEs8R+qugF9+dDRIeeXt
z<bJy}iGOU}eyLyn`ZsMJTE=xi^)Ass{I#6zm{XhLAp{4eFB`B!hud=LaN1fbJWcnr
zNzj5IHlc<=7TlBah&{rV?kjb8#NUuS^4{fcRNPig8Wf`ETV~>dXMt=E0E1zD+D{j~
z3tac7BNu}&;xXbRgs$6Bm&d&C$!My3nfrT>#QDJzP&KHi_rp^elN~l>0h6)&yRmz%
z`!~P~?HL%NDErH!J@Q|LA98!ysm6{gj%A8Br}+2Xrwk~UU;&Ea=zYnVE@gOHk$g==
z>dEj1sxDxSsFpf%to@`q3<7bV!?~*$t|0_u?HzpZw|?#T-AEOtRSwZ|K1#o=7UV^@
zS%5#vs;}1KtdK`ZXsM}phyDvKM%aDNSxhb?GpZ68b)>tM3#D8op#IP%x7Ey61+c`F
za&0?Qy++|BJm89e`N+FBDxl~oxaL+BfCAA8n+7e(Aicsyk|?>+jxws*y?r8(p=)w*
z@O)PrTnLXgrXYa?aJBhVd@FJegJ4|@-88WhmkeCX8$8hTvUNnu6xVWWWZ@tP<>r}v
ztEixl_2Wr6#^IK+v8(%TgBb>-4H}nQJv0mf)}EMY(}7gC%s&%u27-XP{<&Hc%9llE
zxer$NKa;ewl&OOwX1eG$dTVgpZu)CqQS4|o)+eQ7_Kg3Hs67aY8|O=#2}Mx=(9dv6
zY8M^ZkUG+2MCKg8sc*#|m-V9jo#QHzv<fAy)DL2Z0$H@~>1F<s?`f+yy?)!!Pk$mp
zBcm2kr0}7#S^>GDJe2&zxBy!Tm%V0y#VZz8>?tw>*vebQx81JyZuIwoa@fAK0%VxU
z!&h2_<i&InJD(T{4zU;vn!sao6ML8p(k0gqph`@AI(Q-$6@rJ}OM)4p%g8z@CEH1}
z;Y*$7;!c1def-8Rd3w5ms@#M8r&?!D*1Y8brUz!p2$+9{Bpdj%oL<1j;b1K3?aIC3
zr18N~S9*>tw5XXYogCouV_Wjtar<rj;vouF(x3{b+_v~}=bWm+fyM{T5i*Kcx#v7N
z);QXdIUd5+K;{Z$-<EJ$1<RC(32YCS^+dqDMkF5oLLySWLRYx%tAAL_Fylq=a3H2K
z=A(008GR})ZsyD_YCPX@Po)82h|2-i9gA$H0V#H=0VQ0UK@Qra1uu8?9D>(`suh=J
zVHTAS{@)9?-sp=<?*tS|1{&%C(ID7rP-8~Wcm9gx;Nc;8k7lJQE74PS!r_g}jaWiN
z3V7mIxfhzG?IRK~NbKYXh7YQ-+ouWpk>cktWqB4xOACaWoZ3!rz-`~aqw>#QSWy>f
z{(k0LA=Cf$fY;a}D}S!FZ)kp0A6cyz@iXbmS#rFDK$O6Aa=3#))+e~H31+ogUf+IE
z#E^MG0pJ!X*JR$~r(6WAHrVZig&~=AQnK+&4+;H6eqAkjRzGUyRqq}30ssb{h7>t2
z7eng|;09=~b1*M?w;6Fx|5ZpSb{IWDi&_i@vZU$kT#qdAo!XidL9!HykiGuIZidWP
zX-k{>bTHTW)PC-Ym3=D%x1N&rD85N_3UJrAI44aez#)gKKU=r@adL;e4Dun2^s-wc
z|KxeE%$9dAxeyq2_9(~@RB}(sw{7Tv?Of|tuQ$>=uKBQY)z)zvYuvJt`y|?e)_u}{
zkUTyDSWWRv%;@`4qs+966Xby@iw@DFLC8cZ2B?_rojS)s+!D7RF%@jaG`_fA+y|Ao
zRD@AY5bqqs#OW6w7TH3L<QTn+VlZfP6^?ae1RJID`%luNMJ*DS3pDB;WP)iJo<sf@
z*<i#pl4MI4P~V#_J-5848+v$Mp#JxDIG=8t(&~wR4zcOY*nhK)IK9(6y<ysTzj992
zfNQGC6(YGqvm||kgavW)A#NyG)iGu0W6bCcjv+HcC6}Ar-g<LT6#fEaoQhr^n{YRP
zOv~E(y(fqEeEPW)CNqMWQEkLNxX7rtm>u`eR^3AZ^aMR1(6Iftbmq4c4!2DLtM9K=
z9zQ`%Z8K@P3J9o#hsa>Z?y^5Knm>6`k{B16LJ|QLt($lt`Wgq>H0{;{Y!b#i_$3Qg
z+dTZJ(X;R9(1OcYaH6Ep<K#(L$?PX9L(C61XI(2#aJq096Oj-DXU7R>8!^m5lV~IB
zOwNjfkZmGLvQOcb#VzHJ#A)Jn&7U?fp@DaxIZ;_uV$Ao^`t$1^g6`qD`9t1MoPxJ^
zW^I_NuxZ!15`N-Vm>x9PALqM(oc|$`ZDnC5mflk~c(M{tT4Y_8wW;m}a%*p`kboZ!
z65Fe$>ySbb#S!c%8lnYU-{4-0S-0N*PINls#*E6R9RmV|1@AYs7rUKQV@IFAXB85o
z^<n`VCS%nV(=~6<M{nLJUeG(~Z#($c{Y}|V{08I8TKV8w_{;T`bQRPLtLf8Esh7;c
zHD4@S$3p+P61PiZZYE}7NtV{SG5SG{DKGiXH{}OJB!`k%-Es!#BF&tOJ4uHIB2B#-
zR-TfT%-;KQA?d7y^MsToIv&y3M&=$sTyfEVCouwE6J%Ue_OhcKQc<C74Fr9xT7e;j
z$6ybpmzy}SV&LN<z>!xVg1MEG*ticecA&Ed*ow&Z)=MOD7w-G{Y0A%WDfOX(&?9CU
zRXa4&AD&ve4kE^j=8J3+f`TXV3}Q&$Q8c4evgiCX9bRXe9a%#WqB8!XX{CsNe}-vS
zKrS6Tn~GlN1-Au0R>3Hc)eT>D$y)0KSWm}V_oXcvWJ(#lz|-#PlAWU#U>=MP`}sYY
z^Zs$sCs2n;?<k64eDj=R-(f6zm)BV{CSP#vlj>Y*X3>MTAH0d)A}h?_W5o6I>7@-@
ztwp2WW6_bAreVx3^?4d_=GrTp46mWi7C}6@<-7P5>FE)7LP*#>Wo<8JRQ_;_4oY)I
zH&`C}+w2RFe$9?)HOt-uTjUDB^3J{OkK~>2annpJeyFVizTzergd%m~o!yBjMPmrQ
zj|*>>{s2;6Vn!klyw%RCT^F?gZF(=l{GRPvWHJu`zo~lD4JrW8Sje)gx-$-nV_u`!
zrk<4YAAl8(BFG$MqonNK@)I2!g<@4Ht6bSxJZE3NELkHl`|hZ87VdBWH@EJZ{Nd@P
ztxPL1aw41rcR-r%yC$0^h1aA5Gb`T#*=%&OWqDrH&>r(ulyD=?yc@)qX<hV-C+d{R
zTbE}V(4pCtH#^yu|F-CV)d;k&v`t8%AS$jHK0a}CkwKyGRb68;XN5P(d)Mes9L;-I
z5xY%72tzl{q5g(xQYfU6tM4mrQaa^Q3}09xE%jgLXv=J6#Pz1Gww6o@%yF#~eFo&)
z;u^j=c!K2;qr8R{^wIY9ULEzB+RrqtPzkLzD4UKZ_2%T@Tgq0HaJc?xqoXJIB-O_3
zO9(pQbGQbN9y)cULksJlQNoZ+m0x8m-v~e8VuJhKatLZ<nrpUF$J8@gTMyCxP%C;3
za}tRb60SJb2UG@!PV6|Lvh5scYY>lm+$;y!YNL*H(Hd7P2DUT|QNt|yiE$pNGZ$_Z
z9%Cd{C>O2FOux*EyD*DTAu>K(j7dj|U%vRMj^#bLUDG)vRf=Ijzr>SY>etbLFMNZ>
zArUw|aXg7gfd@!T9W~jS<vZxDC0bifj6N|fM$}|SkyU9*o|Lj)A{4a>TUD?vTvX+h
zOAuV8jaTwI*d{O?Gm3eg2)X_`gNzBBsQQnPsY|#qcMoV{xGEb_zCxt=($4biLW!t7
zY}@VDzsB$)<~d3}Z4uWNW<6EKptJ0dNkPw;RIa0OBg9dG*fvqeSQYJZ_lQ&1Lh3p%
z`6t5c!uaLYSEb<xW47NPnDhjw+ScB;N3;Z3g;)QKNDpp10J_T)yXSfgV7L@*IaXN2
zg*~|=uD+E{vuSy`BP8zESo}K#KpnpO;#7&@;?L9hhF^SLez8%}TFg|}^;)uNnnvN{
zn<r=LrQ~SDX+OHiA)&J`y29HmZBJGD9+O;KcbBIr1}DF^slN2-{F$)mUG>+KpQ@|h
z6|JxKKi!kC$zOHTgP$AEyj#lRR1ji)Q=$5ctNr(1gMWxxEt6kjyaIT(C5suGC)+II
z@nxb%w4nuG9tIDsZOq@=+3xk&Y+WUS!h64J@kP;l+DME5RbZ(Zm#hp-AT?{kP>xC4
z&FP^fK)$zz;CyxeS>yn!yl7(LiAJr)#C2i36)P8#F5X*2l^qwJsxCsESAKi=q(`C7
z19VTTUO6#)i@{PFmqTgoF>S87xhpRY=4&*>H`Rx*^(WqaBe$vPYiHpHX@9hQq<$Q+
zxf#cGV?Rhd@L{!edS19~D17&s&5zHCActSfXv8GdzI?X9a4xoT*o-#6?I`x5Y?Z$y
zz|#9$HS==T4J8*7FLPGR|EZitJa_-FbI1P8n6;o!uQX%Qoyy;S@7c_jK!?2dN}N=J
zD@H9@Td6kkq_ugexT8qI<_xuXoZMTq#S|(C&LsKc){H`<BBIrWG2jIUg0#Q|%Q^l2
zQetb3Sww_BApvQw{Jflxn<s<|EDxw49@%p}=SRCE4exFVoj&(q!(slV8<$>1(nvzT
z|HsRBNupTp#UCnN{mq?N+>wcf#c6L+C)w9*hYw+=pUp7)7WB>Ff}>B2XpuI?5El$I
zB<*7#o>aVa0Vp_FP_O-h((;bP&QoxSdUVN?alO*EH3P*Mx{cOFWS#x~{6oR1b9ZT6
zI))p2V4i#}F?9p7w||nc?A9ak#~=25A_ZZ+$;3rlo)<no_UU`Z!7~^SBt~Y(LVT(e
z*Yz+WFBte+48VdaG=g)^eFnjn0LPZD+c_#Ckx}NPwq^MsanJgZr1%PJPkIw!iovm?
z0piyZ=a;xnabw62a-H2*lJVOSqgXQ*8CK8aT`)h&9Q(Z1O=S+967DLJ4f`xY;zV=B
z%GW*!k>>V#JCD6JB}{nNvK_>XJ|Xbyv$oR6fyOr6iUh39L4o`OtSRCmWP()xjp{Uy
zC_oK&eM;{ygOlUW?{5y1$=nB}U1Z>LpWyRAnfMGoQsZnh7)YJcwYXx>q#ZRmuRKt#
zMkfXjS(0f=sgVFO^?(4RA&)Fiy(W-&^zNG@xTv+0SM8||cONM3ZoKKaMM!GvX4&#N
zUAbNz=N|OH-YnvSZm21d|N7qgk$!Bw1a(OIFwN*E;)6RB+$j&G+%I9QL3uENqp~)i
zT@iloU|#UtYT`tZA`sx-(UUNZ>Vnh3anY{(;esgj?sp*q8qyr?SDV+x3<Vw~G?_}H
zvj+dgqK_;!HG<po!?C?{l;=#HJ1!^kq)zDMvXy1^WuOXA@tmjK&Gb9pMopjm?X?fm
zOf@OtY1wryv5X#r<c}J!kvu!n7X2^JCPx+tY*X`xW5;GVjhw%I7x8BuKj^r#q@_&w
zMm0@X+#fM|A&nes4sGOjR40=KiE8Ybi`Mn_8e;R5<1MJZ+XofIFWc(3&4h1XOSC+Z
znf6A-MPfW-g@Z+`WS}fWfe)$yrh@5C)uxM##=R6p2=!=Gf1N%hygsp1pO0#-iec93
zqU#SujAvy0{^u#;Sd>~>@!1N$bwMdbhOL75?avc!MiZ6M)p`#&0e=RPp0pvpjjvgn
z`w1o}-{t01{aYCO?OPpA@+@_3B^i%<^Ai9vdQQXYePhTP_l!Y6$R!NO4#NdhK9gYf
zMP5uEgNl{_Z_Cafk^ARID7?DlBJwciP|e@SpcW<f7vqyOPcE4B+ry|OnR%w_DGVzb
zRIV=LEa-dbggDN1NwfU7XFR$R^v@T?P2xj5_td%LdqNy`42n;&J3QyL%>8ZQwHH#!
z+!utYS~vaUOX7DkPtnCqv&g+6Lh%lBbN|*HJkQs&W<LLIBmDj{{ItKUvb&eg*`qJ4
zwhY6I1!tb_lAp_z1F8mPsarzm_-#7!5lRy2XCn9kCzKYyeXjA}mYI<iE#FpSN}GiE
z*AF?2LcZ;1a8(LXE5}Ezh|k=Z^rx5dn&J6@3@q4J7q(XFzquHP$gZrpnV9z5!<!+5
zb5N#K$HwR}vt>m~+q2}YfBI9Il$Ixiq8-JyZhGMO?C`f|&Y?Ypwt_PYrjDQfr@fM_
zU`yH~XC?<ptYCSP>EA~NL7U=5U%)v##O3XN+t%a1r8?1Wdnf-I>2aSv>OHwcdKZRt
z2~lnbhMfO<SipOxwlj?!vy_wT9{e*oM2)hAe5Bc5Cy4zvD41g{K2v#Y%smcm9UN?N
zWq054L3Lzzm8_<JCh0=$_TNYkzilG5fGH${={4WvJMKAz>}2*!Y=3brc@vAUZpeUI
zN}B8B-;Or<mST&oPJ%~m-vLLPaj)>(ADY0J#YKWgR%nX)xZhLT%kpJ~oVw@<dqEBj
z{3Q-P_VKtH{kJ*xhK$6-*)dw9$j@(wY3y5v`6LWG$}Q;_`=^m@v(E-Q{m4nLNsRB&
z_n|!B7~A!yGb3z8PXQmTWBl~TcQbZ)oG#EZ=mu8%5!3t8&!!fQ1&nWHP4Z{5p_Bxc
zFvqCP`hDw?H)k^Xr_yB>svQ2?qK&>B`9YK}IaIPor@mh}e*dk({?kAE%t?~P$q*~}
z70~h@p49|~#E}&Y-MZKBpU;#Utl`O@dQf^M;I}g{0e@@i1ob71bA+*gp`s^~|D>QO
zL-3~M(9(4V>A!x4v32`v*&DEh(vxqTZT;(?XkQ7h!n4|MbY>b;GlsO5JnvEO{L=z9
z%Z!AImZ8nz0)Kihf4ywle)6pJxz_I;8+k_iVOj!mg0&v^+GQE#pkLtdF_HMw`c%l`
zSdv6w>UEAe<3qsjA4{zpr#;=iTf=4i?Om}22YBPxpNj&ROhX1HJY^X7AV**25k0fj
zP`s2t<}3g83I6t#S_ADtePl4=`0J5Vg;~J%Uqzg$Kue567gra#DVY|i)2{RSWOeK1
z1?!h8OLec1Q~sQL;QyiRyTiF|`@geEM7pv|qG=^1BT<r4DlIY#MK+n4X^;^rQg%vR
zibz&g%dD&<WRxAU$$nm^>b}&~ao@jxp8r%1eSJRX`JS)gehU>zy78ut|Je8a{2cg?
z<=4*tUS-FAu2PKrEtvoII@Huiez_~KGJH0^Dc_vubNMP^W*Wrikxep&BEd=aun=9j
zFT%7)r9g4zl`XvO8_I4tL~<x!bfy*uQb@k-^~T@e)IV<Sl^@sNx$V}m|Nc2x9+5b`
zU2|vA_uG$$=5v`UF946yitSr<?zA8}l(ctl&3=-_DaMsTj-59OTax0MooTt)dfBLo
z0<OrnkvtfdaWeK#&%Pg3&}<$obw2p(O`P-Ke!R9#B$69-t>#>_%5tD(S15)pShK5P
zY`2ntjv<(;f@WCFT<8qYCOGFYC+tfXW^d=?r4b*?xmu*ZufRm=|KkHvmhvF8kthXE
zp{WG*q!_Yq$Bf*tXcf63iYxMBJ{t{fW``c{ih!cg@I6b<+9g~+wB^#M_kC%bA#RTk
z43^8h!UZ?}DMlLH{qe?5>Rj>tQ=0lGPx3L$o??c@5}f)TV6XUy3=K9)1Z>CNZ0hkY
zQYY3d(K}#8`e}TWH9tPeCee^<vVL&DA!*R##(aE~HLI@w@uNij_$ahGz6*Z6k-sGf
zc|NJxk0QmD!lmK?cjIk26JPO;2!;`}=G_3_2#*P{C>wfUwLm<K$l65>iiiol0((~C
zyb}WsgLL_ojrQV>-Jxe{2IbaYJ$WmMzX>%+Ud;uqU&HB-JNx{v?<b1ho_5Wjij+C$
zN}e=96=KE5tuh&@iBd1Iz<9U{I)lNWtkez^$@Be1Lv;T_wCzHT#ONB|!_|cBltqj5
zDXG;Dh0Qa#iQn1h$0rQ-;d{e|>@mo2(t-XwbP4zt2weVGlp%+iMe#>SFO0RC^?Up0
z_5J>phdP3W$_sV>z|5c(Vm57i2rR+SV*<L<YeA0Nqk75cFOSXi#0?h9J=6Ocm|YV`
zE|sO;-{wEMw>c;09V%2ik3BiP1_K{@j&2)oLy6$h9vrxwB>_#F<()scKjNmkKmL*6
z2Au<cg**T2H<Eu%vd;0>bh|I?V%==n<W5dL#;?n<vLM5cm3Nj&T$axw<Pg!C!Y_sg
ztPFInDWQt{aWAh-7w-z^LWY;N*|Cq+D+Hg)p-4^i53ySQF8%iiYEb-F;pYDyxq?JU
z-Ex(AI~{_ahZy<aY1Jq%<B?DlO5jE=`SD1E*%fZC!hhbz_PURSemlY@hw@?~WAIx<
zMuFz_<KzeIYG(Pvr<<Eyl4d*gB%T$s(1w1}xMdlvh!5HED$wzO^b=|V=_BWC+lBcA
zXD<l+4(r0a=Kex#bo`tUHUu$pBiTktYhUato~^E00Ce%=y8^Kh(tIo3^*}k^(rnD(
zqdf_Kxl<#zK<wI2^o9u;6l%OL-!!42A`3e3Q`7r(?~T&r_`lTe&dGE<b4C7MmqPD4
zQ<OmsWKwh1m6P`mzmNRnkAn2`*3pzX*JkeT`7@@w&sK>C4V~0ELcfJxes%6hk8KRQ
z%6ROe?0dsdTPN1KT#R-lkZ_BR7|e0~(v0z?Az{aTe#xL;o+vWpI~@F_bUAh(qnDLO
zZ1ZD2t8d+KUzC+{+{vzh*Pdh4$Se^)AOo-><Q%QT@UaM-7y}STGzy;h(%KkRCxnAE
zrC{Cu_Mrwv8n6Y0P>CaomBDzBm){#V478Vu!k>^pH+H9)*9%xQWVOAZ81)?1+*c9B
z2DgE`<s%^Cy~_HaNhO6;rbJd?V2dz($OkC11)=L|quWgV>f6m<A@oc@E)=C!(v~0q
zXApP?`qg|B4M0j-f;G(p>6SX=-FHxQOiJ%H%Daf2`F<ktOOUzA`6gezZg$E(F)Dw4
zBM!e&=)#>Tgp2`CpVWK3(6}=Mrco(?>7im`96pZ<K!ta(#w7Z_?zX4wy9xY+%rA*P
zIAEVj0hd7_C;{(<GWC^53&Z*GTji4<Rdqy-5t~1Q#K{+t`R-4JPXNgiWC9dwIlTUV
z7{naKwNu%Cw28WdnsLRB=NcUM8KrFdTaaD+Ufs@QIZ#PN_HLU#x)7bDUdC1WD(KmG
zf~ViBm?ImG-5Zhk_B`FH*`OkJSH7s(#M`PJnc!9Ce1sxlKL1|rmtDD?&v$Dd8i$ZV
z31)MC7lL!d$Gdt6Um$5-DyRcF-CoNMj!DJZ5OUyU+Lp4ti5A*?W({YaC5d1vG2|NG
z4#)Abnf)J(wS6Bj1`M;;*(Rhs1n_fLqE7bngzb!{`Pj7Or?DyH^tnacJK-Xm`tIuP
zY7IN~zjmHiI*z8c=}i6UhQE$#EeFv~Y{^Hu(`xLWL0wo;c06>w{zd3i9~}8w&;7`E
zOdd<47EB#OAo5%t)|0;MUSqYEjUPDcujnftSu|oi=@L5bG=x{Z8w>*%jo<<>W3|n5
zr|eWBXvNCT>JklPjHkIcw)fh$^<3mO7dxkUV=(z~Z%b5SAZFQ&@tI)0aE{i)e(2(1
z9Y$@3PZx-6IyVLTbGahT{dHwcoIueo=#pS^h)!Gw=>e9wJk}b{37m#I7<32cu1nd#
zu4U<Al(%rxC_|rIxr@!VD=f+>h^vY|l|S_Y=7w?kOosH44Vx@dPHy^l*!wjUD!aRn
zs_ftQXl7B_g=?puKc#<jFh<ck>g36Nz0Z}e*himH*PjF-%WT+*y_mDziM_`WuqAUD
zb90r^Y{nH^8eUCM)D4OlrBr>=j%x3xm7;g>ucYYxtfIN4;X+_=jB0bgh-J#P^{c{)
zyS^>tm2M$Amk+lLe#mA#_U3TS2qK;4K9n+^?~f<!8_o6++=5G){ZJ&%SKg&MIq9a1
z&C_SKl^Oja#S5j5^or)5oh9=<i2(ZIl?@c9FUlV68l8UrwDqGU+x=;tiuRD1CyRE~
zPQ;IXQj-e41!eh%E-npc+4sr>FTB=yMzn0=x}(aSuGjYypM8~Vvr4`qpG+N=PzQR&
zS(0wgM%?P%0C=X@*J{-%Imi<ga7(P|7;DRXJpn1uepqo<-9M%haWEGIsV;~BpMBXQ
zhlN;hri(Y$U^4HqJ+m0Qe17-?B$+k0)RwB7sxp82XrHCvAPX?`a7*#^lJ}B>&eXM7
zx~H`-oB4KWXBWe%L0aEF_!voRnmJM}bj*5|V){l=dIavrwYS|Yr2SwCemr%&$nuhV
zrE}=u$8r@Xq`UsbVq4q8N1AIk+5T}>7{E-a8W|rt7_>1z*XTuq|D!+Ee!rZVFd8lH
zsV#4t{ZpK#hR2C%12(rSwj9o4T;Zo1K3&&1y<|u$PFHe7#CLIT%x*$Z^^E9pD|;%Y
zMyxSrOf>)&)UvBFOm)W58JlRFF)vwotvt)orhNLQB>u_~O+|8YSUy{K$lu|`XW3IO
z+b6}mj@-I6UT224_uP?w+7jZ^Ikt?@G7A9}F)WKx_^j<^B(rqwi>K;p@ZlB0LWIR^
zRax7+QT5XB470cq?9=rpcc)FO*{5qFq#EEhT0<R`Gj}qf`kthvl6qWx0pT4$qMRY#
zIaBtMj#H-~HoW`P_3hTi<|<7C-@KW3{<>0b=k2H8-8;Tc1{AZp#J6?xL&dXvZnXm7
zEY|8p0tWFHe2{f0^KIBTSGVicjmmZ`$}#Shwp~p80${E&)U0#alG_rbBGfu$n}B`}
zw#aHKIZ_9r<SEK0l5*r|Yvsl}RuGy})QkMC+-Y9DhM{2tAlcg<JQ~dfo%?p<^eU}0
z<vTX{5S>>!P>g0`{5u<6C-xGh)CPCD;O23snwTAoDmr$HR0r9=<k+BXGOkI6p;l{1
zjd6vwDt`DYZmq(g_DGx=PY2H-FG;<SndnEA4SMQoxvnlMW6JqC1kZ$(7Qg~|YL`~R
z%`E(3?nAS`0$JI~1nt80-K_8!!US!j_L*PlD&PBoybC`PjYyVy$D%*f8UNH1@=m1K
z-7abTSgb;t`;&;TuR&mBYSSUPn=}u-wS+p~jXDUO9)5HF;926)3Iyf&<?~Yx&<8F7
z<Xwbtq)`%|cZpc?`s%7HrVUvd#UQ^7Wu0+ieiJ5j&^-Oq#i(W`yZuzpiEa3^%jae9
z`dDY@9<f*x=VOqa>tJ5Qc<sXzlMv|8G^D0$B*p{Iw6ZJZ`zt)HVEbOaMuBe^rE2v<
zx1&S7De(!Jy9vlBlg|~cj8Y)>>71qdHe;|J<78}p<CvM~;ZJ~SLit<U!)%DHUyQ<W
z9hc$5j_}`^^zotHum0rp;k<cDpN3>m*#Tn1Uhckd{Hz=aX?wypA8?(Sve!JzC9%zO
z=Qj()gX<t$9!uVsSApD5wqn7~aW4tFfH3r0%fOTo&Yd2QH=2p5598TZZx8yHT0A6n
z$Si%w%*C=X2$dxDkCp>2Vpd;evfa}T_AYgHdo~O`NyBnfiFcZYRO?(>otM`o1R%d1
z9dfFWs+OuHV*2T{RTWKaZNeXuEFiL*kTtNoCQ1L&*ctsuKITn}stu85Y!7a`#G;&z
zG;{uHCEp)jwEWE<!<g8b3;CA+s;_@ZG#}TI^1WUl|G9FP>G`K(;vaXo=G2B4Pxy~5
z9d=@}S1{pcSGia@s*VF8b-prU`b(xr&;vMuCape5#nR#U$sEPrUsF#knBG4!C|VSX
z$PNMa4|d~h`(&2`{hQi)YePk+R4ilEPCqWeiP2uki|NEy64~A{T`lvBap(Jow*>dG
zIWcv_oX`&7eRH>m*aZf!9m<GiJPG!uL6QA#UtXCbUo+a5mMmxLegw7mlIJ}?56cr8
zcWTOAES8z(Z|d*Cz7GY$4vGCf7L_!kPBZs49^Vx$agc?AOA?>1r*LQ*=qQD-2V0kE
z*=UUaW;7P?2qjCp!G6%SI1?$_WE)-YxtcVl=rH9QG!T3%$`EnNd7*<gErrX8Ck!^b
zrD4KcS@Kl-jQHpW?><{){<LS4O>lG0bao%1GWrACxKGEqe+Qxgn=13OuQDEO%DUY7
zzMT4u(x19Qaz;*P>pJ&e5M9@Gz;i2cfYUG@^na@W8q*t%{G1$BBEz(4t5YAM`~ZqI
zZVHK^VbwHJs!%NReK|AD{`LdBO3bM5egj_ebsX0Owbi4*baBs7?}V@=(ih~n(WEaF
zxOOZ5Pw%~Af8LYGB*$EL{ohuMzmF8=c~H|k<kD4DSy9ZW$g^wJzl3+|mWCZQo0cJC
zlmeJSIpPA)PXuJ8>fR2D_%3jGi*nO^okavU`39`R&Du?L(3zT{I8ky!cW<#S_xKCi
zE2AYRG#&9Rni_EndMVxNGEk4C&ekJ}Dc1)LH(!^$()cv0Y7n@JiObwq00$p-a7(HU
ziu0BqY||F%>=)Uea(&d$fnCE1>fc;--%^$}yb8)Wky^0d<x17ZF)rq+h|%iAEvNMx
z03T{5{&3A7UcZHk{GbCn^R0awM*tI-*#};_X?xe;i>G~+M;by{jYQPOq*}Od$pIeO
zc=EIMJyVfIgHu5FIcDYVwlJ{=ZtVDzDy+ThG6-d{=i|?Oqt7kGNiU#~*O@t7e`VOU
zi^-YAE)keaqq91oO4cJ26b0BW;1J8RGe<eKWjym!3QgB2sPiUSv@+pwjs%DOP?*;y
zDwn8Z8?9@Eb+>%gsn%ZLMsB;&jmf%PX~<L5FpR0UvZl9d1dl7=;Z`TnQQE`wHGjE;
zo1q`;apnyfd0{+AYn^<3l(wfI4wm)z(*-T?|K&GsAv2YoGn)n)xA|$8W<W>HtjEMH
zTY+aXenpQOMdRSgvfC<8_okukzDVg2F_SaNv1!NQpY7&9DFa`{-f{1JoFadvh$}@&
z?bQF_08x)8mxV;Si=<~VEKxcZw<_pDVDyqP4U-WLMpl>V@EQGB#$#;R>J`Qha@N0e
zIuM^pW73N{y>Akr!RaiGPPnJOvpRlfBn{HzN_yuWJa?7h@_H#LGT030!qo2R&^^`d
z0T$<rpn_$V23zb+>|A03@9t8_g;8c;nEc%C^jhsxsxC~aQMH@Y#aWckp(=<UJ;Ed}
zhNkq<*;oo2u!HoTt^xs7_<=MUkFGTUviD5NGfq<tYkG?CY;lq>xJ3}|Pc}lly<iX#
z{nWa6g;0s8y^qWQNlYD-c(g-Ea@8gI(Qs%5I6ybcpsNy|TBDDi4#AB^inhPJ+GWNJ
zi6J$0s1`5yhFiJm9Z$b0DB3Hr&<{2;<hY#MY~UkfylSjXxJ<u7TImI*1!5lZrEa>e
z&!xBhx_A2dI%4nrd=pT;3liM-^pEfBKUE~>Dwd9tZMv@c9ye%TKUZeUpRrq^!NF-7
z7@6Ic_l$vGyX<Dit03tmaY<inRxkBA=J}*Lb{E@~>C^U)-WVTvFtQPkD+1h@wIA*)
zxqCf+D_xSh`3wp?|MokBHieQ+fI5AH9f9d!)L=S_7~N`~Wx#<r);W4m%2Q`1)Z2x3
z%7>4087Vwju^{yOWvOcKM=Q<$<qZaarlnoVo=c3x=9$$^Mm98tZHjX}3J>Dj6w9C2
zNZfGG;H~eiiVtbNKYlX;&{)keFlvjEvP2&vm!{ef(qiMM8?w?;)S11nBevANU&cEM
zVR`;9CB%j|zWp4d$fGL4cGk?m6u|M_h9#O&NpNDWtxVG2saYp|<lN?$+vJjLzir(&
zRq5Ts^l-Vmyg2n_<mhy#!I1Z+KMuc?1u14q-%VB)aWVeio`2JfV6>Zbv-}>algDEj
zn%$lT3i&QbDJ<3La<f4PHq6m^?!BUD@Vf??;Vn8o$Ecp*0c<2ucG`!tWrkE6&6FjW
z<j_fVm2L;u=yCW~U<bD{-WaJMVe4cvlukt>LO_Gr#2y_X3Yy&G@WIy`O21vycRYhP
z;1F&c_x%kR24E6Q%Hwznj-fH$q7k5&Ya^<Guk?<czYpwUQ(Kfo<w?wOt~&>uZ%G&j
zJ$#6z%;wZPp-!)7dMj}S=!9|s@h>F?6LnIMi=cN<wPRNN+_ac}-3JcJ;ZsPPsaCB;
zR|{@aEPkztLfWjAP<TWZRk=nbje<e@KH)8|0#D3m)!L%zQkj`EKTkcyOnbJQ^l{$k
z<$vss)ehy&Y}hz+Ci&tpyV-|2&i~m%dp&lykJc<(B0d)X6o9YYK+X%vA>Dk|?n4P0
zoIcuc))4gaV1E1^ajHBvAkm-bHl+e5KJ%hc5#MWof@us;3N|=(32SCIyt*O|ba%nQ
z+ElZ3z?mhr-mpIq4LV_IP`io))jVa)7dMLKqTU!nCuUJbsS>f0pZCf^N3>9+1#@6u
zQ~PD%wNF#m&HF_JQh1m*iKr+Xe9q0XE#c``>4W|pQ$%-{NVEj?q5yxZEzuJ$Wur6z
z-*0^IWZ&bM;;DhTjW+i)x~4!3r;agiA>=8Tx#9n+n39Us*LtMq(J^e;08JjjhF7U=
z57N6w;eVLr<0))=6n5Zbzot4mf|18%NfysLSvn=aT?ZnZY7T4+TwAS?;_`$kYytHm
z0hTv5uRp$ui*VXKBxl+n0!POTpdH_6D2zT4ogRHyu*^-1)Z8+nwXSZcVJb-blieVH
z;4_o2e$k*Hx^Jp}+U0$Memk=ZI<reUZM(oPxP}Q{sag&=vH4TAL7ayCi|FEV>4q(?
z1Aw{noex5&GI~YRcG6M|P50Ij))wSvo`@RDqDQ+eex_spY*nSd@KA`PmuFA!MR~&_
z`w;ia#zk#Q$8%SSEz<{xy@CvvonEiF{jZ`1!)E1(AH(H=g#RUYF}pi8?zm{IcRbd1
z3BNW+XG)={x}~RbiRRhSQ)|n81fR4TO)Tw<d2du#`6_}d{aMxpz(UzQn0YprIsYw7
zWN=Qrr2qM9_MGIq1f4AfR86<*ADL!4q!H~u*}V&v$!RThY++TIM)@dAN<+cY5asJ#
zdnUNebY1k@SUnWyzx<MkTW_MMaQ8W{eiY5O@{zM>h18zfnW;#gqd8VKZ?|FbwLp~H
z$KDdk(MPwLDg0zDaFPf~`Pe{#ANMNwT_v~dxaIYdcxw`_&9R%L`n&^XS8&#T2JOv}
zgkK0c_ni*#J}t_dm~wh5R79@>6FDA~f15Tl6K_x0v1adPzb*T<e=KikrT66c5ZJ}%
zqG%I(0cAY-b5X{BqHpk-{e)eN1nKMj_k?9W3%fA4JTd3NvqEMsTV3rsKX3fuBLWpW
z9+s!Z<-W_Y>p7+(`JAbx?>VFBhb3>BQEAaH(z{=FOL??Cz+W=dgC}JQ#F%2Y>A_|7
zZ)(4$H#2i?b#*>cn|3h%h%lk6;BY~-?G_P~Q;878B?ujntj!tEf4zdzd$CuQZC#WC
z74Z*y{d5|>O(h@-nhq<~Sulx3dC<Prd?e3&ECF-ln9+5EfcS6GK<_7|LY4@J|31XZ
z<j_<{d|!dS41ie6MDmFN4)5(m(+MR(upRF8TEhAia{MGfI$QurA1@Avf-HzSo7R=C
z(W}IOTycik=T)1;i2QIWgtdsq3kG5Iht4~P%tteLhi!Wri*gj2LF(@|;=eBt<d31(
zCz#XN*4F%}KQ#sa{N2aZv(2F1tLKGVPc4^WToJxxOb{e+Y#9Dli`AbxIFfrdP~@~<
zYPKScb>aE;{mE3wI~5Z%F=ycrV}RjXj8BNmEfK!=?H3(OVA#}M`$UD<=)8SfyALN?
zhR00zG3^^>N(S;^&zA3CFT%SyQEM|bvENhR-9ydlBNi!B{zDu0$@C#y5UjPSOARI_
zfbVZy*Vq~H;qG2;w2n$gLe+Z|KktWR1SW)6?ikNBF$Q}oEOIK9gY)5?m+0(PVM<Y0
zybfX~;Al&(dniLh@}OUEY#{lTCD#PGYFvxI0GynBSjMP~&vBq=mD!j<x+d@&Un7<|
zU6SVq!a!hK=%#<3&OcvOX#ChKh9^qrU;J5>{ioiVd$^m{l5Khit-ul`r8Kb$<3L@0
zN+|)Sq*$H!K>w<lasM+1Oj3FT*8(>wdnt&wP=RU_p@l&a%`D@=q?@BlVZibjM&mb{
z=nq@#?Hf0q(il0|DqX3%v+ig~-Y&iXJ`Z*6G%!&D2xq;b3zYvOnfATrMLq%tFjzu3
zz1+N}d_<<ATDIpUlr3UG!I-6!MPg#H5*riCBhbMZ1{p<A!?>QfXSHInlGO5wpVj*X
zLEuW855kSyfoL+m);Qk7zIwmB|I#g|*TDX$@$%bHi>FY{F~o_^iS7Re)*0FP2q8#N
zb3LUsDdSU}o&qfT((9WWRfdiks_+tg+gq*jgVdj)`^u-2{O(UX&j5lCon4t&(}=77
zu@?CGE<9pqv0$+;9AxsI7j`UBqCM6U$M`xtRk=i_D%#OoN9ewpQr)OB`xB))ofrsO
zKDSw211|ZvNqy!M=kd<C%v(>MSTyR8Y4L-U%9sWsmyFgk@^HS}s>2BzdhAnNhL_10
zQ!iT165o|dOvu54RSBV6z`+N7$+VVNYgx^5SGc?x9DYhDvJf!Qt#Ei(J&Cb1_XJUS
zz~ofJAGXAYB%1c-IFAX9B`GU@E=$msAxhDAG?<*4mgyC!<t<MFbem}N^b`_pUMPEN
zjHxCv?%OhaZ2`|_vG_d_GSBj-AvUNC7_?cH$&Bl_PjsMJ2K8`oeqG$;5+l=oL1%e@
zG6sa+@9ql!;h}k)!LF2GK$HvO{nr2Yv2*$?EGCn}=R#n)u<wG@!ct?`6~euTF0~2b
z=%#X-HLlQp$7Ocb8>*}frOPEDaOBQ}8&{^X-lTFz2|<E`s8vW7dO<28MbdX=h+Lp+
zjT~4K!Ok}P<jLBkxNN*~%Y}N8AfD_IPi~_Cs=@aeHm8V}nbf+I6W<T5Zox*q?HIy1
z7d<Ed8D5#h&X-=S1AQ;tXrn7AVHOYoR))cKzgATN=FvvC+$>~*;I^k{I{884N^QBY
zTSz<e(3kfQP1b<MyS^sPB2q9BR;mL?Z5jh#AzL>~+{KAK&xKh~y&-rEsb9LP;#Ywi
zLky-q_y>NH!?De@Q0-u)^acVnz{slhPu;Jw<PWUXnWq&T3;!^e`zKJio6G{pxbxIG
zIP0Gm?)RCTY~NflFx@^;L5^(CZ%4L}$%6;#ZXyJJ16_uR2-^JH(`z$D?pEpQEt5WT
zul0t=057cH1xvTi<}VrKuPfmal-?`|voauoW^~@YC8Yg6j~1E)!Ta6XHB{$s_<n9T
z*vJ3i*4}*F0!Q*|n2_&6tzn4zB3Pf0X|-&t4hLtc3-Vi)Ma_3=+-N~~{aE~Z$z6OZ
zMPpQ!$4s<{?p!7+u1g)7C{%QS4r;-lf?>pZY~b-#`i&z5c!I6paq2|_iMpP$e$YqR
zdMkyIW?TJM*`-$)&+ZMu{2>@S@)P%WY0nLxKeJgVGSe)$*8AuR{r?U{RlYonn45o5
z_qLF)ulaF<dplfRKd#sG3MWmjzjDEsts`^eLE)MJPZ6&7uU>e6JafDoD^#JqIz%M*
zi6hxp_}IUpgKN9ccuB{z)I%9>&nze(*?5DhfW<3B{y~CfTs2G2;|5QN@a5mI-&*S`
zGn|I%HfFSPKb$^av_LWL$>PeWzeu|f<~BVRJc$uOA@;d~V9^Fol81T&Y6x0bVWi%L
zwV5cfxD#$%v)ywTUTO!RZhk__83xxy2<s5;bDFB6T;|;8kz&!<ptecI+cH1Lb;^kh
zby^|)o4WGuo=|?WbZx!*NrW6(6FCr<iy$%(u)wzVPw+tHRX>kxW--T~uyk{dZ2b2l
zTNbv2>vS&8ZNu|I%Ozu>?NSv3Z8mQvW<%}HI{P>i;SZ!DRyo8+7iV?<sPvXg7`h5T
z_p$7_q1m`kKWIb6yH}mq>E;Jp!jD;<pIb4bQvrAXbxr<>cGRXC17VBqTZbDr-tLL7
z{8qG`uYoct)10_95}*q<uNs+JgxsHb=Dwil5^tl4o}pTn?v14%PcgmTk4b;OgKV2=
z2GW=>A5GiqOG}-Um3%plznz~x1HVY_#Jz;BgNt2=3d59IhknG~T7a4Op(9ey5`A6L
zJ7QL~t&IGs1>iHgyn3&3N@E~=(UmPDH;BgJBc5Wc3LofXA0u=xm1)4L{6>aq#Vj!X
z4FSqQ>71WR$uc=-^^`4Bx4vK}{)DE8-oBe_a|{*?Z6X{KAe$~`S{?h=^y^>rw&CBl
zF07YX<b+zH224UwyWajO!pJw}{>Y44(SN!nbN)_1d6tS`^iAQ`@#C?IPreWKcX_2`
z^~%l^kXggX?Vhu-KF4mGi5>uv;*(Sg2J82=GLl=xg&<S{hG&8L0#hj#fxz$`FQ==d
z-o(?|5VGN&j)3LcTL`VUMM_W&l}HXMk+}Y=m*lf(PB_-10GB6gN6iqzlS<4QxD26=
zcXqGW=p&%SP@E>VtWe53vv$$&pfHM(8C4163PnQciZoL(kwJ)0X`MI~ChMp+G(84M
zbamzn^Ql=oC%1qAHjqcPP`$S7{D>7MyGmlrbqW?28(v_iAS#;Gcdo}0hQ@~A+od|c
z>41DOM3VCDx>1q@LrGADO!{Q*a&oE1pjc1Ft!e%S6dW)sBEiXlzS+9DuHul%DjB<Z
zPbWK5NF!O1%<^QEF3NyZ*x~OO#ZE795Y3JFurc?aAWOW|S`(bNB#&+2{;zQ#*GWJ=
zZ{k(lbgWe#R{JzZB|mAuY&U8~bJ;JF#r(8)>tvs2duxd9hy%L|ooxGH>DII-wph^8
z3iZ0_VZx9)n4N3cqbsDBiS<|NV0Z#k@AE=;+ZziNr6ZI}Zq*GmE<*#@te%CHJ((N1
z^*;b!A#MvH(KTmaQm)ZDX&-|(Z^LeOw&B&))b$|mRk^AA8~1KIgke0*OMk<S8|ovs
zuTl-`bC~JB@P?>;Q0$#tXvB-{n*T%QCq4$;a?a@wnd90Y*(X`6SKY0!U1FyDf-;(g
za{UsoR@-<tT?h7;G7VIOe@Db#PJxb{%Ie@r?U411TNXfyz7CVv-iUtU<XSu~9@AqV
zo9LKh=W+!s!_qHtSg>CS(X#eVbSNx<>h0U496W`s9T|brL}TROeKo;?)HD7VM75Xa
zvS)h;`r3gQ`;}~z;*cQy+dG_lYoDysTQ)q2<%CRxCi=Wm-27hBNLb}eo5fZWn`6Co
zGdRR`wO$6@4N<BQfK5I>lK55wR7F$UtDlx)7IodNcFUGKs{kVu{XUt}>G>mCZXRVP
zf749pO*d(c=1HGT-8~Eb=OK8hk@uv$v03)3W&Eg!d58~456sKtOl4j@?NH5n#z|nu
z)+FHgcO8vrxNw<h_!VKEER5JG?thkDshH?~sc7i(2Vw6YuJ7tQ!k(aujVEE!zd2kx
z0tkqoRSolIgV|FdBN2MNRa;_p1I>-k+8=ISZ?7wNtke}56<I6cN+W}kELp#Rx%p^x
zw$Exm-B%;mhqru5+|ea!3npr9h@qozYH#3M?UbHQ(JRmvd@oL${eIn{dwNASyE{bB
zVTU>uy!}fj6TcqDJ_sX+z$xz+xjW2M;~;eS?`uT2i>S*=fk|j3dPBxOfrwC$3EP_Z
zTxA_}%$Sc1ap|O)hYN4rf6;z<f&)7PcW%Sdd%lwCK7d>Dl^s`c#T>tUq)Oa!5k?Gr
z@2h*r)x|bW;jAv?;`?a_15s*ljS6S)JWD0};Vl7*wW4=PZZ%PwZ6`}}*$6=tW8uq^
z^!_u^4^kw>E;x98GPvW}<BZM^VldpAfp~WN$eyZX)X<>dIB-;CB%u43|CNR&#uzty
zs9@Cfr;8q(iFlX0&eM#jfH_c!(X8Vq?f7`+dWaAZp@%!DFcRFZRKc+q8wHd5JK02Y
zzQsT+t2B6@(f}b?OiD%3(F<rInk1icANkXOSxx&#e_53%H2yyb7UIYV6=i$fg?uM%
zHt}Pu-^rh=TryL$u02~(B$&Lr?+rVxby>|W%tUDA%jH+AYJ#cnXWA?JfcpwYzw2dR
z8$N#O^+D^BMdL(CPYlHhN;9qpu~=YvFl!DR?3DINO$_$9D$l1t*`IEX(q#=59rHEF
zlkeC#pXppw;+U%D_vd_nAxg)qh&M94;*@L{<4HYeQatty$)uI2`BmLfaFQrH7$q#j
zj79u<N++hJvLEON*zSGR8$&+4wq8G988x`)UmN^+fr6laE(QQcfMNLj#p{mjnTF_s
z7wx?D7B!qtr7;y8{ya0GIxc;m;$j#jS7@EV)_y`M$&wQ0Z`y<|$ov-T_@C%!0jGbY
z(EE0R<+^|9ll}?s9@l_z4!LMfWrz%#B_>|6!!<B?p{<{Qb&^4f%a!k)JM5O<9+uvE
zX_NWK)hK#h)qSjcv8smA(%AT$uly%$cyp%g8(peLN;-4&9S2(0FA+pN`7z|Lq4n1j
z9ve+?Ek5T-|8f_fPIV9Xud$~}Ba$sYJvSZB0yvdus+9<aXo1E2<N2}?)pCM9Q@zyk
zcYHz+Gy*I7ZydZ!yY?P--92T<93s%UN@|@DkFiomg=D0+EIeL074zUC2;cfuhgp_Z
z<}pCvJ4_Z6)#KT!V_&NB=0nybH4K*aIS(Ebbo*<2`Yzy=mWW^4sEV2M4o|M63>7jL
zxw+8E6T1=2MYf0kaWPV{>*rh~K|KH5Kegi@zG|AKH+LvWU@nsS^U;oS?Rt5l=;usT
zMTw6tJn@Vxb7x|Fm)mr%#8<cQ2HR?FvK7}%Cg)YHXl4=8`E_!!r{_PNvHL-fn0!oi
z_xq`(U<E^vhi0fjH_cod!%f8_>;A)0fBf|JGv9oFUXza&@Ym)}z6pusT@YR+`j{2T
zhAQQGq8_P66WK)12t@&pl(j&-H{gy%(A2$po%x9k)EkV6tjw{LLNy5p*2X9-#d9_+
zT6jI3^jo6^zy}vX_mPs;r))gNwrpY;qy^u+1=wjb)xIJ@A4<lAOHMB>`E2chc!FjX
ztL%DVTml07B$0c8Jd_vhQTh}^`);seHYFrM?W|DE7K1j-d)EZZE-GRNXP<DK`2tLF
z=!+sKYX3yuRJQ$5=O4;V?jVkOKl`eGe<fc$3qnn!OcoQLybu^DGT@WqEKseS#jf(R
zD{a8~dL7yDEjdoce!x<~UOw<amb?t{kw~Ji<ayMD@EZi6g-qF+;7RfdMe~TbEF|Ad
zIUkEg=UOLvEFNO7no8>y;vY$T5%T^*X6e?MEYZ;6dp7n!u${WIy(f0IuGp?&$>ZU`
zKz70;O_kGp1r*{s(AhF=dyS`X)&t?&T>{8oD*zKDY3{-(>kgy@)Zo^zV%{DBrk!;b
zJuiOJ9?~t3uR+EWcysD{ReyJ5Zew31AN8?)kN6gAtdj2T$kv>Qea45m#gg|XCsLSm
z%(0GFBG^VUXI#H?eAE#VJ$^H1TvN6t)u#B7Lf`J|M`JBE0#0np1<bqx<Ki{ykMz{{
z?$|Jbv*tY=SM4G(xa`QzHP_70l%2fg&&I0EjhAJ-;rfD~3Kc)+StR%X`t*JLwJVd)
z=ub$me*ipugR`8iwXcB}-d?H}>odtOyeyM$WKsVyq5lkNxMa7vF=?}Befj(b!n$=y
zlr02erapQ98PD!(tCvHlXyJZw2fIwUOYiJ3mL|oqeqb_}Hf+zWJMV<-estE{ZV9Ul
z15uj-SldJ@)+g1rJVE4)zX2edLbW+cpSNjz52O{^d><Nxr0s0>C!R^ubEY@=8-wXR
z;g0#gBBM7mlJz|Of?mcwT39(gqIw}+tt>+6&0ZE&P4-cMopVwoZVa%#S1P}#vE*gi
zy~f91M(f5c)({i$qU-BISgA8{6RnE`Llmo`_vG52&&&xL){ybwiOGB*_21bt^E~=U
zCwu(cil|ws40jFrne}h*%3o_(0|8%%)x5Byz|R-rAy3GI0^8*%LkXwZrtN0nuY|p5
z>L2axqs3hStnZ1CtqIymNg>IIE^0r)t}>|AL1)hqvUK$yd(uaGfR`V5$+oW$M|d5w
zpp~hL+XFq=#HitD7BoK-&#+f6lwVE_v<Hn9@ecgc7VF2~c}TD{Eo+N=CLm6Y?`v|S
zRGr~aTG+WY!T8-hR@I+(TV9>~3O?T3=wl&}<&hGhL;EXXM^T{f=h{=M9V5o?drxJ)
zrjs6|HODkhNV^+!m@F(zgR}PRnqAL1W?`KkupPt7)D+ViHPr~&HF#}gTU3nAWGrOo
zD=^F>^JjhDk227|a5x_DtBmQtf{^@5hQ5V^8;nyqK~al+DvwELE0l>JEqIL996>+#
z{N$5`CA)sU8@}75nkug>ZDJKeQnLKbS&9A|wYqQ5)u#2=X%?{-9!Ja+H~k={fjl6;
zcIvetv|fU+@WNE^1}k+m*gV!Xh9NYvtCrGD_s&f3!|W&g(ccl077-HWAJwty<oBdy
zN)9`ibYchTBM4$RD3ReSFPx#Y7aRQRSNS#L1sGFKzL?x|%PJq^?#{KAeN74oLSe`#
zFU4OH-;sm+wip(zApEu2pG3`us4m9qzdV2wGcj@f2u{IYVm6EaJfEZ&2V#8d$glPZ
z`aOZr^f~1dgPNft)lAc|Bl=P-UQf;6D?VN&SBB&-BDijc6nB^yzbT1b#vqZqpdGST
zWO;y*6!$aE$$b^_%;U|J)7+!b!271l_B$2+_rqAYnPU35YyMs-D)5XX{&E2Fy9L|q
zWIt86{#L-#(^{|o^GhwlFU7|>z*@+I;1{}~+V>avfC}9vAUDZsLJBrjMBTGESeUy!
zO2e0zmHmSD+5j(kW~{A=(UGFy{uo6DV$hHRDB<5u=fXb{!kDCa*d55m>@Wbu%zC)$
zDEAmw%O{+gYdDj;K~ab$y0fqTI$DTw?39!!?p`v4t_mjRlV4U{WXYnR7v|?{`0cG4
z#LSx!+sbw-94Zj<(({Rr02#uUlmYCv3i{NkErI{JtJ}Z_SbvSBfK}NIzij8$0^d1Y
zR|;zJ?RsFhk`w1mYtv_z>2(%!iyVW(olEpD3B<_Hr;pvuN4Koii*-dy^rfW(g8Jhh
z&_ijYTSnh`YguaovqsXk1P@9iLA-@*4TK}lLPpP&ord#&{Sr6HFHw0ite>C!62x%?
zm&=g577zap=HjW&;y+)qhaQO=cVDc>$3?Sp^7V$>b7zPiL+nU%9=IMK#LHmBSA<k4
z<oIlnP(JhB7fFX$m5=bzF~!Erbhh<7v=A^OFaw1|)0)eu70=H9Jc{8!5-w=iwVE7c
zZ@yl8_-`^A6+R4Su3f|<qiAp+%!dyW@Cd^K8vrzl5j&@RxQxlQv&<4JocnS95Gnbt
zQh&QN7AjHa3;*|u!%c`K0f<+}E*?dwZC*AHccD0F-+v?+=mnwkuzwrRPQEEXC}U@M
zHf)<4eG3;rGQ=QXc0+L)WzqZ=G2;q>&e|vO`gHc)AC{vRx()&BXY^m2bUVPEB7$Gz
zr*>X{ueS&u7uT_hI9YXDVFiZx5!abhZJP2ZNE8X%Y_*MG$T1LGpdKa<0|R;oE6ab<
z`yapZ*u#pTf2w%T2CT-EPk9gvjG4kyh;R8@v+e@|OD=^-ZduS<e_y5_Pi1;t-h^TT
zl0=hS9uXp$b4X$@L8GWbe<|mVAxiTFElu22P*iI{w{{gSbKjDamrBA*ggmL;JD)x_
zsTNst@UHJ2uDvGvz11zKt--f@khtc{ABUCye0siC@*>tq8WP6v&*O!!yy8J(Z!T78
zrH?(2%yjL>O6gxy#Cc-4r_4X!z{~v>R?UkSER|I!CTVr1#A}VH`InTM=VKIf2ZeWV
zB}zzV1o&`_g&8QNSPGXfsb;nP?H2$3$AAhFo3BO4{ibB_lE^NsUv+02wVEnUfAL;U
z-nn}R4{AmtvXpAkD1wnu5J2TBfRqZnJA0qnzT8Np919x|22tr>C!`6289x<ndqnHv
z!=6cSjI2op0M-z~#cG&LvV^3*zP|K3`Eb!_e`uX=E0G;zO2c%eUaV4T_sb;~l2-~=
zU_rGp|8BJcz^o*H1xIIW3Gi&#m(n<02R_7Vgp|S3w|2E_0$$)w@nJyn#xC4L)z_Af
zu2B}lQB6XV{&*IKuUz-WOxrOfQ-P?w{>NNo1wq`#B0KG99riPOiatz#Hrp&Eyh=YM
z<=6M|`@?gWUo=}wE!t)DD_gqIh?KfRLmM2Z7a#y?(a$+?d;#8+1P~$fHGTWgfkyVu
z#P^_C+|r<+lrYMJaCi&3r`VsC0JD{R<4m*#>B;aUD*)Kj543u(Q0YJYWr}K{te(*3
zYCRURj`~4D+jD1I?`b}&o-P;w3SfX4Koo9pF<QwK7@&ZoXFg?_!|Q}_6_2fbw@DKR
zsHpP1zecIqhey$^>i1rKAaf!t6iM4=fL-GWouv;vcnD{iw2d%9Qvc#75IsWr91?+v
zd5t$Me_vzDL+A-}=~p@yhKNSGu@<C=)E+6pkfLgHf={RJ*`AI1f$1kZVyv>6p*?RT
zWOt?D8;{Hq&pk|b>#?32Lvdq>5-=IceI=6}wzsP@&^4TwZow&k5PY@gZm9>3LkWI^
zQ@(p6>2icIf>A>u1E$l{{z$p4^-BvCCK9IJ>2bI9z*LktQr&#aU%M4xh>vifux=@2
zZI%U#t#tX|-O?#&)D?vpcI#sC8PhYd$wkCS+ex>4B2unhqUTIP1lyN}Pw|o;ZZ}*R
z&KO14#XTxM=yJ4_z&ZM}EMW0uH3KaIi$`C0u5`Mn__?7uDnFYtHXT!X34q;039uK;
z=t-?>fl!#$gBAPlepf(=;#~2+^hG&lSNR+ZuYQ>mB1lGkj{WwD8~Yb;=kp_hxxYpH
z7nxo9)lDfK4sbLe8vZNX-9@acFZ4xkV$zE8DG8@!Fk<l{LI3Zv3k8|#8R|(Vf+zL7
z|HBmwMBw1`;4se~KBD3(z`*1lrNH?5*3}Z6K_hDVGON=8v`ic<rj_e+#=pc9Pd~j^
zJXcUL6+$!Ks#&R^Sv3@KKROHGI0DCtdl4)f*Kgdgv06)?`fFLg@uSOnni9>7M=b2F
z+^-^>2<-0^j_ki;`Q>j6Y#@VPtIo0PZ)fp(WWG~Ee^<+|j84T}{*Zom<Nlqb@_r?Q
z^e1~%IP`#k#ElXhXrYZp?pG}nCJ<LP)0T5!6i?vgYNrIe2?AujrOIYoqo{>U)Sv3Y
z+CJKg_FZ4URq0rRbHdls8|z=B>z?lOw%n9W2ji&;Yj7$q9@I){z__`Hv^^e<l;L6U
zGI0bjV#Upc-<3n;AMcTyn=t+r1Lk4uPv&Inm#(b(4OVb|4N|ID^y-W9TtqLexyaCH
zZXjUUpJ!a2G`(#CRhke|D3e1(rAIDYsq;k(+1IGOv{vtmKXCgZ=pmT_>Qp<<KgD`z
zPtLc`j4D=4ELNzVb^*vMG!`fTWk(C4Y01dumY8R2Q920e=liZ_XUdp?ykPN$Qdzd-
zuq9;|dNRGV*^fm^+pc|J^ZEiP%)&s-ZCaw;4$P%<VRElT4zhMJY7;XBVWCyh0{onj
zFFMy3;l(G$VlRnn6lhvK3VW|&K#O^XCJi!X=}wuQD^ao<lud^&JRQZvR4ax**rW!=
zJySmMWCvT}aZ>3$qhbG$NlXLE^%!dEhn$CtzuLjOw110RSexw9@5urM`IT=u@%r!1
zCeZqg5nkV!LK4J4JyIjDE2?#J;W%Ad-`1~@cC$c^zQ{Edb06ogiGB7xy~IQg`dP;5
zhkgUvun?u1<?fq(<!Om-i2JUSUUPqx9-HN{SZB1?g{V)=3||!8gA%thFh9gkRK;ql
z>Kg*I=D<w}V}1hA6cpW0+{V=4CRk4M9=d>TaCyr3bdmdY4Qo%==gw1+F7K`Vuln4P
zKkrRuCXj4o84}WtZN!+31V7JU#&$jY07FdODSV{;5;4DlF5y@Ac4`@xw{VSE){}&M
zx;y0=1C~ns#E=f{4NbSI_(xQpf(X=X+8xV$_Ik=FniziFDFH!yII|KZ?W3o<P5Q`!
zr7(Dzc|OqQ3m#jR+$uSY?K(huLDyR)+LK`Iqo6r7{RqN>5A3Gw$*NF`imO|&Yy|uJ
zUATmlyi1re2R_=AN-agg8E*tHjMuZUV~R|Y{88s+&I0Vdl`b}(TiyJp8TcoheHI`9
zyJiWkzkd^LqhF4T!XnS3#7Biz)*w${gW}vD^@w4!u-b@dY;OnwAE$k^_USnB8BUxf
z*n5yPHNC@ch&X-0#U+`Cd*kVAL`yeV_MttQ&0%4tl5_9`It**febt$b_2{%MT`;^X
zA0O>)wBW&LG#q4x7iy_^0N-JdE1{FVhS1$QKAR0L@%M?S18L5<AuGc{c4_m`uQ-*f
zb7!WXW~|lMWk1@*-=$ZF&@bRJK4c{nV@A9^0k~PmWvM`ACZv<4o)QTlR7&{m)Xddd
zR3@9viZ~(<Zar}wKfk&NtAWRVSN7`yWHLK1z;!<qf8-5)Ss!9TA&Ba&-m4-4kbDiV
zkM_t;4OI?&(7!Sklv-er-f3eDENyBIT>2V0^ikEsqLYbvws6QX86#Uw4V#OauzMjY
z+8D~p+P|Ix5Ch+}DEvjYITwclm7>u+j(UmB&onKv4ezUZzWLU8HANl|%+GafzRVlP
z2UJpUK~n_}v&Tfd>p)^1RA-cNms^q8)}MRdA)AZ9Ky(reuMhSZ^)WI{BJDEk9(mu4
zS>qi{0yp*I3m=_P!Vy1k#Ix0?ESzNV^9*dn5tGP#&km%uqwTi632zX+O7vSfzx<Di
zD<F{Ud&E|(@>KYhUo64`5E$c14-viq15{kQBIezkI|=rPrN<H4WUt7+i1@VnQ&33E
zdydF*Ld3jV^7Okx`Vl?3KCBoyeY}Q`*1xGEG~YaTiZ-Ewsg-+ge)#fEH^i+i^pt5U
zyO25``ta12u5<mwDR<zU;I@o+4>S%!QGZJYz*oXmXaw`ssdF3$PpagIJkWs&1KBME
zJL`T1N`)ky%hI8eOiR|or*6fE^HPmzi9Qc(4~((vI)5rDc6!&ekoYMasm;3Dfl}DN
z!1c<~EiH(+5|lO{ow^>1TQ<qBNLPqTeW#IF8R<G+q^3w^=#2e#u&fP#xj9$t_C!u?
zEsF+t0<DGh)>RdejF(|%)EgbY+>bMWD-gk-0Rl+OulC7}#j^U!Q;B@D;99NJzB2fh
z^q1_{S;kG2E~*!65t)NMcqq4wZj>(|oIz~gZMtI`+z7mq+Fp$P2FMX-dTOgsOPN@5
z_oO*K9jP_(i<0$_W2<cUrlrFaIEsM4+YWkurTe?oevt={^CH<_r-yT6?riuDxI6oP
z_hxV|`6cN#IT1jQ>*WqUN-LuGWq;yVbw~F3BtqB@VvrZ_JXF3DrWRJZ!L0|_L&&O3
z9JVnkIV+DOXszMu&dciLKcT(Hv?0Yem<ci+se93R8%zj}p$+}8B-vz3KLVmpB#;Q(
z*8mBk4+G^9&5VKKb!G)RsANfTXF7H>nz0U`8B?iXoZ~T=jVQ98p{pYxHN>DG5GQIb
zyTD(uf)MUqA337lHRbiPx~Q`PE~~6eItL(4`{;D_JVCB^d0E%d1xb|2M+ujk6Bp&A
zfQFBeXhOLtq$SdKMeU`UXC%)!legxW95CQ;x;19F8$+?vb*z<r4E_>3Jj21$dY9Bi
zz<<6xQWgE`5~1eIEq#!!EdlPvj8lhT20?}+E^R6~Zl96BgW(^jPE7Gmt53MS9mY7;
zi*`)R0xq`xR=r~FuTC`Pn$1YVGe>mWuYJwGV!DUKELPn(|Mjm#TaRn3C|Zm6{KdP2
z&m2R%&;NK1&A#WR6CU!6l|<kf@5eSzk0xh1Xku0L$xSK~OVNa^gAdyP*b__=mhb`8
z7?Ak1c<7l{#|3sB7VRdphI2ITUZ?mj0mFOwCNgp9N8UQuA28nFNdzh{k=l_)?4LoP
zJ;bWrZQd{qPlpl`>41Yud`4Fcg?i0&w--#3g&512%7~U_73e#s=4%H<`xI^DC_pF^
z+JeDLz~Z%sKwDQ4sOW*Zj2c`%G|N0mCp9F+dioZSnjITka*fD~**5%U#sPq}7GnEA
zcpfwR_f(iOwM11P1ERA94m6K^5B6T?`nK{Nnr6B3JSH5Ul*RQRJzMqM*f6~Gi(5UR
zi&x~jx!EM$=zco9AvM<_azVrm-+<l!WY9l<#dnWj{ub04GyitI;EOlvZFT$332HSZ
z(#4ba@y}dho(Bi6&7MT4h_Rk;>V90x9DfICM~k?!s??l(bF+wWT8m^rieNR>!#Fy4
zl7fnsNFWG1w!I`WCItE@r9uEt#Aid;F-H-y_8sRs(O~X|1*ywk8s~TYvT(u<ZpP~g
z0%ixVay_v{h2Dm+8!%C12v{>fWe7~4&h%KR+c9Qc%V#vngx3(UGOTTjPV?i9Ut@g9
zwh+WB{myc1Dq;cI43^d94>Wgr)Agh`x6|e!gpQ_b{5w&}d(*P4w=joje02`(W@3<%
zach0Oy$K&(b|gXAi@(RXvMynl3Pr9WKzGE>E~y2qS6!}+rwxjTGRLg3l&qNv<hh9{
zql#+^!}m_W=v46FpWq5}-U<CdQg2?Yak>0kC-738j8kauNK>or$BUS+%BB6wIOQHq
z*pj<lqR+ZSpI>?kC^7MNGFr3LXuWk~jtjAW2u{WjDhOnA@5%}BIGPVwxCOvmfJ0%q
z&|g!ctcmaxNhd1#vqY}UYD~ciz#6Rs@r@<EjX{l|@9KLz5!|470%H$rR393<K~via
zIQ!buCmUkI?c@48&kb;HIC<li)s!o8hhNl^Hh);=spFBIdRtt(FSZU8p4F(oRF2eT
z+I+geUW|Vc#?aF#8?E76wU=fdJQ8{r+{CC<F6jqYW?rG5Z~eJZE7oS|uCbpCw7qse
z<pk$a&>*OgQtvhk>V`D=ov@VzR_KU=o7eN=hxwO6Kb;YkjQrB?@Ve-H4_o2#<L@N?
zhlx_>Y#ym>ztI0{olr!^@X0O4>=dIY{4Gnh%Gl;sn#y7F0zUp!!^QXgQNk3#HqSJU
z`q-XHVB%G-uH0+ddTHS@kt?zTUV<}di-Lh|c^-61HGO%&3)J{4(HM)_tKcLF5WmhA
z<LAc9Q)%ru08M^q`pEYuziS7vC*~4*ktOv;L%?G#+N_LvgRjeAl3~~V;dq7!&<J*f
zF?UC^e?be2cO3+_W?4!(!^EF-&ka40_b^?)M22oS6pHNKCS}*b)u!n_3HOjzpv6L9
zq(d#l)<cm-Ywdu%--06Y(Sm)RwjH*dU+<Q-lEUTk*6DA};^6hFG@vYNMHdo+Y4ouH
z7lv%zdZ3zOp6%V{qSPj{s!adhAVUj_vcL1(2lcN^fDtp~<{&1?Ao}9hf5ZS0ihpGK
z>2U*Xu$R=DzvhN1jQbUzM(JKkcOJ98e>gtaEG5%22aAl<$)Cg}m1$o4kIRWqOpEch
zuB!1jb+F4^gI(<gutDls3XlIJgXMx^yt-C$=HX9@S28K3>iAt%C@!5B@qdfq6lTuP
zt%n+Crqd+MYTpI1_hju9ms5Gcu$xTX8yuRM%9Bj`Xs3D+u<7H69cZn+UOogmGF5LW
zfxeOo{J7tCM!|IfFNmWhW{etB@Ay`5J!mD$U630***amjJ6LE95leta%eUkc;~no|
z$|8XQ;|>4A>$>1n20e%OM|w#hBhCTnMC#V-UZid60L_=^RSnE^*E3`tJ~;EOzqj&%
zV}wLdOwfW|2exr;Fj%|mfa1k~MZ4K96kOXDvu)QRHE&c{p=$XO2M+DN-6mx_J#lur
zG$Z%%+uBp<qkASICTbGd+aElsK6dsYwwLv1zZ$zh&hrjcO%41%8b3_8?g`2<^*jIk
zUxnCj&2+#DibTk5-+s2A;g=D%JfAs*+^YP&%2f0%P84k(T5j_>f0?hh$n!POOgR<e
zjqD;mZnp`8W_r9!l#J>F6TLw}s-3U4AE-I95kR_K9X|;^W-H2<(%oIRZu?8`CeR`X
zv4{<Lvgv^9zAt-gXYAE;PX#|A)o|E8cvgqJE(j<3g>Nr+x{zWAAS@*f?bmLQpW$}w
zj~MOtE!}Jx#D?BxP*ktFzx7O!oj8@0K|v1$NLYCH^=$%eg+K>AAM@vo?&iBBzY5dk
zHIg4KJ@-V)Z3(zO4DR7k2r`-;kq)%T;wp^B{TV<-E?|9cI+bAliDiRUT{y6wXd!oi
zWZIIsH2_pNtInFoaTrg@g9OG?=-*it??xoz_rzV<_CFr@OVOq(bjr&(g@2njwJ5rm
zdP^O+%}g08m?s;c(SEgj?sv$WQud2<RNsD}o2&CT>O0;(ECo_1WsR0IeTec{Mo^aT
zLZN8d@p(9J6D~jJW=lQ3PPZPcxA)wXC@N&!+mb5J%+-AYdT{w-AJ2($&^#BA$b0p*
z{;6L!ac%|@|9zp)Xx6D!d>&R!0xLnql_Cy~pXJUSrIbnmMQ`XBa|05zU|PMMrsa{B
zqdLgN;Z=HMd%1)ufhDW6(GWzml{gSgZ&eG=Cem8p)6LBRO+{k4{sM^@e<vnM?3Mxy
zv%m0sgq;&f0uLQGr}QC)b%?Ta9a@#5Wm)_ELH22U-ttSE@Y|`yA9pS8`vU!Y9`U`u
z<^g(`589ld=CNBIu#@@*-9VxD$s2Rui5~q+1SM{a$O9$N9~z~>wTSqK7hw(@xJ;R#
zI8VL}<5V8kg)+ctlz6+3=wL<uok->~2>GT$AlFW`Tv%zX$#?0hs`M@iv2A^u0J6r^
z`mE)VJ-z6~*P_zhQ0xk2A?tk3Qye&}_K7ssvuk81cKsDuddl}88wUJ~HYNZ_Fq2Cn
zev`#KECjBi65zDWT2!dud6O83chnJE5wzzU2KWWDI4}>FLC+(Fg7RTP*T%$Z*BD7W
z==7#@j0Q;l`Icl=L|8GtS=|CdI*Hi-?>YQ#)*cFEx0)Q6vmc}Z1_^`7&_dWiCOtAd
zXHWe#u_>!*)O^&qZ`vRm-V(_M$fEx>CwQ=ssKHUOH)!so#amgk2oL{ql+*&srAzdd
z_Vp3=?3`m9AmQP_K4elLH~meY^wfTGsQnIN)MeO0vr0nm1prvwVS<o83&f$?4ZxGO
z=XOFHBypvjk2Ns$<C4N?WW*T8ubydSx{Ko7^KncLVL(8YYT!r)@nu2}<~!+7`ZR=6
z3S*iN%ak$TT#?U#lfMq5oItqz?MBb*=i>1aR0*MFE<*=!Q-%>A$mWJf+}ySx{Tiv-
zm2_gyAElBqiqr)09MG_ga+h!_pz2Jt6P2QQ1Q0VmVeJ(H$p+39UjWXRMebG7NoM0Z
za|(N>Sv(Bht@gsW>Sm<x-b_)6%I_z<Bb5JF?9iO2^!>2Q2T^AM&J%P|<?-A{h<`jl
z{Vj}x#;cybg#wXT)-s5bf9_>bpN$l>ktp6XfHU_^#aUkt*~g6o=h*0&c>>5j8tw79
z$A%x%t)}D3F)Y))$jp1X`zdt53_wCtiqPawqp>Dw>&fO^pkIiMY1*d?Z#1!}2d&}d
z2VD}@VH_M7=l%UT4}GoIjBp({*7qeSHYoHelY!G@+4&GEDcO+^`U7R#7jkO9cV83j
zvC5ZnIfE#j^$jVTyMzxMTfk4)fEy##FC>S4s*-1Rs{b;~on_no;Q)3@9AVd!N4XY|
zO4+(S+`brvn{P7+$=al8Xt|)`Fl+(pAH98k;oPavA5JgdGfEu{n4Z0emHPE)dvMU;
zajsbWiltBwk8^=s#)6;g<=F<{`TTU@j;%Nlt?0U$G*_J|fab(X(jW!$s_;>!#g#{5
z3N+0}MeY^Be;L_e<z<ak1Gl3&Lz+A@p%=Iv&((BHla&l1GRAjR!@nH)SX`nSMdebQ
zO?#id0UC!QuvKGs@w2Vml=84?FO>`#N<Y(>9=*>KY1>DAsv~0xeVp$RvS#6g-b!G5
zaGy2brT*puP(8GL&Dqx;QYPgm<Xa^MUI@;r&XwQ3lr%w>_NW7qF9jrwd>U+?)j>*C
z1Z*nix5*6?&T=T<{U@<p%a_X@4H4B9DW6u=KXf^$?=(QF`-o20?NM~nS)h&xL^n$^
zvur<stzpcdcE{an9qenjKYn;7#U`#@j+ZzOM894C-y5lcSofnaxw`tVG2c=|?HD>$
zN^N<*H<YibjO8eJmd<^P75q)9QMxh%=<AN{yKLXp1Z-@}c{bHYc1%2*_oFN}Xr()4
z<nDoWCGDwQAf(QUY?g>TeO;KSnU>&Yl@`dkvLCsfkgt8SnYcGyBhpijfs4#VfO_a7
z+Yb`gZBM~?kGM6hYZTT8ql)a_2u3n#g6vw4!uGMSQgLm;DPGxUoG3}Il3l2gEkLWY
z$Q-<Ioa}Ev2;D*CJ(9L+?@a;FM;{Gr-yqTFO>0>5)IKq(WF5%+2Z<gVL=U;AD$nag
za=jBga0_uJoyNI*y@@+rn*4n40k1pZC?&@|i>QmgG`?}2=#iAdO`_5`uA4nBNz9hE
z7<T_!kk4sYzyAYY((Iv6x*famw?nT^4t@OMR+d679QxN=(tLlx<UMD^p|7qO=iEg#
z59YAKriLzmXNs(L!=dyO$AErR^xH#hP6B&_%lKHh7VtY%>M>(^%kVacXqdz0)f9IV
z^HQbL?;jX<iyVkK58&Jm!aZuaq>`Q?fHE7)hSnSU{(%^{*LA3p0W`rhAzE^enR~-s
zsR%?vUu^=P$C*~f^v&0KZ>+WR`)6!n<p2$>Q!c=2uNAsVUJk_4mh3YOK*&*&Vsf{f
zy?h>uYwJ-L386X-LKjQR-0N1=w2G@G8aQ+%SIZM9mkAttX5+H3#NKS!rGuaogdWe!
zf1^Vs9}k!BirIHUOb#XVo<k+qFB+;m_W@(ywnDUuAzTt;M`IoWEuEg`3Kd~Uoc4np
zuE;7Mp<kc%++#fZhJu!p8u04+BfoPOy+rcbUv?;6ljp0Xxj`pBwP1tX+#hAhW~q5&
zsg^nB*Tgik>A{21Sr0KN;%Ic{-NGV?0EwuIyb|}x_MeyS%cEi~=}^_)|KJ((i5g~z
zq3X6Ic&t2^hKZuCNbcl7(}PF7^;$vj+mng*|H7FW?(o?=hJx-)Qd-?z3Q;Twgt2*j
z@MMByI06DHuJ%22rSu6KL-9LzCQw5!mLCI--Vk-iw7WpQ|BPUR%cyn+a`g`Oic97?
z3@uSp74uetQn8>XMlrL1$Wdl|&FnG&<^RZ*H+^b+bUmenijSD)k@Kb&`zbzfN#hQI
z`?5|nLQ)4qHhq#*C!y_`dNEGHnnblVAm2PG6>i4BB7m&i$l|Cf{;1?-T)M3le;;1w
zw*Lm*GUxqoWe4}@yBTe(BkLXN(r9TqW4=r3o4&uKq_8c;|4O*Ov^RVK^JW_(_vmu!
zxycna49rNsH>f23<_kEENX2Qljgxuqcf1rpq}rNt0tt{f5cZ2wzrWB+ro-CwMBJz>
zW_k!knh=*eU+`XbT8IR1m*7)K-^<SH1}N<(yQghklzpgvxF*&fy2nQx6Cq5ca_0DB
zk)av%PE*f#5r6fs+1l(n$wz04#x#MfB|0vx#FD60xBFqcg0^ko1lELxm>Gzlez)2U
z)||Tmh90HMy$6&)hlrbWRkKZx;yLk+UONCEeDIz~5ORWU4RE4_Nu&-N*A2z`=Vm(d
zTcX-EdKYHvY)=!>&iq!O_W0!9nFLVL#;7Q>@RnJ7SD~oAvB~UaNA_Qtone)^ZZl=w
zJa@8rp!RSCCfvH%MVR^R?VZ8~<N$j9wuR4GwB$X7CWk8bO*`(P3MCAKTurKtaJv&q
zzelo=H5#syt;UaL4fQY-UX2t}Qe5rm0G&bW65%ugq<cIt>__{oTyJxdsm>HP(#{E~
zv)BHpxc<*Xl#gOI5K?Tv%`o>yLD`*$zE5gBozWo$z87d0#JkUn^UVDYuU&a&3U29_
zx#XKL9?d^kyS?NK<#BELu<*U?Y$oy-EN_Wq=3;a3yvOcMF2;1NV5->FAAO{!;5Lm~
zWipYh4FJ61FvLycEL2#Sz}SbYqjY;&v^VL2C6u>!@OPD2X9j4ppSe_5u|CJr#amLB
zii5y&B{-Bij@;S(ms^uRAWmbR1|8`Om&$S+d#!vH<Zip}@vg)W$wueO4t>iQRKL^f
zdp>4t*euh17Vo8u?=M`75F|Pdm-q9QaJetvT;_g&kDDZ@vf~|-mB0jet@!)clfQWY
zbqU()=IR!}#WXJo6<PdSGM2$1%?ixL?8Z%dMddEwm-3g%D@}pMv<{fiVEpV$O9q$O
zpbV?Xv^6n_)Vz3f28JwjX^(gOKi1wm5X<)c1D4E0N?F+{Wk$*#MWR6}WS5dHJA0&}
z5-Ou)6UofpBQne0J0qLy&3l~c>7hr@^ZmW=`=9Qv`@XL8ypHpjpW}0EM-jdGa7htw
zgl^=_mo}%1;EEzT(0}BfNAHV1A~?S1_5vD&u<4j~GsDTsgA7vW>4WWo#1pNh>s%<_
zc@2Z(XZsc;&CR+{?gjuCai74{FoH+r3vNQ@_#WRVf;3Q7ESZvmtrp0Qgn;0g23UTg
zJ|h4T)nM!f#DsfPmxRl@?@>TbJFp@~^0x7qY%@2Gk!~r=uz`5!U}`Z`BrDdVbM5}3
z=R^fXP$op(4N%S#>2GAaM!NW*ADbQ!8qcA6cPu))36oE|gnO_QNaO<>O7aipNLN8E
zTa>FLUTsfb@J<^rQYifg!RH#4$5j>>cfX1P266*LbV!?AQ2JToOk17Bt&Z#sb6;fr
z-vcT9H>AAsrY!YVqeJI{X)FDio~nVkln(j0{I7RoZ@xl<6fZ3}57J?^5Fw(WiMMn|
zfIw}IL%cEalh(buhr>{Bc@E?V*X5E(1nH9#am1_FZH3hb*$;Juo!f(&0G?$Vu|ueE
zvi#u%MBr^wE5$^SRRv0TvF9m=n_yZI)V$xq1&JV0X&pDGrG*!7TLJcUWe4j=(xonU
zGj|T9E|bi4U}>4eWpjO-i^E(KOaTtpTL7+<yz5C3C2Hj(lmw`f0wQCrX)<~Y8m|m^
z4Z<O0Mw1(>a7{qJJzUQu%u#?i=O{$&yLdjonl^>G!_;n_(_pAy0#29f+Xrki(3ugC
z15yQZzu$k%A>K3@XRfqT9vxdpFJK2_7HBB?5vZj~AA0!jJ}q%+#)eb7==$zAR3Whk
zMi5DFO&B}oYkJ7(Pz*em-6YPAqH|abY=d3kU7}PHbZm`K-ahSd4d`u#!0MTexAWQH
z5*@RHP@@cCeO|CkRSHa$$|<O7_LumfV#_+_0-VDv`I?7Sl>Vok>9Fonx`qJz7QiqI
zVnzWSRt_w@zTWOLcOY|S0<A_*DREK}*>k7u^cT_LdxT&9;sW66iN;B*K08YJ>MNWm
zx9jp%rvd)CG7xe(<jmd*9E%7WTPJ(QB{}EgP;xj2N_e0(*?#UWa6zN!4pl=@ICu@>
zGDJ7Yp$<S~#Z>**_3z~>QtbIVxJwJ59&-$}vm<Mj;sELGDfJ0d+5{r<R`m#|)^dT3
zjG4*|b^uON#UaZslyNKepbctpV_yqG-59zJ%q$!}JOFXaF?SeTjhI)>eiJxw%bl<I
zRY@hNfkiNnVW$E*OanWnLa2F7`drtk^o%HSNW%<Zs-`ER$VCla2-YU<x0Bn`O-;mF
z2$-B$^y{pNP8yYp?Ga*ZwYT8;;UcsMz4ZF8kkV!2xVaRDY@OHcDC!T#&5gu!81J?t
z#<VDozxbmcG5t?1*h1>Mzm&fE70!WbM?A{iHz6}|CGoaY6ewc=nN-u3(qCo|Y=T;)
z5<#jt{Ss6|#mSmVvUF}>+qJHOmx>0@z5t6L1tQ)T{!)6^@{FZwJ9Hdb!+7xoFJ`&(
zX3uRcQUQ2{u$uiMY5?I~xziFwdldL%t{^I}uC<B0^||>pU`+xfJ6|>SFkeeHSFZnz
zkWv6RALT&g+?Up6KaKnG@;W+Z6k7gY9SGxSjjBcQp_<?Wq+hrbshzmxkt3Pa_PY5`
z;KQZ(c<}WWO0EH35IJ3g3@hE2Dn@4GSKSBrcN?CVSp~V#L@^aml%ob3D)V~Tkh2=+
z5hFAL9D}WY38PdOChA@>7hjfiPK;4btEBZmTQv!IC2FzQb#=z90l}>PLUZwfAP5Xx
zmxgoVO`lXgg-Wq6DEx&9huO;Uh^BGK@ec;=NijHy2D>`2*DQ+fe7*nJ6~?SDcf98z
zI}XDP&7W9q*G0SOHHn&Fqx_4>xl1FPjSyV%73E)~VM^ANJG|XazTY}N)@0~>M2XfF
z_^l>T#R$@YTtw9nQBpv;=sC1dfqRkP29kqZmu$O^YJtuE#y<i&r``LIwWJXLe0fWF
z=cIBNjHQZs<{&Gk)ZQC3lDk-*Z8oQlCD_@-rC*_-@*E1jidPV11U$)#a;BXdPmuN3
z6WDyIKJgz<K(g@!|9}Fh970cE`hM^zrW6T00VNyZ=3K0!3wnYOKsY0ADX}Oh<t74+
z>f<@ZFYm-Rwt;R8v*w#YP?fKb@V-)dwo$4roikgOi=LT|W0xh&GRn4OCqJ>-<EHDz
zhFXR7SJrA=tXdXOv}z9STgGY*?UlSo=K?v8^8lC7gnBoQC`5|LJLjB|+|bCD>iHY$
z-g!Qr+WNOikDFuyonKG-jZ-gXNrAKFAv{M=TreUHczU|85E$Syy*<5oUrVtuhXB3-
z{7g~+{9%m(p}hbmDfVY$N;X?}Lmoh-3fh_obN;>W0Bks_*_<xgYQ%|4<&c$^i+ozi
z08>=IIfW<G9^L$imMU}|PE^uDv)2~E)h2uk`7BjnD7pKyzo(S?r%0HT%!+aRyUpVg
z&5u^6yo13ewpCzF-h#z5?bivb!(+VWXhj#|&%K35E<>Cyz=;Yqp7H^>hAj&4Dy6FT
zIy8UwQ+3<z_@R9}^DGAGH$VGkySbr7VOAv)hTUH^$+sL&(Kr%FLB|Q8s^kOkA{qU?
za5k@njtmd=Sb!xJ42wkWMK*vsB-FU}q2xQ1?I<q!!kA`+C_2D=?f8_`#*O=kem+-5
z?^1~Gmizrw>3@GlL5HII6OKN=_3dR`fdWs)RB{cFt}%m<UY4fZ{F<Fq7|{3^D=5Ar
z<l_{WVMH1lj+(ZSljG5iG&ZdbO!q+@l~MAseyCOfb$Z{KTc#LD(wzEWOHevgvT`v(
zZnNxdeMuV!ixtU`wz=2tt=Bnup?K)<XDX~6GO&A=OEcBK<x9~?#Bfr9RU&bvCsPqL
zgGq4zIINpuuM-=I>^p;8yS7|L2g7CfIOhXSSi6aaLcU`k+xczB=2~6)aW=)&vi4DI
zH`eR#S0LKDajRBQ-s4+@$Fl^|U9U<wFdlh09?BQY*Zl2F|M;h<?au?yBM@J|8O5>9
z^62iM{Ue|L$6Q#?%b<&nc)ctpF7*a2@Z*JVvpak6f4r;j)o0HrZ1ew{wu(!6<FKXj
zhiw*^9az~c`2*q;fd{Q0iN1%}^~&Kd#@qc6m)y_&i%>Im+kP79UpLD)MY^rp^s?F3
zv2vw{kY+i=ga*T&hGUIMl(pTRh+R+rywz?d4~D;w#I}Dio<XZRqM@JstCjM{A@{U{
z9lX=!Mu~UZ5A171fNb%0rrJ~Y=b4Pk-N*ayA*X2I0c63f<{lh3`fI5h@(H<STj`6_
z_}$;}EJYoy@^>-;&U72u==<wl8B2+w)7viR*Vi(n(7L`I5nI|i1^o64P>m;s^tni}
zOpd_$Kf$B>8<yzDPY^?Nj9e6AUa{=W2K{d-|MeFjVv!-}yICi=Mcf@l$zcm0Inf_<
z5@dv1xk#T9+YAr;OP+yJZX;-S3HLYrecbl%DkGm1ZJ~us*>9N^$%nA4t0A8`iNB{}
zOgpBbk(qy={a=55&;4;L+zEAonJqn|{}3cPv5iCg=sVX|-9*=f^hHwu7r&e6N2D*L
z&pU7TDy%O46hfN+o(;RSajPY9Ual=0=$VG}MOu|e*f~+SQ##Q<9uh6gRp#f7YzMqs
zw|kOJpB1qpt8j}31ADVEf?ok!Fi`Zsq?Q~L41^V(Lf|R>?Fxh4^gpGY=D+7i-yuMj
z@$*0CNZ*4+4R)A{RS`>i!@+tV&bK9T`BQ>LBf!f-R;09^#oc&Y5&d;SETxgGTN6F`
zZBDcm0y9|kiJ!QL#ejEwk9ht~vMu}j*AH<1G+3@*`((GHjQAx!U~$AZ?7g_#v)`#}
z6FFc9N8iZqbh-#$Xle3^a?9;r*Oi|R7JY*PX32IU`nSZFCT+NQRe{7fTl`48a79?x
zkJ{88SeIotV>R{`h4SNpomBq1mx8*y*_JxiKc0+^2Moc5!uH#e>RTkk=^`qn9|^cr
zd|$K@w7bbSiS5`=cVJ=uG+LNit_s^c%U@?5t?tE!5WW=D-vXhBEAWsffBg9EL!K@O
zGf`c3=iMgR{iWKZ_Wn>!o?dmA!nV6$(`U^amSp@G-=@#d4@Dt(2D29*4rV8*>#04K
zS@SbQo1gL522Btw*ybjiy$~hRxj7L0+pqlgPb*ovR*}y#O`zi^RgT};vJLqJpyOvg
zDhBVt3s<0P74+=vG&b9E#a~ipk|&zEb_qsLn5?H1;{AG%?XP_-n(Xi@D2{{B6rZ+Z
zYr{5|G)M`>z4H{rFe?|<csuJf^RGSJ^5l;6o^cK@&(h)ca(n(Qe%t;<XN6e<oCKM<
z+!s4t(QwAi&qs%B%b}r!Jx<y5#0E5_2Y0#c+j_-M!R*5iAL+UcZkyGBm(Lb?`{(7y
zu|ii1SIq2mw9Q`pUIQsdu@09m(E~><?9eYe1IKu&w>?sHzIZ=x=&M3!M)J>_Y+5Vb
zJc0PK%i_>d?eX~vF3hb9ytfN<|GhvS>t~`2=#GyCNe#BuO`c@fwX=5<A;c$(KUM7Z
z%U1rSayIY3C#jP*um-3CxDngg`8Q*W7etZ)LI4uu_=jn&MNB{5p36ez!Ib~lOjpIf
zB$RWH<NeJK0S|_M0M5i2KAqi8D#y_4^L7rc`|Xk+|Fo2(Q(ZsuRFsG&6K`<Wwkq|n
z?}^^|wVO!|zs+auRz>mheEYuz{ylr4MP1OkemlQm^GiF3$y!~8Lwa`0`ke$(CjTGE
zl+A}gG!XT`2MUhu{`XD<I%>Fql#6wz&^Jhk-A3KS(4Sf}k8(bcPrBfQi9N->>(IPg
zKIA)Z1Fk0#q9`=8wmw=xW)~^z;On)Ty;xTu{~EAM(;c~*mg;akJ4Md@iyPwPj|W}$
zOuF^IIXjyczHy-FaM9{_P|BIGeg}k0*)0#DJN3|)()1SaHWD=WSXbD-1JoXkeAk4^
ziK0s`0%}3cL{X=c&XWx5Rpj(X#XD~B&P2`Nl}PY}{&vtdi*JT3)E+Y?V`{Xv9F4!U
z*I)VoWGEjqK*g%n#2|7b{|<O&w>6-^P-V}VxpK$@3p0vzqs3>mxrTQX5vxI9nfv@4
zgRm3OBA*3<7;1psK81XlcqMP|?hNP#>(%pW#Xn&5CQ|G3D|*yeGia>MYJ=G>&@vs$
z4P_f-fENgLcKrExbL=86EqJQ+P<Y!j`|E7p4#yxcdts{NLB+>_s3!n<Q2}tRk2K+s
zheo5l5xoqc{S|;{k^Pb%H@>nk&jc^Zc|xlVq9xpYYYZ~o&?j^dsE#C4M1dWkA4Wo)
zA`$>hq}9S#muK0Q`}iRb{gHqR66ipit==4g3eTv%0kQN201Jdez6=2a5ZhHVf5y!c
zV3Y6O(~j6}!m=Qrnj+Fw%J`hr$!K~8Wmof#)YTp%Vr~~u!Gh!v5xm9-v4C5@DU(f-
z?`aEs8;o7A4sH7E*9L6*tefnHle7u@1L&=g0`hs2dP{_nKq!=~@yrWmr|(7Rqe2`c
zhbapYVLD9BE+{?ZB+8EkeSsTTHRD0V`i|&;`#!y_?q;{XJ8-+skqF@GUI6>N0NDkT
zTJhEq==G9AV+EsWp*-V_2t&)64$Wd3PAB6Ws1YU7A0p}QlnN3$Cqc(S07{r6$gX&+
zSG4J&J<Bk`aqLT~X#Ptax0eaQ!yu!5{_eoOFAe<Vdlh=2lI;T=9Fl;_=+$MC^+rft
z#4`_U9g>h3eyfsU(xPj<vM>n<5wft#vt+7_h^+t;s!rFyQxIOai)ga^9&uWNyIC^y
z6yfbqg8tqXGvQajbaY_<;lNW@<q%*n5oY;~7$NYMze)Kqd3!Fdo`8}UHB1~SL#Je5
z)|UL8&#%6|HT;&5Yx%Xa%ye)!Z$IYes$)zZ)}nAc4~K3(kAG>}pPzyI4##sNzw>Wv
zcWOThOD_saL6~xZ($nc`{=m49c%uLl5U39649Ci71uSCBCOSK4i|CC+3Av1c@(1^Y
ztU5sX3@JD@9>~T)oo))|Qc-S*cs{BJK*Hz*kqwA_1t6M{h@&wkSKgkdn7=ID7X7ou
zdCifT&)^XCEP#rRKr!?LQS6SJ+@cqmzom<Rz9lhm;`rB7GJAf2sl1%Rz6gn(pU-(|
z6F#V8c;JTp+_%@grlCAl1UYAZS2(}kFr6Amb`guD-M{Lm9atLsL9*R0w;Y{=qj?@4
z<r(Ee0x};!sE}Jh;f|PaDM7o2K{3xh=K}{X>ZXKQihwvS?@|Y1v?)MpQI-S%Ns<{T
zu+pbqydyD5eE=Qeeu8+sYGDGw-#X5mO&tcrLVtSl19LQ56%~QV$G~*Rz%WmoLw7+v
zOo4tz-7beJ37XVwViC{TyOG%kIL?SaKQU7g7@2AoFO|G_!nSw%I`h)&%3JAQ>R_I$
zFLVZT2XvHAmq^1l4ai_%NRi#Jm*zKiCjxCq^+EPGJpJZft0DkxI34Nwdso+{gM>7~
z?I)lO_c<I{U!8Ec0PRB09zkOhGn%aToJ<l#9&r%Q3^!Z=)HP{orFATjVL^562upxF
z&(EGB3Kuot%0vZz3AOYhu%F?8V@}knq6tGY43MJxVWg^oX9|%cJS!vuCK{aG<TpTL
zSOP8t&9UY5(}r#<AY<o{P%}C+@DO3MpmIlx>ZbF3pnqa)<F#T`GHOM95ud3$-#2ow
zzn0~hR1<<eiS7|YqF=wkoTNr6NWRzAvk)C5mK?|-8Qd9<$9TbUkxwykGNl6S$I83y
z;@%&c(NO^#nARFjh~@p;(E|P{X-91S$0Bc`X}}lCS8dwqK%+SY1!zjE5PU)}gbS;H
zcP(~rkxW?O-Q|LEnX|=kV$M2CVcpGKum)ffaHr*Vk0wl<925g}`)n0av!iw_bk<x%
zp|i3NU`9MdtAGP{6{Z)IRYshZA0OaLnEK5F9OIx>$YVIz<uGwzl=Ac0o!JjM--OIc
ztvfiu8GCIAN(Nf%6@iU6Xh=-U7VL$@j&B3pJ?HqigGj<KK5WOq{X=k^RG3BSTvV)k
z22cIF+c!=EhpVSb5pXBsLH^X)&w$+!2v|w|`<LutD3}Q#_>`Byabg1A9v|<ZCvFH0
z2791I^&S9`sG;@{$CVyLRddiVz2GRK5Jscrf9cs{s+q~7?HP5v$?R%PXGb5m0_Oo8
z&{cB!9R%gn4m_p+AJ%{xt_h4GRmwswhYp^&dbkumq72_Sqic3ofFA+vRT{7POz%8;
z=Jl1%(a{&{RZx!Esn5CJ`D$k99{y3uibf!}>}n-r@9&P-=F^2m{5=~T9GfSAay|lh
zr4-6%<^{7*vV03O4CGgVMCTgnLSqp*Kmn#b0TIR)hnUGOLXrJe-`VrpR;q3^*|RWS
zj0O%)H|zQ`wCR_@r~`UvB=$lx#g<8)hdiC1#jge`2X6p2od<TdeKq<Q4X}z8fRzBz
zCtNgtAty<V<|QFsp3t*TQR(R|crj)C{(4zuGk7%Wz%kkfYTO6dRLNHtpb_mYOstUz
z<VRJ{(x*%t6lef;Lu`85-w-wu-WouPAHz_w3WKzcJ7Y^A*nKb!FL`SEd@&H^t)4cW
z8MqF%;817D<5wlwJ(RYpOjEh$Zz3BEVBF3ziY0&uu9(({tgrM2`FijVzYC}3c|9#t
zD?nSaBhhq1od2TQl42%M)vZ@OgwmxXl#LlaTwlg)dX0l?vP@m`-qYEw-FBVP)V{H+
zVjVE~>wMAB?Gvt{R#q@Ht$zT5in;n~&OfXFo_*l+tJq%qujIPL4${FQ!vhz}07+g7
zol{!-8y-$(dS(IBokXs34;<~6Ak#`&!qCf|F&O%EWf6e=dsXx8y7PCehb{vM7$G0D
z8t{E5LA;-`H#gkUIS9S0KF~BT4!{;+2k*xuOo$RC={>@<A|r6;?d&u5^LJr>&htoo
zRbmks%Epoh(!+JPQ}B=jp!rQ;?DEnyw1Wva1moH@Je<uey~acMts20n>t}jEmo845
zOyz-hRSK+dXLsr%wtXLV)nO5)R-ewD@cLf@&(gv&;5}sczCIvHG$&Wpj+aSl>K$Y#
zuo=%d2V-CTVorUl3}8Xb+ADx&u%FMrZ%@ct4d52&=LrVrJ!7;jSJ&W&St=PnZmAjT
z20gljsE0WFP{#)v`3o!&W-z(%)yjO{x;N+{)T@iioO^{Lzowd54=w=k<I=iANH1bN
z>dFG>IO4@{_`b`0u=zt6rxzkRijWC1ME>G(3g`m~l2%`+n=UQi7&Vg8uJ+1brO37c
zWMe47Rntu9G*v~83!mlp;Qj0>&9tn=V{3%ao{O=;R1P~|rP~Yh%VyN8Ld+3%v&KfP
zB&LAdEU?aoOQqWr75@NNrZRDCEQx)kL6n?7Z8#zj!Q}^EcFbm8Di(~GkHTrFIoraI
z6Yxe2403fyW&}V)wT7DFgY?_eYee#($G&y~U?wb0Qz5UT((iVB)0xc%TP2qD(sKsz
z)%^hZc!}I#Xyf!TXHS{T&yHz&jKf*WC+93*l=xO-t*p<6>4kB`D(f?gSRkQZe-6zY
zWuO=6K-rgv)^t(K*e}9%j_o1#N;x$6Y%OlG{;EtFQ1EMlTdJ7Xq<!24_>W}KwD`#t
zwU=)uf^L7V+OY(Hh;l@Q3g!P7AVrVY{C9WG%wJno49UawH!h|q$g-SZfcg#D{=`4q
zv{I}cOd10JNxL5|n!jjYLpGGR-;6$t=qu}RlAsJ<zt^jhK`eoU;gLmS(s0d`uNJxf
z*mH`@VL4WS1d6@I`}Ff_gFln>vOji5qkY<~%zinTPRgH6;#Dxfx|Y_el4G4xrPw|Q
z4UYjS9LWY+&r_@LX>O!YvlUt;W>qO_k41xrS$_-5eF&@lJ6{2pEg-O;G}<<<fr>mh
zCh*jDb(xf1We~ACe&8_j?^Y1{1`x`m)(eeq0FJ%nkL@PA9A+mlTYuFKkb3ic1oW{}
z(l?x3RWJIwkQgz3o8%)XvXd5t%obe%WfxFPGq`hP3Cy<R1l_fpVC{RB@fslH%Ci+P
zqutal?Y=a|p6E+V@9xV?yuO|W?d;bL{O#si<jn(i`giY=S7dtUCMSO)c~51iU08PK
zRY1^8LECNN04wXsIc$CNK%>Cy@e3evV>s?4mTbUbr?=k0br<NfqSqj*|0Z8we&Lhd
zLBqgC8)k|&n89INLS^cvxi5jx&isDog8vOZPV>Vp(=tc81im_+nAOq=*E{SJ<Si{q
z+67cFgr{=U%`2fzRlR6Lc_8}w07fr{Gy_kUNZG5PhI4IW&^s#Q;Qp|Z_G>Nll4u=e
z8zOcxvT$1Q>|xxJSNY{?Z`eCXuB}{H9$F~~+S0JHojAQ<&3`FRLeL%}hP!@sz*#v5
z#>n(R>`zp=Yq?WVbll;*srC*nCqKXrl?MTHz7CqsF54F!pZjF7tP&7GK8TUz-3GA&
z!0t6+P>Bg-3MiHnUsmTrM&Qf4P#JMy&}wekU^tKh?1!{NYO%)t-BPHN1^AC%daBHa
zyMKTf6FkYhbTZc$$^aC~yNZmAk4n;=XF#@6cQj09(c->>Byc8uQk>BnCGf4bnFNxO
zmjEB*Q+b{~1$5!#6CO#H!rv^+Q~;2;U+Y7ZY>8qzv$W%`LmZxxz9nd$1hgo=F<-@A
z>YsZ*AX()>LYLo+JhmU5d!Jq!k<OL_U5(ojLqOtML45TL@QKi!eS|5%?8RtXT2*uG
z{18)+twx_FQ^-1Nq2hPuN=$<=KjAO!iZ|{;X9E??GfLF>wCmRVFQjawOvGJqlb=V*
zyuF9{klf0t=%>Ts6uxc_4QOkNh}B2JURwi@I8)$6GTGA9>Gc2{<RlTWLG5L_{`5H+
zd&soBcFXm+q-z~%hSO#+T<kW~`@cT&3bSQ$^cXyeFmhB44uJaaFK3D0t;};^L*k@w
z?5sF-)xqacmi&y?+9jTj{nJ(BkOg>LX?%{_+-set$%>(I`AC+?{E4q?QpFRmfsA@D
zleQnPlUB?K&lL$sdL*hcU9owc?osvlwWT}-$)9~pcSjtP8{6i3H~C;reFTF|rq9<d
z*ze&Q!80GO{~&no8Rda#x`0m=@^)G~5@hp$N=2T6>F9mgBPt&i^|c!srYBhO<`vMS
zKGpeinG*Ns`JS%Dij7>Ycdi&*1_x>h@mKY`_z+xzKD<SF2!<<sIp%@zinDFZg(e<`
zIv9RAXjS=9V}uvnfB1G+%u4l~SY0LN<*+gF!$9Mcqd$K*`EG%ZR?Y3^xVMG3plb)8
zI*8;3-ROczVi<%q^6UOLaAn`{19Wg=G^r|ecsOGLu<Bur9`gD4MK@A}fS^meVP$y$
zKr;;u_Iw5MF-y{(<k2xCM+=$BN-ZJr2P%vXkA-8wuYSHVRK<a$>{SU^27T&0u+C2G
z=WK<*R>ngEiu|f}lOpqka+=jr=u`G5ym6!B(2W;Xh3|Qb4=bny`0zkMPTs8GEx2!6
zsWFz|BHgg^GHPD>sQ!_4R&#jI_g?w0j5c-^n2pzW8QhN_xZa6E-H_3^p^F&~`-DVB
zs>%pBvX95036{3dj5OO1bW+S$VbHFE9@n}IUF*pehr?D|3A4AxpBYl!>aB_cPci;;
zAhBd3hK$lBs_%#(3i^B~(>t~66~D7W3wS2<sl5Vtw%K5HLF+8|s--#eoo~Fm!0P~B
zV-<O&u8qz?pguiR#m0FfAH%Be`8hXL;KTC*8UwKzrRqpZ5$MB2aF~UPY#{4N1b}*&
zEi`e3Bljfm2;lzu9BQp0PJ3o2eH_c-jl><vGahz8OjimnHAm$~e&FD|iwKfT2CM1A
z!jtG^P+J1@??M|Vi&mgj#9&>j7*A}xOsnu%a$O%FuxCV!m4n3B;b=Zi!VpSM=PfwW
z2Mwn4z9|-K9ARjr66D&9vQ`)Qg|%dpd7p}}_SWg0sS+Dp@IM<2(RHA!f#$tPYk#v^
zDa6DatqU-(Pl}g0P{g&qB2ciS51f*x&A4_;>1rHbQ?J>?D5abSN_AFX4^7Af5Ur37
z&{3{iS1(TicAS(+QnZLE7TQ#6Lhn3oov~(vpn&)3rDsQGDev7+B)2NS7gD#L&~^|@
zGp*!1Hxu=3(y$U!1I-Pp`3yMset<f0Pqps@cIY%Wa?X#<{aDJLkm-d+wg;8!Fzk_M
z3KD5IDY$OCLRuj`g}Rw+Wf-gUyLy{69c4?lRgxggkX*Cjq?m3-%rM~m5eR2Hbf!bI
z15%qPIqINf`(0S;<_Cye*m)}~gNBzI4Fboz^73Y<KMvkLwGWYxLX&SsWs-Fe$Vih2
zN08A+{530u_(K5~kiSW-30-$RSTzY#ClAR@%S3bv6uf)Amc;F&_epVu15&yqOuR1#
zC-#8xIIx;?)C>*h><8ntZ&m2HWnI_VnR};JGeSk?%_|93><=7O0KUHUl>?GlId^JT
zUk)xVLY`@Eorr2UV&IYkPu=2K2i&XkAW<r6_KTXv_eB;R6!M@k36XK_jy=yWDLTer
z5E5j?;H^pfXGMr-DVe``NaTNV&itya&lnh~?aK7zbTJ~8SN4&n!w5>k;20(kcxaFd
z7W4Y-EdIwz4fU{GlI8+tvZ0~50`HDM24SP_{BDNgWypUimr#n>XUiT@3I%Uc{$N4V
zbugTu8cd>JLwik4N}#Yef4c?5KS4?9Z7G3CuFYuxwrqM|i7C<r$mTvJ)W8sDj`<ow
zom{<~eDYcRDb)HLY}q&hzLm-dP{q<mU(CJ>Epwv+aO6}y>Y}bP1zA5d-0;o{_*F&*
zU&Un(zprMN;P<8hmZKTI0%DbAN<&-eZrmV}=8q}v?i0wK6_ec60Fs@?4-8)jV2as1
zU%^4!rSR|hC-<)?0e3<dbOQ8_`Q+IH$z{wf@vP32&`u6WSS1b%gD)7jYJ!<}U2`^S
z^y?|61XnrX=3oGh3NsN)$eZU7p?Iopq{HAskst`v$!Hk@QdT1YfA0a-mDFK=CyW>q
zCpWWpu${?C+CtTD)78ucD?@cF@)hYu>qZP9-4{1DV&0f-Wg6Ydme$^*q;}%!$@SYy
zIWNiVeI&Bv{V@oHwZ7(=$EN7i6=xoC9$>PIH(hxr8y1uK)xN4ZMbB66l;`uu^D;N|
zmnbJzY=MGBX`mPS<Ms$Yn6-ysOPERfbrrt6#@M04{dhU4Vs&?PuKboK-uzEbWi0OD
zlmCMV(V3l=*5=1nEqw*;Lf<au0T07j>L8pWyf7y{HnX)6S~l)Bp#HGf#1a~t7&&(C
zLW9#8?BT2MA7W4Myaj~09<?4Ch710G4Yn)gJNo*=o{iWo5T2MANRZ9zu48KMj0c??
zrf2nq3dmRG)J8fid!OY({{2C0q(E|fWnFjd5`gD#79nMuG>o-0EQsHl2JIY8OO<t5
zvDOd{T^RiM5#+0r|IyqS7+mwQt8`bL@wL03=JVcH4`5QkeUgzIQCn!WFHMiuW_Gy0
zHBB^v-Or#l{IQ-YVZ!rLw)T^J5O=G0=K+_(GPL%M)Gc3kPQ$Xn&Eo)m?!EbRo%Zf9
z|Bhcic|_HNPp^?RI*?t(g5XT(yf;n|9*s-xI(Sy9S@Hb=OkK+TJ7-T`zMyS%u<6l#
zuHJwW;P1FS8woiC;T#!_7VDY8n+>#rmPXC*qjK*SeyIcFj^9=?I*N*8C+&rT07Y(D
ze<#yoOt&Eib$B&i024ekFP(11@)D#aQy#H>p88$~(cewgmrjdP>gKsM`D%u3XeM`4
z=GXfdbq>C^T*qoFIw7JL%pLI5d3AQ+QhN9Cf|7naeawPX!--pZ5C2U(x=%Ve=(zZc
ztR?+>*$m2uFR=X|PVM!`SGc#=$}M2SF)j#rKEP`2rJ!!tP)fiE*>y(Lv$>msOB{~Q
z0hBO?=0Vhpw=v~L(5{zNgLGi%yF?%fhA86C_9kTSYs`Z@ih?k(<8*5%PJn>z;rsUL
zR$}c^u_7H4m@8r-HP$yIMWXJoS52WP>kQ7BBFy|A%v3)CBu0mUt%7?FVE7mPG4R1!
zC;KN6t3g&wV^pBu>ncjyyJ?fW>lQq%?9ar9`(jPOz!XZj5^8iGtY>wBG)SIKCmXu`
zI~>;hKg~<n(|r{*=$sf5Tom-|cOc{u4!gV?Kj+>o<*u=;BvCoUGG$CEBB6hl+FajS
z75lC<-CpCs+cf^(v%=mql|81Y*WG-Amtgf1{@h-3+sc>e{pr2#BwC#I^6Rr-AgC+C
ze~&WXgD~&x;;ZBIfk#%LsU?a5&De%KcX5Y&@QxggO8$`;(b{#&A?rQ+^&vRU1`b<r
z@@Eq(*k6KIOJU_#J<@{jsnIu*Gl6Ayd<ye1WS*V`iltqO*J8qW!b_0Q0bQPlHNgGv
z#5cN`jQG3O?QY+GBn8Yx*K&cB@YV!$u+pJffk3eEL!DJtNPE&pJG6W;!W>+PR}_jb
zXhS71AWQbr&U4o0V=|I7l0#{iPx#z7`g9n)Hv$8zuP6|G3%RMappK<wm$o@HejLtx
zyM<K9nTNNbMYIQN(xf9^L}qsWUG*i0_-F_lv{P>nuab~!XHQQ<vke=tV(;M|aA@dK
zPF77V!V`ZJw&?$gTubTCwQ;%zr~*o+@cuW<3QGI!QBB>LXryCJFnc{A(e4CrnNSPb
z>N<q)T5As@B_6IHcIO{Hir5%quyvghDXDINM*&+d=Pdj{S#VHLR*gF$zrY#y8Gq(?
zC9)e?28i%b1r57pSIy@AnbHR1?gi8bohhXd2{2{S3ARTNkQ0!xvr>Miv#i7J&LkEz
zvw;e6KQx=1OgwqriTw(co>QOGvVRq1YXd4H=vrCVSck-qa>GWWGZe;SZNIqg|1LT|
z(vt=eijou7$7eDlx3!ibt=k95+0^%g=J#L`%8MZslt)9L_R=HR(%SrX2hB&eaGAFy
zkbgvR9QRVh^lP=CkOF+SGwT-yXXQ}*4YuaPGSv3=5&C()yR=ga5ipc|(NPC_r^7J#
zGurE@EJMjyy4=c7>XTyjoXUVme;2|BAq2rEo1N8Y3zXCIv4DTgOFDobo9uw9|2it2
zSvftxNkbIIm}M~vLTT1xa2mN#UlnAGLB@|j)#C(F(+^&x81UY!7cW#PYv+3!&0F!0
zL%5=}Q$bt3r=!z)gbg)wyWN3Sp>OKist{Nl65$Rx?G6NY<bGp&Ue!acfkotRIZW3(
zJ3z52)SW$DnG_c<`be4BBVRaxg$5=miq={;Scy915p;0pJFf{7XKEt()5@MsTB9;1
za1ASX&)rO<yaG5bw=Pf8?eZHxKKmZ-IK4;yKT6L4>A}XVBzu%Ax&AhofWp`nl;wi|
zC+WRxhxl+OOamW`zkz00abKtg)baZyb#v$kKqIeG|44mQuPMwiiJeB{TFs$27MRfA
zVZlpK9ec_w213qzEQWnB(YyTp2rI@MAHnx$CVtg|<^&e?^N>MJEte=aB7LH4!22}8
zP8wr(lWyhxOVQah^U&{T%tq^<apwzVbli0KcsGM_&OOo!<*_uKx+C9C1?1Do+YLxH
z?H!4%EyS0&IU5g^w7tt@*>(7FHyZ8*y3(jHj?YVb;;94YMNIDlwaDkS;IDaP(LSd%
zg7Gh_ruTy$#MT1NAv&8R=HqN`y~KFVyjrv^G{6+5#+$mmImbHhr5pxtAACFJXc=&e
zv)jJLWJA59<~=GYEB9kk3~eM<nT*Z`Uog%&1ubfnc_dzp`fs3_Ynrb>#MY3=T9w!7
zk}Xtkjbek!7taZJKAx9WTU4YR-yqFbiI466j0{m`#B8fRnENXH_I)b#*6UyZuA8&K
zOcZZ|+qb7y!*03Vv$weKm9aJ+n-|Tyg9p<syO_RvZkm^u!#yHuE1-I7w5qdz!I!3O
zG3>t`az;f*rr0Am3*agEf+hUa(TvaJPlq%ql)=vFS19TqMGV{JyD7O#XNUTw6L65t
zE5fI`;e}?<A|^K|g!Vv_>%7Qoe%=XGD+9$M%*)bab5izAcPw-+PC*vxaT4As-Ztxa
zBFK-OoRSe3<5q;ZGN93+*!?8QwYL1;{k@VvhZ?hDM4@gw`4Bj^J~qe7RCNML2t;Bk
zK$pp7GdBvR=Do~o^$e4@ygTUJR;tA<ipi9Cl$3%<<5%%khwY&FSPO<tyJ2;?<k>+c
zh&4}?^Aoh54GN|<yx?5|MWavn?KjO23G~(_N(b)=Bh2Xpki$(db4FQ#{`o|qO&<)G
z%;pf0$vWjYue8xF)JS=D<nXP!5!Gdi?5NX&Ig!gLc~HP0$;rK+237J)nPx-#z^ZoQ
zrz2vaFTm#pPp~lwy$^ezVgIa>JX3%`VivM-i_Q3~x2T{rn&)$(lDDwb^g?>aIbU`A
z3}iOtC(TfzYX)g#i1Lz_UIpw9zzuLbE$<BomE{3Vt(q*z$H{;@atIOcK>^N{q(~Bc
zncEFDZfFc4Wa${JpQRf_`8i;gR2JDV#jl|B2yh(r<>3T4o_D%Bx-^U>{2Vd0x26HC
zz!;O``eFeEqCA72!O|+@8}LWuy_{3r&xhtc+&TA7PXi;_QEA8B29?cI163USP3_Q{
z-GgOueD1lcDqzPxD9dh(uH4AZLSLR|G2K3@e@C@pI0+u^wtcQGL=BQHbF+xE8ImRZ
zp|=vE+EDds38~khB*&8xYA9sqR|9*eIx<}MMa*M})fIf}oVFBOx+KW8$mf7_fn#yb
zu7SzM7iuB8h4@6oDRHFaf|dz#O^RLvN=>n&$0GfEluliVNXl>Lj}UI*YlrqGzUURm
zUG9xbe?HMTVAOkpx!4m?IW)vWOXyjlfr3kbGx#EZ5YmqX>>4>T-Yz5;KMuc1T4)?#
z#Y<b-&n{MVb<mCz;!~qfrd}|S!)SyYMuhXU)-t7!*u9-UtL=cP!*uz+EYoxRwCNSd
z{G`)p0Ev@4ZL~uCxBz6q^sFE=?|4L>8Zz^GR!oC#dW%p6D@n0seLY!uOWFuKdc6|!
zW)MmHVV}Zf!;F5?=rx$otuhkPn@h+e1H=`rDIP6lwlK*r<^yir&}c6I(@`iFymWPu
zX;>i{6hG1_U?}3tHK`TVmSWPRdJkLLpr{F;5=u=eYeZxfjgd06(c>^hWI*X;-2^SQ
z8CaXtNjTKX5e3svNid4nr_EvUO8#tbFbjMPCz4vIKK)BBKK~Hr(lz4$&XhShwuSn4
zXG5Xp5K#>blyfC0!;HubzPNKIugw1W4x>WCT&P+t`!i)}wRJ;-UWz*~gdB>dR%Jvp
zg`;EGkGLhT!t@e(YE{ozg4kOfJ{;<XDIrAbK(VW#+XFsm(lE3awx*0j%kE9@FJ(Ow
za71c#&$U^!wY!<w!f`ccG3?KHHUZiH)bf7DLW2*3;1p7sUbXJH7IVEO(KXo<!h0xV
zCNEjkoF)4tD6E<tw*wDzUbEXAMrL!S?Y(e1r(~xAt{V9ERH?(zoE9^s6c26ZH%E%H
zdZQu(6Wv>Y6d<|<bpsE?<?~|3f4wiPz0=TC6e#avQg>27G@7&&T2fX<6=jgA`m!aU
zM0l}}Bw7Z@Dt4Rk_2{K9Yk&LvI&QYsxGOUt49ZDaa~xT(6;#c=HObNqHK6v|xkq>*
z_%7t4_U@Umh?^xZ^Y=J0*;z&)a-mR_1(fO8BE5_E9ub@hE6CPpspK1=8VN*UonFJm
zJcss5F^ho#4F5HDY6Iz-d<_^Lpc1?*o^zMD>9Y?MzZ3j!wla<-SO7!$Kyv2ZpM4O7
zOyE*LFciKt!-9P?#|j_?!n0N(P4_@js0~uQDcZ3>VId#<5LiL3_apB~89M&Mzu6(D
zRZb@8OaM@eFg;RJr=$`k<0(m`hQy!zD%gxn`37w3kP>UaRj5=*l$#O6RRdc?6Ekqc
zd~lDlYoy!pLEo?!LVyX4D9V(JkRpxAB_uyir<)_XgD`dPaMTb0S+Jg5kB^$EW-4--
zgL1`b`Jj)QpWpCQO$5tKV_>j`ONpPp_C9_LB13i+Z>s*Lc=axR>EwsusA*pA+;S@P
zf>X`XE4_v7&*xVI=vJ&QBnXnry@4Ji4W?*C>(ljuqn$jAtd~<q;|LHE#ZEywH@W<i
z-*Z>1QAAj~wEr0SIis$R9q0HRvJ5tSiy<(^U9mPTm35B9A@e%P$(8-U!th>Z=j32Y
z1|+UtxxXo2WU3Cin0;^RW7U;3Jv11Lb9K~$;1r^UuSkOsCtq{#oSXyCQ2?lz%dCLK
zi8JR~U6U$Wq1w#@AhITmYaynUKq_@VdQ$j_=sR_7{`2M*n;VvY!<{&uL!hG?60x-Z
zc-t~RvNO{X{tJ77fu$~zVDZct#Bps!w5;#Wrk%A12j??$$PN;BfpOkwb;y_+$-g$b
zJA*$M(hJm~3h6y@i!kjh70s%1(QB#x4CmmwIRPZL1z;wCwk$xumSYlIp`O0p!F=$T
z7@6X5U6)wNDufQ{@;2R0xG{;D3%ovG3~WJ4=BNcNkNXVERt#b;R~v)IJp}LH3EIG0
z-qRYKwqq7*CoQxGx6Wz&oLvsUFJmU<0J5T+GrGST0~b;_N$%6286@=Y0Mc5XTF4jJ
zc|ERkL<#^*f@j1Mh!l?dwdL5P#Y62^J_8(M%KkGXh=odS04DnFV)OBC<U2BxIR*&1
z#<kI0sUpD$CzUhH``O6P1qXp|k~BPgf6_?#M<~(J)A8|31vp!CI^XeDCpzwR&Duc(
zmWu}dM4q+=K0E*3!U&T!d^4`P@c^R%);fEiV$;}k>^C(StRU=72*awfQ}TfT*=w_P
zMh$x+bJ(k__`6-X!i3jJ`S0RJo_vL?47nD={~^}`;6=+Zw`$ry!&g8KoZJo=DSMIl
z`+nFd)15E8zGA9Za2>*EJ2>Ss-Lj(*mBP#VBk4;IzURZ#gm}ee7%AUfmrqi2%8tS2
zyLs@0`Hou_sgS5I(Y*zj63&yHrr$0`k8fD$d*EWpD~{b|?cHtoJ#T$AUQYm=9zu4X
z*<7GFk+)KPMOE?bRBoNL0GrQy*$OItv(l3ZrGSSGYiz)@?#RSWPKN@?p++#)IMaAJ
zr%qCjEYz2<9W4N2yM4yCEggISWr+?#^lf7;NlY47DK8fpoL-Mj>!kdkCVfvgpNx=U
zpZ<F%9W#348+DF^1<SKJICFUHBu5Xfv;xE;S}okB$Uypyu${Hh%NFY(SF6u#<nDoO
z+REw6gHY7H7wZ>?GXM6#u%oj4D&RB7w(@`)nF5m@q|Qj%*Dq$>n<}rWwTwHEcH;zy
zv%_~dva5)_zk|ejJ;T5Xs*^=4qjw^$0O&ZSLr0%><NVU}2n+<@rQp3n>{K1J8Zz{s
zHF1E5`T8Ht{Eds{H;A?8GWHp=z6TMe>v%6k?n4=<SjoQ$^;y#{z$r~|UxsAEl1CO4
zo75rO8M}H&!p$`b{h2h(uAVCxk*Op}7B=!SLM%QKoYJEDV(?93GSLJZy$+}(H!>?j
zx}YiXITz5hUx^NYeC@;79@0@!vX8}yV0=^C44=N$t>=PJEv-0YfW9%ZLj(w)<7Uk{
z9m3h<Xh5@?j&*^z!eGv!*M$HGN)AB#rl`4Z@(y6@8jXPvoBV1W<u!(xPMFoLbJ5)N
z$~(X}H%_AY5e`?ag)%xSYVIcvQe6gOk?0PidCBxC>s-^!PXZn~)r5q)du`ddb3nS@
zsn$k*#BCzTc7Vd$U(VG!I2dtupQR((vQ;9b14g-@$0mY<eMZ1rYTmt#uc9dE?W76|
zSd--O5pC0(-F`4m*F!Gt-bANUM{#spMrragQ|Dp}2Eoe(D5Kf}w*Esq6l*8eg-%{P
zndpp#DU$Uu2j@UP+rLy-`WPVhsUE<A@!#Qm8D?c@z3aK|=qy*kwE#r81>i19>@4$<
zBWPaM)V=_6G&%?v6eW^?Jk}fBu5u81W*`Gf{4fD?cf7%eJ`aZ&5tdFfY9q4p=c)kw
zMIZ?AEHzATe@DU|!7^dbnLNO#OQafh=mH2p#rV?!fEIJ#ng`faAFc26A;#RrZwGat
zy+b<{MJ-SEAJgd1(&RnOjF>8`U5LoDP{jp-mjGg>6_P|{`k7NAH|eOX{|!t4=ssv$
z0Fcvx3@BI@%01>oQxw)dJdigc4Y~lO1+k=OGf+9{cRNw(301p;i+&3br+u=zpXwb?
z0VZ899aXSJ{JRfdf&3AR2yY?2G+ktXg&z<Y(($&f+k^AggEgHnW$IT06O0jwl57qj
zz|g243V;<NI^FI9=Vz)=S3Mx)Qb|($Y(<<g4oHx#K)*kW0M=4w>j2<sQR}o{Hn`QL
zs1<M+pbRIb&XB+$l{aVxL#YL6)f1A-MUZZ8FIRYfO$t&4#7AGmHUaomWaYav!rT`d
z>(#%bZ?y>jqWA!%7VqVF{DtC+8UEo5jDEh=NOzwLvd&j^z+_x@mjuQ53IFrWzQYX9
z!3{{xdu=w+_i#f9mzJGK>WfA8;Hu+V(+*82)T9F6w+wiXxAxm0b}7>scr8EO09{gc
zA>e5?#?(~IYKl=j1nMbms_Zy+=xQhJl1-B0oBjVHkCt)(?^hE#G4t;o^mM&g*Mmvc
zk2PRmWa9By+5j`wTfiU}DB?TB6GR0>iX9vq?)eIgsw&7(upBKwjAg*_AYqkO1iTb)
zAy-fw2Y&w#Vve)td1c<Ey1#u&E&|z4$%e$ap`J_tPXSi(3qS-`o_>FJkqOP~i-PtD
z?Nn`}N7WDe<ms7u(Lf?_drKkE))|e1IUoq1eR81y4bwuTWwL6{RN4v7*H5KQ4uwv_
zpwW^mLK+n2rpo^B2$;<GTBl!fKIo6gE-fcDniw)M)X93X45DVPr|5du$EmgrOC(RP
zB8Cb?atH~?v|-&x?8loPe%lNO{wr4kaA3rb3g}6tEVIBUh4?^k+Pe~vf6|1ayu=B_
zRp}4RlP0iL2N#!4h)tZ$k!LW-6kUX2ttXy{XP#>>*UR5YXzTeBAo$IIaE}QlfSAUl
zSx<5~YW_Kn1D?>l6X&4qThj}JKJLf>83N?&bD)qwvQ3M07Pd-_o)%R8IB}aF&jv^N
zJDgeel7i&af+A&AN~p*!RJ!QhSKWP@Qv7IZ#M2H4PFUK~?(#g*#drgNO>!&0(wSwZ
zz0*k#Nskm77_!0WAQ8HA$@zC~`hC9I0<;KMBi|JKvE8Td@Qwij*8e3goE8e_sC76t
zcqUG`2!L&N$QRok5R5BD+XRlE0d*RXn+a=$gm_P|2D@j7RBD(;=O(=p__m~!0L@+U
z2a!F!Xn?waAOV>=J=T>cqQ4503a*>e0H`9N-}&CmXiMGu==mb>wP{s8TXvzOV5RdO
zVQnvbU!#LTN$;xf0dbg-T%vdU7%#y@%-SWWf#?fBFUg!L=sNj;{PxQbRV0|3pJNvC
zKbsV{3Mm4f!Z~Abx^g#C>w)ecVCL%0k#Dz4Aa|}Z@DMoEnIpdW-=K^Gm>hM{HIb}`
zGp%_3Bl&uOwYNjKqbVp#6~FSE00^8MPpwdoe5WEKZQY0VU64GwI^lS(dlpJGrUL#V
zG817?2v#)5G!MXFxrx#M0BokU>~>19`&>!H6#S6GVQt0dAqmwXfiFeZPC?>X+l6ba
zvutKVPoOy@aRX4y;+c72bO|Qoe0)EG<@B8zz||?X5nUpWUj?atRpAQ2(D2D2sV{rW
zsA!1k6r}!Dtz+2p-#{{*WYtWC3IAE1+lBR!?K6H~WH3R-5;4T!DWt9kj$Y1>Bq+K1
zK`u_Eh2OrreKl`x99!P~{ceo!$F5@~*Pl^VJK&6sz5C=9PjSip>iew}#YO2JVaQdJ
zW9=r!mL+z3eEKpAqvH`#@qBLGYn_aj!aYgxE+73Wt8_lZ@hx8R+iD-<$|E)>lHd)}
z=Clc9buoSc+@2{-*K58kn>W*UXU(CVstgzsl?*Gv=T}p>6BAi;zcBQEZ%?uAYC{t^
zW+_MYR9nFnoygP^nSNbW-(PWL6wue^O-1&whYWAn(r8NtVQ^ej7JDks>ci0prvtew
z{TxlIzH7Dj;#MSZ&#nq|JMN^^oH%L+7(nx}Wq{ZQfsls8Wdp4*Bj}Il3U%tV1?$h8
zCH1Plmj+YL?auZzP;X!VNS5nqP+M)EJr9^aHD3{%RVZHwdpTSkMJU~<XbyH=bpQ}v
z_*8OsmD5rx^_g*=BHaU@U51`IV9L$S8ycxS&?r7e98Voh7QdZ_S!1b>d}&Nd>CFvZ
zozRjlTqz7_-ambPDSFb};7RWS0F~?Iy2?%WJHO>_vtDga6_EJSs%FMs8J03%1E_Ch
z-aQ9YY3oQi)cir2Ce{p{w~ikO(!aGd)-p6dUu!t-9a=@eNq(NC4Uh#_=2>@>VTA?-
z1sQ%DfZ&VoanNfRAfI_t#J9N0>TJ+m`|r-Q$z1wV;g~6=mSdcZm{@qkbfS(JJAV5k
z*dG&<&&#wu+3z3!_3Pr`2bd`YCb}LMn%@SVTI9)Fywt%EDdVtIC!{A<6I#^DxzbC!
z-Yjo}0Opy(pSa@Fd!Cdtj039k2pI4SfDg(>!voAUOC0x1`bOxr?Ir6=O;oiWdTZVU
zr0zqPIM|jn7q2_0kW-U+!~_0maEYhW^3Y>~<azKNGOc?~Lbls(3cQg${T!{}Se)rd
z0pnmR+nAXgx)xyd?X~%0rj#o|vEFL$Bt$wdU^;vDm6>%I-Nm@ZJO?QPg6TW)d83IH
zey1XvXr3+cKq;Paj-@jUrtQdRWHHGE3hIOZQWtb;Og)P!9$;XZ{JofS7O66J!&l8U
zsVtgrj_t28=|pw0xVX`Ul2x|UW(Guluf&{GLsdOUq2uKMjfHFt3SLodR;)f5+BKnP
ze96svV9s&uHAL;5x<a+uLzZ_`clfeW*s?WaV4K9ZXT{p@o*RtX_8#myIwJrmsFx65
zllmk>uA$tH)Y<16St7zQJ+z8`oD=Bqezv&D-o|2SH7EIv&0!p|_W@+<Q@(k(@47r%
zo}9iimckj^(pXg`!!vC<mosS^w5&1(6X;@(^L~b6XAl)AGV}4t9R2b3j4y)$6?pZX
zX(=VIV=VBu6$?iUr+xTtj}0!Zf4_TZ8?pH%8#f+c19fg)GS0erlNXcia$6S>yt`ZM
zb&#i{jDxbPN~<LG%6Au$-H=|BE9}XeONoETt$n(1`YcoJr`E3*)R(T*L=jK;#-_G1
zBdRR#X$bB2>=Bx|%zl?Lfj*qT$W&*jrz`<7Ln=yP&H%N4WOh?+d<J%w8v2LbRFvlE
zR4*Q^&EI_!D*G*cA*<kiKY~zS2(|0ct5OMjsbx)n-Q&QEW_x~33R=}4iwfyA3o(5K
z*0<WYUCLSH0hFPId>aRY-#H9%TyTK^IJV}nL1Y*7AAE(jYXeJFDi5<A){<nVP5n|k
zX&*p|g3Z?+`1rY5@0wa7>Bl=qYiAbstiid}CfQHTt^<_8nQZmMwk{Atqk7qMnlCe3
zIsC|KKP`NPgLAc_AGQ8`g;^b6`nxoJRX>7<?{%T=8ZI1RJxerxhV&jF3?HbkP5Ba5
zY02#Wvg1qO^UxK6(FC#U877*k&8c%)0VNumL-83d3o+LQvh;WM*par852S<+XahQ#
z0KAL}wS9)*IUk1@1vV+zx60tTRT5`;YEr{0Dh7l*J=?Wf`>EcWZxI$?Ne*yH%{4bD
zo>mD%i$}9(bPr;F1fJTf?6KAfeBXj}I~tnotu2-oHw4h0RC?#NhMapv%PDiy(zCgx
zxe=$+n1k37b<mhE-tqL_M5&BbdwYBSfS{&WWwO)`i?)TGvV76JG80hRya_JbO=u<Z
zwnCFnG6Wl(p?#8fWmHQfn#P)1B7eT<rlbFnKS<u;1Hk(izB97r4@wP5XFo}TXk-@W
z1qTY=@VwSQ7Nmi`+H@_eu^WUF#pTSkeh-+ijZ9cbPZ)!oA)fK%JppAE!PO5QcC0ZF
z#5Gqu!8}{^X~1}|wZ+q?<Z#r-^{*7JreiEYGW&A?w9?x1g5fdINx=m;PLP<&_qHKY
zXWk-=9WGQRl64Ho8f1R*?U<W@a(^tVrJ<7XL8e}Y67uGr*II!(z3++bXToS7f-2`)
z8CJ+ESDgn@;koF~%jqjG)J#PKt;xFq*RE~}y%f=3w1;N1x>gM}Cs)Ca237!T@)R-I
zt-J4Iwbd4&d6PC5ENGg_CzOZjxnI{h1|`FHtViQ6P0aD)+<p91?`i77=-WI)8Vyk|
z`BaMm$Iun^#d3u+yY;Jw8pi<XHV<Jm5s=ks&L+HVg2HstzHn4Wuju!1+67}wy)IPH
z%uHplaoyFTsYuOgSp$(JWrrsBJBp8l$X2TWu=3IN&8@X41Iaugfz>6o)-O5xx-N(;
zmdSj31f1lP^pz!rcJA(EpXiQp-nD=-%)s#dGn!Q;tM+yq@)4s*(xa6koyJ?z0o6}>
zoEINbjvQ}aRXunuEagf+!ym)|>r|HL(mDHszuUB5$_f1;L@c>(dVDR$pwK}``;1Ps
zq{A$a6q$2Fix4PWp^-Wmm^nNg?><1t(_U>X5C^8~QNMhJNBq$J^-O-CeO<}ayQRdU
zoLQYQ-fA{!`C@ICDT^<Y!$KkL3$udTsuV4!n#LoY2wnh}HX!z&_zXAKueQOs)t6AM
zO1pMX;59%B{lW{VloDO<=v?+ZKj}hatv&9?ogh)C%i~)DApV!!*QMfOQ*J2>b%NBq
zFypj*OAE02YvX<a7Ohb_>#ag9+G?P_aa);(As4L|7toN)3Q<C=iP9H;oS<Vd9ID)J
z!t?mMbUD3k$2v*QE^@%Up44I1^7b!+@@yQ08ZW|63-2;`f9r}JHJtLPwyw2QlObp!
z_0PJKKJZRSog%PJS6SK04l@5Qpon^CF&|ghbNZbh;k+Ntf%8zqZb*0^YQq$fl6WAe
z%Z!V5zBj4PW8!T`I}5Obn|7?e!LifrKXswUMsCcGrYlRSWeFg$X7<uMxf-k*ml|mt
z2E}O4vu6&5_Kl@=89r1EbQ6u$95X%=t@@ic^6Sk7i(`g}8#G+|)BmFj-;%yl^}K4M
zGF<u0sJFMLHocTj&u2};wBj`Gsk75GHlrU@RvS`*K07J8hX>fs6+QjetZi>Dayow6
zdyYyr{V4nDfu;77#y6qe)Kpow3a|qHMJo;VMNP2FBW$Y5U-px!Rr#G{X}R;pz@-0n
zh1)WC`KbT+RUd3#j9P2$NW^&8hZ35$D>(tTtpT0S(pz8JbBK0Pa%wtgUC;`m7ndg<
zPp>A)i>yt0xKth()0g%@ZfV!bHy+cmFR4N}Mhj&!_Aiw2XeQe913~p>y;G)+{A5MX
zPWh#A^0AC91DuC@t<2Lf$Kbria#DZMr#wzO##HEAWaNA&m73e^F{3cr78loH6WXy7
zd+OaAD^}%mR0uQ3srAzQ$;3|1@8Y(d@7vOHXJb{?(wp-d43b6CV&@LMxmy!@whVgr
zUY{tVyffcxldq7s)X;Y?Z&@`F91wj@^W~0qfwL1l0iR1&7e>Vg?QJ;5zNtkK#+-oi
zen>EzIgjS0kB~_SC~~~kVzc6PRXVDyVI_ff!zG0y9y!(W&6#%PM&(I8s`*|mfAV)M
z!Xv4bSh|<=JOo%8<o=|)^S_ajlN|J1``RV(OB}ZOBOi|-{ArC}V|U)T-_!#DpYwzh
z6rEX$F1rc(<|gE&bnOd{2p${}9t?K=2m?d>S!Tm%SG6T~3O+EaxyPRK(%EdU!F6D|
zv;;=Xk5DE%lc~)fo+nL9B=|i)0SXtVSh>F0L5k;0rX?r1j_mZRuiB2^9eWJv5&7Zs
zWQ%ea-!xpcF_`1+%re%mRWBq$b-wF>42OOMlV-pJFt68&619vPjCnj-OR4LIqRX{n
zW2k!Ege)&TA@4q@!;=EJwTCB8f9Zh(9lp?I-*XatWm8GAsDVl{YJEeG=1#;F9!_va
zZ0TtmJ)uAG95^-!Fg5KFH}C1VvJVcZ(YgQivsEw`D&s2=yslMuD2?c0h?{VD0XegK
zfL9h!j+P5iRkT>N8L(7cF>v5NmR56e(#iqSCCt;U^2!EsR*k(Uy2={Ig=!4fEL!V_
z1m%^_B|n~5T=8W$L5TQR!E=*VPfS%UY@m0BWnPi0(2Uc#ZjsAI)xv>WB(23FHHmg$
z9#Wi-xbZIZpR7EmlkfRNmXi1ADD+|+XtEfeHXIMQG;3cMUnNb|W^aweNNQ*1wN8;<
zn>#D4dk5y^(;!vm0`}){krloDsi*VSOv)m{A|0QrjO|r{Gu;v<BD35gsqlkoY5gZe
zSEz_f${GV4Cmwe!X8b8`ag*QSL?tm&s!jH1^B~%Ve?9nWCw;(4J^6G1o0(9%K2?W?
zdu9CZp3r8uIU-`)wGi0M;HCcusSUo3!YKWGZr{yc+In$tHYVwym+2_-iV;&l)5$eJ
zvlAl7vuY;RW=)sO+-;)~ePwFzb`cx6<|g=feV5}VYn3*CX{#Uaw2uImmRn#0Y4?w(
zajEwM#NStMg)DF3lM$p_T~GkyJHBO!V|cJccc1L7PyOqXU_rD*t(au5e_3K5TB06?
z0sw%a?}7%|KD`A<L4Jf11E~dc^r>X--2BkVEU?lnjVY2ptrQPd>TX<ne)E@pTd6QH
zczS})3&1(-0ta2N-MEB0L!yLpr73Ui01>#%zunV}0q)798%O^Kw<1g(m=J12B$On7
zt^FT=R5PK4YSF>@>oF=j>|1|wq208KFmG)(ex&yB<_8QW+0jKJAZtqa2QegHg>7|P
z<E%aNZ%f<@OQhOq1u&C`FlYAxqm&6Dn&(gm-pfa5hBO@_+8o|jN5R<$D9>>J!<%w5
zL}z1uL7Vyy4+U>Slop*0_X0G8xljb)A1b$)F1{irWxY-enR>qq_crgIlOdL^%&0_F
z${%Fs#>_6fzT?&<5*3>tbMr62gDhPH;u@3PgFE*K5<`NW1aR5ZFtp%c#KqV9q<OzS
zNN6+6@($anB=z^(5Y>VEvVOU&bMy}j(Y^rhGBy47pCSqLyww?irU+=uc(Hl6R*!;^
zh`7HO{?i#G#xgM?+x(*ceySh8Gx!{SR_MO`mKT2>$(|I>`4fK#yG7rCG$i965%|M;
z&@pk6#%8S^x>jkvd9w#)K+Kj+Rc4=TUd5&>0BS(9{UQ5ttLj5L2+XXpVktL^p_e!W
zVlQ@k-PyPOSDj8_W0`1^T(sEyrQbItVAu8Fx+wpV&9A;X2s4*yZXkX5z#kS01U5py
z$;amJiw2*=EM%)5l09)vY|F+U-YfYcoRO{J;-8*~js++p<O_xWaNms$-n$=W17jH<
zxU}h~orK}TkUh9^zFVH-*A+MDu&_+XU2ir@{prr|=4$BZJPWnSu?;qFr>GTt^Jvv|
zonwD^rcyXOo(%~yzwY(VKW<>biy%L(a(1&K4c-GE4i=tnIK2HMk>i0Hz|nYRzD4pk
z$qc;78{D9Qk5fvVKDc?=mucXe<tOBo&j0gKHa<xNF}{6rbo0~vekJ*7q=bkSYBt|+
zPz*j?GP-R3czZF%IRod}iTq9g<DZ3toDgZ$N3zv8o0si%4!$|%=I!$*QI3K7n<C-z
z=BL^GmeBJ^T!!D+TO`N5`9&?E4~40&scs*pIyn)8Fws3yWctIA-+rN+cYE_voz&oc
zR1;{L|9;B;?YaN-V>>W+(t|*X?oN{a!>(*z=%0V(*S&LKtsyYn{r8RgAHMqQH~#+P
ze|g#ehdWurTE#fL8czQ%iT{V+_<JF{AAr2<<owLP$sGOPUgeg(ok4=Fb~M`g|HE(m
zEfW9B$^Jj=!r((#YtChP+%1p%_wD{~8+e=;&gbr;<0Fj!zx-S^0jLb?kQk!Pr~Uu!
zvx)vMYW)B3*uiA5*8gW;oFZirsTz@lNOt4_2T)-waCpsRsu<^>53G1zv~IGZz)6VY
sI$F??kS<U&-)KQcZVIG&L3iLk;{kE4Bad%X9A^LmPgg&ebxsLQ0D<l)hyVZp

diff --git a/0.1/search/search_index.json b/0.1/search/search_index.json
index 0dfcf54c..8e7445b6 100644
--- a/0.1/search/search_index.json
+++ b/0.1/search/search_index.json
@@ -1 +1 @@
-{"config":{"lang":["en"],"min_search_length":3,"prebuild_index":false,"separator":"[\\s\\-]+"},"docs":[{"location":"","text":"Welcome to RAPIDS documentation \u00b6 Warning The functionality outlined in these docs is implemented in the branch time_segments which we will merge to master soon as release 0.1 . The previous (first) release of RAPIDS along with the old docs will be labeled beta . If you landed on this page feel free to look around, just have in mind that we are polishing the last rough patches before we advertise 0.1 (Nov 16, 2020) Reproducible Analysis Pipeline for Data Streams (RAPIDS) allows you to process smartphone and wearable data to extract behavioral features (a.k.a. digital biomarkers/phenotypes). RAPIDS is open source, documented, modular, tested, and reproducible. At the moment we support smartphone data collected with AWARE and wearable data from Fitbit devices. Questions or feedback can be posted on #rapids in AWARE Framework's slack . Bugs should be reported on Github . Join our discussions on our algorithms and assumptions for feature processing . Ready to start? Go to Installation , then to Configuration , and then to Execution How does it work? \u00b6 RAPIDS is formed by R and Python scripts orchestrated by Snakemake . We suggest you read Snakemake\u2019s docs but in short: every link in the analysis chain is atomic and has files as input and output. Behavioral features are processed per sensor and per participant. What are the benefits of using RAPIDS? \u00b6 Consistent analysis . Every participant sensor dataset is analyzed in the exact same way and isolated from each other. Efficient analysis . Every analysis step is executed only once. Whenever your data or configuration changes only the affected files are updated. Parallel execution . Thanks to Snakemake, your analysis can be executed over multiple cores without changing your code. Extensible code . You can easily add your own behavioral features in R or Python and keep authorship and citations. Timezone aware . Your data is adjusted to the specified timezone (multiple timezones suport coming soon ). Flexible time segments . You can extract behavioral features on time windows of any length (e.g. 5 minutes, 3 hours, 2 days), on every day or particular days (e.g. weekends, Mondays, the 1 st of each month, etc.) or around events of interest (e.g. surveys or clinical relapses). Tested code . We are constantly adding tests to make sure our behavioral features are correct. Reproducible code . You can be sure your code will run in other computers as intended thanks to R and Python virtual environments. You can share your analysis code along your publications without any overhead. Private . All your data is processed locally. How is it organized? \u00b6 In broad terms the config.yaml , .env file , participants files , time segment files are the only ones that you will have to modify. All data is stored in data/ and all scripts are stored in src/ . For more information see RAPIDS\u2019 File Structure .","title":"Home"},{"location":"#welcome-to-rapids-documentation","text":"Warning The functionality outlined in these docs is implemented in the branch time_segments which we will merge to master soon as release 0.1 . The previous (first) release of RAPIDS along with the old docs will be labeled beta . If you landed on this page feel free to look around, just have in mind that we are polishing the last rough patches before we advertise 0.1 (Nov 16, 2020) Reproducible Analysis Pipeline for Data Streams (RAPIDS) allows you to process smartphone and wearable data to extract behavioral features (a.k.a. digital biomarkers/phenotypes). RAPIDS is open source, documented, modular, tested, and reproducible. At the moment we support smartphone data collected with AWARE and wearable data from Fitbit devices. Questions or feedback can be posted on #rapids in AWARE Framework's slack . Bugs should be reported on Github . Join our discussions on our algorithms and assumptions for feature processing . Ready to start? Go to Installation , then to Configuration , and then to Execution","title":"Welcome to RAPIDS documentation"},{"location":"#how-does-it-work","text":"RAPIDS is formed by R and Python scripts orchestrated by Snakemake . We suggest you read Snakemake\u2019s docs but in short: every link in the analysis chain is atomic and has files as input and output. Behavioral features are processed per sensor and per participant.","title":"How does it work?"},{"location":"#what-are-the-benefits-of-using-rapids","text":"Consistent analysis . Every participant sensor dataset is analyzed in the exact same way and isolated from each other. Efficient analysis . Every analysis step is executed only once. Whenever your data or configuration changes only the affected files are updated. Parallel execution . Thanks to Snakemake, your analysis can be executed over multiple cores without changing your code. Extensible code . You can easily add your own behavioral features in R or Python and keep authorship and citations. Timezone aware . Your data is adjusted to the specified timezone (multiple timezones suport coming soon ). Flexible time segments . You can extract behavioral features on time windows of any length (e.g. 5 minutes, 3 hours, 2 days), on every day or particular days (e.g. weekends, Mondays, the 1 st of each month, etc.) or around events of interest (e.g. surveys or clinical relapses). Tested code . We are constantly adding tests to make sure our behavioral features are correct. Reproducible code . You can be sure your code will run in other computers as intended thanks to R and Python virtual environments. You can share your analysis code along your publications without any overhead. Private . All your data is processed locally.","title":"What are the benefits of using RAPIDS?"},{"location":"#how-is-it-organized","text":"In broad terms the config.yaml , .env file , participants files , time segment files are the only ones that you will have to modify. All data is stored in data/ and all scripts are stored in src/ . For more information see RAPIDS\u2019 File Structure .","title":"How is it organized?"},{"location":"citation/","text":"Cite RAPIDS and providers \u00b6 RAPIDS and the community RAPIDS is a community effort and as such we want to continue recognizing the contributions from other researchers. Besides citing RAPIDS, we ask you to cite any of the authors listed below if you used those sensor providers in your analysis, thank you! RAPIDS \u00b6 If you used RAPIDS, please cite this paper . RAPIDS et al. citation Vega J, Li M, Aguillera K, Goel N, Joshi E, Durica KC, Kunta AR, Low CA RAPIDS: Reproducible Analysis Pipeline for Data Streams Collected with Mobile Devices JMIR Preprints. 18/08/2020:23246 DOI: 10.2196/preprints.23246 URL: https://preprints.jmir.org/preprint/23246 Panda (accelerometer) \u00b6 If you computed accelerometer features using the provider [PHONE_ACCLEROMETER][PANDA] cite this paper in addition to RAPIDS. Panda et al. citation Panda N, Solsky I, Huang EJ, Lipsitz S, Pradarelli JC, Delisle M, Cusack JC, Gadd MA, Lubitz CC, Mullen JT, Qadan M, Smith BL, Specht M, Stephen AE, Tanabe KK, Gawande AA, Onnela JP, Haynes AB. Using Smartphones to Capture Novel Recovery Metrics After Cancer Surgery. JAMA Surg. 2020 Feb 1;155(2):123-129. doi: 10.1001/jamasurg.2019.4702. PMID: 31657854; PMCID: PMC6820047. Stachl (applications foreground) \u00b6 If you computed applications foreground features using the app category (genre) catalogue in [PHONE_APPLICATIONS_FOREGROUND][RAPIDS] cite this paper in addition to RAPIDS. Stachl et al. citation Clemens Stachl, Quay Au, Ramona Schoedel, Samuel D. Gosling, Gabriella M. Harari, Daniel Buschek, Sarah Theres V\u00f6lkel, Tobias Schuwerk, Michelle Oldemeier, Theresa Ullmann, Heinrich Hussmann, Bernd Bischl, Markus B\u00fchner. Proceedings of the National Academy of Sciences Jul 2020, 117 (30) 17680-17687; DOI: 10.1073/pnas.1920484117 Barnett (locations) \u00b6 If you computed locations features using the provider [PHONE_LOCATIONS][BARNETT] cite this paper and this paper in addition to RAPIDS. Barnett et al. citation Ian Barnett, Jukka-Pekka Onnela, Inferring mobility measures from GPS traces with missing data, Biostatistics, Volume 21, Issue 2, April 2020, Pages e98\u2013e112, https://doi.org/10.1093/biostatistics/kxy059 Canzian et al. citation Luca Canzian and Mirco Musolesi. 2015. Trajectories of depression: unobtrusive monitoring of depressive states by means of smartphone mobility traces analysis. In Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp \u201815). Association for Computing Machinery, New York, NY, USA, 1293\u20131304. DOI: https://doi.org/10.1145/2750858.2805845 Doryab (locations) \u00b6 If you computed locations features using the provider [PHONE_LOCATIONS][DORYAB] cite this paper and this paper in addition to RAPIDS. Doryab et al. citation Doryab, A., Chikarsel, P., Liu, X., & Dey, A. K. (2019). Extraction of Behavioral Features from Smartphone and Wearable Data. ArXiv:1812.10394 [Cs, Stat]. http://arxiv.org/abs/1812.10394 Canzian et al. citation Luca Canzian and Mirco Musolesi. 2015. Trajectories of depression: unobtrusive monitoring of depressive states by means of smartphone mobility traces analysis. In Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp \u201815). Association for Computing Machinery, New York, NY, USA, 1293\u20131304. DOI: https://doi.org/10.1145/2750858.2805845","title":"Citation"},{"location":"citation/#cite-rapids-and-providers","text":"RAPIDS and the community RAPIDS is a community effort and as such we want to continue recognizing the contributions from other researchers. Besides citing RAPIDS, we ask you to cite any of the authors listed below if you used those sensor providers in your analysis, thank you!","title":"Cite RAPIDS and providers"},{"location":"citation/#rapids","text":"If you used RAPIDS, please cite this paper . RAPIDS et al. citation Vega J, Li M, Aguillera K, Goel N, Joshi E, Durica KC, Kunta AR, Low CA RAPIDS: Reproducible Analysis Pipeline for Data Streams Collected with Mobile Devices JMIR Preprints. 18/08/2020:23246 DOI: 10.2196/preprints.23246 URL: https://preprints.jmir.org/preprint/23246","title":"RAPIDS"},{"location":"citation/#panda-accelerometer","text":"If you computed accelerometer features using the provider [PHONE_ACCLEROMETER][PANDA] cite this paper in addition to RAPIDS. Panda et al. citation Panda N, Solsky I, Huang EJ, Lipsitz S, Pradarelli JC, Delisle M, Cusack JC, Gadd MA, Lubitz CC, Mullen JT, Qadan M, Smith BL, Specht M, Stephen AE, Tanabe KK, Gawande AA, Onnela JP, Haynes AB. Using Smartphones to Capture Novel Recovery Metrics After Cancer Surgery. JAMA Surg. 2020 Feb 1;155(2):123-129. doi: 10.1001/jamasurg.2019.4702. PMID: 31657854; PMCID: PMC6820047.","title":"Panda (accelerometer)"},{"location":"citation/#stachl-applications-foreground","text":"If you computed applications foreground features using the app category (genre) catalogue in [PHONE_APPLICATIONS_FOREGROUND][RAPIDS] cite this paper in addition to RAPIDS. Stachl et al. citation Clemens Stachl, Quay Au, Ramona Schoedel, Samuel D. Gosling, Gabriella M. Harari, Daniel Buschek, Sarah Theres V\u00f6lkel, Tobias Schuwerk, Michelle Oldemeier, Theresa Ullmann, Heinrich Hussmann, Bernd Bischl, Markus B\u00fchner. Proceedings of the National Academy of Sciences Jul 2020, 117 (30) 17680-17687; DOI: 10.1073/pnas.1920484117","title":"Stachl (applications foreground)"},{"location":"citation/#barnett-locations","text":"If you computed locations features using the provider [PHONE_LOCATIONS][BARNETT] cite this paper and this paper in addition to RAPIDS. Barnett et al. citation Ian Barnett, Jukka-Pekka Onnela, Inferring mobility measures from GPS traces with missing data, Biostatistics, Volume 21, Issue 2, April 2020, Pages e98\u2013e112, https://doi.org/10.1093/biostatistics/kxy059 Canzian et al. citation Luca Canzian and Mirco Musolesi. 2015. Trajectories of depression: unobtrusive monitoring of depressive states by means of smartphone mobility traces analysis. In Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp \u201815). Association for Computing Machinery, New York, NY, USA, 1293\u20131304. DOI: https://doi.org/10.1145/2750858.2805845","title":"Barnett (locations)"},{"location":"citation/#doryab-locations","text":"If you computed locations features using the provider [PHONE_LOCATIONS][DORYAB] cite this paper and this paper in addition to RAPIDS. Doryab et al. citation Doryab, A., Chikarsel, P., Liu, X., & Dey, A. K. (2019). Extraction of Behavioral Features from Smartphone and Wearable Data. ArXiv:1812.10394 [Cs, Stat]. http://arxiv.org/abs/1812.10394 Canzian et al. citation Luca Canzian and Mirco Musolesi. 2015. Trajectories of depression: unobtrusive monitoring of depressive states by means of smartphone mobility traces analysis. In Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp \u201815). Association for Computing Machinery, New York, NY, USA, 1293\u20131304. DOI: https://doi.org/10.1145/2750858.2805845","title":"Doryab (locations)"},{"location":"code_of_conduct/","text":"Contributor Covenant Code of Conduct \u00b6 Our Pledge \u00b6 We as members, contributors, and leaders pledge to make participation in our community a harassment-free experience for everyone, regardless of age, body size, visible or invisible disability, ethnicity, sex characteristics, gender identity and expression, level of experience, education, socio-economic status, nationality, personal appearance, race, religion, or sexual identity and orientation. We pledge to act and interact in ways that contribute to an open, welcoming, diverse, inclusive, and healthy community. Our Standards \u00b6 Examples of behavior that contributes to a positive environment for our community include: Demonstrating empathy and kindness toward other people Being respectful of differing opinions, viewpoints, and experiences Giving and gracefully accepting constructive feedback Accepting responsibility and apologizing to those affected by our mistakes, and learning from the experience Focusing on what is best not just for us as individuals, but for the overall community Examples of unacceptable behavior include: The use of sexualized language or imagery, and sexual attention or advances of any kind Trolling, insulting or derogatory comments, and personal or political attacks Public or private harassment Publishing others\u2019 private information, such as a physical or email address, without their explicit permission Other conduct which could reasonably be considered inappropriate in a professional setting Enforcement Responsibilities \u00b6 Community leaders are responsible for clarifying and enforcing our standards of acceptable behavior and will take appropriate and fair corrective action in response to any behavior that they deem inappropriate, threatening, offensive, or harmful. Community leaders have the right and responsibility to remove, edit, or reject comments, commits, code, wiki edits, issues, and other contributions that are not aligned to this Code of Conduct, and will communicate reasons for moderation decisions when appropriate. Scope \u00b6 This Code of Conduct applies within all community spaces, and also applies when an individual is officially representing the community in public spaces. Examples of representing our community include using an official e-mail address, posting via an official social media account, or acting as an appointed representative at an online or offline event. Enforcement \u00b6 Instances of abusive, harassing, or otherwise unacceptable behavior may be reported to the community leaders responsible for enforcement at moshi@pitt.edu . All complaints will be reviewed and investigated promptly and fairly. All community leaders are obligated to respect the privacy and security of the reporter of any incident. Enforcement Guidelines \u00b6 Community leaders will follow these Community Impact Guidelines in determining the consequences for any action they deem in violation of this Code of Conduct: 1. Correction \u00b6 Community Impact : Use of inappropriate language or other behavior deemed unprofessional or unwelcome in the community. Consequence : A private, written warning from community leaders, providing clarity around the nature of the violation and an explanation of why the behavior was inappropriate. A public apology may be requested. 2. Warning \u00b6 Community Impact : A violation through a single incident or series of actions. Consequence : A warning with consequences for continued behavior. No interaction with the people involved, including unsolicited interaction with those enforcing the Code of Conduct, for a specified period of time. This includes avoiding interactions in community spaces as well as external channels like social media. Violating these terms may lead to a temporary or permanent ban. 3. Temporary Ban \u00b6 Community Impact : A serious violation of community standards, including sustained inappropriate behavior. Consequence : A temporary ban from any sort of interaction or public communication with the community for a specified period of time. No public or private interaction with the people involved, including unsolicited interaction with those enforcing the Code of Conduct, is allowed during this period. Violating these terms may lead to a permanent ban. 4. Permanent Ban \u00b6 Community Impact : Demonstrating a pattern of violation of community standards, including sustained inappropriate behavior, harassment of an individual, or aggression toward or disparagement of classes of individuals. Consequence : A permanent ban from any sort of public interaction within the community. Attribution \u00b6 This Code of Conduct is adapted from the Contributor Covenant , version 2.0, available at https://www.contributor-covenant.org/version/2/0/code_of_conduct.html . Community Impact Guidelines were inspired by Mozilla\u2019s code of conduct enforcement ladder . For answers to common questions about this code of conduct, see the FAQ at https://www.contributor-covenant.org/faq . Translations are available at https://www.contributor-covenant.org/translations .","title":"Code of Conduct"},{"location":"code_of_conduct/#contributor-covenant-code-of-conduct","text":"","title":"Contributor Covenant Code of Conduct"},{"location":"code_of_conduct/#our-pledge","text":"We as members, contributors, and leaders pledge to make participation in our community a harassment-free experience for everyone, regardless of age, body size, visible or invisible disability, ethnicity, sex characteristics, gender identity and expression, level of experience, education, socio-economic status, nationality, personal appearance, race, religion, or sexual identity and orientation. We pledge to act and interact in ways that contribute to an open, welcoming, diverse, inclusive, and healthy community.","title":"Our Pledge"},{"location":"code_of_conduct/#our-standards","text":"Examples of behavior that contributes to a positive environment for our community include: Demonstrating empathy and kindness toward other people Being respectful of differing opinions, viewpoints, and experiences Giving and gracefully accepting constructive feedback Accepting responsibility and apologizing to those affected by our mistakes, and learning from the experience Focusing on what is best not just for us as individuals, but for the overall community Examples of unacceptable behavior include: The use of sexualized language or imagery, and sexual attention or advances of any kind Trolling, insulting or derogatory comments, and personal or political attacks Public or private harassment Publishing others\u2019 private information, such as a physical or email address, without their explicit permission Other conduct which could reasonably be considered inappropriate in a professional setting","title":"Our Standards"},{"location":"code_of_conduct/#enforcement-responsibilities","text":"Community leaders are responsible for clarifying and enforcing our standards of acceptable behavior and will take appropriate and fair corrective action in response to any behavior that they deem inappropriate, threatening, offensive, or harmful. Community leaders have the right and responsibility to remove, edit, or reject comments, commits, code, wiki edits, issues, and other contributions that are not aligned to this Code of Conduct, and will communicate reasons for moderation decisions when appropriate.","title":"Enforcement Responsibilities"},{"location":"code_of_conduct/#scope","text":"This Code of Conduct applies within all community spaces, and also applies when an individual is officially representing the community in public spaces. Examples of representing our community include using an official e-mail address, posting via an official social media account, or acting as an appointed representative at an online or offline event.","title":"Scope"},{"location":"code_of_conduct/#enforcement","text":"Instances of abusive, harassing, or otherwise unacceptable behavior may be reported to the community leaders responsible for enforcement at moshi@pitt.edu . All complaints will be reviewed and investigated promptly and fairly. All community leaders are obligated to respect the privacy and security of the reporter of any incident.","title":"Enforcement"},{"location":"code_of_conduct/#enforcement-guidelines","text":"Community leaders will follow these Community Impact Guidelines in determining the consequences for any action they deem in violation of this Code of Conduct:","title":"Enforcement Guidelines"},{"location":"code_of_conduct/#1-correction","text":"Community Impact : Use of inappropriate language or other behavior deemed unprofessional or unwelcome in the community. Consequence : A private, written warning from community leaders, providing clarity around the nature of the violation and an explanation of why the behavior was inappropriate. A public apology may be requested.","title":"1. Correction"},{"location":"code_of_conduct/#2-warning","text":"Community Impact : A violation through a single incident or series of actions. Consequence : A warning with consequences for continued behavior. No interaction with the people involved, including unsolicited interaction with those enforcing the Code of Conduct, for a specified period of time. This includes avoiding interactions in community spaces as well as external channels like social media. Violating these terms may lead to a temporary or permanent ban.","title":"2. Warning"},{"location":"code_of_conduct/#3-temporary-ban","text":"Community Impact : A serious violation of community standards, including sustained inappropriate behavior. Consequence : A temporary ban from any sort of interaction or public communication with the community for a specified period of time. No public or private interaction with the people involved, including unsolicited interaction with those enforcing the Code of Conduct, is allowed during this period. Violating these terms may lead to a permanent ban.","title":"3. Temporary Ban"},{"location":"code_of_conduct/#4-permanent-ban","text":"Community Impact : Demonstrating a pattern of violation of community standards, including sustained inappropriate behavior, harassment of an individual, or aggression toward or disparagement of classes of individuals. Consequence : A permanent ban from any sort of public interaction within the community.","title":"4. Permanent Ban"},{"location":"code_of_conduct/#attribution","text":"This Code of Conduct is adapted from the Contributor Covenant , version 2.0, available at https://www.contributor-covenant.org/version/2/0/code_of_conduct.html . Community Impact Guidelines were inspired by Mozilla\u2019s code of conduct enforcement ladder . For answers to common questions about this code of conduct, see the FAQ at https://www.contributor-covenant.org/faq . Translations are available at https://www.contributor-covenant.org/translations .","title":"Attribution"},{"location":"faq/","text":"Frequently Asked Questions \u00b6 Cannot connect to your MySQL server \u00b6 Problem **Error in .local ( drv, \\. .. ) :** **Failed to connect to database: Error: Can \\' t initialize character set unknown ( path: compiled \\_ in ) ** : Calls: dbConnect -> dbConnect -> .local -> .Call Execution halted [ Tue Mar 10 19 :40:15 2020 ] Error in rule download_dataset: jobid: 531 output: data/raw/p60/locations_raw.csv RuleException: CalledProcessError in line 20 of /home/ubuntu/rapids/rules/preprocessing.snakefile: Command 'set -euo pipefail; Rscript --vanilla /home/ubuntu/rapids/.snakemake/scripts/tmp_2jnvqs7.download_dataset.R' returned non-zero exit status 1 . File \"/home/ubuntu/rapids/rules/preprocessing.snakefile\" , line 20 , in __rule_download_dataset File \"/home/ubuntu/anaconda3/envs/moshi-env/lib/python3.7/concurrent/futures/thread.py\" , line 57 , in run Shutting down, this might take some time. Exiting because a job execution failed. Look above for error message Solution Please make sure the DATABASE_GROUP in config.yaml matches your DB credentials group in .env . Cannot start mysql in linux via brew services start mysql \u00b6 Problem Cannot start mysql in linux via brew services start mysql Solution Use mysql.server start Every time I run force the download_dataset rule all rules are executed \u00b6 Problem When running snakemake -j1 -R download_phone_data or ./rapids -j1 -R download_phone_data all the rules and files are re-computed Solution This is expected behavior. The advantage of using snakemake under the hood is that every time a file containing data is modified every rule that depends on that file will be re-executed to update their results. In this case, since download_dataset updates all the raw data, and you are forcing the rule with the flag -R every single rule that depends on those raw files will be executed. Error Table XXX doesn't exist while running the download_phone_data or download_fitbit_data rule. \u00b6 Problem Error in .local ( conn, statement, ... ) : could not run statement: Table 'db_name.table_name' doesn ' t exist Calls: colnames ... .local -> dbSendQuery -> dbSendQuery -> .local -> .Call Execution halted Solution Please make sure the sensors listed in [PHONE_VALID_SENSED_BINS][PHONE_SENSORS] and the [TABLE] of each sensor you activated in config.yaml match your database tables. How do I install RAPIDS on Ubuntu 16.04 \u00b6 Solution Install dependencies (Homebrew - if not installed): sudo apt-get install libmariadb-client-lgpl-dev libxml2-dev libssl-dev Install brew for linux and add the following line to ~/.bashrc : export PATH=$HOME/.linuxbrew/bin:$PATH source ~/.bashrc Install MySQL brew install mysql brew services start mysql Install R, pandoc and rmarkdown: brew install r brew install gcc@6 (needed due to this bug ) HOMEBREW_CC=gcc-6 brew install pandoc Install miniconda using these instructions Clone our repo: git clone https://github.com/carissalow/rapids Create a python virtual environment: cd rapids conda env create -f environment.yml -n MY_ENV_NAME conda activate MY_ENV_NAME Install R packages and virtual environment: snakemake renv_install snakemake renv_init snakemake renv_restore This step could take several minutes to complete. Please be patient and let it run until completion. mysql.h cannot be found \u00b6 Problem -------------------------- [ ERROR MESSAGE ] ---------------------------- <stdin>:1:10: fatal error: mysql.h: No such file or directory compilation terminated. ----------------------------------------------------------------------- ERROR: configuration failed for package 'RMySQL' Solution sudo apt install libmariadbclient-dev No package libcurl found \u00b6 Problem libcurl cannot be found Solution Install libcurl sudo apt install libcurl4-openssl-dev Configuration failed because openssl was not found. \u00b6 Problem openssl cannot be found Solution Install openssl sudo apt install libssl-dev Configuration failed because libxml-2.0 was not found \u00b6 Problem libxml-2.0 cannot be found Solution Install libxml-2.0 sudo apt install libxml2-dev SSL connection error when running RAPIDS \u00b6 Problem You are getting the following error message when running RAPIDS: Error: Failed to connect: SSL connection error: error:1425F102:SSL routines:ssl_choose_client_version:unsupported protocol. Solution This is a bug in Ubuntu 20.04 when trying to connect to an old MySQL server with MySQL client 8.0. You should get the same error message if you try to connect from the command line. There you can add the option --ssl-mode=DISABLED but we can't do this from the R connector. If you can't update your server, the quickest solution would be to import your database to another server or to a local environment. Alternatively, you could replace mysql-client and libmysqlclient-dev with mariadb-client and libmariadbclient-dev and reinstall renv. More info about this issue here DB_TABLES key not found \u00b6 Problem If you get the following error KeyError in line 43 of preprocessing.smk: 'PHONE_SENSORS' , it means that the indentation of the key [PHONE_SENSORS] is not matching the other child elements of PHONE_VALID_SENSED_BINS Solution You need to add or remove any leading whitespaces as needed on that line. PHONE_VALID_SENSED_BINS : COMPUTE : False # This flag is automatically ignored (set to True) if you are extracting PHONE_VALID_SENSED_DAYS or screen or Barnett's location features BIN_SIZE : &bin_size 5 # (in minutes) PHONE_SENSORS : [] Error while updating your conda environment in Ubuntu \u00b6 Problem You get the following error: CondaMultiError: CondaVerificationError: The package for tk located at /home/ubuntu/miniconda2/pkgs/tk-8.6.9-hed695b0_1003 appears to be corrupted. The path 'include/mysqlStubs.h' specified in the package manifest cannot be found. ClobberError: This transaction has incompatible packages due to a shared path. packages: conda-forge/linux-64::llvm-openmp-10.0.0-hc9558a2_0, anaconda/linux-64::intel-openmp-2019.4-243 path: 'lib/libiomp5.so' Solution Reinstall conda","title":"Frequently Asked Questions"},{"location":"faq/#frequently-asked-questions","text":"","title":"Frequently Asked Questions"},{"location":"faq/#cannot-connect-to-your-mysql-server","text":"Problem **Error in .local ( drv, \\. .. ) :** **Failed to connect to database: Error: Can \\' t initialize character set unknown ( path: compiled \\_ in ) ** : Calls: dbConnect -> dbConnect -> .local -> .Call Execution halted [ Tue Mar 10 19 :40:15 2020 ] Error in rule download_dataset: jobid: 531 output: data/raw/p60/locations_raw.csv RuleException: CalledProcessError in line 20 of /home/ubuntu/rapids/rules/preprocessing.snakefile: Command 'set -euo pipefail; Rscript --vanilla /home/ubuntu/rapids/.snakemake/scripts/tmp_2jnvqs7.download_dataset.R' returned non-zero exit status 1 . File \"/home/ubuntu/rapids/rules/preprocessing.snakefile\" , line 20 , in __rule_download_dataset File \"/home/ubuntu/anaconda3/envs/moshi-env/lib/python3.7/concurrent/futures/thread.py\" , line 57 , in run Shutting down, this might take some time. Exiting because a job execution failed. Look above for error message Solution Please make sure the DATABASE_GROUP in config.yaml matches your DB credentials group in .env .","title":"Cannot connect to your MySQL server"},{"location":"faq/#cannot-start-mysql-in-linux-via-brew-services-start-mysql","text":"Problem Cannot start mysql in linux via brew services start mysql Solution Use mysql.server start","title":"Cannot start mysql in linux via brew services start mysql"},{"location":"faq/#every-time-i-run-force-the-download_dataset-rule-all-rules-are-executed","text":"Problem When running snakemake -j1 -R download_phone_data or ./rapids -j1 -R download_phone_data all the rules and files are re-computed Solution This is expected behavior. The advantage of using snakemake under the hood is that every time a file containing data is modified every rule that depends on that file will be re-executed to update their results. In this case, since download_dataset updates all the raw data, and you are forcing the rule with the flag -R every single rule that depends on those raw files will be executed.","title":"Every time I run force the download_dataset rule all rules are executed"},{"location":"faq/#error-table-xxx-doesnt-exist-while-running-the-download_phone_data-or-download_fitbit_data-rule","text":"Problem Error in .local ( conn, statement, ... ) : could not run statement: Table 'db_name.table_name' doesn ' t exist Calls: colnames ... .local -> dbSendQuery -> dbSendQuery -> .local -> .Call Execution halted Solution Please make sure the sensors listed in [PHONE_VALID_SENSED_BINS][PHONE_SENSORS] and the [TABLE] of each sensor you activated in config.yaml match your database tables.","title":"Error Table XXX doesn't exist while running the download_phone_data or download_fitbit_data rule."},{"location":"faq/#how-do-i-install-rapids-on-ubuntu-1604","text":"Solution Install dependencies (Homebrew - if not installed): sudo apt-get install libmariadb-client-lgpl-dev libxml2-dev libssl-dev Install brew for linux and add the following line to ~/.bashrc : export PATH=$HOME/.linuxbrew/bin:$PATH source ~/.bashrc Install MySQL brew install mysql brew services start mysql Install R, pandoc and rmarkdown: brew install r brew install gcc@6 (needed due to this bug ) HOMEBREW_CC=gcc-6 brew install pandoc Install miniconda using these instructions Clone our repo: git clone https://github.com/carissalow/rapids Create a python virtual environment: cd rapids conda env create -f environment.yml -n MY_ENV_NAME conda activate MY_ENV_NAME Install R packages and virtual environment: snakemake renv_install snakemake renv_init snakemake renv_restore This step could take several minutes to complete. Please be patient and let it run until completion.","title":"How do I install RAPIDS on Ubuntu 16.04"},{"location":"faq/#mysqlh-cannot-be-found","text":"Problem -------------------------- [ ERROR MESSAGE ] ---------------------------- <stdin>:1:10: fatal error: mysql.h: No such file or directory compilation terminated. ----------------------------------------------------------------------- ERROR: configuration failed for package 'RMySQL' Solution sudo apt install libmariadbclient-dev","title":"mysql.h cannot be found"},{"location":"faq/#no-package-libcurl-found","text":"Problem libcurl cannot be found Solution Install libcurl sudo apt install libcurl4-openssl-dev","title":"No package libcurl found"},{"location":"faq/#configuration-failed-because-openssl-was-not-found","text":"Problem openssl cannot be found Solution Install openssl sudo apt install libssl-dev","title":"Configuration failed because openssl was not found."},{"location":"faq/#configuration-failed-because-libxml-20-was-not-found","text":"Problem libxml-2.0 cannot be found Solution Install libxml-2.0 sudo apt install libxml2-dev","title":"Configuration failed because libxml-2.0 was not found"},{"location":"faq/#ssl-connection-error-when-running-rapids","text":"Problem You are getting the following error message when running RAPIDS: Error: Failed to connect: SSL connection error: error:1425F102:SSL routines:ssl_choose_client_version:unsupported protocol. Solution This is a bug in Ubuntu 20.04 when trying to connect to an old MySQL server with MySQL client 8.0. You should get the same error message if you try to connect from the command line. There you can add the option --ssl-mode=DISABLED but we can't do this from the R connector. If you can't update your server, the quickest solution would be to import your database to another server or to a local environment. Alternatively, you could replace mysql-client and libmysqlclient-dev with mariadb-client and libmariadbclient-dev and reinstall renv. More info about this issue here","title":"SSL connection error when running RAPIDS"},{"location":"faq/#db_tables-key-not-found","text":"Problem If you get the following error KeyError in line 43 of preprocessing.smk: 'PHONE_SENSORS' , it means that the indentation of the key [PHONE_SENSORS] is not matching the other child elements of PHONE_VALID_SENSED_BINS Solution You need to add or remove any leading whitespaces as needed on that line. PHONE_VALID_SENSED_BINS : COMPUTE : False # This flag is automatically ignored (set to True) if you are extracting PHONE_VALID_SENSED_DAYS or screen or Barnett's location features BIN_SIZE : &bin_size 5 # (in minutes) PHONE_SENSORS : []","title":"DB_TABLES key not found"},{"location":"faq/#error-while-updating-your-conda-environment-in-ubuntu","text":"Problem You get the following error: CondaMultiError: CondaVerificationError: The package for tk located at /home/ubuntu/miniconda2/pkgs/tk-8.6.9-hed695b0_1003 appears to be corrupted. The path 'include/mysqlStubs.h' specified in the package manifest cannot be found. ClobberError: This transaction has incompatible packages due to a shared path. packages: conda-forge/linux-64::llvm-openmp-10.0.0-hc9558a2_0, anaconda/linux-64::intel-openmp-2019.4-243 path: 'lib/libiomp5.so' Solution Reinstall conda","title":"Error while updating your conda environment in Ubuntu"},{"location":"file-structure/","text":"File Structure \u00b6 Tip Read this page if you want to learn more about how RAPIDS is structured. If you want to start using it go to Installation , then to Configuration , and then to Execution All paths mentioned in this page are relative to RAPIDS\u2019 root folder. If you want to extract the behavioral features that RAPIDS offers, you will only have to create or modify the .env file , participants files , time segment files , and the config.yaml file. The config.yaml file is the heart of RAPIDS and includes parameters to manage participants, data sources, sensor data, visualizations and more. All data is saved in data/ . The data/external/ folder stores any data imported or created by the user, data/raw/ stores sensor data as imported from your database, data/interim/ has intermediate files necessary to compute behavioral features from raw data, and data/processed/ has all the final files with the behavioral features in folders per participant and sensor. All the source code is saved in src/ . The src/data/ folder stores scripts to download, clean and pre-process sensor data, src/features has scripts to extract behavioral features organized in their respective subfolders , src/models/ can host any script to create models or statistical analyses with the behavioral features you extract, and src/visualization/ has scripts to create plots of the raw and processed data. There are other important files and folders but only relevant if you are interested in extending RAPIDS (e.g. virtual env files, docs, tests, Dockerfile, the Snakefile, etc.). In the figure below, we represent the interactions between users and files. After a user modifies config.yaml and .env the Snakefile file will decide what Snakemake rules have to be executed to produce the required output files (behavioral features) and what scripts are in charge of producing such files. In addition, users can add or modifiy files in the data folder (for example to configure the participants files or the time segment files ). Interaction diagram between the user, and important files in RAPIDS","title":"File Structure"},{"location":"file-structure/#file-structure","text":"Tip Read this page if you want to learn more about how RAPIDS is structured. If you want to start using it go to Installation , then to Configuration , and then to Execution All paths mentioned in this page are relative to RAPIDS\u2019 root folder. If you want to extract the behavioral features that RAPIDS offers, you will only have to create or modify the .env file , participants files , time segment files , and the config.yaml file. The config.yaml file is the heart of RAPIDS and includes parameters to manage participants, data sources, sensor data, visualizations and more. All data is saved in data/ . The data/external/ folder stores any data imported or created by the user, data/raw/ stores sensor data as imported from your database, data/interim/ has intermediate files necessary to compute behavioral features from raw data, and data/processed/ has all the final files with the behavioral features in folders per participant and sensor. All the source code is saved in src/ . The src/data/ folder stores scripts to download, clean and pre-process sensor data, src/features has scripts to extract behavioral features organized in their respective subfolders , src/models/ can host any script to create models or statistical analyses with the behavioral features you extract, and src/visualization/ has scripts to create plots of the raw and processed data. There are other important files and folders but only relevant if you are interested in extending RAPIDS (e.g. virtual env files, docs, tests, Dockerfile, the Snakefile, etc.). In the figure below, we represent the interactions between users and files. After a user modifies config.yaml and .env the Snakefile file will decide what Snakemake rules have to be executed to produce the required output files (behavioral features) and what scripts are in charge of producing such files. In addition, users can add or modifiy files in the data folder (for example to configure the participants files or the time segment files ). Interaction diagram between the user, and important files in RAPIDS","title":"File Structure"},{"location":"team/","text":"RAPIDS Team \u00b6 If you are interested in contributing feel free to submit a pull request or contact us. Core Team \u00b6 Julio Vega (Designer and Lead Developer) \u00b6 About Julio Vega is a postdoctoral associate at the Mobile Sensing + Health Institute. He is interested in personalized methodologies to monitor chronic conditions that affect daily human behavior using mobile and wearable data. vegaju at upmc . edu Personal Website Meng Li \u00b6 About Meng Li received her Master of Science degree in Information Science from the University of Pittsburgh. She is interested in applying machine learning algorithms to the medical field. lim11 at upmc . edu Linkedin Profile Github Profile Abhineeth Reddy Kunta \u00b6 About Abhineeth Reddy Kunta is a Senior Software Engineer with the Mobile Sensing + Health Institute. He is experienced in software development and specializes in building solutions using machine learning. Abhineeth likes exploring ways to leverage technology in advancing medicine and education. Previously he worked as a Computer Programmer at Georgia Department of Public Health. He has a master\u2019s degree in Computer Science from George Mason University. Kwesi Aguillera \u00b6 About Kwesi Aguillera is currently in his first year at the University of Pittsburgh pursuing a Master of Sciences in Information Science specializing in Big Data Analytics. He received his Bachelor of Science degree in Computer Science and Management from the University of the West Indies. Kwesi considers himself a full stack developer and looks forward to applying this knowledge to big data analysis. Linkedin Profile Echhit Joshi \u00b6 About Echhit Joshi is a Masters student at the School of Computing and Information at University of Pittsburgh. His areas of interest are Machine/Deep Learning, Data Mining, and Analytics. Linkedin Profile Nicolas Leo \u00b6 About Nicolas is a rising senior studying computer science at the University of Pittsburgh. His academic interests include databases, machine learning, and application development. After completing his undergraduate degree, he plans to attend graduate school for a MS in Computer Science with a focus on Intelligent Systems. Nikunj Goel \u00b6 About Nik is a graduate student at the University of Pittsburgh pursuing Master of Science in Information Science. He earned his Bachelor of Technology degree in Information Technology from India. He is a Data Enthusiasts and passionate about finding the meaning out of raw data. In a long term, his goal is to create a breakthrough in Data Science and Deep Learning. Linkedin Profile Community Contributors \u00b6 Agam Kumar \u00b6 About Agam is a junior at Carnegie Mellon University studying Statistics and Machine Learning and pursuing an additional major in Computer Science. He is a member of the Data Science team in the Health and Human Performance Lab at CMU and has keen interests in software development and data science. His research interests include ML applications in medicine. Linkedin Profile Github Profile Yasaman S. Sefidgar \u00b6 About Linkedin Profile Advisors \u00b6 Afsaneh Doryab \u00b6 About Personal Website Carissa Low \u00b6 About Profile","title":"Team"},{"location":"team/#rapids-team","text":"If you are interested in contributing feel free to submit a pull request or contact us.","title":"RAPIDS Team"},{"location":"team/#core-team","text":"","title":"Core Team"},{"location":"team/#julio-vega-designer-and-lead-developer","text":"About Julio Vega is a postdoctoral associate at the Mobile Sensing + Health Institute. He is interested in personalized methodologies to monitor chronic conditions that affect daily human behavior using mobile and wearable data. vegaju at upmc . edu Personal Website","title":"Julio Vega (Designer and Lead Developer)"},{"location":"team/#meng-li","text":"About Meng Li received her Master of Science degree in Information Science from the University of Pittsburgh. She is interested in applying machine learning algorithms to the medical field. lim11 at upmc . edu Linkedin Profile Github Profile","title":"Meng Li"},{"location":"team/#abhineeth-reddy-kunta","text":"About Abhineeth Reddy Kunta is a Senior Software Engineer with the Mobile Sensing + Health Institute. He is experienced in software development and specializes in building solutions using machine learning. Abhineeth likes exploring ways to leverage technology in advancing medicine and education. Previously he worked as a Computer Programmer at Georgia Department of Public Health. He has a master\u2019s degree in Computer Science from George Mason University.","title":"Abhineeth Reddy Kunta"},{"location":"team/#kwesi-aguillera","text":"About Kwesi Aguillera is currently in his first year at the University of Pittsburgh pursuing a Master of Sciences in Information Science specializing in Big Data Analytics. He received his Bachelor of Science degree in Computer Science and Management from the University of the West Indies. Kwesi considers himself a full stack developer and looks forward to applying this knowledge to big data analysis. Linkedin Profile","title":"Kwesi Aguillera"},{"location":"team/#echhit-joshi","text":"About Echhit Joshi is a Masters student at the School of Computing and Information at University of Pittsburgh. His areas of interest are Machine/Deep Learning, Data Mining, and Analytics. Linkedin Profile","title":"Echhit Joshi"},{"location":"team/#nicolas-leo","text":"About Nicolas is a rising senior studying computer science at the University of Pittsburgh. His academic interests include databases, machine learning, and application development. After completing his undergraduate degree, he plans to attend graduate school for a MS in Computer Science with a focus on Intelligent Systems.","title":"Nicolas Leo"},{"location":"team/#nikunj-goel","text":"About Nik is a graduate student at the University of Pittsburgh pursuing Master of Science in Information Science. He earned his Bachelor of Technology degree in Information Technology from India. He is a Data Enthusiasts and passionate about finding the meaning out of raw data. In a long term, his goal is to create a breakthrough in Data Science and Deep Learning. Linkedin Profile","title":"Nikunj Goel"},{"location":"team/#community-contributors","text":"","title":"Community Contributors"},{"location":"team/#agam-kumar","text":"About Agam is a junior at Carnegie Mellon University studying Statistics and Machine Learning and pursuing an additional major in Computer Science. He is a member of the Data Science team in the Health and Human Performance Lab at CMU and has keen interests in software development and data science. His research interests include ML applications in medicine. Linkedin Profile Github Profile","title":"Agam Kumar"},{"location":"team/#yasaman-s-sefidgar","text":"About Linkedin Profile","title":"Yasaman S. Sefidgar"},{"location":"team/#advisors","text":"","title":"Advisors"},{"location":"team/#afsaneh-doryab","text":"About Personal Website","title":"Afsaneh Doryab"},{"location":"team/#carissa-low","text":"About Profile","title":"Carissa Low"},{"location":"developers/documentation/","text":"Documentation \u00b6 We use mkdocs with the material theme to write these docs. Whenever you make any changes, just push them back to the repo and the documentation will be deployed automatically. Set up development environment \u00b6 Make sure your conda environment is active pip install mkdocs pip install mkdocs-material Preview \u00b6 Run the following command in RAPIDS root folder and go to http://127.0.0.1:8000 : mkdocs serve File Structure \u00b6 The documentation config file is /mkdocs.yml , if you are adding new .md files to the docs modify the nav attribute at the bottom of that file. You can use the hierarchy there to find all the files that appear in the documentation. Reference \u00b6 Check this page to get familiar with the different visual elements we can use in the docs (admonitions, code blocks, tables, etc.) You can also refer to /docs/setup/installation.md and /docs/setup/configuration.md to see practical examples of these elements. Hint Any links to internal pages should be relative to the current page. For example, any link from this page (documentation) which is inside ./developers should begin with ../ to go one folder level up like: [ mylink ]( ../setup/installation.md ) Extras \u00b6 You can insert emojis using this syntax :[SOURCE]-[ICON_NAME] from the following sources: https://materialdesignicons.com/ https://fontawesome.com/icons/tasks?style=solid https://primer.style/octicons/ You can use this page to create markdown tables more easily","title":"Documentation"},{"location":"developers/documentation/#documentation","text":"We use mkdocs with the material theme to write these docs. Whenever you make any changes, just push them back to the repo and the documentation will be deployed automatically.","title":"Documentation"},{"location":"developers/documentation/#set-up-development-environment","text":"Make sure your conda environment is active pip install mkdocs pip install mkdocs-material","title":"Set up development environment"},{"location":"developers/documentation/#preview","text":"Run the following command in RAPIDS root folder and go to http://127.0.0.1:8000 : mkdocs serve","title":"Preview"},{"location":"developers/documentation/#file-structure","text":"The documentation config file is /mkdocs.yml , if you are adding new .md files to the docs modify the nav attribute at the bottom of that file. You can use the hierarchy there to find all the files that appear in the documentation.","title":"File Structure"},{"location":"developers/documentation/#reference","text":"Check this page to get familiar with the different visual elements we can use in the docs (admonitions, code blocks, tables, etc.) You can also refer to /docs/setup/installation.md and /docs/setup/configuration.md to see practical examples of these elements. Hint Any links to internal pages should be relative to the current page. For example, any link from this page (documentation) which is inside ./developers should begin with ../ to go one folder level up like: [ mylink ]( ../setup/installation.md )","title":"Reference"},{"location":"developers/documentation/#extras","text":"You can insert emojis using this syntax :[SOURCE]-[ICON_NAME] from the following sources: https://materialdesignicons.com/ https://fontawesome.com/icons/tasks?style=solid https://primer.style/octicons/ You can use this page to create markdown tables more easily","title":"Extras"},{"location":"developers/remote-support/","text":"Remote Support \u00b6 We use the Live Share extension of Visual Studio Code to debug bugs when sharing data or database credentials is not possible. Install Visual Studio Code Open you RAPIDS root folder in a new VSCode window Open a new Terminal Terminal > New terminal Install the Live Share extension pack Press Ctrl + P or Cmd + P and run this command: >live share: start collaboration session 6. Follow the instructions and share the session link you receive","title":"Remote Support"},{"location":"developers/remote-support/#remote-support","text":"We use the Live Share extension of Visual Studio Code to debug bugs when sharing data or database credentials is not possible. Install Visual Studio Code Open you RAPIDS root folder in a new VSCode window Open a new Terminal Terminal > New terminal Install the Live Share extension pack Press Ctrl + P or Cmd + P and run this command: >live share: start collaboration session 6. Follow the instructions and share the session link you receive","title":"Remote Support"},{"location":"developers/test-cases/","text":"Test Cases \u00b6 Along with the continued development and the addition of new sensors and features to the RAPIDS pipeline, tests for the currently available sensors and features are being implemented. Since this is a Work In Progress this page will be updated with the list of sensors and features for which testing is available. For each of the sensors listed a description of the data used for testing (test cases) are outline. Currently for all intent and testing purposes the tests/data/raw/test01/ contains all the test data files for testing android data formats and tests/data/raw/test02/ contains all the test data files for testing iOS data formats. It follows that the expected (verified output) are contained in the tests/data/processed/test01/ and tests/data/processed/test02/ for Android and iOS respectively. tests/data/raw/test03/ and tests/data/raw/test04/ contain data files for testing empty raw data files for android and iOS respectively. The following is a list of the sensors that testing is currently available. Messages (SMS) \u00b6 The raw message data file contains data for 2 separate days. The data for the first day contains records 5 records for every epoch . The second day's data contains 6 records for each of only 2 epoch (currently morning and evening ) The raw message data contains records for both message_types (i.e. recieved and sent ) in both days in all epochs. The number records with each message_types per epoch is randomly distributed There is at least one records with each message_types per epoch. There is one raw message data file each, as described above, for testing both iOS and Android data. There is also an additional empty data file for both android and iOS for testing empty data files Calls \u00b6 Due to the difference in the format of the raw call data for iOS and Android the following is the expected results the calls_with_datetime_unified.csv . This would give a better idea of the use cases being tested since the calls_with_datetime_unified.csv would make both the iOS and Android data comparable. The call data would contain data for 2 days. The data for the first day contains 6 records for every epoch . The second day's data contains 6 records for each of only 2 epoch (currently morning and evening ) The call data contains records for all call_types (i.e. incoming , outgoing and missed ) in both days in all epochs. The number records with each of the call_types per epoch is randomly distributed. There is at least one records with each call_types per epoch. There is one call data file each, as described above, for testing both iOS and Android data. There is also an additional empty data file for both android and iOS for testing empty data files Screen \u00b6 Due to the difference in the format of the raw screen data for iOS and Android the following is the expected results the screen_deltas.csv . This would give a better idea of the use cases being tested since the screen_eltas.csv would make both the iOS and Android data comparable These files are used to calculate the features for the screen sensor The screen delta data file contains data for 1 day. The screen delta data contains 1 record to represent an unlock episode that falls within an epoch for every epoch . The screen delta data contains 1 record to represent an unlock episode that falls across the boundary of 2 epochs. Namely the unlock episode starts in one epoch and ends in the next, thus there is a record for unlock episodes that fall across night to morning , morning to afternoon and finally afternoon to night The testing is done for unlock episode_type. There is one screen data file each for testing both iOS and Android data formats. There is also an additional empty data file for both android and iOS for testing empty data files Battery \u00b6 Due to the difference in the format of the raw battery data for iOS and Android as well as versions of iOS the following is the expected results the battery_deltas.csv . This would give a better idea of the use cases being tested since the battery_deltas.csv would make both the iOS and Android data comparable. These files are used to calculate the features for the battery sensor. The battery delta data file contains data for 1 day. The battery delta data contains 1 record each for a charging and discharging episode that falls within an epoch for every epoch . Thus, for the daily epoch there would be multiple charging and discharging episodes Since either a charging episode or a discharging episode and not both can occur across epochs, in order to test episodes that occur across epochs alternating episodes of charging and discharging episodes that fall across night to morning , morning to afternoon and finally afternoon to night are present in the battery delta data. This starts with a discharging episode that begins in night and end in morning . There is one battery data file each, for testing both iOS and Android data formats. There is also an additional empty data file for both android and iOS for testing empty data files Bluetooth \u00b6 The raw Bluetooth data file contains data for 1 day. The raw Bluetooth data contains at least 2 records for each epoch . Each epoch has a record with a timestamp for the beginning boundary for that epoch and a record with a timestamp for the ending boundary for that epoch . (e.g. For the morning epoch there is a record with a timestamp for 6:00AM and another record with a timestamp for 11:59:59AM . These are to test edge cases) An option of 5 Bluetooth devices are randomly distributed throughout the data records. There is one raw Bluetooth data file each, for testing both iOS and Android data formats. There is also an additional empty data file for both android and iOS for testing empty data files. WIFI \u00b6 There are 2 data files ( wifi_raw.csv and sensor_wifi_raw.csv ) for each fake participant for each phone platform. The raw WIFI data files contain data for 1 day. The sensor_wifi_raw.csv data contains at least 2 records for each epoch . Each epoch has a record with a timestamp for the beginning boundary for that epoch and a record with a timestamp for the ending boundary for that epoch . (e.g. For the morning epoch there is a record with a timestamp for 6:00AM and another record with a timestamp for 11:59:59AM . These are to test edge cases) The wifi_raw.csv data contains 3 records with random timestamps for each epoch to represent visible broadcasting WIFI network. This file is empty for the iOS phone testing data. An option of 10 access point devices is randomly distributed throughout the data records. 5 each for sensor_wifi_raw.csv and wifi_raw.csv . There data files for testing both iOS and Android data formats. There are also additional empty data files for both android and iOS for testing empty data files. Light \u00b6 The raw light data file contains data for 1 day. The raw light data contains 3 or 4 rows of data for each epoch except night . The single row of data for night is for testing features for single values inputs. (Example testing the standard deviation of one input value) Since light is only available for Android there is only one file that contains data for Android. All other files (i.e. for iPhone) are empty data files. Application Foreground \u00b6 The raw application foreground data file contains data for 1 day. The raw application foreground data contains 7 - 9 rows of data for each epoch . The records for each epoch contains apps that are randomly selected from a list of apps that are from the MULTIPLE_CATEGORIES and SINGLE_CATEGORIES (See testing_config.yaml ). There are also records in each epoch that have apps randomly selected from a list of apps that are from the EXCLUDED_CATEGORIES and EXCLUDED_APPS . This is to test that these apps are actually being excluded from the calculations of features. There are also records to test SINGLE_APPS calculations. Since application foreground is only available for Android there is only one file that contains data for Android. All other files (i.e. for iPhone) are empty data files. Activity Recognition \u00b6 The raw Activity Recognition data file contains data for 1 day. The raw Activity Recognition data each epoch period contains rows that records 2 - 5 different activity_types . The is such that durations of activities can be tested. Additionally, there are records that mimic the duration of an activity over the time boundary of neighboring epochs. (For example, there a set of records that mimic the participant in_vehicle from afternoon into evening ) There is one file each with raw Activity Recognition data for testing both iOS and Android data formats. (plugin_google_activity_recognition_raw.csv for android and plugin_ios_activity_recognition_raw.csv for iOS) There is also an additional empty data file for both android and iOS for testing empty data files. Conversation \u00b6 The raw conversation data file contains data for 2 day. The raw conversation data contains records with a sample of both datatypes (i.e. voice/noise = 0 , and conversation = 2 ) as well as rows with for samples of each of the inference values (i.e. silence = 0 , noise = 1 , voice = 2 , and unknown = 3 ) for each epoch . The different datatype and inference records are randomly distributed throughout the epoch . Additionally there are 2 - 5 records for conversations ( datatype = 2, and inference = -1) in each epoch and for each epoch except night, there is a conversation record that has a double_convo_start timestamp that is from the previous epoch . This is to test the calculations of features across epochs . There is a raw conversation data file for both android and iOS platforms ( plugin_studentlife_audio_android_raw.csv and plugin_studentlife_audio_raw.csv respectively). Finally, there are also additional empty data files for both android and iOS for testing empty data files","title":"Test cases"},{"location":"developers/test-cases/#test-cases","text":"Along with the continued development and the addition of new sensors and features to the RAPIDS pipeline, tests for the currently available sensors and features are being implemented. Since this is a Work In Progress this page will be updated with the list of sensors and features for which testing is available. For each of the sensors listed a description of the data used for testing (test cases) are outline. Currently for all intent and testing purposes the tests/data/raw/test01/ contains all the test data files for testing android data formats and tests/data/raw/test02/ contains all the test data files for testing iOS data formats. It follows that the expected (verified output) are contained in the tests/data/processed/test01/ and tests/data/processed/test02/ for Android and iOS respectively. tests/data/raw/test03/ and tests/data/raw/test04/ contain data files for testing empty raw data files for android and iOS respectively. The following is a list of the sensors that testing is currently available.","title":"Test Cases"},{"location":"developers/test-cases/#messages-sms","text":"The raw message data file contains data for 2 separate days. The data for the first day contains records 5 records for every epoch . The second day's data contains 6 records for each of only 2 epoch (currently morning and evening ) The raw message data contains records for both message_types (i.e. recieved and sent ) in both days in all epochs. The number records with each message_types per epoch is randomly distributed There is at least one records with each message_types per epoch. There is one raw message data file each, as described above, for testing both iOS and Android data. There is also an additional empty data file for both android and iOS for testing empty data files","title":"Messages (SMS)"},{"location":"developers/test-cases/#calls","text":"Due to the difference in the format of the raw call data for iOS and Android the following is the expected results the calls_with_datetime_unified.csv . This would give a better idea of the use cases being tested since the calls_with_datetime_unified.csv would make both the iOS and Android data comparable. The call data would contain data for 2 days. The data for the first day contains 6 records for every epoch . The second day's data contains 6 records for each of only 2 epoch (currently morning and evening ) The call data contains records for all call_types (i.e. incoming , outgoing and missed ) in both days in all epochs. The number records with each of the call_types per epoch is randomly distributed. There is at least one records with each call_types per epoch. There is one call data file each, as described above, for testing both iOS and Android data. There is also an additional empty data file for both android and iOS for testing empty data files","title":"Calls"},{"location":"developers/test-cases/#screen","text":"Due to the difference in the format of the raw screen data for iOS and Android the following is the expected results the screen_deltas.csv . This would give a better idea of the use cases being tested since the screen_eltas.csv would make both the iOS and Android data comparable These files are used to calculate the features for the screen sensor The screen delta data file contains data for 1 day. The screen delta data contains 1 record to represent an unlock episode that falls within an epoch for every epoch . The screen delta data contains 1 record to represent an unlock episode that falls across the boundary of 2 epochs. Namely the unlock episode starts in one epoch and ends in the next, thus there is a record for unlock episodes that fall across night to morning , morning to afternoon and finally afternoon to night The testing is done for unlock episode_type. There is one screen data file each for testing both iOS and Android data formats. There is also an additional empty data file for both android and iOS for testing empty data files","title":"Screen"},{"location":"developers/test-cases/#battery","text":"Due to the difference in the format of the raw battery data for iOS and Android as well as versions of iOS the following is the expected results the battery_deltas.csv . This would give a better idea of the use cases being tested since the battery_deltas.csv would make both the iOS and Android data comparable. These files are used to calculate the features for the battery sensor. The battery delta data file contains data for 1 day. The battery delta data contains 1 record each for a charging and discharging episode that falls within an epoch for every epoch . Thus, for the daily epoch there would be multiple charging and discharging episodes Since either a charging episode or a discharging episode and not both can occur across epochs, in order to test episodes that occur across epochs alternating episodes of charging and discharging episodes that fall across night to morning , morning to afternoon and finally afternoon to night are present in the battery delta data. This starts with a discharging episode that begins in night and end in morning . There is one battery data file each, for testing both iOS and Android data formats. There is also an additional empty data file for both android and iOS for testing empty data files","title":"Battery"},{"location":"developers/test-cases/#bluetooth","text":"The raw Bluetooth data file contains data for 1 day. The raw Bluetooth data contains at least 2 records for each epoch . Each epoch has a record with a timestamp for the beginning boundary for that epoch and a record with a timestamp for the ending boundary for that epoch . (e.g. For the morning epoch there is a record with a timestamp for 6:00AM and another record with a timestamp for 11:59:59AM . These are to test edge cases) An option of 5 Bluetooth devices are randomly distributed throughout the data records. There is one raw Bluetooth data file each, for testing both iOS and Android data formats. There is also an additional empty data file for both android and iOS for testing empty data files.","title":"Bluetooth"},{"location":"developers/test-cases/#wifi","text":"There are 2 data files ( wifi_raw.csv and sensor_wifi_raw.csv ) for each fake participant for each phone platform. The raw WIFI data files contain data for 1 day. The sensor_wifi_raw.csv data contains at least 2 records for each epoch . Each epoch has a record with a timestamp for the beginning boundary for that epoch and a record with a timestamp for the ending boundary for that epoch . (e.g. For the morning epoch there is a record with a timestamp for 6:00AM and another record with a timestamp for 11:59:59AM . These are to test edge cases) The wifi_raw.csv data contains 3 records with random timestamps for each epoch to represent visible broadcasting WIFI network. This file is empty for the iOS phone testing data. An option of 10 access point devices is randomly distributed throughout the data records. 5 each for sensor_wifi_raw.csv and wifi_raw.csv . There data files for testing both iOS and Android data formats. There are also additional empty data files for both android and iOS for testing empty data files.","title":"WIFI"},{"location":"developers/test-cases/#light","text":"The raw light data file contains data for 1 day. The raw light data contains 3 or 4 rows of data for each epoch except night . The single row of data for night is for testing features for single values inputs. (Example testing the standard deviation of one input value) Since light is only available for Android there is only one file that contains data for Android. All other files (i.e. for iPhone) are empty data files.","title":"Light"},{"location":"developers/test-cases/#application-foreground","text":"The raw application foreground data file contains data for 1 day. The raw application foreground data contains 7 - 9 rows of data for each epoch . The records for each epoch contains apps that are randomly selected from a list of apps that are from the MULTIPLE_CATEGORIES and SINGLE_CATEGORIES (See testing_config.yaml ). There are also records in each epoch that have apps randomly selected from a list of apps that are from the EXCLUDED_CATEGORIES and EXCLUDED_APPS . This is to test that these apps are actually being excluded from the calculations of features. There are also records to test SINGLE_APPS calculations. Since application foreground is only available for Android there is only one file that contains data for Android. All other files (i.e. for iPhone) are empty data files.","title":"Application Foreground"},{"location":"developers/test-cases/#activity-recognition","text":"The raw Activity Recognition data file contains data for 1 day. The raw Activity Recognition data each epoch period contains rows that records 2 - 5 different activity_types . The is such that durations of activities can be tested. Additionally, there are records that mimic the duration of an activity over the time boundary of neighboring epochs. (For example, there a set of records that mimic the participant in_vehicle from afternoon into evening ) There is one file each with raw Activity Recognition data for testing both iOS and Android data formats. (plugin_google_activity_recognition_raw.csv for android and plugin_ios_activity_recognition_raw.csv for iOS) There is also an additional empty data file for both android and iOS for testing empty data files.","title":"Activity Recognition"},{"location":"developers/test-cases/#conversation","text":"The raw conversation data file contains data for 2 day. The raw conversation data contains records with a sample of both datatypes (i.e. voice/noise = 0 , and conversation = 2 ) as well as rows with for samples of each of the inference values (i.e. silence = 0 , noise = 1 , voice = 2 , and unknown = 3 ) for each epoch . The different datatype and inference records are randomly distributed throughout the epoch . Additionally there are 2 - 5 records for conversations ( datatype = 2, and inference = -1) in each epoch and for each epoch except night, there is a conversation record that has a double_convo_start timestamp that is from the previous epoch . This is to test the calculations of features across epochs . There is a raw conversation data file for both android and iOS platforms ( plugin_studentlife_audio_android_raw.csv and plugin_studentlife_audio_raw.csv respectively). Finally, there are also additional empty data files for both android and iOS for testing empty data files","title":"Conversation"},{"location":"developers/testing/","text":"Testing \u00b6 The following is a simple guide to testing RAPIDS. All files necessary for testing are stored in the /tests directory Steps for Testing \u00b6 To begin testing RAPIDS place the fake raw input data csv files in tests/data/raw/ . The fake participant files should be placed in tests/data/external/ . The expected output files of RAPIDS after processing the input data should be placed in tests/data/processesd/ . The Snakemake rule(s) that are to be tested must be placed in the tests/Snakemake file. The current tests/Snakemake is a good example of how to define them. (At the time of writing this documentation the snakefile contains rules messages (SMS), calls and screen) Edit the tests/settings/config.yaml . Add and/or remove the rules to be run for testing from the forcerun list. Edit the tests/settings/testing_config.yaml with the necessary configuration settings for running the rules to be tested. Add any additional testscripts in tests/scripts . Uncomment or comment off lines in the testing shell script tests/scripts/run_tests.sh . Run the testing shell script. tests/scripts/run_tests.sh The following is a snippet of the output you should see after running your test. test_sensors_files_exist ( test_sensor_features.TestSensorFeatures ) ... ok test_sensors_features_calculations ( test_sensor_features.TestSensorFeatures ) ... FAIL ====================================================================== FAIL: test_sensors_features_calculations ( test_sensor_features.TestSensorFeatures ) ---------------------------------------------------------------------- The results above show that the first test test_sensors_files_exist passed while test_sensors_features_calculations failed. In addition you should get the traceback of the failure (not shown here). For more information on how to implement test scripts and use unittest please see Unittest Documentation Testing of the RAPIDS sensors and features is a work-in-progress. Please see test-cases for a list of sensors and features that have testing currently available. Currently the repository is set up to test a number of sensors out of the box by simply running the tests/scripts/run_tests.sh command once the RAPIDS python environment is active.","title":"Testing"},{"location":"developers/testing/#testing","text":"The following is a simple guide to testing RAPIDS. All files necessary for testing are stored in the /tests directory","title":"Testing"},{"location":"developers/testing/#steps-for-testing","text":"To begin testing RAPIDS place the fake raw input data csv files in tests/data/raw/ . The fake participant files should be placed in tests/data/external/ . The expected output files of RAPIDS after processing the input data should be placed in tests/data/processesd/ . The Snakemake rule(s) that are to be tested must be placed in the tests/Snakemake file. The current tests/Snakemake is a good example of how to define them. (At the time of writing this documentation the snakefile contains rules messages (SMS), calls and screen) Edit the tests/settings/config.yaml . Add and/or remove the rules to be run for testing from the forcerun list. Edit the tests/settings/testing_config.yaml with the necessary configuration settings for running the rules to be tested. Add any additional testscripts in tests/scripts . Uncomment or comment off lines in the testing shell script tests/scripts/run_tests.sh . Run the testing shell script. tests/scripts/run_tests.sh The following is a snippet of the output you should see after running your test. test_sensors_files_exist ( test_sensor_features.TestSensorFeatures ) ... ok test_sensors_features_calculations ( test_sensor_features.TestSensorFeatures ) ... FAIL ====================================================================== FAIL: test_sensors_features_calculations ( test_sensor_features.TestSensorFeatures ) ---------------------------------------------------------------------- The results above show that the first test test_sensors_files_exist passed while test_sensors_features_calculations failed. In addition you should get the traceback of the failure (not shown here). For more information on how to implement test scripts and use unittest please see Unittest Documentation Testing of the RAPIDS sensors and features is a work-in-progress. Please see test-cases for a list of sensors and features that have testing currently available. Currently the repository is set up to test a number of sensors out of the box by simply running the tests/scripts/run_tests.sh command once the RAPIDS python environment is active.","title":"Steps for Testing"},{"location":"developers/virtual-environments/","text":"Virtual Environments \u00b6 Add new packages \u00b6 Try to install any new package using conda . If a package is not available in one of conda \u2018s channels you can install it with pip but make sure your virtual environment is active. Update your conda environment.yaml \u00b6 After installing a new package you can use the following command in your terminal to update your environment.yaml before publishing your pipeline. Note that we ignore the package version for libfortran to keep compatibility with Linux: conda env export --no-builds | sed 's/^.*libgfortran.*$/ - libgfortran/' > environment.yml Update and prune your conda environment from a environment.yaml file \u00b6 Execute the following command in your terminal, see these docs for more information conda env update --prefix ./env --file environment.yml --prune","title":"Virtual Environments"},{"location":"developers/virtual-environments/#virtual-environments","text":"","title":"Virtual Environments"},{"location":"developers/virtual-environments/#add-new-packages","text":"Try to install any new package using conda . If a package is not available in one of conda \u2018s channels you can install it with pip but make sure your virtual environment is active.","title":"Add new packages"},{"location":"developers/virtual-environments/#update-your-conda-environmentyaml","text":"After installing a new package you can use the following command in your terminal to update your environment.yaml before publishing your pipeline. Note that we ignore the package version for libfortran to keep compatibility with Linux: conda env export --no-builds | sed 's/^.*libgfortran.*$/ - libgfortran/' > environment.yml","title":"Update your conda environment.yaml"},{"location":"developers/virtual-environments/#update-and-prune-your-conda-environment-from-a-environmentyaml-file","text":"Execute the following command in your terminal, see these docs for more information conda env update --prefix ./env --file environment.yml --prune","title":"Update and prune your conda environment from a environment.yaml file"},{"location":"features/add-new-features/","text":"Add New Features \u00b6 Hint We recommend reading the Behavioral Features Introduction before reading this page Hint You won\u2019t have to deal with time zones, dates, times, data cleaning or preprocessing. The data that RAPIDS pipes to your feature extraction code is ready to process. New Features for Existing Sensors \u00b6 You can add new features to any existing sensors (see list below) by adding a new provider in three steps: Modify the config.yaml file Create a provider folder, script and function Implement your features extraction code As a tutorial, we will add a new provider for PHONE_ACCELEROMETER called VEGA that extracts feature1 , feature2 , feature3 in Python and that it requires a parameter from the user called MY_PARAMETER . Existing Sensors An existing sensor is any of the phone or Fitbit sensors with a configuration entry in config.yaml : Phone Accelerometer Phone Activity Recognition Phone Applications Foreground Phone Battery Phone Bluetooth Phone Calls Phone Conversation Phone Data Yield Phone Light Phone Locations Phone Messages Phone Screen Phone WiFI Connected Phone WiFI Visible Fitbit Heart Rate Summary Fitbit Heart Rate Intraday Fitbit Sleep Summary Fitbit Steps Summary Fitbit Steps Intraday Modify the config.yaml file \u00b6 In this step you need to add your provider configuration section under the relevant sensor in config.yaml . See our example for our tutorial\u2019s VEGA provider for PHONE_ACCELEROMETER : Example configuration for a new accelerometer provider VEGA PHONE_ACCELEROMETER : TABLE : accelerometer PROVIDERS : RAPIDS : COMPUTE : False ... PANDA : COMPUTE : False ... VEGA : COMPUTE : False FEATURES : [ \"feature1\" , \"feature2\" , \"feature3\" ] MY_PARAMTER : a_string SRC_FOLDER : \"vega\" SRC_LANGUAGE : \"python\" Key Description [COMPUTE] Flag to activate/deactivate your provider [FEATURES] List of features your provider supports. Your provider code should only return the features on this list [MY_PARAMTER] An arbitrary parameter that our example provider VEGA needs. This can be a boolean, integer, float, string or an array of any of such types. [SRC_LANGUAGE] The programming language of your provider script, it can be python or r , in our example python [SRC_FOLDER] The name of your provider in lower case, in our example vega (this will be the name of your folder in the next step) Create a provider folder, script and function \u00b6 In this step you need to add a folder, script and function for your provider. Create your provider folder under src/feature/DEVICE_SENSOR/YOUR_PROVIDER , in our example src/feature/phone_accelerometer/vega (same as [SRC_FOLDER] in the step above). Create your provider script inside your provider folder, it can be a Python file called main.py or an R file called main.R . Add your provider function in your provider script. The name of such function should be [providername]_features , in our example vega_features Python function def [ providername ] _features ( sensor_data_files , time_segment , provider , filter_data_by_segment , * args , ** kwargs ): R function [ providername ] _ features <- function ( sensor_data , time_segment , provider ) Implement your feature extraction code \u00b6 The provider function that you created in the step above will receive the following parameters: Parameter Description sensor_data_files Path to the CSV file containing the data of a single participant. This data has been cleaned and preprocessed. Your function will be automatically called for each participant in your study (in the [PIDS] array in config.yaml ) time_segment The label of the time segment that should be processed. provider The parameters you configured for your provider in config.yaml will be available in this variable as a dictionary in Python or a list in R. In our example this dictionary contains {MY_PARAMETER:\"a_string\"} filter_data_by_segment Python only. A function that you will use to filter your data. In R this function is already available in the environment. *args Python only. Not used for now **kwargs Python only. Not used for now The code to extract your behavioral features should be implemented in your provider function and in general terms it will have three stages: 1. Read a participant\u2019s data by loading the CSV data stored in the file pointed by sensor_data_files acc_data = pd . read_csv ( sensor_data_files [ \"sensor_data\" ]) Note that phone\u2019s battery, screen, and activity recognition data is given as episodes instead of event rows (for example, start and end timestamps of the periods the phone screen was on) 2. Filter your data to process only those rows that belong to time_segment This step is only one line of code, but to undersand why we need it, keep reading. acc_data = filter_data_by_segment ( acc_data , time_segment ) You should use the filter_data_by_segment() function to process and group those rows that belong to each of the time segments RAPIDS could be configured with . Let\u2019s understand the filter_data_by_segment() function with an example. A RAPIDS user can extract features on any arbitrary time segment . A time segment is a period of time that has a label and one or more instances. For example, the user (or you) could have requested features on a daily, weekly, and week-end basis for p01 . The labels are arbritrary and the instances depend on the days a participant was monitored for: the daily segment could be named my_days and if p01 was monitored for 14 days, it would have 14 instances the weekly segment could be named my_weeks and if p01 was monitored for 14 days, it would have 2 instances. the weekend segment could be named my_weekends and if p01 was monitored for 14 days, it would have 2 instances. For this example, RAPIDS will call your provider function three times for p01 , once where time_segment is my_days , once where time_segment is my_weeks and once where time_segment is my_weekends . In this example not every row in p01 \u2018s data needs to take part in the feature computation for either segment and the rows need to be grouped differently. Thus filter_data_by_segment() comes in handy, it will return a data frame that contains the rows that were logged during a time segment plus an extra column called local_segment . This new column will have as many unique values as time segment instances exist (14, 2, and 2 for our p01 \u2018s my_days , my_weeks , and my_weekends examples). After filtering, you should group the data frame by this column and compute any desired features , for example: acc_features [ \"maxmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . max () The reason RAPIDS does not filter the participant\u2019s data set for you is because your code might need to compute something based on a participant\u2019s complete dataset before computing their features. For example, you might want to identify the number that called a participant the most throughout the study before computing a feature with the number of calls the participant received from this number. 3. Return a data frame with your features After filtering, grouping your data, and computing your features, your provider function should return a data frame that has: One row per time segment instance (e.g. 14 our p01 \u2018s my_days example) The local_segment column added by filter_data_by_segment() One column per feature. By convention the name of your features should only contain letters or numbers ( feature1 ). RAPIDS will automatically add the right sensor and provider prefix ( phone_accelerometr_vega_ ) PHONE_ACCELEROMETER Provider Example For your reference, this a short example of our own provider ( RAPIDS ) for PHONE_ACCELEROMETER that computes five acceleration features def rapids_features ( sensor_data_files , time_segment , provider , filter_data_by_segment , * args , ** kwargs ): acc_data = pd . read_csv ( sensor_data_files [ \"sensor_data\" ]) requested_features = provider [ \"FEATURES\" ] # name of the features this function can compute base_features_names = [ \"maxmagnitude\" , \"minmagnitude\" , \"avgmagnitude\" , \"medianmagnitude\" , \"stdmagnitude\" ] # the subset of requested features this function can compute features_to_compute = list ( set ( requested_features ) & set ( base_features_names )) acc_features = pd . DataFrame ( columns = [ \"local_segment\" ] + features_to_compute ) if not acc_data . empty : acc_data = filter_data_by_segment ( acc_data , time_segment ) if not acc_data . empty : acc_features = pd . DataFrame () # get magnitude related features: magnitude = sqrt(x^2+y^2+z^2) magnitude = acc_data . apply ( lambda row : np . sqrt ( row [ \"double_values_0\" ] ** 2 + row [ \"double_values_1\" ] ** 2 + row [ \"double_values_2\" ] ** 2 ), axis = 1 ) acc_data = acc_data . assign ( magnitude = magnitude . values ) if \"maxmagnitude\" in features_to_compute : acc_features [ \"maxmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . max () if \"minmagnitude\" in features_to_compute : acc_features [ \"minmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . min () if \"avgmagnitude\" in features_to_compute : acc_features [ \"avgmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . mean () if \"medianmagnitude\" in features_to_compute : acc_features [ \"medianmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . median () if \"stdmagnitude\" in features_to_compute : acc_features [ \"stdmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . std () acc_features = acc_features . reset_index () return acc_features New Features for Non-Existing Sensors \u00b6 If you want to add features for a device or a sensor that we do not support at the moment (those that do not appear in the \"Existing Sensors\" list above), contact us or request it on Slack and we can add the necessary code so you can follow the instructions above.","title":"Add New Features"},{"location":"features/add-new-features/#add-new-features","text":"Hint We recommend reading the Behavioral Features Introduction before reading this page Hint You won\u2019t have to deal with time zones, dates, times, data cleaning or preprocessing. The data that RAPIDS pipes to your feature extraction code is ready to process.","title":"Add New Features"},{"location":"features/add-new-features/#new-features-for-existing-sensors","text":"You can add new features to any existing sensors (see list below) by adding a new provider in three steps: Modify the config.yaml file Create a provider folder, script and function Implement your features extraction code As a tutorial, we will add a new provider for PHONE_ACCELEROMETER called VEGA that extracts feature1 , feature2 , feature3 in Python and that it requires a parameter from the user called MY_PARAMETER . Existing Sensors An existing sensor is any of the phone or Fitbit sensors with a configuration entry in config.yaml : Phone Accelerometer Phone Activity Recognition Phone Applications Foreground Phone Battery Phone Bluetooth Phone Calls Phone Conversation Phone Data Yield Phone Light Phone Locations Phone Messages Phone Screen Phone WiFI Connected Phone WiFI Visible Fitbit Heart Rate Summary Fitbit Heart Rate Intraday Fitbit Sleep Summary Fitbit Steps Summary Fitbit Steps Intraday","title":"New Features for Existing Sensors"},{"location":"features/add-new-features/#modify-the-configyaml-file","text":"In this step you need to add your provider configuration section under the relevant sensor in config.yaml . See our example for our tutorial\u2019s VEGA provider for PHONE_ACCELEROMETER : Example configuration for a new accelerometer provider VEGA PHONE_ACCELEROMETER : TABLE : accelerometer PROVIDERS : RAPIDS : COMPUTE : False ... PANDA : COMPUTE : False ... VEGA : COMPUTE : False FEATURES : [ \"feature1\" , \"feature2\" , \"feature3\" ] MY_PARAMTER : a_string SRC_FOLDER : \"vega\" SRC_LANGUAGE : \"python\" Key Description [COMPUTE] Flag to activate/deactivate your provider [FEATURES] List of features your provider supports. Your provider code should only return the features on this list [MY_PARAMTER] An arbitrary parameter that our example provider VEGA needs. This can be a boolean, integer, float, string or an array of any of such types. [SRC_LANGUAGE] The programming language of your provider script, it can be python or r , in our example python [SRC_FOLDER] The name of your provider in lower case, in our example vega (this will be the name of your folder in the next step)","title":"Modify the config.yaml file"},{"location":"features/add-new-features/#create-a-provider-folder-script-and-function","text":"In this step you need to add a folder, script and function for your provider. Create your provider folder under src/feature/DEVICE_SENSOR/YOUR_PROVIDER , in our example src/feature/phone_accelerometer/vega (same as [SRC_FOLDER] in the step above). Create your provider script inside your provider folder, it can be a Python file called main.py or an R file called main.R . Add your provider function in your provider script. The name of such function should be [providername]_features , in our example vega_features Python function def [ providername ] _features ( sensor_data_files , time_segment , provider , filter_data_by_segment , * args , ** kwargs ): R function [ providername ] _ features <- function ( sensor_data , time_segment , provider )","title":"Create a provider folder, script and function"},{"location":"features/add-new-features/#implement-your-feature-extraction-code","text":"The provider function that you created in the step above will receive the following parameters: Parameter Description sensor_data_files Path to the CSV file containing the data of a single participant. This data has been cleaned and preprocessed. Your function will be automatically called for each participant in your study (in the [PIDS] array in config.yaml ) time_segment The label of the time segment that should be processed. provider The parameters you configured for your provider in config.yaml will be available in this variable as a dictionary in Python or a list in R. In our example this dictionary contains {MY_PARAMETER:\"a_string\"} filter_data_by_segment Python only. A function that you will use to filter your data. In R this function is already available in the environment. *args Python only. Not used for now **kwargs Python only. Not used for now The code to extract your behavioral features should be implemented in your provider function and in general terms it will have three stages: 1. Read a participant\u2019s data by loading the CSV data stored in the file pointed by sensor_data_files acc_data = pd . read_csv ( sensor_data_files [ \"sensor_data\" ]) Note that phone\u2019s battery, screen, and activity recognition data is given as episodes instead of event rows (for example, start and end timestamps of the periods the phone screen was on) 2. Filter your data to process only those rows that belong to time_segment This step is only one line of code, but to undersand why we need it, keep reading. acc_data = filter_data_by_segment ( acc_data , time_segment ) You should use the filter_data_by_segment() function to process and group those rows that belong to each of the time segments RAPIDS could be configured with . Let\u2019s understand the filter_data_by_segment() function with an example. A RAPIDS user can extract features on any arbitrary time segment . A time segment is a period of time that has a label and one or more instances. For example, the user (or you) could have requested features on a daily, weekly, and week-end basis for p01 . The labels are arbritrary and the instances depend on the days a participant was monitored for: the daily segment could be named my_days and if p01 was monitored for 14 days, it would have 14 instances the weekly segment could be named my_weeks and if p01 was monitored for 14 days, it would have 2 instances. the weekend segment could be named my_weekends and if p01 was monitored for 14 days, it would have 2 instances. For this example, RAPIDS will call your provider function three times for p01 , once where time_segment is my_days , once where time_segment is my_weeks and once where time_segment is my_weekends . In this example not every row in p01 \u2018s data needs to take part in the feature computation for either segment and the rows need to be grouped differently. Thus filter_data_by_segment() comes in handy, it will return a data frame that contains the rows that were logged during a time segment plus an extra column called local_segment . This new column will have as many unique values as time segment instances exist (14, 2, and 2 for our p01 \u2018s my_days , my_weeks , and my_weekends examples). After filtering, you should group the data frame by this column and compute any desired features , for example: acc_features [ \"maxmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . max () The reason RAPIDS does not filter the participant\u2019s data set for you is because your code might need to compute something based on a participant\u2019s complete dataset before computing their features. For example, you might want to identify the number that called a participant the most throughout the study before computing a feature with the number of calls the participant received from this number. 3. Return a data frame with your features After filtering, grouping your data, and computing your features, your provider function should return a data frame that has: One row per time segment instance (e.g. 14 our p01 \u2018s my_days example) The local_segment column added by filter_data_by_segment() One column per feature. By convention the name of your features should only contain letters or numbers ( feature1 ). RAPIDS will automatically add the right sensor and provider prefix ( phone_accelerometr_vega_ ) PHONE_ACCELEROMETER Provider Example For your reference, this a short example of our own provider ( RAPIDS ) for PHONE_ACCELEROMETER that computes five acceleration features def rapids_features ( sensor_data_files , time_segment , provider , filter_data_by_segment , * args , ** kwargs ): acc_data = pd . read_csv ( sensor_data_files [ \"sensor_data\" ]) requested_features = provider [ \"FEATURES\" ] # name of the features this function can compute base_features_names = [ \"maxmagnitude\" , \"minmagnitude\" , \"avgmagnitude\" , \"medianmagnitude\" , \"stdmagnitude\" ] # the subset of requested features this function can compute features_to_compute = list ( set ( requested_features ) & set ( base_features_names )) acc_features = pd . DataFrame ( columns = [ \"local_segment\" ] + features_to_compute ) if not acc_data . empty : acc_data = filter_data_by_segment ( acc_data , time_segment ) if not acc_data . empty : acc_features = pd . DataFrame () # get magnitude related features: magnitude = sqrt(x^2+y^2+z^2) magnitude = acc_data . apply ( lambda row : np . sqrt ( row [ \"double_values_0\" ] ** 2 + row [ \"double_values_1\" ] ** 2 + row [ \"double_values_2\" ] ** 2 ), axis = 1 ) acc_data = acc_data . assign ( magnitude = magnitude . values ) if \"maxmagnitude\" in features_to_compute : acc_features [ \"maxmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . max () if \"minmagnitude\" in features_to_compute : acc_features [ \"minmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . min () if \"avgmagnitude\" in features_to_compute : acc_features [ \"avgmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . mean () if \"medianmagnitude\" in features_to_compute : acc_features [ \"medianmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . median () if \"stdmagnitude\" in features_to_compute : acc_features [ \"stdmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . std () acc_features = acc_features . reset_index () return acc_features","title":"Implement your feature extraction code"},{"location":"features/add-new-features/#new-features-for-non-existing-sensors","text":"If you want to add features for a device or a sensor that we do not support at the moment (those that do not appear in the \"Existing Sensors\" list above), contact us or request it on Slack and we can add the necessary code so you can follow the instructions above.","title":"New Features for Non-Existing Sensors"},{"location":"features/feature-introduction/","text":"Behavioral Features Introduction \u00b6 Every phone or Fitbit sensor has a corresponding config section in config.yaml , these sections follow a similar structure and we\u2019ll use PHONE_ACCELEROMETER as an example to explain this structure. Hint We recommend reading this page if you are using RAPIDS for the first time All computed sensor features are stored under /data/processed/features on files per sensor, per participant and per study (all participants). Every time you change any sensor parameters, provider parameters or provider features, all the necessary files will be updated as soon as you execute RAPIDS. Config section example for PHONE_ACCELEROMETER # 1) Config section PHONE_ACCELEROMETER : # 2) Parameters for PHONE_ACCELEROMETER TABLE : accelerometer # 3) Providers for PHONE_ACCELEROMETER PROVIDERS : # 4) RAPIDS provider RAPIDS : # 4.1) Parameters of RAPIDS provider of PHONE_ACCELEROMETER COMPUTE : False # 4.2) Features of RAPIDS provider of PHONE_ACCELEROMETER FEATURES : [ \"maxmagnitude\" , \"minmagnitude\" , \"avgmagnitude\" , \"medianmagnitude\" , \"stdmagnitude\" ] SRC_FOLDER : \"rapids\" # inside src/features/phone_accelerometer SRC_LANGUAGE : \"python\" # 5) PANDA provider PANDA : # 5.1) Parameters of PANDA provider of PHONE_ACCELEROMETER COMPUTE : False VALID_SENSED_MINUTES : False # 5.2) Features of PANDA provider of PHONE_ACCELEROMETER FEATURES : exertional_activity_episode : [ \"sumduration\" , \"maxduration\" , \"minduration\" , \"avgduration\" , \"medianduration\" , \"stdduration\" ] nonexertional_activity_episode : [ \"sumduration\" , \"maxduration\" , \"minduration\" , \"avgduration\" , \"medianduration\" , \"stdduration\" ] SRC_FOLDER : \"panda\" # inside src/features/phone_accelerometer SRC_LANGUAGE : \"python\" Sensor Parameters \u00b6 Each sensor configuration section has a \u201cparameters\u201d subsection (see #2 in the example). These are parameters that affect different aspects of how the raw data is downloaded, and processed. The TABLE parameter exists for every sensor, but some sensors will have extra parameters like [PHONE_LOCATIONS] . We explain these parameters in a table at the top of each sensor documentation page. Sensor Providers \u00b6 Each sensor configuration section can have zero, one or more behavioral feature providers (see #3 in the example). A provider is a script created by the core RAPIDS team or other researchers that extracts behavioral features for that sensor. In this example, accelerometer has two providers: RAPIDS (see #4 ) and PANDA (see #5 ). Provider Parameters \u00b6 Each provider has parameters that affect the computation of the behavioral features it offers (see #4.1 or #5.1 in the example). These parameters will include at least a [COMPUTE] flag that you switch to True to extract a provider\u2019s behavioral features. We explain every provider\u2019s parameter in a table under the Parameters description heading on each provider documentation page. Provider Features \u00b6 Each provider offers a set of behavioral features (see #4.2 or #5.2 in the example). For some providers these features are grouped in an array (like those for RAPIDS provider in #4.2 ) but for others they are grouped in a collection of arrays depending on the meaning and purpose of those features (like those for PANDAS provider in #5.2 ). In either case, you can delete the features you are not interested in and they will not be included in the sensor\u2019s output feature file. We explain each behavioral feature in a table under the Features description heading on each provider documentation page.","title":"Introduction"},{"location":"features/feature-introduction/#behavioral-features-introduction","text":"Every phone or Fitbit sensor has a corresponding config section in config.yaml , these sections follow a similar structure and we\u2019ll use PHONE_ACCELEROMETER as an example to explain this structure. Hint We recommend reading this page if you are using RAPIDS for the first time All computed sensor features are stored under /data/processed/features on files per sensor, per participant and per study (all participants). Every time you change any sensor parameters, provider parameters or provider features, all the necessary files will be updated as soon as you execute RAPIDS. Config section example for PHONE_ACCELEROMETER # 1) Config section PHONE_ACCELEROMETER : # 2) Parameters for PHONE_ACCELEROMETER TABLE : accelerometer # 3) Providers for PHONE_ACCELEROMETER PROVIDERS : # 4) RAPIDS provider RAPIDS : # 4.1) Parameters of RAPIDS provider of PHONE_ACCELEROMETER COMPUTE : False # 4.2) Features of RAPIDS provider of PHONE_ACCELEROMETER FEATURES : [ \"maxmagnitude\" , \"minmagnitude\" , \"avgmagnitude\" , \"medianmagnitude\" , \"stdmagnitude\" ] SRC_FOLDER : \"rapids\" # inside src/features/phone_accelerometer SRC_LANGUAGE : \"python\" # 5) PANDA provider PANDA : # 5.1) Parameters of PANDA provider of PHONE_ACCELEROMETER COMPUTE : False VALID_SENSED_MINUTES : False # 5.2) Features of PANDA provider of PHONE_ACCELEROMETER FEATURES : exertional_activity_episode : [ \"sumduration\" , \"maxduration\" , \"minduration\" , \"avgduration\" , \"medianduration\" , \"stdduration\" ] nonexertional_activity_episode : [ \"sumduration\" , \"maxduration\" , \"minduration\" , \"avgduration\" , \"medianduration\" , \"stdduration\" ] SRC_FOLDER : \"panda\" # inside src/features/phone_accelerometer SRC_LANGUAGE : \"python\"","title":"Behavioral Features Introduction"},{"location":"features/feature-introduction/#sensor-parameters","text":"Each sensor configuration section has a \u201cparameters\u201d subsection (see #2 in the example). These are parameters that affect different aspects of how the raw data is downloaded, and processed. The TABLE parameter exists for every sensor, but some sensors will have extra parameters like [PHONE_LOCATIONS] . We explain these parameters in a table at the top of each sensor documentation page.","title":"Sensor Parameters"},{"location":"features/feature-introduction/#sensor-providers","text":"Each sensor configuration section can have zero, one or more behavioral feature providers (see #3 in the example). A provider is a script created by the core RAPIDS team or other researchers that extracts behavioral features for that sensor. In this example, accelerometer has two providers: RAPIDS (see #4 ) and PANDA (see #5 ).","title":"Sensor Providers"},{"location":"features/feature-introduction/#provider-parameters","text":"Each provider has parameters that affect the computation of the behavioral features it offers (see #4.1 or #5.1 in the example). These parameters will include at least a [COMPUTE] flag that you switch to True to extract a provider\u2019s behavioral features. We explain every provider\u2019s parameter in a table under the Parameters description heading on each provider documentation page.","title":"Provider Parameters"},{"location":"features/feature-introduction/#provider-features","text":"Each provider offers a set of behavioral features (see #4.2 or #5.2 in the example). For some providers these features are grouped in an array (like those for RAPIDS provider in #4.2 ) but for others they are grouped in a collection of arrays depending on the meaning and purpose of those features (like those for PANDAS provider in #5.2 ). In either case, you can delete the features you are not interested in and they will not be included in the sensor\u2019s output feature file. We explain each behavioral feature in a table under the Features description heading on each provider documentation page.","title":"Provider Features"},{"location":"features/fitbit-heartrate-intraday/","text":"Fitbit Heart Rate Intraday \u00b6 Sensor parameters description for [FITBIT_HEARTRATE_INTRADAY] : Key Description [TABLE] Database table name or file path where the heart rate intraday data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-07\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:1200.6102,\u201dmax\u201d:88,\u201dmin\u201d:31,\u201dminutes\u201d:1058,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:760.3020,\u201dmax\u201d:120,\u201dmin\u201d:86,\u201dminutes\u201d:366,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:15.2048,\u201dmax\u201d:146,\u201dmin\u201d:120,\u201dminutes\u201d:2,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:221,\u201dmin\u201d:148,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:72}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:68},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:67},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:67},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-08\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:1100.1120,\u201dmax\u201d:89,\u201dmin\u201d:30,\u201dminutes\u201d:921,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:660.0012,\u201dmax\u201d:118,\u201dmin\u201d:82,\u201dminutes\u201d:361,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:23.7088,\u201dmax\u201d:142,\u201dmin\u201d:108,\u201dminutes\u201d:3,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:221,\u201dmin\u201d:148,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:70}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:77},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:75},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:73},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-09\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:750.3615,\u201dmax\u201d:77,\u201dmin\u201d:30,\u201dminutes\u201d:851,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:734.1516,\u201dmax\u201d:107,\u201dmin\u201d:77,\u201dminutes\u201d:550,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:131.8579,\u201dmax\u201d:130,\u201dmin\u201d:107,\u201dminutes\u201d:29,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:220,\u201dmin\u201d:130,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:69}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:90},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:89},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:88},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} PLAIN_TEXT device_id local_date_time heartrate heartrate_zone a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:00:00 68 outofrange a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:01:00 67 outofrange a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:02:00 67 outofrange RAPIDS provider \u00b6 Available time segments Available for all time segments File Sequence - data/raw/ { pid } /fitbit_heartrate_intraday_raw.csv - data/raw/ { pid } /fitbit_heartrate_intraday_parsed.csv - data/raw/ { pid } /fitbit_heartrate_intraday_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_heartrate_intraday_features/fitbit_heartrate_intraday_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_heartrate_intraday.csv Parameters description for [FITBIT_HEARTRATE_INTRADAY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_HEARTRATE_INTRADAY features from the RAPIDS provider [FEATURES] Features to be computed from heart rate intraday data, see table below Features description for [FITBIT_HEARTRATE_INTRADAY][PROVIDERS][RAPIDS] : Feature Units Description maxhr beats/mins The maximum heart rate during a time segment. minhr beats/mins The minimum heart rate during a time segment. avghr beats/mins The average heart rate during a time segment. medianhr beats/mins The median of heart rate during a time segment. modehr beats/mins The mode of heart rate during a time segment. stdhr beats/mins The standard deviation of heart rate during a time segment. diffmaxmodehr beats/mins The difference between the maximum and mode heart rate during a time segment. diffminmodehr beats/mins The difference between the mode and minimum heart rate during a time segment. entropyhr nats Shannon\u2019s entropy measurement based on heart rate during a time segment. minutesonZONE minutes Number of minutes the user\u2019s heart rate fell within each heartrate_zone during a time segment. Assumptions/Observations There are four heart rate zones (ZONE): outofrange , fatburn , cardio , and peak . Please refer to Fitbit documentation for more information about the way they are computed.","title":"Fitbit Heart Rate Intraday"},{"location":"features/fitbit-heartrate-intraday/#fitbit-heart-rate-intraday","text":"Sensor parameters description for [FITBIT_HEARTRATE_INTRADAY] : Key Description [TABLE] Database table name or file path where the heart rate intraday data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-07\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:1200.6102,\u201dmax\u201d:88,\u201dmin\u201d:31,\u201dminutes\u201d:1058,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:760.3020,\u201dmax\u201d:120,\u201dmin\u201d:86,\u201dminutes\u201d:366,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:15.2048,\u201dmax\u201d:146,\u201dmin\u201d:120,\u201dminutes\u201d:2,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:221,\u201dmin\u201d:148,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:72}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:68},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:67},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:67},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-08\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:1100.1120,\u201dmax\u201d:89,\u201dmin\u201d:30,\u201dminutes\u201d:921,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:660.0012,\u201dmax\u201d:118,\u201dmin\u201d:82,\u201dminutes\u201d:361,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:23.7088,\u201dmax\u201d:142,\u201dmin\u201d:108,\u201dminutes\u201d:3,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:221,\u201dmin\u201d:148,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:70}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:77},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:75},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:73},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-09\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:750.3615,\u201dmax\u201d:77,\u201dmin\u201d:30,\u201dminutes\u201d:851,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:734.1516,\u201dmax\u201d:107,\u201dmin\u201d:77,\u201dminutes\u201d:550,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:131.8579,\u201dmax\u201d:130,\u201dmin\u201d:107,\u201dminutes\u201d:29,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:220,\u201dmin\u201d:130,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:69}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:90},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:89},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:88},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} PLAIN_TEXT device_id local_date_time heartrate heartrate_zone a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:00:00 68 outofrange a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:01:00 67 outofrange a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:02:00 67 outofrange","title":"Fitbit Heart Rate Intraday"},{"location":"features/fitbit-heartrate-intraday/#rapids-provider","text":"Available time segments Available for all time segments File Sequence - data/raw/ { pid } /fitbit_heartrate_intraday_raw.csv - data/raw/ { pid } /fitbit_heartrate_intraday_parsed.csv - data/raw/ { pid } /fitbit_heartrate_intraday_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_heartrate_intraday_features/fitbit_heartrate_intraday_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_heartrate_intraday.csv Parameters description for [FITBIT_HEARTRATE_INTRADAY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_HEARTRATE_INTRADAY features from the RAPIDS provider [FEATURES] Features to be computed from heart rate intraday data, see table below Features description for [FITBIT_HEARTRATE_INTRADAY][PROVIDERS][RAPIDS] : Feature Units Description maxhr beats/mins The maximum heart rate during a time segment. minhr beats/mins The minimum heart rate during a time segment. avghr beats/mins The average heart rate during a time segment. medianhr beats/mins The median of heart rate during a time segment. modehr beats/mins The mode of heart rate during a time segment. stdhr beats/mins The standard deviation of heart rate during a time segment. diffmaxmodehr beats/mins The difference between the maximum and mode heart rate during a time segment. diffminmodehr beats/mins The difference between the mode and minimum heart rate during a time segment. entropyhr nats Shannon\u2019s entropy measurement based on heart rate during a time segment. minutesonZONE minutes Number of minutes the user\u2019s heart rate fell within each heartrate_zone during a time segment. Assumptions/Observations There are four heart rate zones (ZONE): outofrange , fatburn , cardio , and peak . Please refer to Fitbit documentation for more information about the way they are computed.","title":"RAPIDS provider"},{"location":"features/fitbit-heartrate-summary/","text":"Fitbit Heart Rate Summary \u00b6 Sensor parameters description for [FITBIT_HEARTRATE_SUMMARY] : Key Description [TABLE] Database table name or file path where the heart rate summary data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-07\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:1200.6102,\u201dmax\u201d:88,\u201dmin\u201d:31,\u201dminutes\u201d:1058,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:760.3020,\u201dmax\u201d:120,\u201dmin\u201d:86,\u201dminutes\u201d:366,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:15.2048,\u201dmax\u201d:146,\u201dmin\u201d:120,\u201dminutes\u201d:2,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:221,\u201dmin\u201d:148,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:72}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:68},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:67},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:67},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-08\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:1100.1120,\u201dmax\u201d:89,\u201dmin\u201d:30,\u201dminutes\u201d:921,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:660.0012,\u201dmax\u201d:118,\u201dmin\u201d:82,\u201dminutes\u201d:361,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:23.7088,\u201dmax\u201d:142,\u201dmin\u201d:108,\u201dminutes\u201d:3,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:221,\u201dmin\u201d:148,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:70}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:77},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:75},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:73},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-09\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:750.3615,\u201dmax\u201d:77,\u201dmin\u201d:30,\u201dminutes\u201d:851,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:734.1516,\u201dmax\u201d:107,\u201dmin\u201d:77,\u201dminutes\u201d:550,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:131.8579,\u201dmax\u201d:130,\u201dmin\u201d:107,\u201dminutes\u201d:29,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:220,\u201dmin\u201d:130,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:69}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:90},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:89},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:88},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} PLAIN_TEXT device_id local_date_time heartrate_daily_restinghr heartrate_daily_caloriesoutofrange heartrate_daily_caloriesfatburn heartrate_daily_caloriescardio heartrate_daily_caloriespeak a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 72 1200.6102 760.3020 15.2048 0 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-08 70 1100.1120 660.0012 23.7088 0 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-09 69 750.3615 734.1516 131.8579 0 RAPIDS provider \u00b6 Available time segments Only available for segments that span 1 or more complete days (e.g. Jan 1 st 00:00 to Jan 3 rd 23:59) File Sequence - data/raw/ { pid } /fitbit_heartrate_summary_raw.csv - data/raw/ { pid } /fitbit_heartrate_summary_parsed.csv - data/raw/ { pid } /fitbit_heartrate_summary_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_heartrate_summary_features/fitbit_heartrate_summary_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_heartrate_summary.csv Parameters description for [FITBIT_HEARTRATE_SUMMARY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_HEARTRATE_SUMMARY features from the RAPIDS provider [FEATURES] Features to be computed from heart rate summary data, see table below Features description for [FITBIT_HEARTRATE_SUMMARY][PROVIDERS][RAPIDS] : Feature Units Description maxrestinghr beats/mins The maximum daily resting heart rate during a time segment. minrestinghr beats/mins The minimum daily resting heart rate during a time segment. avgrestinghr beats/mins The average daily resting heart rate during a time segment. medianrestinghr beats/mins The median of daily resting heart rate during a time segment. moderestinghr beats/mins The mode of daily resting heart rate during a time segment. stdrestinghr beats/mins The standard deviation of daily resting heart rate during a time segment. diffmaxmoderestinghr beats/mins The difference between the maximum and mode daily resting heart rate during a time segment. diffminmoderestinghr beats/mins The difference between the mode and minimum daily resting heart rate during a time segment. entropyrestinghr nats Shannon\u2019s entropy measurement based on daily resting heart rate during a time segment. sumcaloriesZONE cals The total daily calories burned within heartrate_zone during a time segment. maxcaloriesZONE cals The maximum daily calories burned within heartrate_zone during a time segment. mincaloriesZONE cals The minimum daily calories burned within heartrate_zone during a time segment. avgcaloriesZONE cals The average daily calories burned within heartrate_zone during a time segment. mediancaloriesZONE cals The median of daily calories burned within heartrate_zone during a time segment. stdcaloriesZONE cals The standard deviation of daily calories burned within heartrate_zone during a time segment. entropycaloriesZONE nats Shannon\u2019s entropy measurement based on daily calories burned within heartrate_zone during a time segment. Assumptions/Observations There are four heart rate zones (ZONE): outofrange , fatburn , cardio , and peak . Please refer to Fitbit documentation for more information about the way they are computed. Calories\u2019 accuracy depends on the users\u2019 Fitbit profile (weight, height, etc.).","title":"Fitbit Heart Rate Summary"},{"location":"features/fitbit-heartrate-summary/#fitbit-heart-rate-summary","text":"Sensor parameters description for [FITBIT_HEARTRATE_SUMMARY] : Key Description [TABLE] Database table name or file path where the heart rate summary data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-07\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:1200.6102,\u201dmax\u201d:88,\u201dmin\u201d:31,\u201dminutes\u201d:1058,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:760.3020,\u201dmax\u201d:120,\u201dmin\u201d:86,\u201dminutes\u201d:366,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:15.2048,\u201dmax\u201d:146,\u201dmin\u201d:120,\u201dminutes\u201d:2,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:221,\u201dmin\u201d:148,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:72}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:68},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:67},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:67},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-08\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:1100.1120,\u201dmax\u201d:89,\u201dmin\u201d:30,\u201dminutes\u201d:921,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:660.0012,\u201dmax\u201d:118,\u201dmin\u201d:82,\u201dminutes\u201d:361,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:23.7088,\u201dmax\u201d:142,\u201dmin\u201d:108,\u201dminutes\u201d:3,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:221,\u201dmin\u201d:148,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:70}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:77},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:75},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:73},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-09\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:750.3615,\u201dmax\u201d:77,\u201dmin\u201d:30,\u201dminutes\u201d:851,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:734.1516,\u201dmax\u201d:107,\u201dmin\u201d:77,\u201dminutes\u201d:550,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:131.8579,\u201dmax\u201d:130,\u201dmin\u201d:107,\u201dminutes\u201d:29,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:220,\u201dmin\u201d:130,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:69}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:90},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:89},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:88},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} PLAIN_TEXT device_id local_date_time heartrate_daily_restinghr heartrate_daily_caloriesoutofrange heartrate_daily_caloriesfatburn heartrate_daily_caloriescardio heartrate_daily_caloriespeak a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 72 1200.6102 760.3020 15.2048 0 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-08 70 1100.1120 660.0012 23.7088 0 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-09 69 750.3615 734.1516 131.8579 0","title":"Fitbit Heart Rate Summary"},{"location":"features/fitbit-heartrate-summary/#rapids-provider","text":"Available time segments Only available for segments that span 1 or more complete days (e.g. Jan 1 st 00:00 to Jan 3 rd 23:59) File Sequence - data/raw/ { pid } /fitbit_heartrate_summary_raw.csv - data/raw/ { pid } /fitbit_heartrate_summary_parsed.csv - data/raw/ { pid } /fitbit_heartrate_summary_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_heartrate_summary_features/fitbit_heartrate_summary_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_heartrate_summary.csv Parameters description for [FITBIT_HEARTRATE_SUMMARY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_HEARTRATE_SUMMARY features from the RAPIDS provider [FEATURES] Features to be computed from heart rate summary data, see table below Features description for [FITBIT_HEARTRATE_SUMMARY][PROVIDERS][RAPIDS] : Feature Units Description maxrestinghr beats/mins The maximum daily resting heart rate during a time segment. minrestinghr beats/mins The minimum daily resting heart rate during a time segment. avgrestinghr beats/mins The average daily resting heart rate during a time segment. medianrestinghr beats/mins The median of daily resting heart rate during a time segment. moderestinghr beats/mins The mode of daily resting heart rate during a time segment. stdrestinghr beats/mins The standard deviation of daily resting heart rate during a time segment. diffmaxmoderestinghr beats/mins The difference between the maximum and mode daily resting heart rate during a time segment. diffminmoderestinghr beats/mins The difference between the mode and minimum daily resting heart rate during a time segment. entropyrestinghr nats Shannon\u2019s entropy measurement based on daily resting heart rate during a time segment. sumcaloriesZONE cals The total daily calories burned within heartrate_zone during a time segment. maxcaloriesZONE cals The maximum daily calories burned within heartrate_zone during a time segment. mincaloriesZONE cals The minimum daily calories burned within heartrate_zone during a time segment. avgcaloriesZONE cals The average daily calories burned within heartrate_zone during a time segment. mediancaloriesZONE cals The median of daily calories burned within heartrate_zone during a time segment. stdcaloriesZONE cals The standard deviation of daily calories burned within heartrate_zone during a time segment. entropycaloriesZONE nats Shannon\u2019s entropy measurement based on daily calories burned within heartrate_zone during a time segment. Assumptions/Observations There are four heart rate zones (ZONE): outofrange , fatburn , cardio , and peak . Please refer to Fitbit documentation for more information about the way they are computed. Calories\u2019 accuracy depends on the users\u2019 Fitbit profile (weight, height, etc.).","title":"RAPIDS provider"},{"location":"features/fitbit-sleep-summary/","text":"Fitbit Sleep Summary \u00b6 Sensor parameters description for [FITBIT_SLEEP_SUMMARY] : Key Description [TABLE] Database table name or file path where the sleep summary data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data with Fitbit\u2019s sleep API Version 1 JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d: [{\u201cawakeCount\u201d: 2, \u201cawakeDuration\u201d: 3, \u201cawakeningsCount\u201d: 10, \u201cdateOfSleep\u201d: \u201c2020-10-07\u201d, \u201cduration\u201d: 8100000, \u201cefficiency\u201d: 91, \u201cendTime\u201d: \u201c2020-10-07T18:10:00.000\u201d, \u201cisMainSleep\u201d: true, \u201clogId\u201d: 14147921940, \u201cminuteData\u201d: [{\u201cdateTime\u201d: \u201c15:55:00\u201d, \u201cvalue\u201d: \u201c3\u201d}, {\u201cdateTime\u201d: \u201c15:56:00\u201d, \u201cvalue\u201d: \u201c3\u201d}, {\u201cdateTime\u201d: \u201c15:57:00\u201d, \u201cvalue\u201d: \u201c2\u201d},\u2026], \u201cminutesAfterWakeup\u201d: 0, \u201cminutesAsleep\u201d: 123, \u201cminutesAwake\u201d: 12, \u201cminutesToFallAsleep\u201d: 0, \u201crestlessCount\u201d: 8, \u201crestlessDuration\u201d: 9, \u201cstartTime\u201d: \u201c2020-10-07T15:55:00.000\u201d, \u201ctimeInBed\u201d: 135}, {\u201cawakeCount\u201d: 0, \u201cawakeDuration\u201d: 0, \u201cawakeningsCount\u201d: 1, \u201cdateOfSleep\u201d: \u201c2020-10-07\u201d, \u201cduration\u201d: 3780000, \u201cefficiency\u201d: 100, \u201cendTime\u201d: \u201c2020-10-07T10:52:30.000\u201d, \u201cisMainSleep\u201d: false, \u201clogId\u201d: 14144903977, \u201cminuteData\u201d: [{\u201cdateTime\u201d: \u201c09:49:00\u201d, \u201cvalue\u201d: \u201c1\u201d}, {\u201cdateTime\u201d: \u201c09:50:00\u201d, \u201cvalue\u201d: \u201c1\u201d}, {\u201cdateTime\u201d: \u201c09:51:00\u201d, \u201cvalue\u201d: \u201c1\u201d},\u2026], \u201cminutesAfterWakeup\u201d: 1, \u201cminutesAsleep\u201d: 62, \u201cminutesAwake\u201d: 0, \u201cminutesToFallAsleep\u201d: 0, \u201crestlessCount\u201d: 1, \u201crestlessDuration\u201d: 1, \u201cstartTime\u201d: \u201c2020-10-07T09:49:00.000\u201d, \u201ctimeInBed\u201d: 63}], \u201csummary\u201d: {\u201ctotalMinutesAsleep\u201d: 185, \u201ctotalSleepRecords\u201d: 2, \u201ctotalTimeInBed\u201d: 198}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d: [{\u201cawakeCount\u201d: 3, \u201cawakeDuration\u201d: 21, \u201cawakeningsCount\u201d: 16, \u201cdateOfSleep\u201d: \u201c2020-10-08\u201d, \u201cduration\u201d: 19260000, \u201cefficiency\u201d: 89, \u201cendTime\u201d: \u201c2020-10-08T06:01:30.000\u201d, \u201cisMainSleep\u201d: true, \u201clogId\u201d: 14150613895, \u201cminuteData\u201d: [{\u201cdateTime\u201d: \u201c00:40:00\u201d, \u201cvalue\u201d: \u201c3\u201d}, {\u201cdateTime\u201d: \u201c00:41:00\u201d, \u201cvalue\u201d: \u201c3\u201d}, {\u201cdateTime\u201d: \u201c00:42:00\u201d, \u201cvalue\u201d: \u201c3\u201d},\u2026], \u201cminutesAfterWakeup\u201d: 0, \u201cminutesAsleep\u201d: 275, \u201cminutesAwake\u201d: 33, \u201cminutesToFallAsleep\u201d: 0, \u201crestlessCount\u201d: 13, \u201crestlessDuration\u201d: 25, \u201cstartTime\u201d: \u201c2020-10-08T00:40:00.000\u201d, \u201ctimeInBed\u201d: 321}], \u201csummary\u201d: {\u201ctotalMinutesAsleep\u201d: 275, \u201ctotalSleepRecords\u201d: 1, \u201ctotalTimeInBed\u201d: 321}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d: [{\u201cawakeCount\u201d: 1, \u201cawakeDuration\u201d: 3, \u201cawakeningsCount\u201d: 8, \u201cdateOfSleep\u201d: \u201c2020-10-09\u201d, \u201cduration\u201d: 19320000, \u201cefficiency\u201d: 96, \u201cendTime\u201d: \u201c2020-10-09T05:57:30.000\u201d, \u201cisMainSleep\u201d: true, \u201clogId\u201d: 14161136803, \u201cminuteData\u201d: [{\u201cdateTime\u201d: \u201c00:35:30\u201d, \u201cvalue\u201d: \u201c2\u201d}, {\u201cdateTime\u201d: \u201c00:36:30\u201d, \u201cvalue\u201d: \u201c1\u201d}, {\u201cdateTime\u201d: \u201c00:37:30\u201d, \u201cvalue\u201d: \u201c1\u201d},\u2026], \u201cminutesAfterWakeup\u201d: 0, \u201cminutesAsleep\u201d: 309, \u201cminutesAwake\u201d: 13, \u201cminutesToFallAsleep\u201d: 0, \u201crestlessCount\u201d: 7, \u201crestlessDuration\u201d: 10, \u201cstartTime\u201d: \u201c2020-10-09T00:35:30.000\u201d, \u201ctimeInBed\u201d: 322}], \u201csummary\u201d: {\u201ctotalMinutesAsleep\u201d: 309, \u201ctotalSleepRecords\u201d: 1, \u201ctotalTimeInBed\u201d: 322}} PLAIN_TEXT device_id local_start_date_time local_end_date_time efficiency minutes_after_wakeup minutes_asleep minutes_awake minutes_to_fall_asleep minutes_in_bed is_main_sleep type count_awake duration_awake count_awakenings count_restless duration_restless a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 15:55:00 2020-10-07 18:10:00 91 0 123 12 0 135 1 classic 2 3 10 8 9 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 09:49:00 2020-10-07 10:52:30 100 1 62 0 0 63 0 classic 0 0 1 1 1 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-08 00:40:00 2020-10-08 06:01:30 89 0 275 33 0 321 1 classic 3 21 16 13 25 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-09 00:35:30 2020-10-09 05:57:30 96 0 309 13 0 322 1 classic 1 3 8 7 10 Example of the structure of source data with Fitbit\u2019s sleep API Version 1.2 JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d:[{\u201cdateOfSleep\u201d:\u201d2020-10-10\u201d,\u201dduration\u201d:3600000,\u201defficiency\u201d:92,\u201dendTime\u201d:\u201d2020-10-10T16:37:00.000\u201d,\u201dinfoCode\u201d:2,\u201disMainSleep\u201d:false,\u201dlevels\u201d:{\u201cdata\u201d:[{\u201cdateTime\u201d:\u201d2020-10-10T15:36:30.000\u201d,\u201dlevel\u201d:\u201drestless\u201d,\u201dseconds\u201d:60},{\u201cdateTime\u201d:\u201d2020-10-10T15:37:30.000\u201d,\u201dlevel\u201d:\u201dasleep\u201d,\u201dseconds\u201d:660},{\u201cdateTime\u201d:\u201d2020-10-10T15:48:30.000\u201d,\u201dlevel\u201d:\u201drestless\u201d,\u201dseconds\u201d:60},\u2026], \u201csummary\u201d:{\u201casleep\u201d:{\u201ccount\u201d:0,\u201dminutes\u201d:56},\u201dawake\u201d:{\u201ccount\u201d:0,\u201dminutes\u201d:0},\u201drestless\u201d:{\u201ccount\u201d:3,\u201dminutes\u201d:4}}},\u201dlogId\u201d:26315914306,\u201dminutesAfterWakeup\u201d:0,\u201dminutesAsleep\u201d:55,\u201dminutesAwake\u201d:5,\u201dminutesToFallAsleep\u201d:0,\u201dstartTime\u201d:\u201d2020-10-10T15:36:30.000\u201d,\u201dtimeInBed\u201d:60,\u201dtype\u201d:\u201dclassic\u201d},{\u201cdateOfSleep\u201d:\u201d2020-10-10\u201d,\u201dduration\u201d:22980000,\u201defficiency\u201d:88,\u201dendTime\u201d:\u201d2020-10-10T08:10:00.000\u201d,\u201dinfoCode\u201d:0,\u201disMainSleep\u201d:true,\u201dlevels\u201d:{\u201cdata\u201d:[{\u201cdateTime\u201d:\u201d2020-10-10T01:46:30.000\u201d,\u201dlevel\u201d:\u201dlight\u201d,\u201dseconds\u201d:420},{\u201cdateTime\u201d:\u201d2020-10-10T01:53:30.000\u201d,\u201dlevel\u201d:\u201ddeep\u201d,\u201dseconds\u201d:1230},{\u201cdateTime\u201d:\u201d2020-10-10T02:14:00.000\u201d,\u201dlevel\u201d:\u201dlight\u201d,\u201dseconds\u201d:360},\u2026], \u201csummary\u201d:{\u201cdeep\u201d:{\u201ccount\u201d:3,\u201dminutes\u201d:92,\u201dthirtyDayAvgMinutes\u201d:0},\u201dlight\u201d:{\u201ccount\u201d:29,\u201dminutes\u201d:193,\u201dthirtyDayAvgMinutes\u201d:0},\u201drem\u201d:{\u201ccount\u201d:4,\u201dminutes\u201d:33,\u201dthirtyDayAvgMinutes\u201d:0},\u201dwake\u201d:{\u201ccount\u201d:28,\u201dminutes\u201d:65,\u201dthirtyDayAvgMinutes\u201d:0}}},\u201dlogId\u201d:26311786557,\u201dminutesAfterWakeup\u201d:0,\u201dminutesAsleep\u201d:318,\u201dminutesAwake\u201d:65,\u201dminutesToFallAsleep\u201d:0,\u201dstartTime\u201d:\u201d2020-10-10T01:46:30.000\u201d,\u201dtimeInBed\u201d:383,\u201dtype\u201d:\u201dstages\u201d}],\u201dsummary\u201d:{\u201cstages\u201d:{\u201cdeep\u201d:92,\u201dlight\u201d:193,\u201drem\u201d:33,\u201dwake\u201d:65},\u201dtotalMinutesAsleep\u201d:373,\u201dtotalSleepRecords\u201d:2,\u201dtotalTimeInBed\u201d:443}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d:[{\u201cdateOfSleep\u201d:\u201d2020-10-11\u201d,\u201dduration\u201d:41640000,\u201defficiency\u201d:89,\u201dendTime\u201d:\u201d2020-10-11T11:47:00.000\u201d,\u201dinfoCode\u201d:0,\u201disMainSleep\u201d:true,\u201dlevels\u201d:{\u201cdata\u201d:[{\u201cdateTime\u201d:\u201d2020-10-11T00:12:30.000\u201d,\u201dlevel\u201d:\u201dwake\u201d,\u201dseconds\u201d:450},{\u201cdateTime\u201d:\u201d2020-10-11T00:20:00.000\u201d,\u201dlevel\u201d:\u201dlight\u201d,\u201dseconds\u201d:870},{\u201cdateTime\u201d:\u201d2020-10-11T00:34:30.000\u201d,\u201dlevel\u201d:\u201dwake\u201d,\u201dseconds\u201d:780},\u2026], \u201csummary\u201d:{\u201cdeep\u201d:{\u201ccount\u201d:4,\u201dminutes\u201d:52,\u201dthirtyDayAvgMinutes\u201d:62},\u201dlight\u201d:{\u201ccount\u201d:32,\u201dminutes\u201d:442,\u201dthirtyDayAvgMinutes\u201d:364},\u201drem\u201d:{\u201ccount\u201d:6,\u201dminutes\u201d:68,\u201dthirtyDayAvgMinutes\u201d:58},\u201dwake\u201d:{\u201ccount\u201d:29,\u201dminutes\u201d:132,\u201dthirtyDayAvgMinutes\u201d:94}}},\u201dlogId\u201d:26589710670,\u201dminutesAfterWakeup\u201d:1,\u201dminutesAsleep\u201d:562,\u201dminutesAwake\u201d:132,\u201dminutesToFallAsleep\u201d:0,\u201dstartTime\u201d:\u201d2020-10-11T00:12:30.000\u201d,\u201dtimeInBed\u201d:694,\u201dtype\u201d:\u201dstages\u201d}],\u201dsummary\u201d:{\u201cstages\u201d:{\u201cdeep\u201d:52,\u201dlight\u201d:442,\u201drem\u201d:68,\u201dwake\u201d:132},\u201dtotalMinutesAsleep\u201d:562,\u201dtotalSleepRecords\u201d:1,\u201dtotalTimeInBed\u201d:694}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d:[{\u201cdateOfSleep\u201d:\u201d2020-10-12\u201d,\u201dduration\u201d:28980000,\u201defficiency\u201d:93,\u201dendTime\u201d:\u201d2020-10-12T09:34:30.000\u201d,\u201dinfoCode\u201d:0,\u201disMainSleep\u201d:true,\u201dlevels\u201d:{\u201cdata\u201d:[{\u201cdateTime\u201d:\u201d2020-10-12T01:31:00.000\u201d,\u201dlevel\u201d:\u201dwake\u201d,\u201dseconds\u201d:600},{\u201cdateTime\u201d:\u201d2020-10-12T01:41:00.000\u201d,\u201dlevel\u201d:\u201dlight\u201d,\u201dseconds\u201d:60},{\u201cdateTime\u201d:\u201d2020-10-12T01:42:00.000\u201d,\u201dlevel\u201d:\u201ddeep\u201d,\u201dseconds\u201d:2340},\u2026], \u201csummary\u201d:{\u201cdeep\u201d:{\u201ccount\u201d:4,\u201dminutes\u201d:63,\u201dthirtyDayAvgMinutes\u201d:59},\u201dlight\u201d:{\u201ccount\u201d:27,\u201dminutes\u201d:257,\u201dthirtyDayAvgMinutes\u201d:364},\u201drem\u201d:{\u201ccount\u201d:5,\u201dminutes\u201d:94,\u201dthirtyDayAvgMinutes\u201d:58},\u201dwake\u201d:{\u201ccount\u201d:24,\u201dminutes\u201d:69,\u201dthirtyDayAvgMinutes\u201d:95}}},\u201dlogId\u201d:26589710673,\u201dminutesAfterWakeup\u201d:0,\u201dminutesAsleep\u201d:415,\u201dminutesAwake\u201d:68,\u201dminutesToFallAsleep\u201d:0,\u201dstartTime\u201d:\u201d2020-10-12T01:31:00.000\u201d,\u201dtimeInBed\u201d:483,\u201dtype\u201d:\u201dstages\u201d}],\u201dsummary\u201d:{\u201cstages\u201d:{\u201cdeep\u201d:63,\u201dlight\u201d:257,\u201drem\u201d:94,\u201dwake\u201d:69},\u201dtotalMinutesAsleep\u201d:415,\u201dtotalSleepRecords\u201d:1,\u201dtotalTimeInBed\u201d:483}} PLAIN_TEXT device_id local_start_date_time local_end_date_time efficiency minutes_after_wakeup minutes_asleep minutes_awake minutes_to_fall_asleep minutes_in_bed is_main_sleep type a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-10 15:36:30 2020-10-10 16:37:00 92 0 55 5 0 60 0 classic a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-10 01:46:30 2020-10-10 08:10:00 88 0 318 65 0 383 1 stages a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-11 00:12:30 2020-10-11 11:47:00 89 1 562 132 0 694 1 stages a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-12 01:31:00 2020-10-12 09:34:30 93 0 415 68 0 483 1 stages RAPIDS provider \u00b6 Available time segments Only available for segments that span 1 or more complete days (e.g. Jan 1 st 00:00 to Jan 3 rd 23:59) File Sequence - data/raw/ { pid } /fitbit_sleep_summary_raw.csv - data/raw/ { pid } /fitbit_sleep_summary_parsed.csv - data/raw/ { pid } /fitbit_sleep_summary_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_sleep_summary_features/fitbit_sleep_summary_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_sleep_summary.csv Parameters description for [FITBIT_SLEEP_SUMMARY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_SLEEP_SUMMARY features from the RAPIDS provider [SLEEP_TYPES] Types of sleep to be included in the feature extraction computation. Fitbit provides 3 types of sleep: main , nap , all . [FEATURES] Features to be computed from sleep summary data, see table below Features description for [FITBIT_SLEEP_SUMMARY][PROVIDERS][RAPIDS] : Feature Units Description countepisodeTYPE episodes Number of sleep episodes for a certain sleep type during a time segment. avgefficiencyTYPE scores Average sleep efficiency for a certain sleep type during a time segment. sumdurationafterwakeupTYPE minutes Total duration the user stayed in bed after waking up for a certain sleep type during a time segment. sumdurationasleepTYPE minutes Total sleep duration for a certain sleep type during a time segment. sumdurationawakeTYPE minutes Total duration the user stayed awake but still in bed for a certain sleep type during a time segment. sumdurationtofallasleepTYPE minutes Total duration the user spent to fall asleep for a certain sleep type during a time segment. sumdurationinbedTYPE minutes Total duration the user stayed in bed (sumdurationtofallasleep + sumdurationawake + sumdurationasleep + sumdurationafterwakeup) for a certain sleep type during a time segment. avgdurationafterwakeupTYPE minutes Average duration the user stayed in bed after waking up for a certain sleep type during a time segment. avgdurationasleepTYPE minutes Average sleep duration for a certain sleep type during a time segment. avgdurationawakeTYPE minutes Average duration the user stayed awake but still in bed for a certain sleep type during a time segment. avgdurationtofallasleepTYPE minutes Average duration the user spent to fall asleep for a certain sleep type during a time segment. avgdurationinbedTYPE minutes Average duration the user stayed in bed (sumdurationtofallasleep + sumdurationawake + sumdurationasleep + sumdurationafterwakeup) for a certain sleep type during a time segment. Assumptions/Observations There are three sleep types (TYPE): main , nap , all . The all type contains both main sleep and naps. There are two versions of Fitbit\u2019s sleep API ( version 1 and version 1.2 ), and each provides raw sleep data in a different format: Count & duration summaries . v1 contains count_awake , duration_awake , count_awakenings , count_restless , and duration_restless fields for every sleep record but v1.2 does not. API columns . Features are computed based on the values provided by Fitbit\u2019s API: efficiency , minutes_after_wakeup , minutes_asleep , minutes_awake , minutes_to_fall_asleep , minutes_in_bed , is_main_sleep and type .","title":"Fitbit Sleep Summary"},{"location":"features/fitbit-sleep-summary/#fitbit-sleep-summary","text":"Sensor parameters description for [FITBIT_SLEEP_SUMMARY] : Key Description [TABLE] Database table name or file path where the sleep summary data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data with Fitbit\u2019s sleep API Version 1 JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d: [{\u201cawakeCount\u201d: 2, \u201cawakeDuration\u201d: 3, \u201cawakeningsCount\u201d: 10, \u201cdateOfSleep\u201d: \u201c2020-10-07\u201d, \u201cduration\u201d: 8100000, \u201cefficiency\u201d: 91, \u201cendTime\u201d: \u201c2020-10-07T18:10:00.000\u201d, \u201cisMainSleep\u201d: true, \u201clogId\u201d: 14147921940, \u201cminuteData\u201d: [{\u201cdateTime\u201d: \u201c15:55:00\u201d, \u201cvalue\u201d: \u201c3\u201d}, {\u201cdateTime\u201d: \u201c15:56:00\u201d, \u201cvalue\u201d: \u201c3\u201d}, {\u201cdateTime\u201d: \u201c15:57:00\u201d, \u201cvalue\u201d: \u201c2\u201d},\u2026], \u201cminutesAfterWakeup\u201d: 0, \u201cminutesAsleep\u201d: 123, \u201cminutesAwake\u201d: 12, \u201cminutesToFallAsleep\u201d: 0, \u201crestlessCount\u201d: 8, \u201crestlessDuration\u201d: 9, \u201cstartTime\u201d: \u201c2020-10-07T15:55:00.000\u201d, \u201ctimeInBed\u201d: 135}, {\u201cawakeCount\u201d: 0, \u201cawakeDuration\u201d: 0, \u201cawakeningsCount\u201d: 1, \u201cdateOfSleep\u201d: \u201c2020-10-07\u201d, \u201cduration\u201d: 3780000, \u201cefficiency\u201d: 100, \u201cendTime\u201d: \u201c2020-10-07T10:52:30.000\u201d, \u201cisMainSleep\u201d: false, \u201clogId\u201d: 14144903977, \u201cminuteData\u201d: [{\u201cdateTime\u201d: \u201c09:49:00\u201d, \u201cvalue\u201d: \u201c1\u201d}, {\u201cdateTime\u201d: \u201c09:50:00\u201d, \u201cvalue\u201d: \u201c1\u201d}, {\u201cdateTime\u201d: \u201c09:51:00\u201d, \u201cvalue\u201d: \u201c1\u201d},\u2026], \u201cminutesAfterWakeup\u201d: 1, \u201cminutesAsleep\u201d: 62, \u201cminutesAwake\u201d: 0, \u201cminutesToFallAsleep\u201d: 0, \u201crestlessCount\u201d: 1, \u201crestlessDuration\u201d: 1, \u201cstartTime\u201d: \u201c2020-10-07T09:49:00.000\u201d, \u201ctimeInBed\u201d: 63}], \u201csummary\u201d: {\u201ctotalMinutesAsleep\u201d: 185, \u201ctotalSleepRecords\u201d: 2, \u201ctotalTimeInBed\u201d: 198}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d: [{\u201cawakeCount\u201d: 3, \u201cawakeDuration\u201d: 21, \u201cawakeningsCount\u201d: 16, \u201cdateOfSleep\u201d: \u201c2020-10-08\u201d, \u201cduration\u201d: 19260000, \u201cefficiency\u201d: 89, \u201cendTime\u201d: \u201c2020-10-08T06:01:30.000\u201d, \u201cisMainSleep\u201d: true, \u201clogId\u201d: 14150613895, \u201cminuteData\u201d: [{\u201cdateTime\u201d: \u201c00:40:00\u201d, \u201cvalue\u201d: \u201c3\u201d}, {\u201cdateTime\u201d: \u201c00:41:00\u201d, \u201cvalue\u201d: \u201c3\u201d}, {\u201cdateTime\u201d: \u201c00:42:00\u201d, \u201cvalue\u201d: \u201c3\u201d},\u2026], \u201cminutesAfterWakeup\u201d: 0, \u201cminutesAsleep\u201d: 275, \u201cminutesAwake\u201d: 33, \u201cminutesToFallAsleep\u201d: 0, \u201crestlessCount\u201d: 13, \u201crestlessDuration\u201d: 25, \u201cstartTime\u201d: \u201c2020-10-08T00:40:00.000\u201d, \u201ctimeInBed\u201d: 321}], \u201csummary\u201d: {\u201ctotalMinutesAsleep\u201d: 275, \u201ctotalSleepRecords\u201d: 1, \u201ctotalTimeInBed\u201d: 321}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d: [{\u201cawakeCount\u201d: 1, \u201cawakeDuration\u201d: 3, \u201cawakeningsCount\u201d: 8, \u201cdateOfSleep\u201d: \u201c2020-10-09\u201d, \u201cduration\u201d: 19320000, \u201cefficiency\u201d: 96, \u201cendTime\u201d: \u201c2020-10-09T05:57:30.000\u201d, \u201cisMainSleep\u201d: true, \u201clogId\u201d: 14161136803, \u201cminuteData\u201d: [{\u201cdateTime\u201d: \u201c00:35:30\u201d, \u201cvalue\u201d: \u201c2\u201d}, {\u201cdateTime\u201d: \u201c00:36:30\u201d, \u201cvalue\u201d: \u201c1\u201d}, {\u201cdateTime\u201d: \u201c00:37:30\u201d, \u201cvalue\u201d: \u201c1\u201d},\u2026], \u201cminutesAfterWakeup\u201d: 0, \u201cminutesAsleep\u201d: 309, \u201cminutesAwake\u201d: 13, \u201cminutesToFallAsleep\u201d: 0, \u201crestlessCount\u201d: 7, \u201crestlessDuration\u201d: 10, \u201cstartTime\u201d: \u201c2020-10-09T00:35:30.000\u201d, \u201ctimeInBed\u201d: 322}], \u201csummary\u201d: {\u201ctotalMinutesAsleep\u201d: 309, \u201ctotalSleepRecords\u201d: 1, \u201ctotalTimeInBed\u201d: 322}} PLAIN_TEXT device_id local_start_date_time local_end_date_time efficiency minutes_after_wakeup minutes_asleep minutes_awake minutes_to_fall_asleep minutes_in_bed is_main_sleep type count_awake duration_awake count_awakenings count_restless duration_restless a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 15:55:00 2020-10-07 18:10:00 91 0 123 12 0 135 1 classic 2 3 10 8 9 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 09:49:00 2020-10-07 10:52:30 100 1 62 0 0 63 0 classic 0 0 1 1 1 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-08 00:40:00 2020-10-08 06:01:30 89 0 275 33 0 321 1 classic 3 21 16 13 25 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-09 00:35:30 2020-10-09 05:57:30 96 0 309 13 0 322 1 classic 1 3 8 7 10 Example of the structure of source data with Fitbit\u2019s sleep API Version 1.2 JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d:[{\u201cdateOfSleep\u201d:\u201d2020-10-10\u201d,\u201dduration\u201d:3600000,\u201defficiency\u201d:92,\u201dendTime\u201d:\u201d2020-10-10T16:37:00.000\u201d,\u201dinfoCode\u201d:2,\u201disMainSleep\u201d:false,\u201dlevels\u201d:{\u201cdata\u201d:[{\u201cdateTime\u201d:\u201d2020-10-10T15:36:30.000\u201d,\u201dlevel\u201d:\u201drestless\u201d,\u201dseconds\u201d:60},{\u201cdateTime\u201d:\u201d2020-10-10T15:37:30.000\u201d,\u201dlevel\u201d:\u201dasleep\u201d,\u201dseconds\u201d:660},{\u201cdateTime\u201d:\u201d2020-10-10T15:48:30.000\u201d,\u201dlevel\u201d:\u201drestless\u201d,\u201dseconds\u201d:60},\u2026], \u201csummary\u201d:{\u201casleep\u201d:{\u201ccount\u201d:0,\u201dminutes\u201d:56},\u201dawake\u201d:{\u201ccount\u201d:0,\u201dminutes\u201d:0},\u201drestless\u201d:{\u201ccount\u201d:3,\u201dminutes\u201d:4}}},\u201dlogId\u201d:26315914306,\u201dminutesAfterWakeup\u201d:0,\u201dminutesAsleep\u201d:55,\u201dminutesAwake\u201d:5,\u201dminutesToFallAsleep\u201d:0,\u201dstartTime\u201d:\u201d2020-10-10T15:36:30.000\u201d,\u201dtimeInBed\u201d:60,\u201dtype\u201d:\u201dclassic\u201d},{\u201cdateOfSleep\u201d:\u201d2020-10-10\u201d,\u201dduration\u201d:22980000,\u201defficiency\u201d:88,\u201dendTime\u201d:\u201d2020-10-10T08:10:00.000\u201d,\u201dinfoCode\u201d:0,\u201disMainSleep\u201d:true,\u201dlevels\u201d:{\u201cdata\u201d:[{\u201cdateTime\u201d:\u201d2020-10-10T01:46:30.000\u201d,\u201dlevel\u201d:\u201dlight\u201d,\u201dseconds\u201d:420},{\u201cdateTime\u201d:\u201d2020-10-10T01:53:30.000\u201d,\u201dlevel\u201d:\u201ddeep\u201d,\u201dseconds\u201d:1230},{\u201cdateTime\u201d:\u201d2020-10-10T02:14:00.000\u201d,\u201dlevel\u201d:\u201dlight\u201d,\u201dseconds\u201d:360},\u2026], \u201csummary\u201d:{\u201cdeep\u201d:{\u201ccount\u201d:3,\u201dminutes\u201d:92,\u201dthirtyDayAvgMinutes\u201d:0},\u201dlight\u201d:{\u201ccount\u201d:29,\u201dminutes\u201d:193,\u201dthirtyDayAvgMinutes\u201d:0},\u201drem\u201d:{\u201ccount\u201d:4,\u201dminutes\u201d:33,\u201dthirtyDayAvgMinutes\u201d:0},\u201dwake\u201d:{\u201ccount\u201d:28,\u201dminutes\u201d:65,\u201dthirtyDayAvgMinutes\u201d:0}}},\u201dlogId\u201d:26311786557,\u201dminutesAfterWakeup\u201d:0,\u201dminutesAsleep\u201d:318,\u201dminutesAwake\u201d:65,\u201dminutesToFallAsleep\u201d:0,\u201dstartTime\u201d:\u201d2020-10-10T01:46:30.000\u201d,\u201dtimeInBed\u201d:383,\u201dtype\u201d:\u201dstages\u201d}],\u201dsummary\u201d:{\u201cstages\u201d:{\u201cdeep\u201d:92,\u201dlight\u201d:193,\u201drem\u201d:33,\u201dwake\u201d:65},\u201dtotalMinutesAsleep\u201d:373,\u201dtotalSleepRecords\u201d:2,\u201dtotalTimeInBed\u201d:443}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d:[{\u201cdateOfSleep\u201d:\u201d2020-10-11\u201d,\u201dduration\u201d:41640000,\u201defficiency\u201d:89,\u201dendTime\u201d:\u201d2020-10-11T11:47:00.000\u201d,\u201dinfoCode\u201d:0,\u201disMainSleep\u201d:true,\u201dlevels\u201d:{\u201cdata\u201d:[{\u201cdateTime\u201d:\u201d2020-10-11T00:12:30.000\u201d,\u201dlevel\u201d:\u201dwake\u201d,\u201dseconds\u201d:450},{\u201cdateTime\u201d:\u201d2020-10-11T00:20:00.000\u201d,\u201dlevel\u201d:\u201dlight\u201d,\u201dseconds\u201d:870},{\u201cdateTime\u201d:\u201d2020-10-11T00:34:30.000\u201d,\u201dlevel\u201d:\u201dwake\u201d,\u201dseconds\u201d:780},\u2026], \u201csummary\u201d:{\u201cdeep\u201d:{\u201ccount\u201d:4,\u201dminutes\u201d:52,\u201dthirtyDayAvgMinutes\u201d:62},\u201dlight\u201d:{\u201ccount\u201d:32,\u201dminutes\u201d:442,\u201dthirtyDayAvgMinutes\u201d:364},\u201drem\u201d:{\u201ccount\u201d:6,\u201dminutes\u201d:68,\u201dthirtyDayAvgMinutes\u201d:58},\u201dwake\u201d:{\u201ccount\u201d:29,\u201dminutes\u201d:132,\u201dthirtyDayAvgMinutes\u201d:94}}},\u201dlogId\u201d:26589710670,\u201dminutesAfterWakeup\u201d:1,\u201dminutesAsleep\u201d:562,\u201dminutesAwake\u201d:132,\u201dminutesToFallAsleep\u201d:0,\u201dstartTime\u201d:\u201d2020-10-11T00:12:30.000\u201d,\u201dtimeInBed\u201d:694,\u201dtype\u201d:\u201dstages\u201d}],\u201dsummary\u201d:{\u201cstages\u201d:{\u201cdeep\u201d:52,\u201dlight\u201d:442,\u201drem\u201d:68,\u201dwake\u201d:132},\u201dtotalMinutesAsleep\u201d:562,\u201dtotalSleepRecords\u201d:1,\u201dtotalTimeInBed\u201d:694}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d:[{\u201cdateOfSleep\u201d:\u201d2020-10-12\u201d,\u201dduration\u201d:28980000,\u201defficiency\u201d:93,\u201dendTime\u201d:\u201d2020-10-12T09:34:30.000\u201d,\u201dinfoCode\u201d:0,\u201disMainSleep\u201d:true,\u201dlevels\u201d:{\u201cdata\u201d:[{\u201cdateTime\u201d:\u201d2020-10-12T01:31:00.000\u201d,\u201dlevel\u201d:\u201dwake\u201d,\u201dseconds\u201d:600},{\u201cdateTime\u201d:\u201d2020-10-12T01:41:00.000\u201d,\u201dlevel\u201d:\u201dlight\u201d,\u201dseconds\u201d:60},{\u201cdateTime\u201d:\u201d2020-10-12T01:42:00.000\u201d,\u201dlevel\u201d:\u201ddeep\u201d,\u201dseconds\u201d:2340},\u2026], \u201csummary\u201d:{\u201cdeep\u201d:{\u201ccount\u201d:4,\u201dminutes\u201d:63,\u201dthirtyDayAvgMinutes\u201d:59},\u201dlight\u201d:{\u201ccount\u201d:27,\u201dminutes\u201d:257,\u201dthirtyDayAvgMinutes\u201d:364},\u201drem\u201d:{\u201ccount\u201d:5,\u201dminutes\u201d:94,\u201dthirtyDayAvgMinutes\u201d:58},\u201dwake\u201d:{\u201ccount\u201d:24,\u201dminutes\u201d:69,\u201dthirtyDayAvgMinutes\u201d:95}}},\u201dlogId\u201d:26589710673,\u201dminutesAfterWakeup\u201d:0,\u201dminutesAsleep\u201d:415,\u201dminutesAwake\u201d:68,\u201dminutesToFallAsleep\u201d:0,\u201dstartTime\u201d:\u201d2020-10-12T01:31:00.000\u201d,\u201dtimeInBed\u201d:483,\u201dtype\u201d:\u201dstages\u201d}],\u201dsummary\u201d:{\u201cstages\u201d:{\u201cdeep\u201d:63,\u201dlight\u201d:257,\u201drem\u201d:94,\u201dwake\u201d:69},\u201dtotalMinutesAsleep\u201d:415,\u201dtotalSleepRecords\u201d:1,\u201dtotalTimeInBed\u201d:483}} PLAIN_TEXT device_id local_start_date_time local_end_date_time efficiency minutes_after_wakeup minutes_asleep minutes_awake minutes_to_fall_asleep minutes_in_bed is_main_sleep type a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-10 15:36:30 2020-10-10 16:37:00 92 0 55 5 0 60 0 classic a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-10 01:46:30 2020-10-10 08:10:00 88 0 318 65 0 383 1 stages a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-11 00:12:30 2020-10-11 11:47:00 89 1 562 132 0 694 1 stages a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-12 01:31:00 2020-10-12 09:34:30 93 0 415 68 0 483 1 stages","title":"Fitbit Sleep Summary"},{"location":"features/fitbit-sleep-summary/#rapids-provider","text":"Available time segments Only available for segments that span 1 or more complete days (e.g. Jan 1 st 00:00 to Jan 3 rd 23:59) File Sequence - data/raw/ { pid } /fitbit_sleep_summary_raw.csv - data/raw/ { pid } /fitbit_sleep_summary_parsed.csv - data/raw/ { pid } /fitbit_sleep_summary_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_sleep_summary_features/fitbit_sleep_summary_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_sleep_summary.csv Parameters description for [FITBIT_SLEEP_SUMMARY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_SLEEP_SUMMARY features from the RAPIDS provider [SLEEP_TYPES] Types of sleep to be included in the feature extraction computation. Fitbit provides 3 types of sleep: main , nap , all . [FEATURES] Features to be computed from sleep summary data, see table below Features description for [FITBIT_SLEEP_SUMMARY][PROVIDERS][RAPIDS] : Feature Units Description countepisodeTYPE episodes Number of sleep episodes for a certain sleep type during a time segment. avgefficiencyTYPE scores Average sleep efficiency for a certain sleep type during a time segment. sumdurationafterwakeupTYPE minutes Total duration the user stayed in bed after waking up for a certain sleep type during a time segment. sumdurationasleepTYPE minutes Total sleep duration for a certain sleep type during a time segment. sumdurationawakeTYPE minutes Total duration the user stayed awake but still in bed for a certain sleep type during a time segment. sumdurationtofallasleepTYPE minutes Total duration the user spent to fall asleep for a certain sleep type during a time segment. sumdurationinbedTYPE minutes Total duration the user stayed in bed (sumdurationtofallasleep + sumdurationawake + sumdurationasleep + sumdurationafterwakeup) for a certain sleep type during a time segment. avgdurationafterwakeupTYPE minutes Average duration the user stayed in bed after waking up for a certain sleep type during a time segment. avgdurationasleepTYPE minutes Average sleep duration for a certain sleep type during a time segment. avgdurationawakeTYPE minutes Average duration the user stayed awake but still in bed for a certain sleep type during a time segment. avgdurationtofallasleepTYPE minutes Average duration the user spent to fall asleep for a certain sleep type during a time segment. avgdurationinbedTYPE minutes Average duration the user stayed in bed (sumdurationtofallasleep + sumdurationawake + sumdurationasleep + sumdurationafterwakeup) for a certain sleep type during a time segment. Assumptions/Observations There are three sleep types (TYPE): main , nap , all . The all type contains both main sleep and naps. There are two versions of Fitbit\u2019s sleep API ( version 1 and version 1.2 ), and each provides raw sleep data in a different format: Count & duration summaries . v1 contains count_awake , duration_awake , count_awakenings , count_restless , and duration_restless fields for every sleep record but v1.2 does not. API columns . Features are computed based on the values provided by Fitbit\u2019s API: efficiency , minutes_after_wakeup , minutes_asleep , minutes_awake , minutes_to_fall_asleep , minutes_in_bed , is_main_sleep and type .","title":"RAPIDS provider"},{"location":"features/fitbit-steps-intraday/","text":"Fitbit Steps Intraday \u00b6 Sensor parameters description for [FITBIT_STEPS_INTRADAY] : Key Description [TABLE] Database table name or file path where the steps intraday data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-07\u201d,\u201dvalue\u201d:\u201d1775\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:5},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:3},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:0},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-08\u201d,\u201dvalue\u201d:\u201d3201\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:14},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:11},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:10},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-09\u201d,\u201dvalue\u201d:\u201d998\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:0},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:0},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:0},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} PLAIN_TEXT device_id local_date_time steps a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:00:00 5 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:01:00 3 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:02:00 0 RAPIDS provider \u00b6 Available time segments Available for all time segments File Sequence - data/raw/ { pid } /fitbit_steps_intraday_raw.csv - data/raw/ { pid } /fitbit_steps_intraday_parsed.csv - data/raw/ { pid } /fitbit_steps_intraday_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_steps_intraday_features/fitbit_steps_intraday_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_steps_intraday.csv Parameters description for [FITBIT_STEPS_INTRADAY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_STEPS_INTRADAY features from the RAPIDS provider [FEATURES] Features to be computed from steps intraday data, see table below [THRESHOLD_ACTIVE_BOUT] Every minute with Fitbit steps data wil be labelled as sedentary if its step count is below this threshold, otherwise, active . [INCLUDE_ZERO_STEP_ROWS] Whether or not to include time segments with a 0 step count during the whole day. Features description for [FITBIT_STEPS_INTRADAY][PROVIDERS][RAPIDS] : Feature Units Description sumsteps steps The total step count during a time segment. maxsteps steps The maximum step count during a time segment. minsteps steps The minimum step count during a time segment. avgsteps steps The average step count during a time segment. stdsteps steps The standard deviation of step count during a time segment. countepisodesedentarybout bouts Number of sedentary bouts during a time segment. sumdurationsedentarybout minutes Total duration of all sedentary bouts during a time segment. maxdurationsedentarybout minutes The maximum duration of any sedentary bout during a time segment. mindurationsedentarybout minutes The minimum duration of any sedentary bout during a time segment. avgdurationsedentarybout minutes The average duration of sedentary bouts during a time segment. stddurationsedentarybout minutes The standard deviation of the duration of sedentary bouts during a time segment. countepisodeactivebout bouts Number of active bouts during a time segment. sumdurationactivebout minutes Total duration of all active bouts during a time segment. maxdurationactivebout minutes The maximum duration of any active bout during a time segment. mindurationactivebout minutes The minimum duration of any active bout during a time segment. avgdurationactivebout minutes The average duration of active bouts during a time segment. stddurationactivebout minutes The standard deviation of the duration of active bouts during a time segment. Assumptions/Observations Active and sedentary bouts . If the step count per minute is smaller than THRESHOLD_ACTIVE_BOUT (default value is 10), that minute is labelled as sedentary, otherwise, is labelled as active. Active and sedentary bouts are periods of consecutive minutes labelled as active or sedentary .","title":"Fitbit Steps Intraday"},{"location":"features/fitbit-steps-intraday/#fitbit-steps-intraday","text":"Sensor parameters description for [FITBIT_STEPS_INTRADAY] : Key Description [TABLE] Database table name or file path where the steps intraday data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-07\u201d,\u201dvalue\u201d:\u201d1775\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:5},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:3},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:0},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-08\u201d,\u201dvalue\u201d:\u201d3201\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:14},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:11},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:10},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-09\u201d,\u201dvalue\u201d:\u201d998\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:0},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:0},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:0},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} PLAIN_TEXT device_id local_date_time steps a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:00:00 5 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:01:00 3 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:02:00 0","title":"Fitbit Steps Intraday"},{"location":"features/fitbit-steps-intraday/#rapids-provider","text":"Available time segments Available for all time segments File Sequence - data/raw/ { pid } /fitbit_steps_intraday_raw.csv - data/raw/ { pid } /fitbit_steps_intraday_parsed.csv - data/raw/ { pid } /fitbit_steps_intraday_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_steps_intraday_features/fitbit_steps_intraday_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_steps_intraday.csv Parameters description for [FITBIT_STEPS_INTRADAY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_STEPS_INTRADAY features from the RAPIDS provider [FEATURES] Features to be computed from steps intraday data, see table below [THRESHOLD_ACTIVE_BOUT] Every minute with Fitbit steps data wil be labelled as sedentary if its step count is below this threshold, otherwise, active . [INCLUDE_ZERO_STEP_ROWS] Whether or not to include time segments with a 0 step count during the whole day. Features description for [FITBIT_STEPS_INTRADAY][PROVIDERS][RAPIDS] : Feature Units Description sumsteps steps The total step count during a time segment. maxsteps steps The maximum step count during a time segment. minsteps steps The minimum step count during a time segment. avgsteps steps The average step count during a time segment. stdsteps steps The standard deviation of step count during a time segment. countepisodesedentarybout bouts Number of sedentary bouts during a time segment. sumdurationsedentarybout minutes Total duration of all sedentary bouts during a time segment. maxdurationsedentarybout minutes The maximum duration of any sedentary bout during a time segment. mindurationsedentarybout minutes The minimum duration of any sedentary bout during a time segment. avgdurationsedentarybout minutes The average duration of sedentary bouts during a time segment. stddurationsedentarybout minutes The standard deviation of the duration of sedentary bouts during a time segment. countepisodeactivebout bouts Number of active bouts during a time segment. sumdurationactivebout minutes Total duration of all active bouts during a time segment. maxdurationactivebout minutes The maximum duration of any active bout during a time segment. mindurationactivebout minutes The minimum duration of any active bout during a time segment. avgdurationactivebout minutes The average duration of active bouts during a time segment. stddurationactivebout minutes The standard deviation of the duration of active bouts during a time segment. Assumptions/Observations Active and sedentary bouts . If the step count per minute is smaller than THRESHOLD_ACTIVE_BOUT (default value is 10), that minute is labelled as sedentary, otherwise, is labelled as active. Active and sedentary bouts are periods of consecutive minutes labelled as active or sedentary .","title":"RAPIDS provider"},{"location":"features/fitbit-steps-summary/","text":"Fitbit Steps Summary \u00b6 Sensor parameters description for [FITBIT_STEPS_SUMMARY] : Key Description [TABLE] Database table name or file path where the steps summary data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-07\u201d,\u201dvalue\u201d:\u201d1775\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:5},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:3},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:0},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-08\u201d,\u201dvalue\u201d:\u201d3201\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:14},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:11},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:10},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-09\u201d,\u201dvalue\u201d:\u201d998\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:0},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:0},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:0},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} PLAIN_TEXT device_id local_date_time steps a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 1775 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-08 3201 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-09 998 RAPIDS provider \u00b6 Available time segments Only available for segments that span 1 or more complete days (e.g. Jan 1 st 00:00 to Jan 3 rd 23:59) File Sequence - data/raw/ { pid } /fitbit_steps_summary_raw.csv - data/raw/ { pid } /fitbit_steps_summary_parsed.csv - data/raw/ { pid } /fitbit_steps_summary_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_steps_summary_features/fitbit_steps_summary_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_steps_summary.csv Parameters description for [FITBIT_STEPS_SUMMARY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_STEPS_SUMMARY features from the RAPIDS provider [FEATURES] Features to be computed from steps summary data, see table below Features description for [FITBIT_STEPS_SUMMARY][PROVIDERS][RAPIDS] : Feature Units Description maxsumsteps steps The maximum daily step count during a time segment. minsumsteps steps The minimum daily step count during a time segment. avgsumsteps steps The average daily step count during a time segment. mediansumsteps steps The median of daily step count during a time segment. stdsumsteps steps The standard deviation of daily step count during a time segment. Assumptions/Observations NA","title":"Fitbit Steps Summary"},{"location":"features/fitbit-steps-summary/#fitbit-steps-summary","text":"Sensor parameters description for [FITBIT_STEPS_SUMMARY] : Key Description [TABLE] Database table name or file path where the steps summary data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-07\u201d,\u201dvalue\u201d:\u201d1775\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:5},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:3},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:0},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-08\u201d,\u201dvalue\u201d:\u201d3201\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:14},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:11},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:10},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-09\u201d,\u201dvalue\u201d:\u201d998\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:0},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:0},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:0},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} PLAIN_TEXT device_id local_date_time steps a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 1775 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-08 3201 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-09 998","title":"Fitbit Steps Summary"},{"location":"features/fitbit-steps-summary/#rapids-provider","text":"Available time segments Only available for segments that span 1 or more complete days (e.g. Jan 1 st 00:00 to Jan 3 rd 23:59) File Sequence - data/raw/ { pid } /fitbit_steps_summary_raw.csv - data/raw/ { pid } /fitbit_steps_summary_parsed.csv - data/raw/ { pid } /fitbit_steps_summary_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_steps_summary_features/fitbit_steps_summary_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_steps_summary.csv Parameters description for [FITBIT_STEPS_SUMMARY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_STEPS_SUMMARY features from the RAPIDS provider [FEATURES] Features to be computed from steps summary data, see table below Features description for [FITBIT_STEPS_SUMMARY][PROVIDERS][RAPIDS] : Feature Units Description maxsumsteps steps The maximum daily step count during a time segment. minsumsteps steps The minimum daily step count during a time segment. avgsumsteps steps The average daily step count during a time segment. mediansumsteps steps The median of daily step count during a time segment. stdsumsteps steps The standard deviation of daily step count during a time segment. Assumptions/Observations NA","title":"RAPIDS provider"},{"location":"features/phone-accelerometer/","text":"Phone Accelerometer \u00b6 Sensor parameters description for [PHONE_ACCELEROMETER] : Key Description [TABLE] Database table where the accelerometer data is stored RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_accelerometer_raw.csv - data/raw/ { pid } /phone_accelerometer_with_datetime.csv - data/interim/ { pid } /phone_accelerometer_features/phone_accelerometer_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_accelerometer.csv Parameters description for [PHONE_ACCELEROMETER][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_ACCELEROMETER features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_ACCELEROMETER][PROVIDERS][RAPIDS] : Feature Units Description maxmagnitude m/s 2 The maximum magnitude of acceleration ( \\(\\|acceleration\\| = \\sqrt{x^2 + y^2 + z^2}\\) ). minmagnitude m/s 2 The minimum magnitude of acceleration. avgmagnitude m/s 2 The average magnitude of acceleration. medianmagnitude m/s 2 The median magnitude of acceleration. stdmagnitude m/s 2 The standard deviation of acceleration. Assumptions/Observations Analyzing accelerometer data is a memory intensive task. If RAPIDS crashes is likely because the accelerometer dataset for a participant is to big to fit in memory. We are considering different alternatives to overcome this problem. PANDA provider \u00b6 These features are based on the work by Panda et al . Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_accelerometer_raw.csv - data/raw/ { pid } /phone_accelerometer_with_datetime.csv - data/interim/ { pid } /phone_accelerometer_features/phone_accelerometer_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_accelerometer.csv Parameters description for [PHONE_ACCELEROMETER][PROVIDERS][PANDA] : Key Description [COMPUTE] Set to True to extract PHONE_ACCELEROMETER features from the PANDA provider [FEATURES] Features to be computed for exertional and non-exertional activity episodes, see table below Features description for [PHONE_ACCELEROMETER][PROVIDERS][PANDA] : Feature Units Description sumduration minutes Total duration of all exertional or non-exertional activity episodes. maxduration minutes Longest duration of any exertional or non-exertional activity episode. minduration minutes Shortest duration of any exertional or non-exertional activity episode. avgduration minutes Average duration of any exertional or non-exertional activity episode. medianduration minutes Median duration of any exertional or non-exertional activity episode. stdduration minutes Standard deviation of the duration of all exertional or non-exertional activity episodes. Assumptions/Observations Analyzing accelerometer data is a memory intensive task. If RAPIDS crashes is likely because the accelerometer dataset for a participant is to big to fit in memory. We are considering different alternatives to overcome this problem. See Panda et al for a definition of exertional and non-exertional activity episodes","title":"Phone Accelerometer"},{"location":"features/phone-accelerometer/#phone-accelerometer","text":"Sensor parameters description for [PHONE_ACCELEROMETER] : Key Description [TABLE] Database table where the accelerometer data is stored","title":"Phone Accelerometer"},{"location":"features/phone-accelerometer/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_accelerometer_raw.csv - data/raw/ { pid } /phone_accelerometer_with_datetime.csv - data/interim/ { pid } /phone_accelerometer_features/phone_accelerometer_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_accelerometer.csv Parameters description for [PHONE_ACCELEROMETER][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_ACCELEROMETER features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_ACCELEROMETER][PROVIDERS][RAPIDS] : Feature Units Description maxmagnitude m/s 2 The maximum magnitude of acceleration ( \\(\\|acceleration\\| = \\sqrt{x^2 + y^2 + z^2}\\) ). minmagnitude m/s 2 The minimum magnitude of acceleration. avgmagnitude m/s 2 The average magnitude of acceleration. medianmagnitude m/s 2 The median magnitude of acceleration. stdmagnitude m/s 2 The standard deviation of acceleration. Assumptions/Observations Analyzing accelerometer data is a memory intensive task. If RAPIDS crashes is likely because the accelerometer dataset for a participant is to big to fit in memory. We are considering different alternatives to overcome this problem.","title":"RAPIDS provider"},{"location":"features/phone-accelerometer/#panda-provider","text":"These features are based on the work by Panda et al . Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_accelerometer_raw.csv - data/raw/ { pid } /phone_accelerometer_with_datetime.csv - data/interim/ { pid } /phone_accelerometer_features/phone_accelerometer_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_accelerometer.csv Parameters description for [PHONE_ACCELEROMETER][PROVIDERS][PANDA] : Key Description [COMPUTE] Set to True to extract PHONE_ACCELEROMETER features from the PANDA provider [FEATURES] Features to be computed for exertional and non-exertional activity episodes, see table below Features description for [PHONE_ACCELEROMETER][PROVIDERS][PANDA] : Feature Units Description sumduration minutes Total duration of all exertional or non-exertional activity episodes. maxduration minutes Longest duration of any exertional or non-exertional activity episode. minduration minutes Shortest duration of any exertional or non-exertional activity episode. avgduration minutes Average duration of any exertional or non-exertional activity episode. medianduration minutes Median duration of any exertional or non-exertional activity episode. stdduration minutes Standard deviation of the duration of all exertional or non-exertional activity episodes. Assumptions/Observations Analyzing accelerometer data is a memory intensive task. If RAPIDS crashes is likely because the accelerometer dataset for a participant is to big to fit in memory. We are considering different alternatives to overcome this problem. See Panda et al for a definition of exertional and non-exertional activity episodes","title":"PANDA provider"},{"location":"features/phone-activity-recognition/","text":"Phone Activity Recognition \u00b6 Sensor parameters description for [PHONE_ACTIVITY_RECOGNITION] : Key Description [TABLE][ANDROID] Database table where the activity data from Android devices is stored (the AWARE client saves this data on different tables for Android and iOS) [TABLE][IOS] Database table where the activity data from iOS devices is stored (the AWARE client saves this data on different tables for Android and iOS) [EPISODE_THRESHOLD_BETWEEN_ROWS] Difference in minutes between any two rows for them to be considered part of the same activity episode RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_activity_recognition_raw.csv - data/raw/ { pid } /phone_activity_recognition_with_datetime.csv - data/raw/ { pid } /phone_activity_recognition_with_datetime_unified.csv - data/interim/ { pid } /phone_activity_recognition_episodes.csv - data/interim/ { pid } /phone_activity_recognition_episodes_resampled.csv - data/interim/ { pid } /phone_activity_recognition_episodes_resampled_with_datetime.csv - data/interim/ { pid } /phone_activity_recognition_features/phone_activity_recognition_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_activity_recognition.csv Parameters description for [PHONE_ACTIVITY_RECOGNITION][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_ACTIVITY_RECOGNITION features from the RAPIDS provider [FEATURES] Features to be computed, see table below [ACTIVITY_CLASSES][STATIONARY] An array of the activity labels to be considered in the STATIONARY category choose any of still , tilting [ACTIVITY_CLASSES][MOBILE] An array of the activity labels to be considered in the MOBILE category choose any of on_foot , walking , running , on_bicycle [ACTIVITY_CLASSES][VEHICLE] An array of the activity labels to be considered in the VEHICLE category choose any of in_vehicule Features description for [PHONE_ACTIVITY_RECOGNITION][PROVIDERS][RAPIDS] : Feature Units Description count rows Number of episodes. mostcommonactivity activity type The most common activity type (e.g. still , on_foot , etc.). If there is a tie, the first one is chosen. countuniqueactivities activity type Number of unique activities. durationstationary minutes The total duration of [ACTIVITY_CLASSES][STATIONARY] episodes durationmobile minutes The total duration of [ACTIVITY_CLASSES][MOBILE] episodes of on foot, running, and on bicycle activities durationvehicle minutes The total duration of [ACTIVITY_CLASSES][VEHICLE] episodes of on vehicle activity Assumptions/Observations iOS Activity Recognition names and types are unified with Android labels: iOS Activity Name Android Activity Name Android Activity Type walking walking 7 running running 8 cycling on_bicycle 1 automotive in_vehicle 0 stationary still 3 unknown unknown 4 In AWARE, Activity Recognition data for Android and iOS are stored in two different database tables, RAPIDS automatically infers what platform each participant belongs to based on their participant file .","title":"Phone Activity Recognition"},{"location":"features/phone-activity-recognition/#phone-activity-recognition","text":"Sensor parameters description for [PHONE_ACTIVITY_RECOGNITION] : Key Description [TABLE][ANDROID] Database table where the activity data from Android devices is stored (the AWARE client saves this data on different tables for Android and iOS) [TABLE][IOS] Database table where the activity data from iOS devices is stored (the AWARE client saves this data on different tables for Android and iOS) [EPISODE_THRESHOLD_BETWEEN_ROWS] Difference in minutes between any two rows for them to be considered part of the same activity episode","title":"Phone Activity Recognition"},{"location":"features/phone-activity-recognition/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_activity_recognition_raw.csv - data/raw/ { pid } /phone_activity_recognition_with_datetime.csv - data/raw/ { pid } /phone_activity_recognition_with_datetime_unified.csv - data/interim/ { pid } /phone_activity_recognition_episodes.csv - data/interim/ { pid } /phone_activity_recognition_episodes_resampled.csv - data/interim/ { pid } /phone_activity_recognition_episodes_resampled_with_datetime.csv - data/interim/ { pid } /phone_activity_recognition_features/phone_activity_recognition_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_activity_recognition.csv Parameters description for [PHONE_ACTIVITY_RECOGNITION][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_ACTIVITY_RECOGNITION features from the RAPIDS provider [FEATURES] Features to be computed, see table below [ACTIVITY_CLASSES][STATIONARY] An array of the activity labels to be considered in the STATIONARY category choose any of still , tilting [ACTIVITY_CLASSES][MOBILE] An array of the activity labels to be considered in the MOBILE category choose any of on_foot , walking , running , on_bicycle [ACTIVITY_CLASSES][VEHICLE] An array of the activity labels to be considered in the VEHICLE category choose any of in_vehicule Features description for [PHONE_ACTIVITY_RECOGNITION][PROVIDERS][RAPIDS] : Feature Units Description count rows Number of episodes. mostcommonactivity activity type The most common activity type (e.g. still , on_foot , etc.). If there is a tie, the first one is chosen. countuniqueactivities activity type Number of unique activities. durationstationary minutes The total duration of [ACTIVITY_CLASSES][STATIONARY] episodes durationmobile minutes The total duration of [ACTIVITY_CLASSES][MOBILE] episodes of on foot, running, and on bicycle activities durationvehicle minutes The total duration of [ACTIVITY_CLASSES][VEHICLE] episodes of on vehicle activity Assumptions/Observations iOS Activity Recognition names and types are unified with Android labels: iOS Activity Name Android Activity Name Android Activity Type walking walking 7 running running 8 cycling on_bicycle 1 automotive in_vehicle 0 stationary still 3 unknown unknown 4 In AWARE, Activity Recognition data for Android and iOS are stored in two different database tables, RAPIDS automatically infers what platform each participant belongs to based on their participant file .","title":"RAPIDS provider"},{"location":"features/phone-applications-foreground/","text":"Phone Applications Foreground \u00b6 Sensor parameters description for [PHONE_APPLICATIONS_FOREGROUND] (these parameters are used by the only provider available at the moment, RAPIDS): Key Description [TABLE] Database table where the applications foreground data is stored [APPLICATION_CATEGORIES][CATALOGUE_SOURCE] FILE or GOOGLE . If FILE , app categories (genres) are read from [CATALOGUE_FILE] . If [GOOGLE] , app categories (genres) are scrapped from the Play Store [APPLICATION_CATEGORIES][CATALOGUE_FILE] CSV file with a package_name and genre column. By default we provide the catalogue created by Stachl et al in data/external/stachl_application_genre_catalogue.csv [APPLICATION_CATEGORIES][UPDATE_CATALOGUE_FILE] if [CATALOGUE_SOURCE] is equal to FILE , this flag signals whether or not to update [CATALOGUE_FILE] , if [CATALOGUE_SOURCE] is equal to GOOGLE all scraped genres will be saved to [CATALOGUE_FILE] [APPLICATION_CATEGORIES][SCRAPE_MISSING_CATEGORIES] This flag signals whether or not to scrape categories (genres) missing from the [CATALOGUE_FILE] . If [CATALOGUE_SOURCE] is equal to GOOGLE , all genres are scraped anyway (this flag is ignored) RAPIDS provider \u00b6 The app category (genre) catalogue used in these features was originally created by Stachl et al . Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_applications_foreground_raw.csv - data/raw/ { pid } /phone_applications_foreground_with_datetime.csv - data/raw/ { pid } /phone_applications_foreground_with_datetime_with_categories.csv - data/interim/ { pid } /phone_applications_foreground_features/phone_applications_foreground_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_applications_foreground.csv Parameters description for [PHONE_APPLICATIONS_FOREGROUND][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_APPLICATIONS_FOREGROUND features from the RAPIDS provider [FEATURES] Features to be computed, see table below [SINGLE_CATEGORIES] An array of app categories to be included in the feature extraction computation. The special keyword all represents a category with all the apps from each participant. By default we use the category catalogue pointed by [APPLICATION_CATEGORIES][CATALOGUE_FILE] (see the Sensor parameters description table above) [MULTIPLE_CATEGORIES] An array of collections representing meta-categories (a group of categories). They key of each element is the name of the meta-category and the value is an array of member app categories. By default we use the category catalogue pointed by [APPLICATION_CATEGORIES][CATALOGUE_FILE] (see the Sensor parameters description table above) [SINGLE_APPS] An array of apps to be included in the feature extraction computation. Use their package name (e.g. com.google.android.youtube ) or the reserved keyword top1global (the most used app by a participant over the whole monitoring study) [EXCLUDED_CATEGORIES] An array of app categories to be excluded from the feature extraction computation. By default we use the category catalogue pointed by [APPLICATION_CATEGORIES][CATALOGUE_FILE] (see the Sensor parameters description table above) [EXCLUDED_APPS] An array of apps to be excluded from the feature extraction computation. Use their package name, for example: com.google.android.youtube Features description for [PHONE_APPLICATIONS_FOREGROUND][PROVIDERS][RAPIDS] : Feature Units Description count apps Number of times a single app or apps within a category were used (i.e. they were brought to the foreground either by tapping their icon or switching to it from another app) timeoffirstuse minutes The time in minutes between 12:00am (midnight) and the first use of a single app or apps within a category during a time_segment timeoflastuse minutes The time in minutes between 12:00am (midnight) and the last use of a single app or apps within a category during a time_segment frequencyentropy nats The entropy of the used apps within a category during a time_segment (each app is seen as a unique event, the more apps were used, the higher the entropy). This is especially relevant when computed over all apps. Entropy cannot be obtained for a single app Assumptions/Observations Features can be computed by app, by apps grouped under a single category (genre) and by multiple categories grouped together (meta-categories). For example, we can get features for Facebook (single app), for Social Network apps (a category including Facebook and other social media apps) or for Social (a meta-category formed by Social Network and Social Media Tools categories). Apps installed by default like YouTube are considered systems apps on some phones. We do an exact match to exclude apps where \u201cgenre\u201d == EXCLUDED_CATEGORIES or \u201cpackage_name\u201d == EXCLUDED_APPS . We provide three ways of classifying and app within a category (genre): a) by automatically scraping its official category from the Google Play Store, b) by using the catalogue created by Stachl et al. which we provide in RAPIDS ( data/external/stachl_application_genre_catalogue.csv ), or c) by manually creating a personalized catalogue. You can choose a, b or c by modifying [APPLICATION_GENRES] keys and values (see the Sensor parameters description table above).","title":"Phone Applications Foreground"},{"location":"features/phone-applications-foreground/#phone-applications-foreground","text":"Sensor parameters description for [PHONE_APPLICATIONS_FOREGROUND] (these parameters are used by the only provider available at the moment, RAPIDS): Key Description [TABLE] Database table where the applications foreground data is stored [APPLICATION_CATEGORIES][CATALOGUE_SOURCE] FILE or GOOGLE . If FILE , app categories (genres) are read from [CATALOGUE_FILE] . If [GOOGLE] , app categories (genres) are scrapped from the Play Store [APPLICATION_CATEGORIES][CATALOGUE_FILE] CSV file with a package_name and genre column. By default we provide the catalogue created by Stachl et al in data/external/stachl_application_genre_catalogue.csv [APPLICATION_CATEGORIES][UPDATE_CATALOGUE_FILE] if [CATALOGUE_SOURCE] is equal to FILE , this flag signals whether or not to update [CATALOGUE_FILE] , if [CATALOGUE_SOURCE] is equal to GOOGLE all scraped genres will be saved to [CATALOGUE_FILE] [APPLICATION_CATEGORIES][SCRAPE_MISSING_CATEGORIES] This flag signals whether or not to scrape categories (genres) missing from the [CATALOGUE_FILE] . If [CATALOGUE_SOURCE] is equal to GOOGLE , all genres are scraped anyway (this flag is ignored)","title":"Phone Applications Foreground"},{"location":"features/phone-applications-foreground/#rapids-provider","text":"The app category (genre) catalogue used in these features was originally created by Stachl et al . Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_applications_foreground_raw.csv - data/raw/ { pid } /phone_applications_foreground_with_datetime.csv - data/raw/ { pid } /phone_applications_foreground_with_datetime_with_categories.csv - data/interim/ { pid } /phone_applications_foreground_features/phone_applications_foreground_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_applications_foreground.csv Parameters description for [PHONE_APPLICATIONS_FOREGROUND][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_APPLICATIONS_FOREGROUND features from the RAPIDS provider [FEATURES] Features to be computed, see table below [SINGLE_CATEGORIES] An array of app categories to be included in the feature extraction computation. The special keyword all represents a category with all the apps from each participant. By default we use the category catalogue pointed by [APPLICATION_CATEGORIES][CATALOGUE_FILE] (see the Sensor parameters description table above) [MULTIPLE_CATEGORIES] An array of collections representing meta-categories (a group of categories). They key of each element is the name of the meta-category and the value is an array of member app categories. By default we use the category catalogue pointed by [APPLICATION_CATEGORIES][CATALOGUE_FILE] (see the Sensor parameters description table above) [SINGLE_APPS] An array of apps to be included in the feature extraction computation. Use their package name (e.g. com.google.android.youtube ) or the reserved keyword top1global (the most used app by a participant over the whole monitoring study) [EXCLUDED_CATEGORIES] An array of app categories to be excluded from the feature extraction computation. By default we use the category catalogue pointed by [APPLICATION_CATEGORIES][CATALOGUE_FILE] (see the Sensor parameters description table above) [EXCLUDED_APPS] An array of apps to be excluded from the feature extraction computation. Use their package name, for example: com.google.android.youtube Features description for [PHONE_APPLICATIONS_FOREGROUND][PROVIDERS][RAPIDS] : Feature Units Description count apps Number of times a single app or apps within a category were used (i.e. they were brought to the foreground either by tapping their icon or switching to it from another app) timeoffirstuse minutes The time in minutes between 12:00am (midnight) and the first use of a single app or apps within a category during a time_segment timeoflastuse minutes The time in minutes between 12:00am (midnight) and the last use of a single app or apps within a category during a time_segment frequencyentropy nats The entropy of the used apps within a category during a time_segment (each app is seen as a unique event, the more apps were used, the higher the entropy). This is especially relevant when computed over all apps. Entropy cannot be obtained for a single app Assumptions/Observations Features can be computed by app, by apps grouped under a single category (genre) and by multiple categories grouped together (meta-categories). For example, we can get features for Facebook (single app), for Social Network apps (a category including Facebook and other social media apps) or for Social (a meta-category formed by Social Network and Social Media Tools categories). Apps installed by default like YouTube are considered systems apps on some phones. We do an exact match to exclude apps where \u201cgenre\u201d == EXCLUDED_CATEGORIES or \u201cpackage_name\u201d == EXCLUDED_APPS . We provide three ways of classifying and app within a category (genre): a) by automatically scraping its official category from the Google Play Store, b) by using the catalogue created by Stachl et al. which we provide in RAPIDS ( data/external/stachl_application_genre_catalogue.csv ), or c) by manually creating a personalized catalogue. You can choose a, b or c by modifying [APPLICATION_GENRES] keys and values (see the Sensor parameters description table above).","title":"RAPIDS provider"},{"location":"features/phone-battery/","text":"Phone Battery \u00b6 Sensor parameters description for [PHONE_BATTERY] : Key Description [TABLE] Database table where the battery data is stored [EPISODE_THRESHOLD_BETWEEN_ROWS] Difference in minutes between any two rows for them to be considered part of the same battery charge or discharge episode RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_battery_raw.csv - data/interim/ { pid } /phone_battery_episodes.csv - data/interim/ { pid } /phone_battery_episodes_resampled.csv - data/interim/ { pid } /phone_battery_episodes_resampled_with_datetime.csv - data/interim/ { pid } /phone_battery_features/phone_battery_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_battery.csv Parameters description for [PHONE_BATTERY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_BATTERY features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_BATTERY][PROVIDERS][RAPIDS] : Feature Units Description countdischarge episodes Number of discharging episodes. sumdurationdischarge minutes The total duration of all discharging episodes. countcharge episodes Number of battery charging episodes. sumdurationcharge minutes The total duration of all charging episodes. avgconsumptionrate episodes/minutes The average of all episodes\u2019 consumption rates. An episode\u2019s consumption rate is defined as the ratio between its battery delta and duration maxconsumptionrate episodes/minutes The highest of all episodes\u2019 consumption rates. An episode\u2019s consumption rate is defined as the ratio between its battery delta and duration Assumptions/Observations We convert battery data collected with iOS client v1 (autodetected because battery status 4 do not exist) to match Android battery format: we swap status 3 for 5 and 1 for 3 We group battery data into discharge or charge episodes considering any contiguous rows with consecutive reductions or increases of the battery level if they are logged within [EPISODE_THRESHOLD_BETWEEN_ROWS] minutes from each other.","title":"Phone Battery"},{"location":"features/phone-battery/#phone-battery","text":"Sensor parameters description for [PHONE_BATTERY] : Key Description [TABLE] Database table where the battery data is stored [EPISODE_THRESHOLD_BETWEEN_ROWS] Difference in minutes between any two rows for them to be considered part of the same battery charge or discharge episode","title":"Phone Battery"},{"location":"features/phone-battery/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_battery_raw.csv - data/interim/ { pid } /phone_battery_episodes.csv - data/interim/ { pid } /phone_battery_episodes_resampled.csv - data/interim/ { pid } /phone_battery_episodes_resampled_with_datetime.csv - data/interim/ { pid } /phone_battery_features/phone_battery_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_battery.csv Parameters description for [PHONE_BATTERY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_BATTERY features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_BATTERY][PROVIDERS][RAPIDS] : Feature Units Description countdischarge episodes Number of discharging episodes. sumdurationdischarge minutes The total duration of all discharging episodes. countcharge episodes Number of battery charging episodes. sumdurationcharge minutes The total duration of all charging episodes. avgconsumptionrate episodes/minutes The average of all episodes\u2019 consumption rates. An episode\u2019s consumption rate is defined as the ratio between its battery delta and duration maxconsumptionrate episodes/minutes The highest of all episodes\u2019 consumption rates. An episode\u2019s consumption rate is defined as the ratio between its battery delta and duration Assumptions/Observations We convert battery data collected with iOS client v1 (autodetected because battery status 4 do not exist) to match Android battery format: we swap status 3 for 5 and 1 for 3 We group battery data into discharge or charge episodes considering any contiguous rows with consecutive reductions or increases of the battery level if they are logged within [EPISODE_THRESHOLD_BETWEEN_ROWS] minutes from each other.","title":"RAPIDS provider"},{"location":"features/phone-bluetooth/","text":"Phone Bluetooth \u00b6 Sensor parameters description for [PHONE_BLUETOOTH] : Key Description [TABLE] Database table where the bluetooth data is stored RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_bluetooth_raw.csv - data/raw/ { pid } /phone_bluetooth_with_datetime.csv - data/interim/ { pid } /phone_bluetooth_features/phone_bluetooth_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_bluetooth.csv \" Parameters description for [PHONE_BLUETOOTH][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_BLUETOOTH features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_BLUETOOTH][PROVIDERS][RAPIDS] : Feature Units Description countscans devices Number of scanned devices during a time_segment , a device can be detected multiple times over time and these appearances are counted separately uniquedevices devices Number of unique devices during a time_segment as identified by their hardware ( bt_address ) address countscansmostuniquedevice scans Number of scans of the most scanned device during a time_segment across the whole monitoring period Assumptions/Observations NA","title":"Phone Bluetooth"},{"location":"features/phone-bluetooth/#phone-bluetooth","text":"Sensor parameters description for [PHONE_BLUETOOTH] : Key Description [TABLE] Database table where the bluetooth data is stored","title":"Phone Bluetooth"},{"location":"features/phone-bluetooth/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_bluetooth_raw.csv - data/raw/ { pid } /phone_bluetooth_with_datetime.csv - data/interim/ { pid } /phone_bluetooth_features/phone_bluetooth_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_bluetooth.csv \" Parameters description for [PHONE_BLUETOOTH][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_BLUETOOTH features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_BLUETOOTH][PROVIDERS][RAPIDS] : Feature Units Description countscans devices Number of scanned devices during a time_segment , a device can be detected multiple times over time and these appearances are counted separately uniquedevices devices Number of unique devices during a time_segment as identified by their hardware ( bt_address ) address countscansmostuniquedevice scans Number of scans of the most scanned device during a time_segment across the whole monitoring period Assumptions/Observations NA","title":"RAPIDS provider"},{"location":"features/phone-calls/","text":"Phone Calls \u00b6 Sensor parameters description for [PHONE_CALLS] : Key Description [TABLE] Database table where the calls data is stored RAPIDS Provider \u00b6 Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_calls_raw.csv - data/raw/ { pid } /phone_calls_with_datetime.csv - data/raw/ { pid } /phone_calls_with_datetime_unified.csv - data/interim/ { pid } /phone_calls_features/phone_calls_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_calls.csv Parameters description for [PHONE_CALLS][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_CALLS features from the RAPIDS provider [CALL_TYPES] The particular call_type that will be analyzed. The options for this parameter are incoming, outgoing or missed. [FEATURES] Features to be computed for outgoing , incoming , and missed calls. Note that the same features are available for both incoming and outgoing calls, while missed calls has its own set of features. See the tables below. Features description for [PHONE_CALLS][PROVIDERS][RAPIDS] incoming and outgoing calls: Feature Units Description count calls Number of calls of a particular call_type occurred during a particular time_segment . distinctcontacts contacts Number of distinct contacts that are associated with a particular call_type for a particular time_segment meanduration seconds The mean duration of all calls of a particular call_type during a particular time_segment . sumduration seconds The sum of the duration of all calls of a particular call_type during a particular time_segment . minduration seconds The duration of the shortest call of a particular call_type during a particular time_segment . maxduration seconds The duration of the longest call of a particular call_type during a particular time_segment . stdduration seconds The standard deviation of the duration of all the calls of a particular call_type during a particular time_segment . modeduration seconds The mode of the duration of all the calls of a particular call_type during a particular time_segment . entropyduration nats The estimate of the Shannon entropy for the the duration of all the calls of a particular call_type during a particular time_segment . timefirstcall minutes The time in minutes between 12:00am (midnight) and the first call of call_type . timelastcall minutes The time in minutes between 12:00am (midnight) and the last call of call_type . countmostfrequentcontact calls The number of calls of a particular call_type during a particular time_segment of the most frequent contact throughout the monitored period. Features description for [PHONE_CALLS][PROVIDERS][RAPIDS] missed calls: Feature Units Description count calls Number of missed calls that occurred during a particular time_segment . distinctcontacts contacts Number of distinct contacts that are associated with missed calls for a particular time_segment timefirstcall minutes The time in hours from 12:00am (Midnight) that the first missed call occurred. timelastcall minutes The time in hours from 12:00am (Midnight) that the last missed call occurred. countmostfrequentcontact calls The number of missed calls during a particular time_segment of the most frequent contact throughout the monitored period. Assumptions/Observations Traces for iOS calls are unique even for the same contact calling a participant more than once which renders countmostfrequentcontact meaningless and distinctcontacts equal to the total number of traces. [CALL_TYPES] and [FEATURES] keys in config.yaml need to match. For example, [CALL_TYPES] outgoing matches the [FEATURES] key outgoing iOS calls data is transformed to match Android calls data format. See our algorithm","title":"Phone Calls"},{"location":"features/phone-calls/#phone-calls","text":"Sensor parameters description for [PHONE_CALLS] : Key Description [TABLE] Database table where the calls data is stored","title":"Phone Calls"},{"location":"features/phone-calls/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_calls_raw.csv - data/raw/ { pid } /phone_calls_with_datetime.csv - data/raw/ { pid } /phone_calls_with_datetime_unified.csv - data/interim/ { pid } /phone_calls_features/phone_calls_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_calls.csv Parameters description for [PHONE_CALLS][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_CALLS features from the RAPIDS provider [CALL_TYPES] The particular call_type that will be analyzed. The options for this parameter are incoming, outgoing or missed. [FEATURES] Features to be computed for outgoing , incoming , and missed calls. Note that the same features are available for both incoming and outgoing calls, while missed calls has its own set of features. See the tables below. Features description for [PHONE_CALLS][PROVIDERS][RAPIDS] incoming and outgoing calls: Feature Units Description count calls Number of calls of a particular call_type occurred during a particular time_segment . distinctcontacts contacts Number of distinct contacts that are associated with a particular call_type for a particular time_segment meanduration seconds The mean duration of all calls of a particular call_type during a particular time_segment . sumduration seconds The sum of the duration of all calls of a particular call_type during a particular time_segment . minduration seconds The duration of the shortest call of a particular call_type during a particular time_segment . maxduration seconds The duration of the longest call of a particular call_type during a particular time_segment . stdduration seconds The standard deviation of the duration of all the calls of a particular call_type during a particular time_segment . modeduration seconds The mode of the duration of all the calls of a particular call_type during a particular time_segment . entropyduration nats The estimate of the Shannon entropy for the the duration of all the calls of a particular call_type during a particular time_segment . timefirstcall minutes The time in minutes between 12:00am (midnight) and the first call of call_type . timelastcall minutes The time in minutes between 12:00am (midnight) and the last call of call_type . countmostfrequentcontact calls The number of calls of a particular call_type during a particular time_segment of the most frequent contact throughout the monitored period. Features description for [PHONE_CALLS][PROVIDERS][RAPIDS] missed calls: Feature Units Description count calls Number of missed calls that occurred during a particular time_segment . distinctcontacts contacts Number of distinct contacts that are associated with missed calls for a particular time_segment timefirstcall minutes The time in hours from 12:00am (Midnight) that the first missed call occurred. timelastcall minutes The time in hours from 12:00am (Midnight) that the last missed call occurred. countmostfrequentcontact calls The number of missed calls during a particular time_segment of the most frequent contact throughout the monitored period. Assumptions/Observations Traces for iOS calls are unique even for the same contact calling a participant more than once which renders countmostfrequentcontact meaningless and distinctcontacts equal to the total number of traces. [CALL_TYPES] and [FEATURES] keys in config.yaml need to match. For example, [CALL_TYPES] outgoing matches the [FEATURES] key outgoing iOS calls data is transformed to match Android calls data format. See our algorithm","title":"RAPIDS Provider"},{"location":"features/phone-conversation/","text":"Phone Conversation \u00b6 Sensor parameters description for [PHONE_CONVERSATION] : Key Description [TABLE][ANDROID] Database table where the conversation data from Android devices is stored (the AWARE client saves this data on different tables for Android and iOS) [TABLE][IOS] Database table where the conversation data from iOS devices is stored (the AWARE client saves this data on different tables for Android and iOS) RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_conversation_raw.csv - data/raw/ { pid } /phone_conversation_with_datetime.csv - data/raw/ { pid } /phone_conversation_with_datetime_unified.csv - data/interim/ { pid } /phone_conversation_features/phone_conversation_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_conversation.csv Parameters description for [PHONE_CONVERSATION][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_CONVERSATION features from the RAPIDS provider [FEATURES] Features to be computed, see table below [RECORDING_MINUTES] Minutes the plugin was recording audio (default 1 min) [PAUSED_MINUTES] Minutes the plugin was NOT recording audio (default 3 min) Features description for [PHONE_CONVERSATION][PROVIDERS][RAPIDS] : Feature Units Description minutessilence minutes Minutes labeled as silence minutesnoise minutes Minutes labeled as noise minutesvoice minutes Minutes labeled as voice minutesunknown minutes Minutes labeled as unknown sumconversationduration minutes Total duration of all conversations maxconversationduration minutes Longest duration of all conversations minconversationduration minutes Shortest duration of all conversations avgconversationduration minutes Average duration of all conversations sdconversationduration minutes Standard Deviation of the duration of all conversations timefirstconversation minutes Minutes since midnight when the first conversation for a time segment was detected timelastconversation minutes Minutes since midnight when the last conversation for a time segment was detected noisesumenergy L2-norm Sum of all energy values when inference is noise noiseavgenergy L2-norm Average of all energy values when inference is noise noisesdenergy L2-norm Standard Deviation of all energy values when inference is noise noiseminenergy L2-norm Minimum of all energy values when inference is noise noisemaxenergy L2-norm Maximum of all energy values when inference is noise voicesumenergy L2-norm Sum of all energy values when inference is voice voiceavgenergy L2-norm Average of all energy values when inference is voice voicesdenergy L2-norm Standard Deviation of all energy values when inference is voice voiceminenergy L2-norm Minimum of all energy values when inference is voice voicemaxenergy L2-norm Maximum of all energy values when inference is voice silencesensedfraction - Ratio between minutessilence and the sum of (minutessilence, minutesnoise, minutesvoice, minutesunknown) noisesensedfraction - Ratio between minutesnoise and the sum of (minutessilence, minutesnoise, minutesvoice, minutesunknown) voicesensedfraction - Ratio between minutesvoice and the sum of (minutessilence, minutesnoise, minutesvoice, minutesunknown) unknownsensedfraction - Ratio between minutesunknown and the sum of (minutessilence, minutesnoise, minutesvoice, minutesunknown) silenceexpectedfraction - Ration between minutessilence and the number of minutes that in theory should have been sensed based on the record and pause cycle of the plugin (1440 / recordingMinutes+pausedMinutes) noiseexpectedfraction - Ration between minutesnoise and the number of minutes that in theory should have been sensed based on the record and pause cycle of the plugin (1440 / recordingMinutes+pausedMinutes) voiceexpectedfraction - Ration between minutesvoice and the number of minutes that in theory should have been sensed based on the record and pause cycle of the plugin (1440 / recordingMinutes+pausedMinutes) unknownexpectedfraction - Ration between minutesunknown and the number of minutes that in theory should have been sensed based on the record and pause cycle of the plugin (1440 / recordingMinutes+pausedMinutes) Assumptions/Observations The timestamp of conversation rows in iOS is in seconds so we convert it to milliseconds to match Android\u2019s format","title":"Phone Conversation"},{"location":"features/phone-conversation/#phone-conversation","text":"Sensor parameters description for [PHONE_CONVERSATION] : Key Description [TABLE][ANDROID] Database table where the conversation data from Android devices is stored (the AWARE client saves this data on different tables for Android and iOS) [TABLE][IOS] Database table where the conversation data from iOS devices is stored (the AWARE client saves this data on different tables for Android and iOS)","title":"Phone Conversation"},{"location":"features/phone-conversation/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_conversation_raw.csv - data/raw/ { pid } /phone_conversation_with_datetime.csv - data/raw/ { pid } /phone_conversation_with_datetime_unified.csv - data/interim/ { pid } /phone_conversation_features/phone_conversation_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_conversation.csv Parameters description for [PHONE_CONVERSATION][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_CONVERSATION features from the RAPIDS provider [FEATURES] Features to be computed, see table below [RECORDING_MINUTES] Minutes the plugin was recording audio (default 1 min) [PAUSED_MINUTES] Minutes the plugin was NOT recording audio (default 3 min) Features description for [PHONE_CONVERSATION][PROVIDERS][RAPIDS] : Feature Units Description minutessilence minutes Minutes labeled as silence minutesnoise minutes Minutes labeled as noise minutesvoice minutes Minutes labeled as voice minutesunknown minutes Minutes labeled as unknown sumconversationduration minutes Total duration of all conversations maxconversationduration minutes Longest duration of all conversations minconversationduration minutes Shortest duration of all conversations avgconversationduration minutes Average duration of all conversations sdconversationduration minutes Standard Deviation of the duration of all conversations timefirstconversation minutes Minutes since midnight when the first conversation for a time segment was detected timelastconversation minutes Minutes since midnight when the last conversation for a time segment was detected noisesumenergy L2-norm Sum of all energy values when inference is noise noiseavgenergy L2-norm Average of all energy values when inference is noise noisesdenergy L2-norm Standard Deviation of all energy values when inference is noise noiseminenergy L2-norm Minimum of all energy values when inference is noise noisemaxenergy L2-norm Maximum of all energy values when inference is noise voicesumenergy L2-norm Sum of all energy values when inference is voice voiceavgenergy L2-norm Average of all energy values when inference is voice voicesdenergy L2-norm Standard Deviation of all energy values when inference is voice voiceminenergy L2-norm Minimum of all energy values when inference is voice voicemaxenergy L2-norm Maximum of all energy values when inference is voice silencesensedfraction - Ratio between minutessilence and the sum of (minutessilence, minutesnoise, minutesvoice, minutesunknown) noisesensedfraction - Ratio between minutesnoise and the sum of (minutessilence, minutesnoise, minutesvoice, minutesunknown) voicesensedfraction - Ratio between minutesvoice and the sum of (minutessilence, minutesnoise, minutesvoice, minutesunknown) unknownsensedfraction - Ratio between minutesunknown and the sum of (minutessilence, minutesnoise, minutesvoice, minutesunknown) silenceexpectedfraction - Ration between minutessilence and the number of minutes that in theory should have been sensed based on the record and pause cycle of the plugin (1440 / recordingMinutes+pausedMinutes) noiseexpectedfraction - Ration between minutesnoise and the number of minutes that in theory should have been sensed based on the record and pause cycle of the plugin (1440 / recordingMinutes+pausedMinutes) voiceexpectedfraction - Ration between minutesvoice and the number of minutes that in theory should have been sensed based on the record and pause cycle of the plugin (1440 / recordingMinutes+pausedMinutes) unknownexpectedfraction - Ration between minutesunknown and the number of minutes that in theory should have been sensed based on the record and pause cycle of the plugin (1440 / recordingMinutes+pausedMinutes) Assumptions/Observations The timestamp of conversation rows in iOS is in seconds so we convert it to milliseconds to match Android\u2019s format","title":"RAPIDS provider"},{"location":"features/phone-data-yield/","text":"Phone Data Yield \u00b6 This is a combinatorial sensor which means that we use the data from multiple sensors to extract data yield features. Data yield features can be used to remove rows ( time segments ) that do not contain enough data. You should decide what is your \u201cenough\u201d threshold depending on the type of sensors you collected (frequency vs event based, e.g. acceleroemter vs calls), the length of your study, and the rates of missing data that your analysis could handle. Why is data yield important? Imagine that you want to extract PHONE_CALL features on daily segments ( 00:00 to 23:59 ). Let\u2019s say that on day 1 the phone logged 10 calls and 23 hours of data from other sensors and on day 2 the phone logged 10 calls and only 2 hours of data from other sensors. It\u2019s more likely that other calls were placed on the 22 hours of data that you didn\u2019t log on day 2 than on the 1 hour of data you didn\u2019t log on day 1, and so including day 2 in your analysis could bias your results. Sensor parameters description for [PHONE_DATA_YIELD] : Key Description [SENSORS] One or more phone sensor config keys (e.g. PHONE_MESSAGE ). The more keys you include the more accurately RAPIDS can approximate the time an smartphone was sensing data. The supported phone sensors you can include in this list are outlined below ( do NOT include Fitbit sensors ). Supported phone sensors for [PHONE_DATA_YIELD][SENSORS] PHONE_ACCELEROMETER PHONE_ACTIVITY_RECOGNITION PHONE_APPLICATIONS_FOREGROUND PHONE_BATTERY PHONE_BLUETOOTH PHONE_CALLS PHONE_CONVERSATION PHONE_MESSAGES PHONE_LIGHT PHONE_LOCATIONS PHONE_SCREEN PHONE_WIFI_VISIBLE PHONE_WIFI_CONNECTED RAPIDS provider \u00b6 Before explaining the data yield features, let\u2019s define the following relevant concepts: A valid minute is any 60 second window when any phone sensor logged at least 1 row of data A valid hour is any 60 minute window with at least X valid minutes. The X or threshold is given by [MINUTE_RATIO_THRESHOLD_FOR_VALID_YIELDED_HOURS] The timestamps of all sensors are concatenated and then grouped per time segment. Minute and hour windows are created from the beginning of each time segment instance and these windows are marked as valid based on the definitions above. The duration of each time segment is taken into account to compute the features described below. Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } / { sensor } _raw.csv # one for every [PHONE_DATA_YIELD][SENSORS] - data/interim/ { pid } /phone_yielded_timestamps.csv - data/interim/ { pid } /phone_yielded_timestamps_with_datetime.csv - data/interim/ { pid } /phone_data_yield_features/phone_data_yield_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_data_yield.csv Parameters description for [PHONE_DATA_YIELD][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_DATA_YIELD features from the RAPIDS provider [FEATURES] Features to be computed, see table below [MINUTE_RATIO_THRESHOLD_FOR_VALID_YIELDED_HOURS] The proportion [0.0 ,1.0] of valid minutes in a 60-minute window necessary to flag that window as valid. Features description for [PHONE_DATA_YIELD][PROVIDERS][RAPIDS] : Feature Units Description ratiovalidyieldedminutes rows The ratio between the number of valid minutes and the duration in minutes of a time segment. ratiovalidyieldedhours lux The ratio between the number of valid hours and the duration in hours of a time segment. If the time segment is shorter than 1 hour this feature will always be 1. Assumptions/Observations We recommend using ratiovalidyieldedminutes on time segments that are shorter than two or three hours and ratiovalidyieldedhours for longer segments. This is because relying on yielded minutes only can be misleading when a big chunk of those missing minutes are clustered together. For example, let\u2019s assume we are working with a 24-hour time segment that is missing 12 hours of data. Two extreme cases can occur: the 12 missing hours are from the beginning of the segment or 30 minutes could be missing from every hour (24 * 30 minutes = 12 hours). ratiovalidyieldedminutes would be 0.5 for both a and b (hinting the missing circumstances are similar). However, ratiovalidyieldedhours would be 0.5 for a and 1.0 for b if [MINUTE_RATIO_THRESHOLD_FOR_VALID_YIELDED_HOURS] is between [0.0 and 0.49] (hinting that the missing circumstances might be more favorable for b . In other words, sensed data for b is more evenly spread compared to a .","title":"Phone Data Yield"},{"location":"features/phone-data-yield/#phone-data-yield","text":"This is a combinatorial sensor which means that we use the data from multiple sensors to extract data yield features. Data yield features can be used to remove rows ( time segments ) that do not contain enough data. You should decide what is your \u201cenough\u201d threshold depending on the type of sensors you collected (frequency vs event based, e.g. acceleroemter vs calls), the length of your study, and the rates of missing data that your analysis could handle. Why is data yield important? Imagine that you want to extract PHONE_CALL features on daily segments ( 00:00 to 23:59 ). Let\u2019s say that on day 1 the phone logged 10 calls and 23 hours of data from other sensors and on day 2 the phone logged 10 calls and only 2 hours of data from other sensors. It\u2019s more likely that other calls were placed on the 22 hours of data that you didn\u2019t log on day 2 than on the 1 hour of data you didn\u2019t log on day 1, and so including day 2 in your analysis could bias your results. Sensor parameters description for [PHONE_DATA_YIELD] : Key Description [SENSORS] One or more phone sensor config keys (e.g. PHONE_MESSAGE ). The more keys you include the more accurately RAPIDS can approximate the time an smartphone was sensing data. The supported phone sensors you can include in this list are outlined below ( do NOT include Fitbit sensors ). Supported phone sensors for [PHONE_DATA_YIELD][SENSORS] PHONE_ACCELEROMETER PHONE_ACTIVITY_RECOGNITION PHONE_APPLICATIONS_FOREGROUND PHONE_BATTERY PHONE_BLUETOOTH PHONE_CALLS PHONE_CONVERSATION PHONE_MESSAGES PHONE_LIGHT PHONE_LOCATIONS PHONE_SCREEN PHONE_WIFI_VISIBLE PHONE_WIFI_CONNECTED","title":"Phone Data Yield"},{"location":"features/phone-data-yield/#rapids-provider","text":"Before explaining the data yield features, let\u2019s define the following relevant concepts: A valid minute is any 60 second window when any phone sensor logged at least 1 row of data A valid hour is any 60 minute window with at least X valid minutes. The X or threshold is given by [MINUTE_RATIO_THRESHOLD_FOR_VALID_YIELDED_HOURS] The timestamps of all sensors are concatenated and then grouped per time segment. Minute and hour windows are created from the beginning of each time segment instance and these windows are marked as valid based on the definitions above. The duration of each time segment is taken into account to compute the features described below. Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } / { sensor } _raw.csv # one for every [PHONE_DATA_YIELD][SENSORS] - data/interim/ { pid } /phone_yielded_timestamps.csv - data/interim/ { pid } /phone_yielded_timestamps_with_datetime.csv - data/interim/ { pid } /phone_data_yield_features/phone_data_yield_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_data_yield.csv Parameters description for [PHONE_DATA_YIELD][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_DATA_YIELD features from the RAPIDS provider [FEATURES] Features to be computed, see table below [MINUTE_RATIO_THRESHOLD_FOR_VALID_YIELDED_HOURS] The proportion [0.0 ,1.0] of valid minutes in a 60-minute window necessary to flag that window as valid. Features description for [PHONE_DATA_YIELD][PROVIDERS][RAPIDS] : Feature Units Description ratiovalidyieldedminutes rows The ratio between the number of valid minutes and the duration in minutes of a time segment. ratiovalidyieldedhours lux The ratio between the number of valid hours and the duration in hours of a time segment. If the time segment is shorter than 1 hour this feature will always be 1. Assumptions/Observations We recommend using ratiovalidyieldedminutes on time segments that are shorter than two or three hours and ratiovalidyieldedhours for longer segments. This is because relying on yielded minutes only can be misleading when a big chunk of those missing minutes are clustered together. For example, let\u2019s assume we are working with a 24-hour time segment that is missing 12 hours of data. Two extreme cases can occur: the 12 missing hours are from the beginning of the segment or 30 minutes could be missing from every hour (24 * 30 minutes = 12 hours). ratiovalidyieldedminutes would be 0.5 for both a and b (hinting the missing circumstances are similar). However, ratiovalidyieldedhours would be 0.5 for a and 1.0 for b if [MINUTE_RATIO_THRESHOLD_FOR_VALID_YIELDED_HOURS] is between [0.0 and 0.49] (hinting that the missing circumstances might be more favorable for b . In other words, sensed data for b is more evenly spread compared to a .","title":"RAPIDS provider"},{"location":"features/phone-light/","text":"Phone Light \u00b6 Sensor parameters description for [PHONE_LIGHT] : Key Description [TABLE] Database table where the light data is stored RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_light_raw.csv - data/raw/ { pid } /phone_light_with_datetime.csv - data/interim/ { pid } /phone_light_features/phone_light_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_light.csv Parameters description for [PHONE_LIGHT][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_LIGHT features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_LIGHT][PROVIDERS][RAPIDS] : Feature Units Description count rows Number light sensor rows recorded. maxlux lux The maximum ambient luminance. minlux lux The minimum ambient luminance. avglux lux The average ambient luminance. medianlux lux The median ambient luminance. stdlux lux The standard deviation of ambient luminance. Assumptions/Observations NA","title":"Phone Light"},{"location":"features/phone-light/#phone-light","text":"Sensor parameters description for [PHONE_LIGHT] : Key Description [TABLE] Database table where the light data is stored","title":"Phone Light"},{"location":"features/phone-light/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_light_raw.csv - data/raw/ { pid } /phone_light_with_datetime.csv - data/interim/ { pid } /phone_light_features/phone_light_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_light.csv Parameters description for [PHONE_LIGHT][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_LIGHT features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_LIGHT][PROVIDERS][RAPIDS] : Feature Units Description count rows Number light sensor rows recorded. maxlux lux The maximum ambient luminance. minlux lux The minimum ambient luminance. avglux lux The average ambient luminance. medianlux lux The median ambient luminance. stdlux lux The standard deviation of ambient luminance. Assumptions/Observations NA","title":"RAPIDS provider"},{"location":"features/phone-locations/","text":"Phone Locations \u00b6 Sensor parameters description for [PHONE_LOCATIONS] : Key Description [TABLE] Database table where the location data is stored [LOCATIONS_TO_USE] Type of location data to use, one of ALL , GPS or FUSED_RESAMPLED . This filter is based on the provider column of the AWARE locations table, ALL includes every row, GPS only includes rows where provider is gps, and FUSED_RESAMPLED only includes rows where provider is fused after being resampled. [FUSED_RESAMPLED_CONSECUTIVE_THRESHOLD] if FUSED_RESAMPLED is used, the original fused data has to be resampled, a location row will be resampled to the next valid timestamp (see the Assumptions/Observations below) only if the time difference between them is less or equal than this threshold (in minutes). [FUSED_RESAMPLED_TIME_SINCE_VALID_LOCATION] if FUSED_RESAMPLED is used, the original fused data has to be resampled, a location row will be resampled at most for this long (in minutes) Assumptions/Observations Types of location data to use AWARE Android and iOS clients can collect location coordinates through the phone's GPS, the network cellular towers around the phone or Google's fused location API. If you want to use only the GPS provider set [LOCATIONS_TO_USE] to GPS , if you want to use all providers (not recommended due to the difference in accuracy) set [LOCATIONS_TO_USE] to ALL , if your AWARE client was configured to use fused location only or want to focus only on this provider, set [LOCATIONS_TO_USE] to RESAMPLE_FUSED . RESAMPLE_FUSED takes the original fused location coordinates and replicates each pair forward in time as long as the phone was sensing data as indicated by the joined timestamps of [PHONE_DATA_YIELD][SENSORS] , this is done because Google's API only logs a new location coordinate pair when it is sufficiently different in time or space from the previous one. There are two parameters associated with resampling fused location. FUSED_RESAMPLED_CONSECUTIVE_THRESHOLD (in minutes, default 30) controls the maximum gap between any two coordinate pairs to replicate the last known pair (for example, participant A's phone did not collect data between 10.30am and 10:50am and between 11:05am and 11:40am, the last known coordinate pair will be replicated during the first period but not the second, in other words, we assume that we cannot longer guarantee the participant stayed at the last known location if the phone did not sense data for more than 30 minutes). FUSED_RESAMPLED_TIME_SINCE_VALID_LOCATION (in minutes, default 720 or 12 hours) stops the last known fused location from being replicated longer that this threshold even if the phone was sensing data continuously (for example, participant A went home at 9pm and their phone was sensing data without gaps until 11am the next morning, the last known location will only be replicated until 9am). If you have suggestions to modify or improve this resampling, let us know. BARNETT provider \u00b6 These features are based on the original open-source implementation by Barnett et al and some features created by Canzian et al . Available time segments and platforms Available only for segments that start at 00:00:00 and end at 23:59:59 of the same day (daily segments) Available for Android and iOS File Sequence - data/raw/ { pid } /phone_locations_raw.csv - data/interim/ { pid } /phone_locations_processed.csv - data/interim/ { pid } /phone_locations_processed_with_datetime.csv - data/interim/ { pid } /phone_locations_features/phone_locations_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_locations.csv Parameters description for [PHONE_LOCATIONS][PROVIDERS][BARNETT] : Key Description [COMPUTE] Set to True to extract PHONE_LOCATIONS features from the BARNETT provider [FEATURES] Features to be computed, see table below [ACCURACY_LIMIT] An integer in meters, any location rows with an accuracy higher than this will be dropped. This number means there\u2019s a 68% probability the true location is within this radius [TIMEZONE] Timezone where the location data was collected. By default points to the one defined in the Configuration [MINUTES_DATA_USED] Set to True to include an extra column in the final location feature file containing the number of minutes used to compute the features on each time segment. Use this for quality control purposes, the more data minutes exist for a period, the more reliable its features should be. For fused location, a single minute can contain more than one coordinate pair if the participant is moving fast enough. Features description for [PHONE_LOCATIONS][PROVIDERS][BARNETT] adapted from Beiwe Summary Statistics : Feature Units Description hometime minutes Time at home. Time spent at home in minutes. Home is the most visited significant location between 8 pm and 8 am including any pauses within a 200-meter radius. disttravelled meters Total distance travelled over a day (flights). rog meters The Radius of Gyration (rog) is a measure in meters of the area covered by a person over a day. A centroid is calculated for all the places (pauses) visited during a day and a weighted distance between all the places and that centroid is computed. The weights are proportional to the time spent in each place. maxdiam meters The maximum diameter is the largest distance between any two pauses. maxhomedist meters The maximum distance from home in meters. siglocsvisited locations The number of significant locations visited during the day. Significant locations are computed using k-means clustering over pauses found in the whole monitoring period. The number of clusters is found iterating k from 1 to 200 stopping until the centroids of two significant locations are within 400 meters of one another. avgflightlen meters Mean length of all flights. stdflightlen meters Standard deviation of the length of all flights. avgflightdur seconds Mean duration of all flights. stdflightdur seconds The standard deviation of the duration of all flights. probpause - The fraction of a day spent in a pause (as opposed to a flight) siglocentropy nats Shannon\u2019s entropy measurement based on the proportion of time spent at each significant location visited during a day. circdnrtn - A continuous metric quantifying a person\u2019s circadian routine that can take any value between 0 and 1, where 0 represents a daily routine completely different from any other sensed days and 1 a routine the same as every other sensed day. wkenddayrtn - Same as circdnrtn but computed separately for weekends and weekdays. Assumptions/Observations Barnett's et al features These features are based on a Pause-Flight model. A pause is defined as a mobiity trace (location pings) within a certain duration and distance (by default 300 seconds and 60 meters). A flight is any mobility trace between two pauses. Data is resampled and imputed before the features are computed. See Barnett et al for more information. In RAPIDS we only expose two parameters for these features (timezone and accuracy limit). You can change other parameters in src/features/phone_locations/barnett/library/MobilityFeatures.R . Significant Locations Significant locations are determined using K-means clustering on pauses longer than 10 minutes. The number of clusters (K) is increased until no two clusters are within 400 meters from each other. After this, pauses within a certain range of a cluster (200 meters by default) will count as a visit to that significant location. This description was adapted from the Supplementary Materials of Barnett et al . The Circadian Calculation For a detailed description of how this is calculated, see Canzian et al . DORYAB provider \u00b6 These features are based on the original implementation by Doryab et al. . Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_locations_raw.csv - data/interim/ { pid } /phone_locations_processed.csv - data/interim/ { pid } /phone_locations_processed_with_datetime.csv - data/interim/ { pid } /phone_locations_features/phone_locations_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_locations.csv Parameters description for [PHONE_LOCATIONS][PROVIDERS][BARNETT] : Key Description [COMPUTE] Set to True to extract PHONE_LOCATIONS features from the BARNETT provider [FEATURES] Features to be computed, see table below [DBSCAN_EPS] The maximum distance in meters between two samples for one to be considered as in the neighborhood of the other. This is not a maximum bound on the distances of points within a cluster. This is the most important DBSCAN parameter to choose appropriately for your data set and distance function. [DBSCAN_MINSAMPLES] The number of samples (or total weight) in a neighborhood for a point to be considered as a core point of a cluster. This includes the point itself. [THRESHOLD_STATIC] It is the threshold value in km/hr which labels a row as Static or Moving. [MAXIMUM_GAP_ALLOWED] The maximum gap (in seconds) allowed between any two consecutive rows for them to be considered part of the same displacement. If this threshold is too high, it can throw speed and distance calculations off for periods when the the phone was not sensing. [MINUTES_DATA_USED] Set to True to include an extra column in the final location feature file containing the number of minutes used to compute the features on each time segment. Use this for quality control purposes, the more data minutes exist for a period, the more reliable its features should be. For fused location, a single minute can contain more than one coordinate pair if the participant is moving fast enough. [SAMPLING_FREQUENCY] Expected time difference between any two location rows in minutes. If set to 0 , the sampling frequency will be inferred automatically as the median of all the differences between any two consecutive row timestamps (recommended if you are using FUSED_RESAMPLED data). This parameter impacts all the time calculations. Features description for [PHONE_LOCATIONS][PROVIDERS][BARNETT] : Feature Units Description locationvariance \\(meters^2\\) The sum of the variances of the latitude and longitude columns. loglocationvariance - Log of the sum of the variances of the latitude and longitude columns. totaldistance meters Total distance travelled in a time segment using the haversine formula. averagespeed km/hr Average speed in a time segment considering only the instances labeled as Moving. varspeed km/hr Speed variance in a time segment considering only the instances labeled as Moving. circadianmovement - \"It encodes the extent to which a person\u2019s location patterns follow a 24-hour circadian cycle.\" Doryab et al. . numberofsignificantplaces places Number of significant locations visited. It is calculated using the DBSCAN clustering algorithm which takes in EPS and MIN_SAMPLES as parameters to identify clusters. Each cluster is a significant place. numberlocationtransitions transitions Number of movements between any two clusters in a time segment. radiusgyration meters Quantifies the area covered by a participant timeattop1location minutes Time spent at the most significant location. timeattop2location minutes Time spent at the 2 nd most significant location. timeattop3location minutes Time spent at the 3 rd most significant location. movingtostaticratio - Ratio between the number of rows labeled Moving versus Static outlierstimepercent - Ratio between the number of rows that belong to non-significant clusters divided by the total number of rows in a time segment. maxlengthstayatclusters minutes Maximum time spent in a cluster (significant location). minlengthstayatclusters minutes Minimum time spent in a cluster (significant location). meanlengthstayatclusters minutes Average time spent in a cluster (significant location). stdlengthstayatclusters minutes Standard deviation of time spent in a cluster (significant location). locationentropy nats Shannon Entropy computed over the row count of each cluster (significant location), it will be higher the more rows belong to a cluster (i.e. the more time a participant spent at a significant location). normalizedlocationentropy nats Shannon Entropy computed over the row count of each cluster (significant location) divided by the number of clusters, it will be higher the more rows belong to a cluster (i.e. the more time a participant spent at a significant location). Assumptions/Observations Significant Locations Identified Significant locations are determined using DBSCAN clustering on locations that a patient visit over the course of the period of data collection. The Circadian Calculation For a detailed description of how this is calculated, see Canzian et al .","title":"Phone Locations"},{"location":"features/phone-locations/#phone-locations","text":"Sensor parameters description for [PHONE_LOCATIONS] : Key Description [TABLE] Database table where the location data is stored [LOCATIONS_TO_USE] Type of location data to use, one of ALL , GPS or FUSED_RESAMPLED . This filter is based on the provider column of the AWARE locations table, ALL includes every row, GPS only includes rows where provider is gps, and FUSED_RESAMPLED only includes rows where provider is fused after being resampled. [FUSED_RESAMPLED_CONSECUTIVE_THRESHOLD] if FUSED_RESAMPLED is used, the original fused data has to be resampled, a location row will be resampled to the next valid timestamp (see the Assumptions/Observations below) only if the time difference between them is less or equal than this threshold (in minutes). [FUSED_RESAMPLED_TIME_SINCE_VALID_LOCATION] if FUSED_RESAMPLED is used, the original fused data has to be resampled, a location row will be resampled at most for this long (in minutes) Assumptions/Observations Types of location data to use AWARE Android and iOS clients can collect location coordinates through the phone's GPS, the network cellular towers around the phone or Google's fused location API. If you want to use only the GPS provider set [LOCATIONS_TO_USE] to GPS , if you want to use all providers (not recommended due to the difference in accuracy) set [LOCATIONS_TO_USE] to ALL , if your AWARE client was configured to use fused location only or want to focus only on this provider, set [LOCATIONS_TO_USE] to RESAMPLE_FUSED . RESAMPLE_FUSED takes the original fused location coordinates and replicates each pair forward in time as long as the phone was sensing data as indicated by the joined timestamps of [PHONE_DATA_YIELD][SENSORS] , this is done because Google's API only logs a new location coordinate pair when it is sufficiently different in time or space from the previous one. There are two parameters associated with resampling fused location. FUSED_RESAMPLED_CONSECUTIVE_THRESHOLD (in minutes, default 30) controls the maximum gap between any two coordinate pairs to replicate the last known pair (for example, participant A's phone did not collect data between 10.30am and 10:50am and between 11:05am and 11:40am, the last known coordinate pair will be replicated during the first period but not the second, in other words, we assume that we cannot longer guarantee the participant stayed at the last known location if the phone did not sense data for more than 30 minutes). FUSED_RESAMPLED_TIME_SINCE_VALID_LOCATION (in minutes, default 720 or 12 hours) stops the last known fused location from being replicated longer that this threshold even if the phone was sensing data continuously (for example, participant A went home at 9pm and their phone was sensing data without gaps until 11am the next morning, the last known location will only be replicated until 9am). If you have suggestions to modify or improve this resampling, let us know.","title":"Phone Locations"},{"location":"features/phone-locations/#barnett-provider","text":"These features are based on the original open-source implementation by Barnett et al and some features created by Canzian et al . Available time segments and platforms Available only for segments that start at 00:00:00 and end at 23:59:59 of the same day (daily segments) Available for Android and iOS File Sequence - data/raw/ { pid } /phone_locations_raw.csv - data/interim/ { pid } /phone_locations_processed.csv - data/interim/ { pid } /phone_locations_processed_with_datetime.csv - data/interim/ { pid } /phone_locations_features/phone_locations_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_locations.csv Parameters description for [PHONE_LOCATIONS][PROVIDERS][BARNETT] : Key Description [COMPUTE] Set to True to extract PHONE_LOCATIONS features from the BARNETT provider [FEATURES] Features to be computed, see table below [ACCURACY_LIMIT] An integer in meters, any location rows with an accuracy higher than this will be dropped. This number means there\u2019s a 68% probability the true location is within this radius [TIMEZONE] Timezone where the location data was collected. By default points to the one defined in the Configuration [MINUTES_DATA_USED] Set to True to include an extra column in the final location feature file containing the number of minutes used to compute the features on each time segment. Use this for quality control purposes, the more data minutes exist for a period, the more reliable its features should be. For fused location, a single minute can contain more than one coordinate pair if the participant is moving fast enough. Features description for [PHONE_LOCATIONS][PROVIDERS][BARNETT] adapted from Beiwe Summary Statistics : Feature Units Description hometime minutes Time at home. Time spent at home in minutes. Home is the most visited significant location between 8 pm and 8 am including any pauses within a 200-meter radius. disttravelled meters Total distance travelled over a day (flights). rog meters The Radius of Gyration (rog) is a measure in meters of the area covered by a person over a day. A centroid is calculated for all the places (pauses) visited during a day and a weighted distance between all the places and that centroid is computed. The weights are proportional to the time spent in each place. maxdiam meters The maximum diameter is the largest distance between any two pauses. maxhomedist meters The maximum distance from home in meters. siglocsvisited locations The number of significant locations visited during the day. Significant locations are computed using k-means clustering over pauses found in the whole monitoring period. The number of clusters is found iterating k from 1 to 200 stopping until the centroids of two significant locations are within 400 meters of one another. avgflightlen meters Mean length of all flights. stdflightlen meters Standard deviation of the length of all flights. avgflightdur seconds Mean duration of all flights. stdflightdur seconds The standard deviation of the duration of all flights. probpause - The fraction of a day spent in a pause (as opposed to a flight) siglocentropy nats Shannon\u2019s entropy measurement based on the proportion of time spent at each significant location visited during a day. circdnrtn - A continuous metric quantifying a person\u2019s circadian routine that can take any value between 0 and 1, where 0 represents a daily routine completely different from any other sensed days and 1 a routine the same as every other sensed day. wkenddayrtn - Same as circdnrtn but computed separately for weekends and weekdays. Assumptions/Observations Barnett's et al features These features are based on a Pause-Flight model. A pause is defined as a mobiity trace (location pings) within a certain duration and distance (by default 300 seconds and 60 meters). A flight is any mobility trace between two pauses. Data is resampled and imputed before the features are computed. See Barnett et al for more information. In RAPIDS we only expose two parameters for these features (timezone and accuracy limit). You can change other parameters in src/features/phone_locations/barnett/library/MobilityFeatures.R . Significant Locations Significant locations are determined using K-means clustering on pauses longer than 10 minutes. The number of clusters (K) is increased until no two clusters are within 400 meters from each other. After this, pauses within a certain range of a cluster (200 meters by default) will count as a visit to that significant location. This description was adapted from the Supplementary Materials of Barnett et al . The Circadian Calculation For a detailed description of how this is calculated, see Canzian et al .","title":"BARNETT provider"},{"location":"features/phone-locations/#doryab-provider","text":"These features are based on the original implementation by Doryab et al. . Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_locations_raw.csv - data/interim/ { pid } /phone_locations_processed.csv - data/interim/ { pid } /phone_locations_processed_with_datetime.csv - data/interim/ { pid } /phone_locations_features/phone_locations_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_locations.csv Parameters description for [PHONE_LOCATIONS][PROVIDERS][BARNETT] : Key Description [COMPUTE] Set to True to extract PHONE_LOCATIONS features from the BARNETT provider [FEATURES] Features to be computed, see table below [DBSCAN_EPS] The maximum distance in meters between two samples for one to be considered as in the neighborhood of the other. This is not a maximum bound on the distances of points within a cluster. This is the most important DBSCAN parameter to choose appropriately for your data set and distance function. [DBSCAN_MINSAMPLES] The number of samples (or total weight) in a neighborhood for a point to be considered as a core point of a cluster. This includes the point itself. [THRESHOLD_STATIC] It is the threshold value in km/hr which labels a row as Static or Moving. [MAXIMUM_GAP_ALLOWED] The maximum gap (in seconds) allowed between any two consecutive rows for them to be considered part of the same displacement. If this threshold is too high, it can throw speed and distance calculations off for periods when the the phone was not sensing. [MINUTES_DATA_USED] Set to True to include an extra column in the final location feature file containing the number of minutes used to compute the features on each time segment. Use this for quality control purposes, the more data minutes exist for a period, the more reliable its features should be. For fused location, a single minute can contain more than one coordinate pair if the participant is moving fast enough. [SAMPLING_FREQUENCY] Expected time difference between any two location rows in minutes. If set to 0 , the sampling frequency will be inferred automatically as the median of all the differences between any two consecutive row timestamps (recommended if you are using FUSED_RESAMPLED data). This parameter impacts all the time calculations. Features description for [PHONE_LOCATIONS][PROVIDERS][BARNETT] : Feature Units Description locationvariance \\(meters^2\\) The sum of the variances of the latitude and longitude columns. loglocationvariance - Log of the sum of the variances of the latitude and longitude columns. totaldistance meters Total distance travelled in a time segment using the haversine formula. averagespeed km/hr Average speed in a time segment considering only the instances labeled as Moving. varspeed km/hr Speed variance in a time segment considering only the instances labeled as Moving. circadianmovement - \"It encodes the extent to which a person\u2019s location patterns follow a 24-hour circadian cycle.\" Doryab et al. . numberofsignificantplaces places Number of significant locations visited. It is calculated using the DBSCAN clustering algorithm which takes in EPS and MIN_SAMPLES as parameters to identify clusters. Each cluster is a significant place. numberlocationtransitions transitions Number of movements between any two clusters in a time segment. radiusgyration meters Quantifies the area covered by a participant timeattop1location minutes Time spent at the most significant location. timeattop2location minutes Time spent at the 2 nd most significant location. timeattop3location minutes Time spent at the 3 rd most significant location. movingtostaticratio - Ratio between the number of rows labeled Moving versus Static outlierstimepercent - Ratio between the number of rows that belong to non-significant clusters divided by the total number of rows in a time segment. maxlengthstayatclusters minutes Maximum time spent in a cluster (significant location). minlengthstayatclusters minutes Minimum time spent in a cluster (significant location). meanlengthstayatclusters minutes Average time spent in a cluster (significant location). stdlengthstayatclusters minutes Standard deviation of time spent in a cluster (significant location). locationentropy nats Shannon Entropy computed over the row count of each cluster (significant location), it will be higher the more rows belong to a cluster (i.e. the more time a participant spent at a significant location). normalizedlocationentropy nats Shannon Entropy computed over the row count of each cluster (significant location) divided by the number of clusters, it will be higher the more rows belong to a cluster (i.e. the more time a participant spent at a significant location). Assumptions/Observations Significant Locations Identified Significant locations are determined using DBSCAN clustering on locations that a patient visit over the course of the period of data collection. The Circadian Calculation For a detailed description of how this is calculated, see Canzian et al .","title":"DORYAB provider"},{"location":"features/phone-messages/","text":"Phone Messages \u00b6 Sensor parameters description for [PHONE_MESSAGES] : Key Description [TABLE] Database table where the messages data is stored RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_messages_raw.csv - data/raw/ { pid } /phone_messages_with_datetime.csv - data/interim/ { pid } /phone_messages_features/phone_messages_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_messages.csv Parameters description for [PHONE_MESSAGES][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_MESSAGES features from the RAPIDS provider [MESSAGES_TYPES] The messages_type that will be analyzed. The options for this parameter are received or sent . [FEATURES] Features to be computed, see table below for [MESSAGES_TYPES] received and sent Features description for [PHONE_MESSAGES][PROVIDERS][RAPIDS] : Feature Units Description count messages Number of messages of type messages_type that occurred during a particular time_segment . distinctcontacts contacts Number of distinct contacts that are associated with a particular messages_type during a particular time_segment . timefirstmessages minutes Number of minutes between 12:00am (midnight) and the first message of a particular messages_type during a particular time_segment . timelastmessages minutes Number of minutes between 12:00am (midnight) and the last message of a particular messages_type during a particular time_segment . countmostfrequentcontact messages Number of messages from the contact with the most messages of messages_type during a time_segment throughout the whole dataset of each participant. Assumptions/Observations [MESSAGES_TYPES] and [FEATURES] keys in config.yaml need to match. For example, [MESSAGES_TYPES] sent matches the [FEATURES] key sent","title":"Phone Messages"},{"location":"features/phone-messages/#phone-messages","text":"Sensor parameters description for [PHONE_MESSAGES] : Key Description [TABLE] Database table where the messages data is stored","title":"Phone Messages"},{"location":"features/phone-messages/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_messages_raw.csv - data/raw/ { pid } /phone_messages_with_datetime.csv - data/interim/ { pid } /phone_messages_features/phone_messages_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_messages.csv Parameters description for [PHONE_MESSAGES][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_MESSAGES features from the RAPIDS provider [MESSAGES_TYPES] The messages_type that will be analyzed. The options for this parameter are received or sent . [FEATURES] Features to be computed, see table below for [MESSAGES_TYPES] received and sent Features description for [PHONE_MESSAGES][PROVIDERS][RAPIDS] : Feature Units Description count messages Number of messages of type messages_type that occurred during a particular time_segment . distinctcontacts contacts Number of distinct contacts that are associated with a particular messages_type during a particular time_segment . timefirstmessages minutes Number of minutes between 12:00am (midnight) and the first message of a particular messages_type during a particular time_segment . timelastmessages minutes Number of minutes between 12:00am (midnight) and the last message of a particular messages_type during a particular time_segment . countmostfrequentcontact messages Number of messages from the contact with the most messages of messages_type during a time_segment throughout the whole dataset of each participant. Assumptions/Observations [MESSAGES_TYPES] and [FEATURES] keys in config.yaml need to match. For example, [MESSAGES_TYPES] sent matches the [FEATURES] key sent","title":"RAPIDS provider"},{"location":"features/phone-screen/","text":"Phone Screen \u00b6 Sensor parameters description for [PHONE_SCREEN] : Key Description [TABLE] Database table where the screen data is stored RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_screen_raw.csv - data/raw/ { pid } /phone_screen_with_datetime.csv - data/raw/ { pid } /phone_screen_with_datetime_unified.csv - data/interim/ { pid } /phone_screen_episodes.csv - data/interim/ { pid } /phone_screen_episodes_resampled.csv - data/interim/ { pid } /phone_screen_episodes_resampled_with_datetime.csv - data/interim/ { pid } /phone_screen_features/phone_screen_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_screen.csv Parameters description for [PHONE_SCREEN][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_SCREEN features from the RAPIDS provider [FEATURES] Features to be computed, see table below [REFERENCE_HOUR_FIRST_USE] The reference point from which firstuseafter is to be computed, default is midnight [IGNORE_EPISODES_SHORTER_THAN] Ignore episodes that are shorter than this threshold (minutes). Set to 0 to disable this filter. [IGNORE_EPISODES_LONGER_THAN] Ignore episodes that are longer than this threshold (minutes). Set to 0 to disable this filter. [EPISODE_TYPES] Currently we only support unlock episodes (from when the phone is unlocked until the screen is off) Features description for [PHONE_SCREEN][PROVIDERS][RAPIDS] : Feature Units Description sumduration minutes Total duration of all unlock episodes. maxduration minutes Longest duration of any unlock episode. minduration minutes Shortest duration of any unlock episode. avgduration minutes Average duration of all unlock episodes. stdduration minutes Standard deviation duration of all unlock episodes. countepisode episodes Number of all unlock episodes |firstuseafter |minutes |Minutes until the first unlock episode. Assumptions/Observations In Android, lock events can happen right after an off event, after a few seconds of an off event, or never happen depending on the phone's settings, therefore, an unlock episode is defined as the time between an unlock and a off event. In iOS, on and off events do not exist, so an unlock episode is defined as the time between an unlock and a lock event. Events in iOS are recorded reliably albeit some duplicated lock events within milliseconds from each other, so we only keep consecutive unlock/lock pairs. In Android you cand find multiple consecutive unlock or lock events, so we only keep consecutive unlock/off pairs. In our experiments these cases are less than 10% of the screen events collected and this happens because ACTION_SCREEN_OFF and ACTION_SCREEN_ON are sent when the device becomes non-interactive which may have nothing to do with the screen turning off . In addition to unlock/off episodes, in Android it is possible to measure the time spent on the lock screen before an unlock event as well as the total screen time (i.e. ON to OFF ) but these are not implemented at the moment. We transform iOS screen events to match Android\u2019s format, we replace lock episodes with off episodes (2 with 0) in iOS. However, as mentioned above this is still computing unlock to lock episodes.","title":"Phone Screen"},{"location":"features/phone-screen/#phone-screen","text":"Sensor parameters description for [PHONE_SCREEN] : Key Description [TABLE] Database table where the screen data is stored","title":"Phone Screen"},{"location":"features/phone-screen/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_screen_raw.csv - data/raw/ { pid } /phone_screen_with_datetime.csv - data/raw/ { pid } /phone_screen_with_datetime_unified.csv - data/interim/ { pid } /phone_screen_episodes.csv - data/interim/ { pid } /phone_screen_episodes_resampled.csv - data/interim/ { pid } /phone_screen_episodes_resampled_with_datetime.csv - data/interim/ { pid } /phone_screen_features/phone_screen_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_screen.csv Parameters description for [PHONE_SCREEN][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_SCREEN features from the RAPIDS provider [FEATURES] Features to be computed, see table below [REFERENCE_HOUR_FIRST_USE] The reference point from which firstuseafter is to be computed, default is midnight [IGNORE_EPISODES_SHORTER_THAN] Ignore episodes that are shorter than this threshold (minutes). Set to 0 to disable this filter. [IGNORE_EPISODES_LONGER_THAN] Ignore episodes that are longer than this threshold (minutes). Set to 0 to disable this filter. [EPISODE_TYPES] Currently we only support unlock episodes (from when the phone is unlocked until the screen is off) Features description for [PHONE_SCREEN][PROVIDERS][RAPIDS] : Feature Units Description sumduration minutes Total duration of all unlock episodes. maxduration minutes Longest duration of any unlock episode. minduration minutes Shortest duration of any unlock episode. avgduration minutes Average duration of all unlock episodes. stdduration minutes Standard deviation duration of all unlock episodes. countepisode episodes Number of all unlock episodes |firstuseafter |minutes |Minutes until the first unlock episode. Assumptions/Observations In Android, lock events can happen right after an off event, after a few seconds of an off event, or never happen depending on the phone's settings, therefore, an unlock episode is defined as the time between an unlock and a off event. In iOS, on and off events do not exist, so an unlock episode is defined as the time between an unlock and a lock event. Events in iOS are recorded reliably albeit some duplicated lock events within milliseconds from each other, so we only keep consecutive unlock/lock pairs. In Android you cand find multiple consecutive unlock or lock events, so we only keep consecutive unlock/off pairs. In our experiments these cases are less than 10% of the screen events collected and this happens because ACTION_SCREEN_OFF and ACTION_SCREEN_ON are sent when the device becomes non-interactive which may have nothing to do with the screen turning off . In addition to unlock/off episodes, in Android it is possible to measure the time spent on the lock screen before an unlock event as well as the total screen time (i.e. ON to OFF ) but these are not implemented at the moment. We transform iOS screen events to match Android\u2019s format, we replace lock episodes with off episodes (2 with 0) in iOS. However, as mentioned above this is still computing unlock to lock episodes.","title":"RAPIDS provider"},{"location":"features/phone-wifi-connected/","text":"Phone WiFi Connected \u00b6 Sensor parameters description for [PHONE_WIFI_CONNECTED] : Key Description [TABLE] Database table where the wifi (connected) data is stored RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_wifi_connected_raw.csv - data/raw/ { pid } /phone_wifi_connected_with_datetime.csv - data/interim/ { pid } /phone_wifi_connected_features/phone_wifi_connected_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_wifi_connected.csv Parameters description for [PHONE_WIFI_CONNECTED][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_WIFI_CONNECTED features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_WIFI_CONNECTED][PROVIDERS][RAPIDS] : Feature Units Description countscans devices Number of scanned WiFi access points connected during a time_segment, an access point can be detected multiple times over time and these appearances are counted separately uniquedevices devices Number of unique access point during a time_segment as identified by their hardware address countscansmostuniquedevice scans Number of scans of the most scanned access point during a time_segment across the whole monitoring period Assumptions/Observations A connected WiFI access point is one that a phone was connected to. By default AWARE stores this data in the sensor_wifi table.","title":"Phone WiFI Connected"},{"location":"features/phone-wifi-connected/#phone-wifi-connected","text":"Sensor parameters description for [PHONE_WIFI_CONNECTED] : Key Description [TABLE] Database table where the wifi (connected) data is stored","title":"Phone WiFi Connected"},{"location":"features/phone-wifi-connected/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_wifi_connected_raw.csv - data/raw/ { pid } /phone_wifi_connected_with_datetime.csv - data/interim/ { pid } /phone_wifi_connected_features/phone_wifi_connected_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_wifi_connected.csv Parameters description for [PHONE_WIFI_CONNECTED][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_WIFI_CONNECTED features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_WIFI_CONNECTED][PROVIDERS][RAPIDS] : Feature Units Description countscans devices Number of scanned WiFi access points connected during a time_segment, an access point can be detected multiple times over time and these appearances are counted separately uniquedevices devices Number of unique access point during a time_segment as identified by their hardware address countscansmostuniquedevice scans Number of scans of the most scanned access point during a time_segment across the whole monitoring period Assumptions/Observations A connected WiFI access point is one that a phone was connected to. By default AWARE stores this data in the sensor_wifi table.","title":"RAPIDS provider"},{"location":"features/phone-wifi-visible/","text":"Phone WiFi Visible \u00b6 Sensor parameters description for [PHONE_WIFI_VISIBLE] : Key Description [TABLE] Database table where the wifi (visible) data is stored RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_wifi_visible_raw.csv - data/raw/ { pid } /phone_wifi_visible_with_datetime.csv - data/interim/ { pid } /phone_wifi_visible_features/phone_wifi_visible_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_wifi_visible.csv Parameters description for [PHONE_WIFI_VISIBLE][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_WIFI_VISIBLE features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_WIFI_VISIBLE][PROVIDERS][RAPIDS] : Feature Units Description countscans devices Number of scanned WiFi access points visible during a time_segment, an access point can be detected multiple times over time and these appearances are counted separately uniquedevices devices Number of unique access point during a time_segment as identified by their hardware address countscansmostuniquedevice scans Number of scans of the most scanned access point during a time_segment across the whole monitoring period Assumptions/Observations A visible WiFI access point is one that a phone sensed around itself but that it was not connected to. Due to API restrictions, this sensor is not available on iOS. By default AWARE stores this data in the wifi table.","title":"Phone WiFI Visible"},{"location":"features/phone-wifi-visible/#phone-wifi-visible","text":"Sensor parameters description for [PHONE_WIFI_VISIBLE] : Key Description [TABLE] Database table where the wifi (visible) data is stored","title":"Phone WiFi Visible"},{"location":"features/phone-wifi-visible/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_wifi_visible_raw.csv - data/raw/ { pid } /phone_wifi_visible_with_datetime.csv - data/interim/ { pid } /phone_wifi_visible_features/phone_wifi_visible_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_wifi_visible.csv Parameters description for [PHONE_WIFI_VISIBLE][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_WIFI_VISIBLE features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_WIFI_VISIBLE][PROVIDERS][RAPIDS] : Feature Units Description countscans devices Number of scanned WiFi access points visible during a time_segment, an access point can be detected multiple times over time and these appearances are counted separately uniquedevices devices Number of unique access point during a time_segment as identified by their hardware address countscansmostuniquedevice scans Number of scans of the most scanned access point during a time_segment across the whole monitoring period Assumptions/Observations A visible WiFI access point is one that a phone sensed around itself but that it was not connected to. Due to API restrictions, this sensor is not available on iOS. By default AWARE stores this data in the wifi table.","title":"RAPIDS provider"},{"location":"setup/configuration/","text":"Configuration \u00b6 You need to follow these steps to configure your RAPIDS deployment before you can extract behavioral features Add your database credentials Choose the timezone of your study Create your participants files Select what time segments you want to extract features on Modify your device data source configuration Select what sensors and features you want to process When you are done with this configuration, go to executing RAPIDS . Hint Every time you see config[\"KEY\"] or [KEY] in these docs we are referring to the corresponding key in the config.yaml file. Database credentials \u00b6 Create an empty file called .env in your RAPIDS root directory Add the following lines and replace your database-specific credentials (user, password, host, and database): [ MY_GROUP ] user=MY_USER password=MY_PASSWORD host=MY_HOST port=3306 database=MY_DATABASE Warning The label MY_GROUP is arbitrary but it has to match the following config.yaml key: DATABASE_GROUP : &database_group MY_GROUP Note You can ignore this step if you are only processing Fitbit data in CSV files. Timezone of your study \u00b6 Single timezone \u00b6 If your study only happened in a single time zone, select the appropriate code form this list and change the following config key. Double check your timezone code pick, for example US Eastern Time is America/New_York not EST TIMEZONE : &timezone America/New_York Multiple timezones \u00b6 Support coming soon. Participant files \u00b6 Participant files link together multiple devices (smartphones and wearables) to specific participants and identify them throughout RAPIDS. You can create these files manually or automatically . Participant files are stored in data/external/participant_files/pxx.yaml and follow a unified structure . Note The list PIDS in config.yaml needs to have the participant file names of the people you want to process. For example, if you created p01.yaml , p02.yaml and p03.yaml files in /data/external/participant_files/ , then PIDS should be: PIDS : [ p01 , p02 , p03 ] Tip Attribute values of the [PHONE] and [FITBIT] sections in every participant file are optional which allows you to analyze data from participants that only carried smartphones, only Fitbit devices, or both. Optional: Migrating participants files with the old format If you were using the pre-release version of RAPIDS with participant files in plain text (as opposed to yaml), you can run the following command and your old files will be converted into yaml files stored in data/external/participant_files/ python tools/update_format_participant_files.py Structure of participants files \u00b6 Example of the structure of a participant file In this example, the participant used an android phone, an ios phone, and a fitbit device throughout the study between Apr 23 rd 2020 and Oct 28 th 2020 PHONE : DEVICE_IDS : [ a748ee1a-1d0b-4ae9-9074-279a2b6ba524 , dsadas-2324-fgsf-sdwr-gdfgs4rfsdf43 ] PLATFORMS : [ android , ios ] LABEL : test01 START_DATE : 2020-04-23 END_DATE : 2020-10-28 FITBIT : DEVICE_IDS : [ fitbit1 ] LABEL : test01 START_DATE : 2020-04-23 END_DATE : 2020-10-28 For [PHONE] Key Description [DEVICE_IDS] An array of the strings that uniquely identify each smartphone, you can have more than one for when participants changed phones in the middle of the study, in this case, data from all their devices will be joined and relabeled with the last 1 on this list. [PLATFORMS] An array that specifies the OS of each smartphone in [DEVICE_IDS] , use a combination of android or ios (we support participants that changed platforms in the middle of your study!). If you have an aware_device table in your database you can set [PLATFORMS]: [multiple] and RAPIDS will infer them automatically. [LABEL] A string that is used in reports and visualizations. [START_DATE] A string with format YYY-MM-DD . Only data collected after this date will be included in the analysis [END_DATE] A string with format YYY-MM-DD . Only data collected before this date will be included in the analysis For [FITBIT] Key Description [DEVICE_IDS] An array of the strings that uniquely identify each Fitbit, you can have more than one in case the participant changed devices in the middle of the study, in this case, data from all devices will be joined and relabeled with the last device_id on this list. [LABEL] A string that is used in reports and visualizations. [START_DATE] A string with format YYY-MM-DD . Only data collected after this date will be included in the analysis [END_DATE] A string with format YYY-MM-DD . Only data collected before this date will be included in the analysis Automatic creation of participant files \u00b6 You have two options a) use the aware_device table in your database or b) use a CSV file. In either case, in your config.yaml , set [PHONE_SECTION][ADD] or [FITBIT_SECTION][ADD] to TRUE depending on what devices you used in your study. Set [DEVICE_ID_COLUMN] to the name of the column that uniquely identifies each device and include any device ids you want to ignore in [IGNORED_DEVICE_IDS] . aware_device table Set the following keys in your config.yaml CREATE_PARTICIPANT_FILES : SOURCE : TYPE : AWARE_DEVICE_TABLE DATABASE_GROUP : *database_group CSV_FILE_PATH : \"\" TIMEZONE : *timezone PHONE_SECTION : ADD : TRUE # or FALSE DEVICE_ID_COLUMN : device_id # column name IGNORED_DEVICE_IDS : [] FITBIT_SECTION : ADD : TRUE # or FALSE DEVICE_ID_COLUMN : fitbit_id # column name IGNORED_DEVICE_IDS : [] Then run snakemake -j1 create_participants_files CSV file Set the following keys in your config.yaml . CREATE_PARTICIPANT_FILES : SOURCE : TYPE : CSV_FILE DATABASE_GROUP : \"\" CSV_FILE_PATH : \"your_path/to_your.csv\" TIMEZONE : *timezone PHONE_SECTION : ADD : TRUE # or FALSE DEVICE_ID_COLUMN : device_id # column name IGNORED_DEVICE_IDS : [] FITBIT_SECTION : ADD : TRUE # or FALSE DEVICE_ID_COLUMN : fitbit_id # column name IGNORED_DEVICE_IDS : [] Your CSV file ( [SOURCE][CSV_FILE_PATH] ) should have the following columns but you can omit any values you don\u2019t have on each column: Column Description phone device id The name of this column has to match [PHONE_SECTION][DEVICE_ID_COLUMN] . Separate multiple ids with ; fitbit device id The name of this column has to match [FITBIT_SECTION][DEVICE_ID_COLUMN] . Separate multiple ids with ; pid Unique identifiers with the format pXXX (your participant files will be named with this string platform Use android , ios or multiple as explained above, separate values with ; label A human readable string that is used in reports and visualizations. start_date A string with format YYY-MM-DD . end_date A string with format YYY-MM-DD . Example device_id,pid,label,platform,start_date,end_date,fitbit_id a748ee1a-1d0b-4ae9-9074-279a2b6ba524;dsadas-2324-fgsf-sdwr-gdfgs4rfsdf43,p01,julio,android;ios,2020-01-01,2021-01-01,fitbit1 4c4cf7a1-0340-44bc-be0f-d5053bf7390c,p02,meng,ios,2021-01-01,2022-01-01,fitbit2 Then run snakemake -j1 create_participants_files Time Segments \u00b6 Time segments (or epochs) are the time windows on which you want to extract behavioral features. For example, you might want to process data on every day, every morning, or only during weekends. RAPIDS offers three categories of time segments that are flexible enough to cover most use cases: frequency (short time windows every day), periodic (arbitrary time windows on any day), and event (arbitrary time windows around events of interest). See also our examples . Frequency Segments These segments are computed on every day and all have the same duration (for example 30 minutes). Set the following keys in your config.yaml TIME_SEGMENTS : &time_segments TYPE : FREQUENCY FILE : \"data/external/your_frequency_segments.csv\" INCLUDE_PAST_PERIODIC_SEGMENTS : FALSE The file pointed by [TIME_SEGMENTS][FILE] should have the following format and can only have 1 row. Column Description label A string that is used as a prefix in the name of your time segments length An integer representing the duration of your time segments in minutes Example label,length thirtyminutes,30 This configuration will compute 48 time segments for every day when any data from any participant was sensed. For example: start_time,length,label 00:00,30,thirtyminutes0000 00:30,30,thirtyminutes0001 01:00,30,thirtyminutes0002 01:30,30,thirtyminutes0003 ... Periodic Segments These segments can be computed every day, or on specific days of the week, month, quarter, and year. Their minimum duration is 1 minute but they can be as long as you want. Set the following keys in your config.yaml . TIME_SEGMENTS : &time_segments TYPE : PERIODIC FILE : \"data/external/your_periodic_segments.csv\" INCLUDE_PAST_PERIODIC_SEGMENTS : FALSE # or TRUE If [INCLUDE_PAST_PERIODIC_SEGMENTS] is set to TRUE , RAPIDS will consider instances of your segments back enough in the past as to include the first row of data of each participant. For example, if the first row of data from a participant happened on Saturday March 7 th 2020 and the requested segment duration is 7 days starting on every Sunday, the first segment to be considered would start on Sunday March 1 st if [INCLUDE_PAST_PERIODIC_SEGMENTS] is TRUE or on Sunday March 8 th if FALSE . The file pointed by [TIME_SEGMENTS][FILE] should have the following format and can have multiple rows. Column Description label A string that is used as a prefix in the name of your time segments. It has to be unique between rows start_time A string with format HH:MM:SS representing the starting time of this segment on any day length A string representing the length of this segment.It can have one or more of the following strings XXD XXH XXM XXS to represent days, hours, minutes and seconds. For example 7D 23H 59M 59S repeats_on One of the follow options every_day , wday , qday , mday , and yday . The last four represent a week, quarter, month and year day repeats_value An integer complementing repeats_on . If you set repeats_on to every_day set this to 0 , otherwise 1-7 represent a wday starting from Mondays, 1-31 represent a mday , 1-91 represent a qday , and 1-366 represent a yday Example label,start_time,length,repeats_on,repeats_value daily,00:00:00,23H 59M 59S,every_day,0 morning,06:00:00,5H 59M 59S,every_day,0 afternoon,12:00:00,5H 59M 59S,every_day,0 evening,18:00:00,5H 59M 59S,every_day,0 night,00:00:00,5H 59M 59S,every_day,0 This configuration will create five segments instances ( daily , morning , afternoon , evening , night ) on any given day ( every_day set to 0). The daily segment will start at midnight and will last 23:59:59 , the other four segments will start at 6am, 12pm, 6pm, and 12am respectively and last for 05:59:59 . Event segments These segments can be computed before or after an event of interest (defined as any UNIX timestamp). Their minimum duration is 1 minute but they can be as long as you want. The start of each segment can be shifted backwards or forwards from the specified timestamp. Set the following keys in your config.yaml . TIME_SEGMENTS : &time_segments TYPE : EVENT FILE : \"data/external/your_event_segments.csv\" INCLUDE_PAST_PERIODIC_SEGMENTS : FALSE # or TRUE The file pointed by [TIME_SEGMENTS][FILE] should have the following format and can have multiple rows. Column Description label A string that is used as a prefix in the name of your time segments. If labels are unique, every segment is independent; if two or more segments have the same label, their data will be grouped when computing auxiliary data for features like the most frequent contact for calls (the most frequent contact will be computed across all these segments). There cannot be two overlaping event segments with the same label (RAPIDS will throw an error) event_timestamp A UNIX timestamp that represents the moment an event of interest happened (clinical relapse, survey, readmission, etc.). The corresponding time segment will be computed around this moment using length , shift , and shift_direction length A string representing the length of this segment. It can have one or more of the following keys XXD XXH XXM XXS to represent a number of days, hours, minutes, and seconds. For example 7D 23H 59M 59S shift A string representing the time shift from event_timestamp . It can have one or more of the following keys XXD XXH XXM XXS to represent a number of days, hours, minutes and seconds. For example 7D 23H 59M 59S . Use this value to change the start of a segment with respect to its event_timestamp . For example, set this variable to 1H to create a segment that starts 1 hour from an event of interest ( shift_direction determines if it\u2019s before or after). shift_direction An integer representing whether the shift is before ( -1 ) or after ( 1 ) an event_timestamp device_id The device id (smartphone or fitbit) to whom this segment belongs to. You have to create a line in this event segment file for each event of a participant that you want to analyse. If you have participants with multiple device ids you can choose any of them Example label,event_timestamp,length,shift,shift_direction,device_id stress1,1587661220000,1H,5M,1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 stress2,1587747620000,4H,4H,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 stress3,1587906020000,3H,5M,1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 stress4,1584291600000,7H,4H,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 stress5,1588172420000,9H,5M,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 mood,1587661220000,1H,0,0,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 mood,1587747620000,1D,0,0,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 mood,1587906020000,7D,0,0,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 This example will create eight segments for a single participant ( a748ee1a... ), five independent stressX segments with various lengths (1,4,3,7, and 9 hours). Segments stress1 , stress3 , and stress5 are shifted forwards by 5 minutes and stress2 and stress4 are shifted backwards by 4 hours (that is, if the stress4 event happened on March 15 th at 1pm EST ( 1584291600000 ), the time segment will start on that day at 9am and end at 4pm). The three mood segments are 1 hour, 1 day and 7 days long and have no shift. In addition, these mood segments are grouped together, meaning that although RAPIDS will compute features on each one of them, some necessary information to compute a few of such features will be extracted from all three segments, for example the phone contact that called a participant the most or the location clusters visited by a participant. Segment Examples \u00b6 5-minutes Use the following Frequency segment file to create 288 (12 * 60 * 24) 5-minute segments starting from midnight of every day in your study label,length fiveminutes,5 Daily Use the following Periodic segment file to create daily segments starting from midnight of every day in your study label,start_time,length,repeats_on,repeats_value daily,00:00:00,23H 59M 59S,every_day,0 Morning Use the following Periodic segment file to create morning segments starting at 06:00:00 and ending at 11:59:59 of every day in your study label,start_time,length,repeats_on,repeats_value morning,06:00:00,5H 59M 59S,every_day,0 Overnight Use the following Periodic segment file to create overnight segments starting at 20:00:00 and ending at 07:59:59 (next day) of every day in your study label,start_time,length,repeats_on,repeats_value morning,20:00:00,11H 59M 59S,every_day,0 Weekly Use the following Periodic segment file to create non-overlapping weekly segments starting at midnight of every Monday in your study label,start_time,length,repeats_on,repeats_value weekly,00:00:00,6D 23H 59M 59S,wday,1 Use the following Periodic segment file to create overlapping weekly segments starting at midnight of every day in your study label,start_time,length,repeats_on,repeats_value weekly,00:00:00,6D 23H 59M 59S,every_day,0 Week-ends Use the following Periodic segment file to create week-end segments starting at midnight of every Saturday in your study label,start_time,length,repeats_on,repeats_value weekend,00:00:00,1D 23H 59M 59S,wday,6 Around surveys Use the following Event segment file to create two 2-hour segments that start 1 hour before surveys answered by 3 participants label,event_timestamp,length,shift,shift_direction,device_id survey1,1587661220000,2H,1H,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 survey2,1587747620000,2H,1H,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 survey1,1587906020000,2H,1H,-1,rqtertsd-43ff-34fr-3eeg-efe4fergregr survey2,1584291600000,2H,1H,-1,rqtertsd-43ff-34fr-3eeg-efe4fergregr survey1,1588172420000,2H,1H,-1,klj34oi2-8frk-2343-21kk-324ljklewlr3 survey2,1584291600000,2H,1H,-1,klj34oi2-8frk-2343-21kk-324ljklewlr3 Device Data Source Configuration \u00b6 You might need to modify the following config keys in your config.yaml depending on what devices your participants used and where you are storing your data. You can ignore [PHONE_DATA_CONFIGURATION] or [FITBIT_DATA_CONFIGURATION] if you are not working with either devices. Phone The relevant config.yaml section looks like this by default: PHONE_DATA_CONFIGURATION : SOURCE : TYPE : DATABASE DATABASE_GROUP : *database_group DEVICE_ID_COLUMN : device_id # column name TIMEZONE : TYPE : SINGLE # SINGLE (MULTIPLE support coming soon) VALUE : *timezone Parameters for [PHONE_DATA_CONFIGURATION] Key Description [SOURCE] [TYPE] Only DATABASE is supported (phone data will be pulled from a database) [SOURCE] [DATABASE_GROUP] *database_group points to the value defined before in Database credentials [SOURCE] [DEVICE_ID_COLUMN] A column that contains strings that uniquely identify smartphones. For data collected with AWARE this is usually device_id [TIMEZONE] [TYPE] Only SINGLE is supported for now [TIMEZONE] [VALUE] *timezone points to the value defined before in Timezone of your study Fitbit The relevant config.yaml section looks like this by default: FITBIT_DATA_CONFIGURATION : SOURCE : TYPE : DATABASE # DATABASE or FILES (set each [FITBIT_SENSOR][TABLE] attribute with a table name or a file path accordingly) COLUMN_FORMAT : JSON # JSON or PLAIN_TEXT DATABASE_GROUP : *database_group DEVICE_ID_COLUMN : device_id # column name TIMEZONE : TYPE : SINGLE # Fitbit devices don't support time zones so we read this data in the timezone indicated by VALUE VALUE : *timezone Parameters for For [FITBIT_DATA_CONFIGURATION] Key Description [SOURCE] [TYPE] DATABASE or FILES (set each [FITBIT_SENSOR] [TABLE] attribute accordingly with a table name or a file path) [SOURCE] [COLUMN_FORMAT] JSON or PLAIN_TEXT . Column format of the source data. If you pulled your data directly from the Fitbit API the column containing the sensor data will be in JSON format [SOURCE] [DATABASE_GROUP] *database_group points to the value defined before in Database credentials . Only used if [TYPE] is DATABASE . [SOURCE] [DEVICE_ID_COLUMN] A column that contains strings that uniquely identify Fitbit devices. [TIMEZONE] [TYPE] Only SINGLE is supported (Fitbit devices always store data in local time). [TIMEZONE] [VALUE] *timezone points to the value defined before in Timezone of your study Sensor and Features to Process \u00b6 Finally, you need to modify the config.yaml section of the sensors you want to extract behavioral features from. All sensors follow the same naming nomenclature ( DEVICE_SENSOR ) and parameter structure which we explain in the Behavioral Features Introduction . Done Head over to Execution to learn how to execute RAPIDS.","title":"Configuration"},{"location":"setup/configuration/#configuration","text":"You need to follow these steps to configure your RAPIDS deployment before you can extract behavioral features Add your database credentials Choose the timezone of your study Create your participants files Select what time segments you want to extract features on Modify your device data source configuration Select what sensors and features you want to process When you are done with this configuration, go to executing RAPIDS . Hint Every time you see config[\"KEY\"] or [KEY] in these docs we are referring to the corresponding key in the config.yaml file.","title":"Configuration"},{"location":"setup/configuration/#database-credentials","text":"Create an empty file called .env in your RAPIDS root directory Add the following lines and replace your database-specific credentials (user, password, host, and database): [ MY_GROUP ] user=MY_USER password=MY_PASSWORD host=MY_HOST port=3306 database=MY_DATABASE Warning The label MY_GROUP is arbitrary but it has to match the following config.yaml key: DATABASE_GROUP : &database_group MY_GROUP Note You can ignore this step if you are only processing Fitbit data in CSV files.","title":"Database credentials"},{"location":"setup/configuration/#timezone-of-your-study","text":"","title":"Timezone of your study"},{"location":"setup/configuration/#single-timezone","text":"If your study only happened in a single time zone, select the appropriate code form this list and change the following config key. Double check your timezone code pick, for example US Eastern Time is America/New_York not EST TIMEZONE : &timezone America/New_York","title":"Single timezone"},{"location":"setup/configuration/#multiple-timezones","text":"Support coming soon.","title":"Multiple timezones"},{"location":"setup/configuration/#participant-files","text":"Participant files link together multiple devices (smartphones and wearables) to specific participants and identify them throughout RAPIDS. You can create these files manually or automatically . Participant files are stored in data/external/participant_files/pxx.yaml and follow a unified structure . Note The list PIDS in config.yaml needs to have the participant file names of the people you want to process. For example, if you created p01.yaml , p02.yaml and p03.yaml files in /data/external/participant_files/ , then PIDS should be: PIDS : [ p01 , p02 , p03 ] Tip Attribute values of the [PHONE] and [FITBIT] sections in every participant file are optional which allows you to analyze data from participants that only carried smartphones, only Fitbit devices, or both. Optional: Migrating participants files with the old format If you were using the pre-release version of RAPIDS with participant files in plain text (as opposed to yaml), you can run the following command and your old files will be converted into yaml files stored in data/external/participant_files/ python tools/update_format_participant_files.py","title":"Participant files"},{"location":"setup/configuration/#structure-of-participants-files","text":"Example of the structure of a participant file In this example, the participant used an android phone, an ios phone, and a fitbit device throughout the study between Apr 23 rd 2020 and Oct 28 th 2020 PHONE : DEVICE_IDS : [ a748ee1a-1d0b-4ae9-9074-279a2b6ba524 , dsadas-2324-fgsf-sdwr-gdfgs4rfsdf43 ] PLATFORMS : [ android , ios ] LABEL : test01 START_DATE : 2020-04-23 END_DATE : 2020-10-28 FITBIT : DEVICE_IDS : [ fitbit1 ] LABEL : test01 START_DATE : 2020-04-23 END_DATE : 2020-10-28 For [PHONE] Key Description [DEVICE_IDS] An array of the strings that uniquely identify each smartphone, you can have more than one for when participants changed phones in the middle of the study, in this case, data from all their devices will be joined and relabeled with the last 1 on this list. [PLATFORMS] An array that specifies the OS of each smartphone in [DEVICE_IDS] , use a combination of android or ios (we support participants that changed platforms in the middle of your study!). If you have an aware_device table in your database you can set [PLATFORMS]: [multiple] and RAPIDS will infer them automatically. [LABEL] A string that is used in reports and visualizations. [START_DATE] A string with format YYY-MM-DD . Only data collected after this date will be included in the analysis [END_DATE] A string with format YYY-MM-DD . Only data collected before this date will be included in the analysis For [FITBIT] Key Description [DEVICE_IDS] An array of the strings that uniquely identify each Fitbit, you can have more than one in case the participant changed devices in the middle of the study, in this case, data from all devices will be joined and relabeled with the last device_id on this list. [LABEL] A string that is used in reports and visualizations. [START_DATE] A string with format YYY-MM-DD . Only data collected after this date will be included in the analysis [END_DATE] A string with format YYY-MM-DD . Only data collected before this date will be included in the analysis","title":"Structure of participants files"},{"location":"setup/configuration/#automatic-creation-of-participant-files","text":"You have two options a) use the aware_device table in your database or b) use a CSV file. In either case, in your config.yaml , set [PHONE_SECTION][ADD] or [FITBIT_SECTION][ADD] to TRUE depending on what devices you used in your study. Set [DEVICE_ID_COLUMN] to the name of the column that uniquely identifies each device and include any device ids you want to ignore in [IGNORED_DEVICE_IDS] . aware_device table Set the following keys in your config.yaml CREATE_PARTICIPANT_FILES : SOURCE : TYPE : AWARE_DEVICE_TABLE DATABASE_GROUP : *database_group CSV_FILE_PATH : \"\" TIMEZONE : *timezone PHONE_SECTION : ADD : TRUE # or FALSE DEVICE_ID_COLUMN : device_id # column name IGNORED_DEVICE_IDS : [] FITBIT_SECTION : ADD : TRUE # or FALSE DEVICE_ID_COLUMN : fitbit_id # column name IGNORED_DEVICE_IDS : [] Then run snakemake -j1 create_participants_files CSV file Set the following keys in your config.yaml . CREATE_PARTICIPANT_FILES : SOURCE : TYPE : CSV_FILE DATABASE_GROUP : \"\" CSV_FILE_PATH : \"your_path/to_your.csv\" TIMEZONE : *timezone PHONE_SECTION : ADD : TRUE # or FALSE DEVICE_ID_COLUMN : device_id # column name IGNORED_DEVICE_IDS : [] FITBIT_SECTION : ADD : TRUE # or FALSE DEVICE_ID_COLUMN : fitbit_id # column name IGNORED_DEVICE_IDS : [] Your CSV file ( [SOURCE][CSV_FILE_PATH] ) should have the following columns but you can omit any values you don\u2019t have on each column: Column Description phone device id The name of this column has to match [PHONE_SECTION][DEVICE_ID_COLUMN] . Separate multiple ids with ; fitbit device id The name of this column has to match [FITBIT_SECTION][DEVICE_ID_COLUMN] . Separate multiple ids with ; pid Unique identifiers with the format pXXX (your participant files will be named with this string platform Use android , ios or multiple as explained above, separate values with ; label A human readable string that is used in reports and visualizations. start_date A string with format YYY-MM-DD . end_date A string with format YYY-MM-DD . Example device_id,pid,label,platform,start_date,end_date,fitbit_id a748ee1a-1d0b-4ae9-9074-279a2b6ba524;dsadas-2324-fgsf-sdwr-gdfgs4rfsdf43,p01,julio,android;ios,2020-01-01,2021-01-01,fitbit1 4c4cf7a1-0340-44bc-be0f-d5053bf7390c,p02,meng,ios,2021-01-01,2022-01-01,fitbit2 Then run snakemake -j1 create_participants_files","title":"Automatic creation of participant files"},{"location":"setup/configuration/#time-segments","text":"Time segments (or epochs) are the time windows on which you want to extract behavioral features. For example, you might want to process data on every day, every morning, or only during weekends. RAPIDS offers three categories of time segments that are flexible enough to cover most use cases: frequency (short time windows every day), periodic (arbitrary time windows on any day), and event (arbitrary time windows around events of interest). See also our examples . Frequency Segments These segments are computed on every day and all have the same duration (for example 30 minutes). Set the following keys in your config.yaml TIME_SEGMENTS : &time_segments TYPE : FREQUENCY FILE : \"data/external/your_frequency_segments.csv\" INCLUDE_PAST_PERIODIC_SEGMENTS : FALSE The file pointed by [TIME_SEGMENTS][FILE] should have the following format and can only have 1 row. Column Description label A string that is used as a prefix in the name of your time segments length An integer representing the duration of your time segments in minutes Example label,length thirtyminutes,30 This configuration will compute 48 time segments for every day when any data from any participant was sensed. For example: start_time,length,label 00:00,30,thirtyminutes0000 00:30,30,thirtyminutes0001 01:00,30,thirtyminutes0002 01:30,30,thirtyminutes0003 ... Periodic Segments These segments can be computed every day, or on specific days of the week, month, quarter, and year. Their minimum duration is 1 minute but they can be as long as you want. Set the following keys in your config.yaml . TIME_SEGMENTS : &time_segments TYPE : PERIODIC FILE : \"data/external/your_periodic_segments.csv\" INCLUDE_PAST_PERIODIC_SEGMENTS : FALSE # or TRUE If [INCLUDE_PAST_PERIODIC_SEGMENTS] is set to TRUE , RAPIDS will consider instances of your segments back enough in the past as to include the first row of data of each participant. For example, if the first row of data from a participant happened on Saturday March 7 th 2020 and the requested segment duration is 7 days starting on every Sunday, the first segment to be considered would start on Sunday March 1 st if [INCLUDE_PAST_PERIODIC_SEGMENTS] is TRUE or on Sunday March 8 th if FALSE . The file pointed by [TIME_SEGMENTS][FILE] should have the following format and can have multiple rows. Column Description label A string that is used as a prefix in the name of your time segments. It has to be unique between rows start_time A string with format HH:MM:SS representing the starting time of this segment on any day length A string representing the length of this segment.It can have one or more of the following strings XXD XXH XXM XXS to represent days, hours, minutes and seconds. For example 7D 23H 59M 59S repeats_on One of the follow options every_day , wday , qday , mday , and yday . The last four represent a week, quarter, month and year day repeats_value An integer complementing repeats_on . If you set repeats_on to every_day set this to 0 , otherwise 1-7 represent a wday starting from Mondays, 1-31 represent a mday , 1-91 represent a qday , and 1-366 represent a yday Example label,start_time,length,repeats_on,repeats_value daily,00:00:00,23H 59M 59S,every_day,0 morning,06:00:00,5H 59M 59S,every_day,0 afternoon,12:00:00,5H 59M 59S,every_day,0 evening,18:00:00,5H 59M 59S,every_day,0 night,00:00:00,5H 59M 59S,every_day,0 This configuration will create five segments instances ( daily , morning , afternoon , evening , night ) on any given day ( every_day set to 0). The daily segment will start at midnight and will last 23:59:59 , the other four segments will start at 6am, 12pm, 6pm, and 12am respectively and last for 05:59:59 . Event segments These segments can be computed before or after an event of interest (defined as any UNIX timestamp). Their minimum duration is 1 minute but they can be as long as you want. The start of each segment can be shifted backwards or forwards from the specified timestamp. Set the following keys in your config.yaml . TIME_SEGMENTS : &time_segments TYPE : EVENT FILE : \"data/external/your_event_segments.csv\" INCLUDE_PAST_PERIODIC_SEGMENTS : FALSE # or TRUE The file pointed by [TIME_SEGMENTS][FILE] should have the following format and can have multiple rows. Column Description label A string that is used as a prefix in the name of your time segments. If labels are unique, every segment is independent; if two or more segments have the same label, their data will be grouped when computing auxiliary data for features like the most frequent contact for calls (the most frequent contact will be computed across all these segments). There cannot be two overlaping event segments with the same label (RAPIDS will throw an error) event_timestamp A UNIX timestamp that represents the moment an event of interest happened (clinical relapse, survey, readmission, etc.). The corresponding time segment will be computed around this moment using length , shift , and shift_direction length A string representing the length of this segment. It can have one or more of the following keys XXD XXH XXM XXS to represent a number of days, hours, minutes, and seconds. For example 7D 23H 59M 59S shift A string representing the time shift from event_timestamp . It can have one or more of the following keys XXD XXH XXM XXS to represent a number of days, hours, minutes and seconds. For example 7D 23H 59M 59S . Use this value to change the start of a segment with respect to its event_timestamp . For example, set this variable to 1H to create a segment that starts 1 hour from an event of interest ( shift_direction determines if it\u2019s before or after). shift_direction An integer representing whether the shift is before ( -1 ) or after ( 1 ) an event_timestamp device_id The device id (smartphone or fitbit) to whom this segment belongs to. You have to create a line in this event segment file for each event of a participant that you want to analyse. If you have participants with multiple device ids you can choose any of them Example label,event_timestamp,length,shift,shift_direction,device_id stress1,1587661220000,1H,5M,1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 stress2,1587747620000,4H,4H,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 stress3,1587906020000,3H,5M,1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 stress4,1584291600000,7H,4H,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 stress5,1588172420000,9H,5M,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 mood,1587661220000,1H,0,0,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 mood,1587747620000,1D,0,0,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 mood,1587906020000,7D,0,0,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 This example will create eight segments for a single participant ( a748ee1a... ), five independent stressX segments with various lengths (1,4,3,7, and 9 hours). Segments stress1 , stress3 , and stress5 are shifted forwards by 5 minutes and stress2 and stress4 are shifted backwards by 4 hours (that is, if the stress4 event happened on March 15 th at 1pm EST ( 1584291600000 ), the time segment will start on that day at 9am and end at 4pm). The three mood segments are 1 hour, 1 day and 7 days long and have no shift. In addition, these mood segments are grouped together, meaning that although RAPIDS will compute features on each one of them, some necessary information to compute a few of such features will be extracted from all three segments, for example the phone contact that called a participant the most or the location clusters visited by a participant.","title":"Time Segments"},{"location":"setup/configuration/#segment-examples","text":"5-minutes Use the following Frequency segment file to create 288 (12 * 60 * 24) 5-minute segments starting from midnight of every day in your study label,length fiveminutes,5 Daily Use the following Periodic segment file to create daily segments starting from midnight of every day in your study label,start_time,length,repeats_on,repeats_value daily,00:00:00,23H 59M 59S,every_day,0 Morning Use the following Periodic segment file to create morning segments starting at 06:00:00 and ending at 11:59:59 of every day in your study label,start_time,length,repeats_on,repeats_value morning,06:00:00,5H 59M 59S,every_day,0 Overnight Use the following Periodic segment file to create overnight segments starting at 20:00:00 and ending at 07:59:59 (next day) of every day in your study label,start_time,length,repeats_on,repeats_value morning,20:00:00,11H 59M 59S,every_day,0 Weekly Use the following Periodic segment file to create non-overlapping weekly segments starting at midnight of every Monday in your study label,start_time,length,repeats_on,repeats_value weekly,00:00:00,6D 23H 59M 59S,wday,1 Use the following Periodic segment file to create overlapping weekly segments starting at midnight of every day in your study label,start_time,length,repeats_on,repeats_value weekly,00:00:00,6D 23H 59M 59S,every_day,0 Week-ends Use the following Periodic segment file to create week-end segments starting at midnight of every Saturday in your study label,start_time,length,repeats_on,repeats_value weekend,00:00:00,1D 23H 59M 59S,wday,6 Around surveys Use the following Event segment file to create two 2-hour segments that start 1 hour before surveys answered by 3 participants label,event_timestamp,length,shift,shift_direction,device_id survey1,1587661220000,2H,1H,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 survey2,1587747620000,2H,1H,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 survey1,1587906020000,2H,1H,-1,rqtertsd-43ff-34fr-3eeg-efe4fergregr survey2,1584291600000,2H,1H,-1,rqtertsd-43ff-34fr-3eeg-efe4fergregr survey1,1588172420000,2H,1H,-1,klj34oi2-8frk-2343-21kk-324ljklewlr3 survey2,1584291600000,2H,1H,-1,klj34oi2-8frk-2343-21kk-324ljklewlr3","title":"Segment Examples"},{"location":"setup/configuration/#device-data-source-configuration","text":"You might need to modify the following config keys in your config.yaml depending on what devices your participants used and where you are storing your data. You can ignore [PHONE_DATA_CONFIGURATION] or [FITBIT_DATA_CONFIGURATION] if you are not working with either devices. Phone The relevant config.yaml section looks like this by default: PHONE_DATA_CONFIGURATION : SOURCE : TYPE : DATABASE DATABASE_GROUP : *database_group DEVICE_ID_COLUMN : device_id # column name TIMEZONE : TYPE : SINGLE # SINGLE (MULTIPLE support coming soon) VALUE : *timezone Parameters for [PHONE_DATA_CONFIGURATION] Key Description [SOURCE] [TYPE] Only DATABASE is supported (phone data will be pulled from a database) [SOURCE] [DATABASE_GROUP] *database_group points to the value defined before in Database credentials [SOURCE] [DEVICE_ID_COLUMN] A column that contains strings that uniquely identify smartphones. For data collected with AWARE this is usually device_id [TIMEZONE] [TYPE] Only SINGLE is supported for now [TIMEZONE] [VALUE] *timezone points to the value defined before in Timezone of your study Fitbit The relevant config.yaml section looks like this by default: FITBIT_DATA_CONFIGURATION : SOURCE : TYPE : DATABASE # DATABASE or FILES (set each [FITBIT_SENSOR][TABLE] attribute with a table name or a file path accordingly) COLUMN_FORMAT : JSON # JSON or PLAIN_TEXT DATABASE_GROUP : *database_group DEVICE_ID_COLUMN : device_id # column name TIMEZONE : TYPE : SINGLE # Fitbit devices don't support time zones so we read this data in the timezone indicated by VALUE VALUE : *timezone Parameters for For [FITBIT_DATA_CONFIGURATION] Key Description [SOURCE] [TYPE] DATABASE or FILES (set each [FITBIT_SENSOR] [TABLE] attribute accordingly with a table name or a file path) [SOURCE] [COLUMN_FORMAT] JSON or PLAIN_TEXT . Column format of the source data. If you pulled your data directly from the Fitbit API the column containing the sensor data will be in JSON format [SOURCE] [DATABASE_GROUP] *database_group points to the value defined before in Database credentials . Only used if [TYPE] is DATABASE . [SOURCE] [DEVICE_ID_COLUMN] A column that contains strings that uniquely identify Fitbit devices. [TIMEZONE] [TYPE] Only SINGLE is supported (Fitbit devices always store data in local time). [TIMEZONE] [VALUE] *timezone points to the value defined before in Timezone of your study","title":"Device Data Source Configuration"},{"location":"setup/configuration/#sensor-and-features-to-process","text":"Finally, you need to modify the config.yaml section of the sensors you want to extract behavioral features from. All sensors follow the same naming nomenclature ( DEVICE_SENSOR ) and parameter structure which we explain in the Behavioral Features Introduction . Done Head over to Execution to learn how to execute RAPIDS.","title":"Sensor and Features to Process"},{"location":"setup/execution/","text":"Execution \u00b6 After you have installed and configured RAPIDS, use the following command to execute it. ./rapids -j1 Ready to extract behavioral features If you are ready to extract features head over to the Behavioral Features Introduction Info The script ./rapids is a wrapper around Snakemake so you can pass any parameters that Snakemake accepts (e.g. -j1 ). Updating RAPIDS output after modifying config.yaml Any changes to the config.yaml file will be applied automatically and only the relevant files will be updated. This means that after modifying the features list for PHONE_MESSAGE for example, RAPIDS will update the output file with the correct features. Multi-core You can run RAPIDS over multiple cores by modifying the -j argument (e.g. use -j8 to use 8 cores). However , take into account that this means multiple sensor datasets for different participants will be load in memory at the same time. If RAPIDS crashes because it ran out of memory reduce the number of cores and try again. As reference, we have run RAPIDS over 12 cores and 32 Gb of RAM without problems for a study with 200 participants with 14 days of low-frequency smartphone data (no accelerometer, gyroscope, or magnetometer). Forcing a complete rerun If you want to update your data from your database or rerun the whole pipeline from scratch run one or both of the following commands depending on the devices you are using: ./rapids -j1 -R download_phone_data ./rapids -j1 -R download_fitbit_data Deleting RAPIDS output If you want to delete all the output files RAPIDS produces you can execute the following command: ./rapids -j1 --delete-all-output","title":"Execution"},{"location":"setup/execution/#execution","text":"After you have installed and configured RAPIDS, use the following command to execute it. ./rapids -j1 Ready to extract behavioral features If you are ready to extract features head over to the Behavioral Features Introduction Info The script ./rapids is a wrapper around Snakemake so you can pass any parameters that Snakemake accepts (e.g. -j1 ). Updating RAPIDS output after modifying config.yaml Any changes to the config.yaml file will be applied automatically and only the relevant files will be updated. This means that after modifying the features list for PHONE_MESSAGE for example, RAPIDS will update the output file with the correct features. Multi-core You can run RAPIDS over multiple cores by modifying the -j argument (e.g. use -j8 to use 8 cores). However , take into account that this means multiple sensor datasets for different participants will be load in memory at the same time. If RAPIDS crashes because it ran out of memory reduce the number of cores and try again. As reference, we have run RAPIDS over 12 cores and 32 Gb of RAM without problems for a study with 200 participants with 14 days of low-frequency smartphone data (no accelerometer, gyroscope, or magnetometer). Forcing a complete rerun If you want to update your data from your database or rerun the whole pipeline from scratch run one or both of the following commands depending on the devices you are using: ./rapids -j1 -R download_phone_data ./rapids -j1 -R download_fitbit_data Deleting RAPIDS output If you want to delete all the output files RAPIDS produces you can execute the following command: ./rapids -j1 --delete-all-output","title":"Execution"},{"location":"setup/installation/","text":"Installation \u00b6 You can install RAPIDS using Docker (the fastest), or native instructions for MacOS and Ubuntu Docker Install Docker Pull our RAPIDS container docker pull agamk/rapids:latest ` Run RAPIDS' container (after this step is done you should see a prompt in the main RAPIDS folder with its python environment active) docker run -it agamk/rapids:latest Pull the latest version of RAPIDS git pull Make RAPIDS script executable chmod +x rapids Check that RAPIDS is working ./rapids -j1 Optional . You can edit RAPIDS files with vim but we recommend using Visual Studio Code and its Remote Containers extension How to configure Remote Containers extension Make sure RAPIDS container is running Install the Remote - Containers extension Go to the Remote Explorer panel on the left hand sidebar On the top right dropdown menu choose Containers Double click on the agamk/rapids container in the CONTAINERS tree A new VS Code session should open on RAPIDS main folder insidethe container. MacOS We tested these instructions in Catalina Install brew Install MySQL brew install mysql brew services start mysql Install R 4.0, pandoc and rmarkdown. If you have other instances of R, we recommend uninstalling them brew install r brew install pandoc Rscript --vanilla -e 'install.packages(\"rmarkdown\", repos=\"http://cran.us.r-project.org\")' Install miniconda (restart your terminal afterwards) brew cask install miniconda conda init zsh # (or conda init bash) Clone our repo git clone https://github.com/carissalow/rapids Create a python virtual environment cd rapids conda env create -f environment.yml -n rapids conda activate rapids Install R packages and virtual environment: snakemake -j1 renv_install snakemake -j1 renv_restore Note This step could take several minutes to complete, especially if you have less than 3Gb of RAM or packages need to be compiled from source. Please be patient and let it run until completion. Make RAPIDS script executable chmod +x rapids Check that RAPIDS is working ./rapids -j1 Ubuntu We tested on Ubuntu 18.04 & 20.04 Install dependencies sudo apt install libcurl4-openssl-dev sudo apt install libssl-dev sudo apt install libxml2-dev Install MySQL sudo apt install libmysqlclient-dev sudo apt install mysql-server Add key for R\u2019s repository. sudo apt-key adv --keyserver keyserver.ubuntu.com --recv-keys E298A3A825C0D65DFD57CBB651716619E084DAB9 Add R\u2019s repository For 18.04 sudo add-apt-repository 'deb https://cloud.r-project.org/bin/linux/ubuntu bionic-cran40/' For 20.04 sudo add-apt-repository 'deb https://cloud.r-project.org/bin/linux/ubuntu focal-cran40/' Install R 4.0. If you have other instances of R, we recommend uninstalling them sudo apt update sudo apt install r-base Install Pandoc and rmarkdown sudo apt install pandoc Rscript --vanilla -e 'install.packages(\"rmarkdown\", repos=\"http://cran.us.r-project.org\")' Install git sudo apt install git Install miniconda Restart your current shell Clone our repo: git clone https://github.com/carissalow/rapids Create a python virtual environment: cd rapids conda env create -f environment.yml -n MY_ENV_NAME conda activate MY_ENV_NAME Install R packages and virtual environment: snakemake -j1 renv_install snakemake -j1 renv_restore Note This step could take several minutes to complete, especially if you have less than 3Gb of RAM or packages need to be compiled from source. Please be patient and let it run until completion. Make RAPIDS script executable chmod +x rapids Check that RAPIDS is working ./rapids -j1","title":"Installation"},{"location":"setup/installation/#installation","text":"You can install RAPIDS using Docker (the fastest), or native instructions for MacOS and Ubuntu Docker Install Docker Pull our RAPIDS container docker pull agamk/rapids:latest ` Run RAPIDS' container (after this step is done you should see a prompt in the main RAPIDS folder with its python environment active) docker run -it agamk/rapids:latest Pull the latest version of RAPIDS git pull Make RAPIDS script executable chmod +x rapids Check that RAPIDS is working ./rapids -j1 Optional . You can edit RAPIDS files with vim but we recommend using Visual Studio Code and its Remote Containers extension How to configure Remote Containers extension Make sure RAPIDS container is running Install the Remote - Containers extension Go to the Remote Explorer panel on the left hand sidebar On the top right dropdown menu choose Containers Double click on the agamk/rapids container in the CONTAINERS tree A new VS Code session should open on RAPIDS main folder insidethe container. MacOS We tested these instructions in Catalina Install brew Install MySQL brew install mysql brew services start mysql Install R 4.0, pandoc and rmarkdown. If you have other instances of R, we recommend uninstalling them brew install r brew install pandoc Rscript --vanilla -e 'install.packages(\"rmarkdown\", repos=\"http://cran.us.r-project.org\")' Install miniconda (restart your terminal afterwards) brew cask install miniconda conda init zsh # (or conda init bash) Clone our repo git clone https://github.com/carissalow/rapids Create a python virtual environment cd rapids conda env create -f environment.yml -n rapids conda activate rapids Install R packages and virtual environment: snakemake -j1 renv_install snakemake -j1 renv_restore Note This step could take several minutes to complete, especially if you have less than 3Gb of RAM or packages need to be compiled from source. Please be patient and let it run until completion. Make RAPIDS script executable chmod +x rapids Check that RAPIDS is working ./rapids -j1 Ubuntu We tested on Ubuntu 18.04 & 20.04 Install dependencies sudo apt install libcurl4-openssl-dev sudo apt install libssl-dev sudo apt install libxml2-dev Install MySQL sudo apt install libmysqlclient-dev sudo apt install mysql-server Add key for R\u2019s repository. sudo apt-key adv --keyserver keyserver.ubuntu.com --recv-keys E298A3A825C0D65DFD57CBB651716619E084DAB9 Add R\u2019s repository For 18.04 sudo add-apt-repository 'deb https://cloud.r-project.org/bin/linux/ubuntu bionic-cran40/' For 20.04 sudo add-apt-repository 'deb https://cloud.r-project.org/bin/linux/ubuntu focal-cran40/' Install R 4.0. If you have other instances of R, we recommend uninstalling them sudo apt update sudo apt install r-base Install Pandoc and rmarkdown sudo apt install pandoc Rscript --vanilla -e 'install.packages(\"rmarkdown\", repos=\"http://cran.us.r-project.org\")' Install git sudo apt install git Install miniconda Restart your current shell Clone our repo: git clone https://github.com/carissalow/rapids Create a python virtual environment: cd rapids conda env create -f environment.yml -n MY_ENV_NAME conda activate MY_ENV_NAME Install R packages and virtual environment: snakemake -j1 renv_install snakemake -j1 renv_restore Note This step could take several minutes to complete, especially if you have less than 3Gb of RAM or packages need to be compiled from source. Please be patient and let it run until completion. Make RAPIDS script executable chmod +x rapids Check that RAPIDS is working ./rapids -j1","title":"Installation"},{"location":"workflow-examples/minimal/","text":"Minimal Working Example \u00b6 This is a quick guide for creating and running a simple pipeline to extract missing, outgoing, and incoming call features for daily and night epochs of one participant monitored on the US East coast. Install RAPIDS and make sure your conda environment is active (see Installation ) Make the changes listed below for the corresponding Configuration step (we provide an example of what the relevant sections in your config.yml will look like after you are done) Things to change on each configuration step 1. Setup your database connection credentials in .env . We assume your credentials group is called MY_GROUP . 2. America/New_York should be the default timezone 3. Create a participant file p01.yaml based on one of your participants and add p01 to [PIDS] in config.yaml . The following would be the content of your p01.yaml participant file: PHONE : DEVICE_IDS : [ aaaaaaaa-1111-bbbb-2222-cccccccccccc ] # your participant's AWARE device id PLATFORMS : [ android ] # or ios LABEL : MyTestP01 # any string START_DATE : 2020-01-01 # this can also be empty END_DATE : 2021-01-01 # this can also be empty 4. [TIME_SEGMENTS][TYPE] should be the default PERIODIC . Change [TIME_SEGMENTS][FILE] with the path of a file containing the following lines: label,start_time,length,repeats_on,repeats_value daily,00:00:00,23H 59M 59S,every_day,0 night,00:00:00,5H 59M 59S,every_day,0 5. If you collected data with AWARE you won\u2019t need to modify the attributes of [DEVICE_DATA][PHONE] 6. Set [PHONE_CALLS][PROVIDERS][RAPIDS][COMPUTE] to True Example of the config.yaml sections after the changes outlined above PIDS: [p01] TIMEZONE: &timezone America/New_York DATABASE_GROUP: &database_group MY_GROUP # ... other irrelevant sections TIME_SEGMENTS: &time_segments TYPE: PERIODIC FILE: \"data/external/timesegments_periodic.csv\" # make sure the three lines specified above are in the file INCLUDE_PAST_PERIODIC_SEGMENTS: FALSE # No need to change this if you collected AWARE data on a database and your credentials are grouped under `MY_GROUP` in `.env` DEVICE_DATA: PHONE: SOURCE: TYPE: DATABASE DATABASE_GROUP: *database_group DEVICE_ID_COLUMN: device_id # column name TIMEZONE: TYPE: SINGLE # SINGLE or MULTIPLE VALUE: *timezone ############## PHONE ########################################################### ################################################################################ # ... other irrelevant sections # Communication call features config, TYPES and FEATURES keys need to match PHONE_CALLS: TABLE: calls # change if your calls table has a different name PROVIDERS: RAPIDS: COMPUTE: True # set this to True! CALL_TYPES: ... Run RAPIDS ./rapids -j1 The call features for daily and morning time segments will be in /data/processed/features/p01/phone_calls.csv","title":"Minimal"},{"location":"workflow-examples/minimal/#minimal-working-example","text":"This is a quick guide for creating and running a simple pipeline to extract missing, outgoing, and incoming call features for daily and night epochs of one participant monitored on the US East coast. Install RAPIDS and make sure your conda environment is active (see Installation ) Make the changes listed below for the corresponding Configuration step (we provide an example of what the relevant sections in your config.yml will look like after you are done) Things to change on each configuration step 1. Setup your database connection credentials in .env . We assume your credentials group is called MY_GROUP . 2. America/New_York should be the default timezone 3. Create a participant file p01.yaml based on one of your participants and add p01 to [PIDS] in config.yaml . The following would be the content of your p01.yaml participant file: PHONE : DEVICE_IDS : [ aaaaaaaa-1111-bbbb-2222-cccccccccccc ] # your participant's AWARE device id PLATFORMS : [ android ] # or ios LABEL : MyTestP01 # any string START_DATE : 2020-01-01 # this can also be empty END_DATE : 2021-01-01 # this can also be empty 4. [TIME_SEGMENTS][TYPE] should be the default PERIODIC . Change [TIME_SEGMENTS][FILE] with the path of a file containing the following lines: label,start_time,length,repeats_on,repeats_value daily,00:00:00,23H 59M 59S,every_day,0 night,00:00:00,5H 59M 59S,every_day,0 5. If you collected data with AWARE you won\u2019t need to modify the attributes of [DEVICE_DATA][PHONE] 6. Set [PHONE_CALLS][PROVIDERS][RAPIDS][COMPUTE] to True Example of the config.yaml sections after the changes outlined above PIDS: [p01] TIMEZONE: &timezone America/New_York DATABASE_GROUP: &database_group MY_GROUP # ... other irrelevant sections TIME_SEGMENTS: &time_segments TYPE: PERIODIC FILE: \"data/external/timesegments_periodic.csv\" # make sure the three lines specified above are in the file INCLUDE_PAST_PERIODIC_SEGMENTS: FALSE # No need to change this if you collected AWARE data on a database and your credentials are grouped under `MY_GROUP` in `.env` DEVICE_DATA: PHONE: SOURCE: TYPE: DATABASE DATABASE_GROUP: *database_group DEVICE_ID_COLUMN: device_id # column name TIMEZONE: TYPE: SINGLE # SINGLE or MULTIPLE VALUE: *timezone ############## PHONE ########################################################### ################################################################################ # ... other irrelevant sections # Communication call features config, TYPES and FEATURES keys need to match PHONE_CALLS: TABLE: calls # change if your calls table has a different name PROVIDERS: RAPIDS: COMPUTE: True # set this to True! CALL_TYPES: ... Run RAPIDS ./rapids -j1 The call features for daily and morning time segments will be in /data/processed/features/p01/phone_calls.csv","title":"Minimal Working Example"}]}
\ No newline at end of file
+{"config":{"lang":["en"],"min_search_length":3,"prebuild_index":false,"separator":"[\\s\\-]+"},"docs":[{"location":"","text":"Welcome to RAPIDS documentation \u00b6 Warning The functionality outlined in these docs is implemented in the branch time_segments which we will merge to master soon as release 0.1 . The previous (first) release of RAPIDS along with the old docs will be labeled beta . If you landed on this page feel free to look around, just have in mind that we are polishing the last rough patches before we advertise 0.1 (Nov 16, 2020) Reproducible Analysis Pipeline for Data Streams (RAPIDS) allows you to process smartphone and wearable data to extract behavioral features (a.k.a. digital biomarkers/phenotypes). RAPIDS is open source, documented, modular, tested, and reproducible. At the moment we support smartphone data collected with AWARE and wearable data from Fitbit devices. Questions or feedback can be posted on #rapids in AWARE Framework's slack . Bugs should be reported on Github . Join our discussions on our algorithms and assumptions for feature processing . Ready to start? Go to Installation , then to Configuration , and then to Execution How does it work? \u00b6 RAPIDS is formed by R and Python scripts orchestrated by Snakemake . We suggest you read Snakemake\u2019s docs but in short: every link in the analysis chain is atomic and has files as input and output. Behavioral features are processed per sensor and per participant. What are the benefits of using RAPIDS? \u00b6 Consistent analysis . Every participant sensor dataset is analyzed in the exact same way and isolated from each other. Efficient analysis . Every analysis step is executed only once. Whenever your data or configuration changes only the affected files are updated. Parallel execution . Thanks to Snakemake, your analysis can be executed over multiple cores without changing your code. Extensible code . You can easily add your own behavioral features in R or Python and keep authorship and citations. Timezone aware . Your data is adjusted to the specified timezone (multiple timezones suport coming soon ). Flexible time segments . You can extract behavioral features on time windows of any length (e.g. 5 minutes, 3 hours, 2 days), on every day or particular days (e.g. weekends, Mondays, the 1 st of each month, etc.) or around events of interest (e.g. surveys or clinical relapses). Tested code . We are constantly adding tests to make sure our behavioral features are correct. Reproducible code . You can be sure your code will run in other computers as intended thanks to R and Python virtual environments. You can share your analysis code along your publications without any overhead. Private . All your data is processed locally. How is it organized? \u00b6 In broad terms the config.yaml , .env file , participants files , time segment files are the only ones that you will have to modify. All data is stored in data/ and all scripts are stored in src/ . For more information see RAPIDS\u2019 File Structure .","title":"Home"},{"location":"#welcome-to-rapids-documentation","text":"Warning The functionality outlined in these docs is implemented in the branch time_segments which we will merge to master soon as release 0.1 . The previous (first) release of RAPIDS along with the old docs will be labeled beta . If you landed on this page feel free to look around, just have in mind that we are polishing the last rough patches before we advertise 0.1 (Nov 16, 2020) Reproducible Analysis Pipeline for Data Streams (RAPIDS) allows you to process smartphone and wearable data to extract behavioral features (a.k.a. digital biomarkers/phenotypes). RAPIDS is open source, documented, modular, tested, and reproducible. At the moment we support smartphone data collected with AWARE and wearable data from Fitbit devices. Questions or feedback can be posted on #rapids in AWARE Framework's slack . Bugs should be reported on Github . Join our discussions on our algorithms and assumptions for feature processing . Ready to start? Go to Installation , then to Configuration , and then to Execution","title":"Welcome to RAPIDS documentation"},{"location":"#how-does-it-work","text":"RAPIDS is formed by R and Python scripts orchestrated by Snakemake . We suggest you read Snakemake\u2019s docs but in short: every link in the analysis chain is atomic and has files as input and output. Behavioral features are processed per sensor and per participant.","title":"How does it work?"},{"location":"#what-are-the-benefits-of-using-rapids","text":"Consistent analysis . Every participant sensor dataset is analyzed in the exact same way and isolated from each other. Efficient analysis . Every analysis step is executed only once. Whenever your data or configuration changes only the affected files are updated. Parallel execution . Thanks to Snakemake, your analysis can be executed over multiple cores without changing your code. Extensible code . You can easily add your own behavioral features in R or Python and keep authorship and citations. Timezone aware . Your data is adjusted to the specified timezone (multiple timezones suport coming soon ). Flexible time segments . You can extract behavioral features on time windows of any length (e.g. 5 minutes, 3 hours, 2 days), on every day or particular days (e.g. weekends, Mondays, the 1 st of each month, etc.) or around events of interest (e.g. surveys or clinical relapses). Tested code . We are constantly adding tests to make sure our behavioral features are correct. Reproducible code . You can be sure your code will run in other computers as intended thanks to R and Python virtual environments. You can share your analysis code along your publications without any overhead. Private . All your data is processed locally.","title":"What are the benefits of using RAPIDS?"},{"location":"#how-is-it-organized","text":"In broad terms the config.yaml , .env file , participants files , time segment files are the only ones that you will have to modify. All data is stored in data/ and all scripts are stored in src/ . For more information see RAPIDS\u2019 File Structure .","title":"How is it organized?"},{"location":"citation/","text":"Cite RAPIDS and providers \u00b6 RAPIDS and the community RAPIDS is a community effort and as such we want to continue recognizing the contributions from other researchers. Besides citing RAPIDS, we ask you to cite any of the authors listed below if you used those sensor providers in your analysis, thank you! RAPIDS \u00b6 If you used RAPIDS, please cite this paper . RAPIDS et al. citation Vega J, Li M, Aguillera K, Goel N, Joshi E, Durica KC, Kunta AR, Low CA RAPIDS: Reproducible Analysis Pipeline for Data Streams Collected with Mobile Devices JMIR Preprints. 18/08/2020:23246 DOI: 10.2196/preprints.23246 URL: https://preprints.jmir.org/preprint/23246 Panda (accelerometer) \u00b6 If you computed accelerometer features using the provider [PHONE_ACCLEROMETER][PANDA] cite this paper in addition to RAPIDS. Panda et al. citation Panda N, Solsky I, Huang EJ, Lipsitz S, Pradarelli JC, Delisle M, Cusack JC, Gadd MA, Lubitz CC, Mullen JT, Qadan M, Smith BL, Specht M, Stephen AE, Tanabe KK, Gawande AA, Onnela JP, Haynes AB. Using Smartphones to Capture Novel Recovery Metrics After Cancer Surgery. JAMA Surg. 2020 Feb 1;155(2):123-129. doi: 10.1001/jamasurg.2019.4702. PMID: 31657854; PMCID: PMC6820047. Stachl (applications foreground) \u00b6 If you computed applications foreground features using the app category (genre) catalogue in [PHONE_APPLICATIONS_FOREGROUND][RAPIDS] cite this paper in addition to RAPIDS. Stachl et al. citation Clemens Stachl, Quay Au, Ramona Schoedel, Samuel D. Gosling, Gabriella M. Harari, Daniel Buschek, Sarah Theres V\u00f6lkel, Tobias Schuwerk, Michelle Oldemeier, Theresa Ullmann, Heinrich Hussmann, Bernd Bischl, Markus B\u00fchner. Proceedings of the National Academy of Sciences Jul 2020, 117 (30) 17680-17687; DOI: 10.1073/pnas.1920484117 Barnett (locations) \u00b6 If you computed locations features using the provider [PHONE_LOCATIONS][BARNETT] cite this paper and this paper in addition to RAPIDS. Barnett et al. citation Ian Barnett, Jukka-Pekka Onnela, Inferring mobility measures from GPS traces with missing data, Biostatistics, Volume 21, Issue 2, April 2020, Pages e98\u2013e112, https://doi.org/10.1093/biostatistics/kxy059 Canzian et al. citation Luca Canzian and Mirco Musolesi. 2015. Trajectories of depression: unobtrusive monitoring of depressive states by means of smartphone mobility traces analysis. In Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp \u201815). Association for Computing Machinery, New York, NY, USA, 1293\u20131304. DOI: https://doi.org/10.1145/2750858.2805845 Doryab (locations) \u00b6 If you computed locations features using the provider [PHONE_LOCATIONS][DORYAB] cite this paper and this paper in addition to RAPIDS. Doryab et al. citation Doryab, A., Chikarsel, P., Liu, X., & Dey, A. K. (2019). Extraction of Behavioral Features from Smartphone and Wearable Data. ArXiv:1812.10394 [Cs, Stat]. http://arxiv.org/abs/1812.10394 Canzian et al. citation Luca Canzian and Mirco Musolesi. 2015. Trajectories of depression: unobtrusive monitoring of depressive states by means of smartphone mobility traces analysis. In Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp \u201815). Association for Computing Machinery, New York, NY, USA, 1293\u20131304. DOI: https://doi.org/10.1145/2750858.2805845","title":"Citation"},{"location":"citation/#cite-rapids-and-providers","text":"RAPIDS and the community RAPIDS is a community effort and as such we want to continue recognizing the contributions from other researchers. Besides citing RAPIDS, we ask you to cite any of the authors listed below if you used those sensor providers in your analysis, thank you!","title":"Cite RAPIDS and providers"},{"location":"citation/#rapids","text":"If you used RAPIDS, please cite this paper . RAPIDS et al. citation Vega J, Li M, Aguillera K, Goel N, Joshi E, Durica KC, Kunta AR, Low CA RAPIDS: Reproducible Analysis Pipeline for Data Streams Collected with Mobile Devices JMIR Preprints. 18/08/2020:23246 DOI: 10.2196/preprints.23246 URL: https://preprints.jmir.org/preprint/23246","title":"RAPIDS"},{"location":"citation/#panda-accelerometer","text":"If you computed accelerometer features using the provider [PHONE_ACCLEROMETER][PANDA] cite this paper in addition to RAPIDS. Panda et al. citation Panda N, Solsky I, Huang EJ, Lipsitz S, Pradarelli JC, Delisle M, Cusack JC, Gadd MA, Lubitz CC, Mullen JT, Qadan M, Smith BL, Specht M, Stephen AE, Tanabe KK, Gawande AA, Onnela JP, Haynes AB. Using Smartphones to Capture Novel Recovery Metrics After Cancer Surgery. JAMA Surg. 2020 Feb 1;155(2):123-129. doi: 10.1001/jamasurg.2019.4702. PMID: 31657854; PMCID: PMC6820047.","title":"Panda (accelerometer)"},{"location":"citation/#stachl-applications-foreground","text":"If you computed applications foreground features using the app category (genre) catalogue in [PHONE_APPLICATIONS_FOREGROUND][RAPIDS] cite this paper in addition to RAPIDS. Stachl et al. citation Clemens Stachl, Quay Au, Ramona Schoedel, Samuel D. Gosling, Gabriella M. Harari, Daniel Buschek, Sarah Theres V\u00f6lkel, Tobias Schuwerk, Michelle Oldemeier, Theresa Ullmann, Heinrich Hussmann, Bernd Bischl, Markus B\u00fchner. Proceedings of the National Academy of Sciences Jul 2020, 117 (30) 17680-17687; DOI: 10.1073/pnas.1920484117","title":"Stachl (applications foreground)"},{"location":"citation/#barnett-locations","text":"If you computed locations features using the provider [PHONE_LOCATIONS][BARNETT] cite this paper and this paper in addition to RAPIDS. Barnett et al. citation Ian Barnett, Jukka-Pekka Onnela, Inferring mobility measures from GPS traces with missing data, Biostatistics, Volume 21, Issue 2, April 2020, Pages e98\u2013e112, https://doi.org/10.1093/biostatistics/kxy059 Canzian et al. citation Luca Canzian and Mirco Musolesi. 2015. Trajectories of depression: unobtrusive monitoring of depressive states by means of smartphone mobility traces analysis. In Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp \u201815). Association for Computing Machinery, New York, NY, USA, 1293\u20131304. DOI: https://doi.org/10.1145/2750858.2805845","title":"Barnett (locations)"},{"location":"citation/#doryab-locations","text":"If you computed locations features using the provider [PHONE_LOCATIONS][DORYAB] cite this paper and this paper in addition to RAPIDS. Doryab et al. citation Doryab, A., Chikarsel, P., Liu, X., & Dey, A. K. (2019). Extraction of Behavioral Features from Smartphone and Wearable Data. ArXiv:1812.10394 [Cs, Stat]. http://arxiv.org/abs/1812.10394 Canzian et al. citation Luca Canzian and Mirco Musolesi. 2015. Trajectories of depression: unobtrusive monitoring of depressive states by means of smartphone mobility traces analysis. In Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp \u201815). Association for Computing Machinery, New York, NY, USA, 1293\u20131304. DOI: https://doi.org/10.1145/2750858.2805845","title":"Doryab (locations)"},{"location":"code_of_conduct/","text":"Contributor Covenant Code of Conduct \u00b6 Our Pledge \u00b6 We as members, contributors, and leaders pledge to make participation in our community a harassment-free experience for everyone, regardless of age, body size, visible or invisible disability, ethnicity, sex characteristics, gender identity and expression, level of experience, education, socio-economic status, nationality, personal appearance, race, religion, or sexual identity and orientation. We pledge to act and interact in ways that contribute to an open, welcoming, diverse, inclusive, and healthy community. Our Standards \u00b6 Examples of behavior that contributes to a positive environment for our community include: Demonstrating empathy and kindness toward other people Being respectful of differing opinions, viewpoints, and experiences Giving and gracefully accepting constructive feedback Accepting responsibility and apologizing to those affected by our mistakes, and learning from the experience Focusing on what is best not just for us as individuals, but for the overall community Examples of unacceptable behavior include: The use of sexualized language or imagery, and sexual attention or advances of any kind Trolling, insulting or derogatory comments, and personal or political attacks Public or private harassment Publishing others\u2019 private information, such as a physical or email address, without their explicit permission Other conduct which could reasonably be considered inappropriate in a professional setting Enforcement Responsibilities \u00b6 Community leaders are responsible for clarifying and enforcing our standards of acceptable behavior and will take appropriate and fair corrective action in response to any behavior that they deem inappropriate, threatening, offensive, or harmful. Community leaders have the right and responsibility to remove, edit, or reject comments, commits, code, wiki edits, issues, and other contributions that are not aligned to this Code of Conduct, and will communicate reasons for moderation decisions when appropriate. Scope \u00b6 This Code of Conduct applies within all community spaces, and also applies when an individual is officially representing the community in public spaces. Examples of representing our community include using an official e-mail address, posting via an official social media account, or acting as an appointed representative at an online or offline event. Enforcement \u00b6 Instances of abusive, harassing, or otherwise unacceptable behavior may be reported to the community leaders responsible for enforcement at moshi@pitt.edu . All complaints will be reviewed and investigated promptly and fairly. All community leaders are obligated to respect the privacy and security of the reporter of any incident. Enforcement Guidelines \u00b6 Community leaders will follow these Community Impact Guidelines in determining the consequences for any action they deem in violation of this Code of Conduct: 1. Correction \u00b6 Community Impact : Use of inappropriate language or other behavior deemed unprofessional or unwelcome in the community. Consequence : A private, written warning from community leaders, providing clarity around the nature of the violation and an explanation of why the behavior was inappropriate. A public apology may be requested. 2. Warning \u00b6 Community Impact : A violation through a single incident or series of actions. Consequence : A warning with consequences for continued behavior. No interaction with the people involved, including unsolicited interaction with those enforcing the Code of Conduct, for a specified period of time. This includes avoiding interactions in community spaces as well as external channels like social media. Violating these terms may lead to a temporary or permanent ban. 3. Temporary Ban \u00b6 Community Impact : A serious violation of community standards, including sustained inappropriate behavior. Consequence : A temporary ban from any sort of interaction or public communication with the community for a specified period of time. No public or private interaction with the people involved, including unsolicited interaction with those enforcing the Code of Conduct, is allowed during this period. Violating these terms may lead to a permanent ban. 4. Permanent Ban \u00b6 Community Impact : Demonstrating a pattern of violation of community standards, including sustained inappropriate behavior, harassment of an individual, or aggression toward or disparagement of classes of individuals. Consequence : A permanent ban from any sort of public interaction within the community. Attribution \u00b6 This Code of Conduct is adapted from the Contributor Covenant , version 2.0, available at https://www.contributor-covenant.org/version/2/0/code_of_conduct.html . Community Impact Guidelines were inspired by Mozilla\u2019s code of conduct enforcement ladder . For answers to common questions about this code of conduct, see the FAQ at https://www.contributor-covenant.org/faq . Translations are available at https://www.contributor-covenant.org/translations .","title":"Code of Conduct"},{"location":"code_of_conduct/#contributor-covenant-code-of-conduct","text":"","title":"Contributor Covenant Code of Conduct"},{"location":"code_of_conduct/#our-pledge","text":"We as members, contributors, and leaders pledge to make participation in our community a harassment-free experience for everyone, regardless of age, body size, visible or invisible disability, ethnicity, sex characteristics, gender identity and expression, level of experience, education, socio-economic status, nationality, personal appearance, race, religion, or sexual identity and orientation. We pledge to act and interact in ways that contribute to an open, welcoming, diverse, inclusive, and healthy community.","title":"Our Pledge"},{"location":"code_of_conduct/#our-standards","text":"Examples of behavior that contributes to a positive environment for our community include: Demonstrating empathy and kindness toward other people Being respectful of differing opinions, viewpoints, and experiences Giving and gracefully accepting constructive feedback Accepting responsibility and apologizing to those affected by our mistakes, and learning from the experience Focusing on what is best not just for us as individuals, but for the overall community Examples of unacceptable behavior include: The use of sexualized language or imagery, and sexual attention or advances of any kind Trolling, insulting or derogatory comments, and personal or political attacks Public or private harassment Publishing others\u2019 private information, such as a physical or email address, without their explicit permission Other conduct which could reasonably be considered inappropriate in a professional setting","title":"Our Standards"},{"location":"code_of_conduct/#enforcement-responsibilities","text":"Community leaders are responsible for clarifying and enforcing our standards of acceptable behavior and will take appropriate and fair corrective action in response to any behavior that they deem inappropriate, threatening, offensive, or harmful. Community leaders have the right and responsibility to remove, edit, or reject comments, commits, code, wiki edits, issues, and other contributions that are not aligned to this Code of Conduct, and will communicate reasons for moderation decisions when appropriate.","title":"Enforcement Responsibilities"},{"location":"code_of_conduct/#scope","text":"This Code of Conduct applies within all community spaces, and also applies when an individual is officially representing the community in public spaces. Examples of representing our community include using an official e-mail address, posting via an official social media account, or acting as an appointed representative at an online or offline event.","title":"Scope"},{"location":"code_of_conduct/#enforcement","text":"Instances of abusive, harassing, or otherwise unacceptable behavior may be reported to the community leaders responsible for enforcement at moshi@pitt.edu . All complaints will be reviewed and investigated promptly and fairly. All community leaders are obligated to respect the privacy and security of the reporter of any incident.","title":"Enforcement"},{"location":"code_of_conduct/#enforcement-guidelines","text":"Community leaders will follow these Community Impact Guidelines in determining the consequences for any action they deem in violation of this Code of Conduct:","title":"Enforcement Guidelines"},{"location":"code_of_conduct/#1-correction","text":"Community Impact : Use of inappropriate language or other behavior deemed unprofessional or unwelcome in the community. Consequence : A private, written warning from community leaders, providing clarity around the nature of the violation and an explanation of why the behavior was inappropriate. A public apology may be requested.","title":"1. Correction"},{"location":"code_of_conduct/#2-warning","text":"Community Impact : A violation through a single incident or series of actions. Consequence : A warning with consequences for continued behavior. No interaction with the people involved, including unsolicited interaction with those enforcing the Code of Conduct, for a specified period of time. This includes avoiding interactions in community spaces as well as external channels like social media. Violating these terms may lead to a temporary or permanent ban.","title":"2. Warning"},{"location":"code_of_conduct/#3-temporary-ban","text":"Community Impact : A serious violation of community standards, including sustained inappropriate behavior. Consequence : A temporary ban from any sort of interaction or public communication with the community for a specified period of time. No public or private interaction with the people involved, including unsolicited interaction with those enforcing the Code of Conduct, is allowed during this period. Violating these terms may lead to a permanent ban.","title":"3. Temporary Ban"},{"location":"code_of_conduct/#4-permanent-ban","text":"Community Impact : Demonstrating a pattern of violation of community standards, including sustained inappropriate behavior, harassment of an individual, or aggression toward or disparagement of classes of individuals. Consequence : A permanent ban from any sort of public interaction within the community.","title":"4. Permanent Ban"},{"location":"code_of_conduct/#attribution","text":"This Code of Conduct is adapted from the Contributor Covenant , version 2.0, available at https://www.contributor-covenant.org/version/2/0/code_of_conduct.html . Community Impact Guidelines were inspired by Mozilla\u2019s code of conduct enforcement ladder . For answers to common questions about this code of conduct, see the FAQ at https://www.contributor-covenant.org/faq . Translations are available at https://www.contributor-covenant.org/translations .","title":"Attribution"},{"location":"faq/","text":"Frequently Asked Questions \u00b6 Cannot connect to your MySQL server \u00b6 Problem **Error in .local ( drv, \\. .. ) :** **Failed to connect to database: Error: Can \\' t initialize character set unknown ( path: compiled \\_ in ) ** : Calls: dbConnect -> dbConnect -> .local -> .Call Execution halted [ Tue Mar 10 19 :40:15 2020 ] Error in rule download_dataset: jobid: 531 output: data/raw/p60/locations_raw.csv RuleException: CalledProcessError in line 20 of /home/ubuntu/rapids/rules/preprocessing.snakefile: Command 'set -euo pipefail; Rscript --vanilla /home/ubuntu/rapids/.snakemake/scripts/tmp_2jnvqs7.download_dataset.R' returned non-zero exit status 1 . File \"/home/ubuntu/rapids/rules/preprocessing.snakefile\" , line 20 , in __rule_download_dataset File \"/home/ubuntu/anaconda3/envs/moshi-env/lib/python3.7/concurrent/futures/thread.py\" , line 57 , in run Shutting down, this might take some time. Exiting because a job execution failed. Look above for error message Solution Please make sure the DATABASE_GROUP in config.yaml matches your DB credentials group in .env . Cannot start mysql in linux via brew services start mysql \u00b6 Problem Cannot start mysql in linux via brew services start mysql Solution Use mysql.server start Every time I run force the download_dataset rule all rules are executed \u00b6 Problem When running snakemake -j1 -R download_phone_data or ./rapids -j1 -R download_phone_data all the rules and files are re-computed Solution This is expected behavior. The advantage of using snakemake under the hood is that every time a file containing data is modified every rule that depends on that file will be re-executed to update their results. In this case, since download_dataset updates all the raw data, and you are forcing the rule with the flag -R every single rule that depends on those raw files will be executed. Error Table XXX doesn't exist while running the download_phone_data or download_fitbit_data rule. \u00b6 Problem Error in .local ( conn, statement, ... ) : could not run statement: Table 'db_name.table_name' doesn ' t exist Calls: colnames ... .local -> dbSendQuery -> dbSendQuery -> .local -> .Call Execution halted Solution Please make sure the sensors listed in [PHONE_VALID_SENSED_BINS][PHONE_SENSORS] and the [TABLE] of each sensor you activated in config.yaml match your database tables. How do I install RAPIDS on Ubuntu 16.04 \u00b6 Solution Install dependencies (Homebrew - if not installed): sudo apt-get install libmariadb-client-lgpl-dev libxml2-dev libssl-dev Install brew for linux and add the following line to ~/.bashrc : export PATH=$HOME/.linuxbrew/bin:$PATH source ~/.bashrc Install MySQL brew install mysql brew services start mysql Install R, pandoc and rmarkdown: brew install r brew install gcc@6 (needed due to this bug ) HOMEBREW_CC=gcc-6 brew install pandoc Install miniconda using these instructions Clone our repo: git clone https://github.com/carissalow/rapids Create a python virtual environment: cd rapids conda env create -f environment.yml -n MY_ENV_NAME conda activate MY_ENV_NAME Install R packages and virtual environment: snakemake renv_install snakemake renv_init snakemake renv_restore This step could take several minutes to complete. Please be patient and let it run until completion. mysql.h cannot be found \u00b6 Problem -------------------------- [ ERROR MESSAGE ] ---------------------------- <stdin>:1:10: fatal error: mysql.h: No such file or directory compilation terminated. ----------------------------------------------------------------------- ERROR: configuration failed for package 'RMySQL' Solution sudo apt install libmariadbclient-dev No package libcurl found \u00b6 Problem libcurl cannot be found Solution Install libcurl sudo apt install libcurl4-openssl-dev Configuration failed because openssl was not found. \u00b6 Problem openssl cannot be found Solution Install openssl sudo apt install libssl-dev Configuration failed because libxml-2.0 was not found \u00b6 Problem libxml-2.0 cannot be found Solution Install libxml-2.0 sudo apt install libxml2-dev SSL connection error when running RAPIDS \u00b6 Problem You are getting the following error message when running RAPIDS: Error: Failed to connect: SSL connection error: error:1425F102:SSL routines:ssl_choose_client_version:unsupported protocol. Solution This is a bug in Ubuntu 20.04 when trying to connect to an old MySQL server with MySQL client 8.0. You should get the same error message if you try to connect from the command line. There you can add the option --ssl-mode=DISABLED but we can't do this from the R connector. If you can't update your server, the quickest solution would be to import your database to another server or to a local environment. Alternatively, you could replace mysql-client and libmysqlclient-dev with mariadb-client and libmariadbclient-dev and reinstall renv. More info about this issue here DB_TABLES key not found \u00b6 Problem If you get the following error KeyError in line 43 of preprocessing.smk: 'PHONE_SENSORS' , it means that the indentation of the key [PHONE_SENSORS] is not matching the other child elements of PHONE_VALID_SENSED_BINS Solution You need to add or remove any leading whitespaces as needed on that line. PHONE_VALID_SENSED_BINS : COMPUTE : False # This flag is automatically ignored (set to True) if you are extracting PHONE_VALID_SENSED_DAYS or screen or Barnett's location features BIN_SIZE : &bin_size 5 # (in minutes) PHONE_SENSORS : [] Error while updating your conda environment in Ubuntu \u00b6 Problem You get the following error: CondaMultiError: CondaVerificationError: The package for tk located at /home/ubuntu/miniconda2/pkgs/tk-8.6.9-hed695b0_1003 appears to be corrupted. The path 'include/mysqlStubs.h' specified in the package manifest cannot be found. ClobberError: This transaction has incompatible packages due to a shared path. packages: conda-forge/linux-64::llvm-openmp-10.0.0-hc9558a2_0, anaconda/linux-64::intel-openmp-2019.4-243 path: 'lib/libiomp5.so' Solution Reinstall conda","title":"Frequently Asked Questions"},{"location":"faq/#frequently-asked-questions","text":"","title":"Frequently Asked Questions"},{"location":"faq/#cannot-connect-to-your-mysql-server","text":"Problem **Error in .local ( drv, \\. .. ) :** **Failed to connect to database: Error: Can \\' t initialize character set unknown ( path: compiled \\_ in ) ** : Calls: dbConnect -> dbConnect -> .local -> .Call Execution halted [ Tue Mar 10 19 :40:15 2020 ] Error in rule download_dataset: jobid: 531 output: data/raw/p60/locations_raw.csv RuleException: CalledProcessError in line 20 of /home/ubuntu/rapids/rules/preprocessing.snakefile: Command 'set -euo pipefail; Rscript --vanilla /home/ubuntu/rapids/.snakemake/scripts/tmp_2jnvqs7.download_dataset.R' returned non-zero exit status 1 . File \"/home/ubuntu/rapids/rules/preprocessing.snakefile\" , line 20 , in __rule_download_dataset File \"/home/ubuntu/anaconda3/envs/moshi-env/lib/python3.7/concurrent/futures/thread.py\" , line 57 , in run Shutting down, this might take some time. Exiting because a job execution failed. Look above for error message Solution Please make sure the DATABASE_GROUP in config.yaml matches your DB credentials group in .env .","title":"Cannot connect to your MySQL server"},{"location":"faq/#cannot-start-mysql-in-linux-via-brew-services-start-mysql","text":"Problem Cannot start mysql in linux via brew services start mysql Solution Use mysql.server start","title":"Cannot start mysql in linux via brew services start mysql"},{"location":"faq/#every-time-i-run-force-the-download_dataset-rule-all-rules-are-executed","text":"Problem When running snakemake -j1 -R download_phone_data or ./rapids -j1 -R download_phone_data all the rules and files are re-computed Solution This is expected behavior. The advantage of using snakemake under the hood is that every time a file containing data is modified every rule that depends on that file will be re-executed to update their results. In this case, since download_dataset updates all the raw data, and you are forcing the rule with the flag -R every single rule that depends on those raw files will be executed.","title":"Every time I run force the download_dataset rule all rules are executed"},{"location":"faq/#error-table-xxx-doesnt-exist-while-running-the-download_phone_data-or-download_fitbit_data-rule","text":"Problem Error in .local ( conn, statement, ... ) : could not run statement: Table 'db_name.table_name' doesn ' t exist Calls: colnames ... .local -> dbSendQuery -> dbSendQuery -> .local -> .Call Execution halted Solution Please make sure the sensors listed in [PHONE_VALID_SENSED_BINS][PHONE_SENSORS] and the [TABLE] of each sensor you activated in config.yaml match your database tables.","title":"Error Table XXX doesn't exist while running the download_phone_data or download_fitbit_data rule."},{"location":"faq/#how-do-i-install-rapids-on-ubuntu-1604","text":"Solution Install dependencies (Homebrew - if not installed): sudo apt-get install libmariadb-client-lgpl-dev libxml2-dev libssl-dev Install brew for linux and add the following line to ~/.bashrc : export PATH=$HOME/.linuxbrew/bin:$PATH source ~/.bashrc Install MySQL brew install mysql brew services start mysql Install R, pandoc and rmarkdown: brew install r brew install gcc@6 (needed due to this bug ) HOMEBREW_CC=gcc-6 brew install pandoc Install miniconda using these instructions Clone our repo: git clone https://github.com/carissalow/rapids Create a python virtual environment: cd rapids conda env create -f environment.yml -n MY_ENV_NAME conda activate MY_ENV_NAME Install R packages and virtual environment: snakemake renv_install snakemake renv_init snakemake renv_restore This step could take several minutes to complete. Please be patient and let it run until completion.","title":"How do I install RAPIDS on Ubuntu 16.04"},{"location":"faq/#mysqlh-cannot-be-found","text":"Problem -------------------------- [ ERROR MESSAGE ] ---------------------------- <stdin>:1:10: fatal error: mysql.h: No such file or directory compilation terminated. ----------------------------------------------------------------------- ERROR: configuration failed for package 'RMySQL' Solution sudo apt install libmariadbclient-dev","title":"mysql.h cannot be found"},{"location":"faq/#no-package-libcurl-found","text":"Problem libcurl cannot be found Solution Install libcurl sudo apt install libcurl4-openssl-dev","title":"No package libcurl found"},{"location":"faq/#configuration-failed-because-openssl-was-not-found","text":"Problem openssl cannot be found Solution Install openssl sudo apt install libssl-dev","title":"Configuration failed because openssl was not found."},{"location":"faq/#configuration-failed-because-libxml-20-was-not-found","text":"Problem libxml-2.0 cannot be found Solution Install libxml-2.0 sudo apt install libxml2-dev","title":"Configuration failed because libxml-2.0 was not found"},{"location":"faq/#ssl-connection-error-when-running-rapids","text":"Problem You are getting the following error message when running RAPIDS: Error: Failed to connect: SSL connection error: error:1425F102:SSL routines:ssl_choose_client_version:unsupported protocol. Solution This is a bug in Ubuntu 20.04 when trying to connect to an old MySQL server with MySQL client 8.0. You should get the same error message if you try to connect from the command line. There you can add the option --ssl-mode=DISABLED but we can't do this from the R connector. If you can't update your server, the quickest solution would be to import your database to another server or to a local environment. Alternatively, you could replace mysql-client and libmysqlclient-dev with mariadb-client and libmariadbclient-dev and reinstall renv. More info about this issue here","title":"SSL connection error when running RAPIDS"},{"location":"faq/#db_tables-key-not-found","text":"Problem If you get the following error KeyError in line 43 of preprocessing.smk: 'PHONE_SENSORS' , it means that the indentation of the key [PHONE_SENSORS] is not matching the other child elements of PHONE_VALID_SENSED_BINS Solution You need to add or remove any leading whitespaces as needed on that line. PHONE_VALID_SENSED_BINS : COMPUTE : False # This flag is automatically ignored (set to True) if you are extracting PHONE_VALID_SENSED_DAYS or screen or Barnett's location features BIN_SIZE : &bin_size 5 # (in minutes) PHONE_SENSORS : []","title":"DB_TABLES key not found"},{"location":"faq/#error-while-updating-your-conda-environment-in-ubuntu","text":"Problem You get the following error: CondaMultiError: CondaVerificationError: The package for tk located at /home/ubuntu/miniconda2/pkgs/tk-8.6.9-hed695b0_1003 appears to be corrupted. The path 'include/mysqlStubs.h' specified in the package manifest cannot be found. ClobberError: This transaction has incompatible packages due to a shared path. packages: conda-forge/linux-64::llvm-openmp-10.0.0-hc9558a2_0, anaconda/linux-64::intel-openmp-2019.4-243 path: 'lib/libiomp5.so' Solution Reinstall conda","title":"Error while updating your conda environment in Ubuntu"},{"location":"file-structure/","text":"File Structure \u00b6 Tip Read this page if you want to learn more about how RAPIDS is structured. If you want to start using it go to Installation , then to Configuration , and then to Execution All paths mentioned in this page are relative to RAPIDS\u2019 root folder. If you want to extract the behavioral features that RAPIDS offers, you will only have to create or modify the .env file , participants files , time segment files , and the config.yaml file as instructed in the Configuration page . The config.yaml file is the heart of RAPIDS and includes parameters to manage participants, data sources, sensor data, visualizations and more. All data is saved in data/ . The data/external/ folder stores any data imported or created by the user, data/raw/ stores sensor data as imported from your database, data/interim/ has intermediate files necessary to compute behavioral features from raw data, and data/processed/ has all the final files with the behavioral features in folders per participant and sensor. RAPIDS source code is saved in src/ . The src/data/ folder stores scripts to download, clean and pre-process sensor data, src/features has scripts to extract behavioral features organized in their respective sensor subfolders , src/models/ can host any script to create models or statistical analyses with the behavioral features you extract, and src/visualization/ has scripts to create plots of the raw and processed data. There are other files and folders but only relevant if you are interested in extending RAPIDS (e.g. virtual env files, docs, tests, Dockerfile, the Snakefile, etc.). In the figure below, we represent the interactions between users and files. After a user modifies the configuration files mentioned above, the Snakefile file will search for and execute the Snakemake rules that contain the Python or R scripts necessary to generate or update the required output files (behavioral features, plots, etc.). Interaction diagram between the user, and important files in RAPIDS","title":"File Structure"},{"location":"file-structure/#file-structure","text":"Tip Read this page if you want to learn more about how RAPIDS is structured. If you want to start using it go to Installation , then to Configuration , and then to Execution All paths mentioned in this page are relative to RAPIDS\u2019 root folder. If you want to extract the behavioral features that RAPIDS offers, you will only have to create or modify the .env file , participants files , time segment files , and the config.yaml file as instructed in the Configuration page . The config.yaml file is the heart of RAPIDS and includes parameters to manage participants, data sources, sensor data, visualizations and more. All data is saved in data/ . The data/external/ folder stores any data imported or created by the user, data/raw/ stores sensor data as imported from your database, data/interim/ has intermediate files necessary to compute behavioral features from raw data, and data/processed/ has all the final files with the behavioral features in folders per participant and sensor. RAPIDS source code is saved in src/ . The src/data/ folder stores scripts to download, clean and pre-process sensor data, src/features has scripts to extract behavioral features organized in their respective sensor subfolders , src/models/ can host any script to create models or statistical analyses with the behavioral features you extract, and src/visualization/ has scripts to create plots of the raw and processed data. There are other files and folders but only relevant if you are interested in extending RAPIDS (e.g. virtual env files, docs, tests, Dockerfile, the Snakefile, etc.). In the figure below, we represent the interactions between users and files. After a user modifies the configuration files mentioned above, the Snakefile file will search for and execute the Snakemake rules that contain the Python or R scripts necessary to generate or update the required output files (behavioral features, plots, etc.). Interaction diagram between the user, and important files in RAPIDS","title":"File Structure"},{"location":"team/","text":"RAPIDS Team \u00b6 If you are interested in contributing feel free to submit a pull request or contact us. Core Team \u00b6 Julio Vega (Designer and Lead Developer) \u00b6 About Julio Vega is a postdoctoral associate at the Mobile Sensing + Health Institute. He is interested in personalized methodologies to monitor chronic conditions that affect daily human behavior using mobile and wearable data. vegaju at upmc . edu Personal Website Meng Li \u00b6 About Meng Li received her Master of Science degree in Information Science from the University of Pittsburgh. She is interested in applying machine learning algorithms to the medical field. lim11 at upmc . edu Linkedin Profile Github Profile Abhineeth Reddy Kunta \u00b6 About Abhineeth Reddy Kunta is a Senior Software Engineer with the Mobile Sensing + Health Institute. He is experienced in software development and specializes in building solutions using machine learning. Abhineeth likes exploring ways to leverage technology in advancing medicine and education. Previously he worked as a Computer Programmer at Georgia Department of Public Health. He has a master\u2019s degree in Computer Science from George Mason University. Kwesi Aguillera \u00b6 About Kwesi Aguillera is currently in his first year at the University of Pittsburgh pursuing a Master of Sciences in Information Science specializing in Big Data Analytics. He received his Bachelor of Science degree in Computer Science and Management from the University of the West Indies. Kwesi considers himself a full stack developer and looks forward to applying this knowledge to big data analysis. Linkedin Profile Echhit Joshi \u00b6 About Echhit Joshi is a Masters student at the School of Computing and Information at University of Pittsburgh. His areas of interest are Machine/Deep Learning, Data Mining, and Analytics. Linkedin Profile Nicolas Leo \u00b6 About Nicolas is a rising senior studying computer science at the University of Pittsburgh. His academic interests include databases, machine learning, and application development. After completing his undergraduate degree, he plans to attend graduate school for a MS in Computer Science with a focus on Intelligent Systems. Nikunj Goel \u00b6 About Nik is a graduate student at the University of Pittsburgh pursuing Master of Science in Information Science. He earned his Bachelor of Technology degree in Information Technology from India. He is a Data Enthusiasts and passionate about finding the meaning out of raw data. In a long term, his goal is to create a breakthrough in Data Science and Deep Learning. Linkedin Profile Community Contributors \u00b6 Agam Kumar \u00b6 About Agam is a junior at Carnegie Mellon University studying Statistics and Machine Learning and pursuing an additional major in Computer Science. He is a member of the Data Science team in the Health and Human Performance Lab at CMU and has keen interests in software development and data science. His research interests include ML applications in medicine. Linkedin Profile Github Profile Yasaman S. Sefidgar \u00b6 About Linkedin Profile Advisors \u00b6 Afsaneh Doryab \u00b6 About Personal Website Carissa Low \u00b6 About Profile","title":"Team"},{"location":"team/#rapids-team","text":"If you are interested in contributing feel free to submit a pull request or contact us.","title":"RAPIDS Team"},{"location":"team/#core-team","text":"","title":"Core Team"},{"location":"team/#julio-vega-designer-and-lead-developer","text":"About Julio Vega is a postdoctoral associate at the Mobile Sensing + Health Institute. He is interested in personalized methodologies to monitor chronic conditions that affect daily human behavior using mobile and wearable data. vegaju at upmc . edu Personal Website","title":"Julio Vega (Designer and Lead Developer)"},{"location":"team/#meng-li","text":"About Meng Li received her Master of Science degree in Information Science from the University of Pittsburgh. She is interested in applying machine learning algorithms to the medical field. lim11 at upmc . edu Linkedin Profile Github Profile","title":"Meng Li"},{"location":"team/#abhineeth-reddy-kunta","text":"About Abhineeth Reddy Kunta is a Senior Software Engineer with the Mobile Sensing + Health Institute. He is experienced in software development and specializes in building solutions using machine learning. Abhineeth likes exploring ways to leverage technology in advancing medicine and education. Previously he worked as a Computer Programmer at Georgia Department of Public Health. He has a master\u2019s degree in Computer Science from George Mason University.","title":"Abhineeth Reddy Kunta"},{"location":"team/#kwesi-aguillera","text":"About Kwesi Aguillera is currently in his first year at the University of Pittsburgh pursuing a Master of Sciences in Information Science specializing in Big Data Analytics. He received his Bachelor of Science degree in Computer Science and Management from the University of the West Indies. Kwesi considers himself a full stack developer and looks forward to applying this knowledge to big data analysis. Linkedin Profile","title":"Kwesi Aguillera"},{"location":"team/#echhit-joshi","text":"About Echhit Joshi is a Masters student at the School of Computing and Information at University of Pittsburgh. His areas of interest are Machine/Deep Learning, Data Mining, and Analytics. Linkedin Profile","title":"Echhit Joshi"},{"location":"team/#nicolas-leo","text":"About Nicolas is a rising senior studying computer science at the University of Pittsburgh. His academic interests include databases, machine learning, and application development. After completing his undergraduate degree, he plans to attend graduate school for a MS in Computer Science with a focus on Intelligent Systems.","title":"Nicolas Leo"},{"location":"team/#nikunj-goel","text":"About Nik is a graduate student at the University of Pittsburgh pursuing Master of Science in Information Science. He earned his Bachelor of Technology degree in Information Technology from India. He is a Data Enthusiasts and passionate about finding the meaning out of raw data. In a long term, his goal is to create a breakthrough in Data Science and Deep Learning. Linkedin Profile","title":"Nikunj Goel"},{"location":"team/#community-contributors","text":"","title":"Community Contributors"},{"location":"team/#agam-kumar","text":"About Agam is a junior at Carnegie Mellon University studying Statistics and Machine Learning and pursuing an additional major in Computer Science. He is a member of the Data Science team in the Health and Human Performance Lab at CMU and has keen interests in software development and data science. His research interests include ML applications in medicine. Linkedin Profile Github Profile","title":"Agam Kumar"},{"location":"team/#yasaman-s-sefidgar","text":"About Linkedin Profile","title":"Yasaman S. Sefidgar"},{"location":"team/#advisors","text":"","title":"Advisors"},{"location":"team/#afsaneh-doryab","text":"About Personal Website","title":"Afsaneh Doryab"},{"location":"team/#carissa-low","text":"About Profile","title":"Carissa Low"},{"location":"developers/documentation/","text":"Documentation \u00b6 We use mkdocs with the material theme to write these docs. Whenever you make any changes, just push them back to the repo and the documentation will be deployed automatically. Set up development environment \u00b6 Make sure your conda environment is active pip install mkdocs pip install mkdocs-material Preview \u00b6 Run the following command in RAPIDS root folder and go to http://127.0.0.1:8000 : mkdocs serve File Structure \u00b6 The documentation config file is /mkdocs.yml , if you are adding new .md files to the docs modify the nav attribute at the bottom of that file. You can use the hierarchy there to find all the files that appear in the documentation. Reference \u00b6 Check this page to get familiar with the different visual elements we can use in the docs (admonitions, code blocks, tables, etc.) You can also refer to /docs/setup/installation.md and /docs/setup/configuration.md to see practical examples of these elements. Hint Any links to internal pages should be relative to the current page. For example, any link from this page (documentation) which is inside ./developers should begin with ../ to go one folder level up like: [ mylink ]( ../setup/installation.md ) Extras \u00b6 You can insert emojis using this syntax :[SOURCE]-[ICON_NAME] from the following sources: https://materialdesignicons.com/ https://fontawesome.com/icons/tasks?style=solid https://primer.style/octicons/ You can use this page to create markdown tables more easily","title":"Documentation"},{"location":"developers/documentation/#documentation","text":"We use mkdocs with the material theme to write these docs. Whenever you make any changes, just push them back to the repo and the documentation will be deployed automatically.","title":"Documentation"},{"location":"developers/documentation/#set-up-development-environment","text":"Make sure your conda environment is active pip install mkdocs pip install mkdocs-material","title":"Set up development environment"},{"location":"developers/documentation/#preview","text":"Run the following command in RAPIDS root folder and go to http://127.0.0.1:8000 : mkdocs serve","title":"Preview"},{"location":"developers/documentation/#file-structure","text":"The documentation config file is /mkdocs.yml , if you are adding new .md files to the docs modify the nav attribute at the bottom of that file. You can use the hierarchy there to find all the files that appear in the documentation.","title":"File Structure"},{"location":"developers/documentation/#reference","text":"Check this page to get familiar with the different visual elements we can use in the docs (admonitions, code blocks, tables, etc.) You can also refer to /docs/setup/installation.md and /docs/setup/configuration.md to see practical examples of these elements. Hint Any links to internal pages should be relative to the current page. For example, any link from this page (documentation) which is inside ./developers should begin with ../ to go one folder level up like: [ mylink ]( ../setup/installation.md )","title":"Reference"},{"location":"developers/documentation/#extras","text":"You can insert emojis using this syntax :[SOURCE]-[ICON_NAME] from the following sources: https://materialdesignicons.com/ https://fontawesome.com/icons/tasks?style=solid https://primer.style/octicons/ You can use this page to create markdown tables more easily","title":"Extras"},{"location":"developers/remote-support/","text":"Remote Support \u00b6 We use the Live Share extension of Visual Studio Code to debug bugs when sharing data or database credentials is not possible. Install Visual Studio Code Open you RAPIDS root folder in a new VSCode window Open a new Terminal Terminal > New terminal Install the Live Share extension pack Press Ctrl + P or Cmd + P and run this command: >live share: start collaboration session 6. Follow the instructions and share the session link you receive","title":"Remote Support"},{"location":"developers/remote-support/#remote-support","text":"We use the Live Share extension of Visual Studio Code to debug bugs when sharing data or database credentials is not possible. Install Visual Studio Code Open you RAPIDS root folder in a new VSCode window Open a new Terminal Terminal > New terminal Install the Live Share extension pack Press Ctrl + P or Cmd + P and run this command: >live share: start collaboration session 6. Follow the instructions and share the session link you receive","title":"Remote Support"},{"location":"developers/test-cases/","text":"Test Cases \u00b6 Along with the continued development and the addition of new sensors and features to the RAPIDS pipeline, tests for the currently available sensors and features are being implemented. Since this is a Work In Progress this page will be updated with the list of sensors and features for which testing is available. For each of the sensors listed a description of the data used for testing (test cases) are outline. Currently for all intent and testing purposes the tests/data/raw/test01/ contains all the test data files for testing android data formats and tests/data/raw/test02/ contains all the test data files for testing iOS data formats. It follows that the expected (verified output) are contained in the tests/data/processed/test01/ and tests/data/processed/test02/ for Android and iOS respectively. tests/data/raw/test03/ and tests/data/raw/test04/ contain data files for testing empty raw data files for android and iOS respectively. The following is a list of the sensors that testing is currently available. Messages (SMS) \u00b6 The raw message data file contains data for 2 separate days. The data for the first day contains records 5 records for every epoch . The second day's data contains 6 records for each of only 2 epoch (currently morning and evening ) The raw message data contains records for both message_types (i.e. recieved and sent ) in both days in all epochs. The number records with each message_types per epoch is randomly distributed There is at least one records with each message_types per epoch. There is one raw message data file each, as described above, for testing both iOS and Android data. There is also an additional empty data file for both android and iOS for testing empty data files Calls \u00b6 Due to the difference in the format of the raw call data for iOS and Android the following is the expected results the calls_with_datetime_unified.csv . This would give a better idea of the use cases being tested since the calls_with_datetime_unified.csv would make both the iOS and Android data comparable. The call data would contain data for 2 days. The data for the first day contains 6 records for every epoch . The second day's data contains 6 records for each of only 2 epoch (currently morning and evening ) The call data contains records for all call_types (i.e. incoming , outgoing and missed ) in both days in all epochs. The number records with each of the call_types per epoch is randomly distributed. There is at least one records with each call_types per epoch. There is one call data file each, as described above, for testing both iOS and Android data. There is also an additional empty data file for both android and iOS for testing empty data files Screen \u00b6 Due to the difference in the format of the raw screen data for iOS and Android the following is the expected results the screen_deltas.csv . This would give a better idea of the use cases being tested since the screen_eltas.csv would make both the iOS and Android data comparable These files are used to calculate the features for the screen sensor The screen delta data file contains data for 1 day. The screen delta data contains 1 record to represent an unlock episode that falls within an epoch for every epoch . The screen delta data contains 1 record to represent an unlock episode that falls across the boundary of 2 epochs. Namely the unlock episode starts in one epoch and ends in the next, thus there is a record for unlock episodes that fall across night to morning , morning to afternoon and finally afternoon to night The testing is done for unlock episode_type. There is one screen data file each for testing both iOS and Android data formats. There is also an additional empty data file for both android and iOS for testing empty data files Battery \u00b6 Due to the difference in the format of the raw battery data for iOS and Android as well as versions of iOS the following is the expected results the battery_deltas.csv . This would give a better idea of the use cases being tested since the battery_deltas.csv would make both the iOS and Android data comparable. These files are used to calculate the features for the battery sensor. The battery delta data file contains data for 1 day. The battery delta data contains 1 record each for a charging and discharging episode that falls within an epoch for every epoch . Thus, for the daily epoch there would be multiple charging and discharging episodes Since either a charging episode or a discharging episode and not both can occur across epochs, in order to test episodes that occur across epochs alternating episodes of charging and discharging episodes that fall across night to morning , morning to afternoon and finally afternoon to night are present in the battery delta data. This starts with a discharging episode that begins in night and end in morning . There is one battery data file each, for testing both iOS and Android data formats. There is also an additional empty data file for both android and iOS for testing empty data files Bluetooth \u00b6 The raw Bluetooth data file contains data for 1 day. The raw Bluetooth data contains at least 2 records for each epoch . Each epoch has a record with a timestamp for the beginning boundary for that epoch and a record with a timestamp for the ending boundary for that epoch . (e.g. For the morning epoch there is a record with a timestamp for 6:00AM and another record with a timestamp for 11:59:59AM . These are to test edge cases) An option of 5 Bluetooth devices are randomly distributed throughout the data records. There is one raw Bluetooth data file each, for testing both iOS and Android data formats. There is also an additional empty data file for both android and iOS for testing empty data files. WIFI \u00b6 There are 2 data files ( wifi_raw.csv and sensor_wifi_raw.csv ) for each fake participant for each phone platform. The raw WIFI data files contain data for 1 day. The sensor_wifi_raw.csv data contains at least 2 records for each epoch . Each epoch has a record with a timestamp for the beginning boundary for that epoch and a record with a timestamp for the ending boundary for that epoch . (e.g. For the morning epoch there is a record with a timestamp for 6:00AM and another record with a timestamp for 11:59:59AM . These are to test edge cases) The wifi_raw.csv data contains 3 records with random timestamps for each epoch to represent visible broadcasting WIFI network. This file is empty for the iOS phone testing data. An option of 10 access point devices is randomly distributed throughout the data records. 5 each for sensor_wifi_raw.csv and wifi_raw.csv . There data files for testing both iOS and Android data formats. There are also additional empty data files for both android and iOS for testing empty data files. Light \u00b6 The raw light data file contains data for 1 day. The raw light data contains 3 or 4 rows of data for each epoch except night . The single row of data for night is for testing features for single values inputs. (Example testing the standard deviation of one input value) Since light is only available for Android there is only one file that contains data for Android. All other files (i.e. for iPhone) are empty data files. Application Foreground \u00b6 The raw application foreground data file contains data for 1 day. The raw application foreground data contains 7 - 9 rows of data for each epoch . The records for each epoch contains apps that are randomly selected from a list of apps that are from the MULTIPLE_CATEGORIES and SINGLE_CATEGORIES (See testing_config.yaml ). There are also records in each epoch that have apps randomly selected from a list of apps that are from the EXCLUDED_CATEGORIES and EXCLUDED_APPS . This is to test that these apps are actually being excluded from the calculations of features. There are also records to test SINGLE_APPS calculations. Since application foreground is only available for Android there is only one file that contains data for Android. All other files (i.e. for iPhone) are empty data files. Activity Recognition \u00b6 The raw Activity Recognition data file contains data for 1 day. The raw Activity Recognition data each epoch period contains rows that records 2 - 5 different activity_types . The is such that durations of activities can be tested. Additionally, there are records that mimic the duration of an activity over the time boundary of neighboring epochs. (For example, there a set of records that mimic the participant in_vehicle from afternoon into evening ) There is one file each with raw Activity Recognition data for testing both iOS and Android data formats. (plugin_google_activity_recognition_raw.csv for android and plugin_ios_activity_recognition_raw.csv for iOS) There is also an additional empty data file for both android and iOS for testing empty data files. Conversation \u00b6 The raw conversation data file contains data for 2 day. The raw conversation data contains records with a sample of both datatypes (i.e. voice/noise = 0 , and conversation = 2 ) as well as rows with for samples of each of the inference values (i.e. silence = 0 , noise = 1 , voice = 2 , and unknown = 3 ) for each epoch . The different datatype and inference records are randomly distributed throughout the epoch . Additionally there are 2 - 5 records for conversations ( datatype = 2, and inference = -1) in each epoch and for each epoch except night, there is a conversation record that has a double_convo_start timestamp that is from the previous epoch . This is to test the calculations of features across epochs . There is a raw conversation data file for both android and iOS platforms ( plugin_studentlife_audio_android_raw.csv and plugin_studentlife_audio_raw.csv respectively). Finally, there are also additional empty data files for both android and iOS for testing empty data files","title":"Test cases"},{"location":"developers/test-cases/#test-cases","text":"Along with the continued development and the addition of new sensors and features to the RAPIDS pipeline, tests for the currently available sensors and features are being implemented. Since this is a Work In Progress this page will be updated with the list of sensors and features for which testing is available. For each of the sensors listed a description of the data used for testing (test cases) are outline. Currently for all intent and testing purposes the tests/data/raw/test01/ contains all the test data files for testing android data formats and tests/data/raw/test02/ contains all the test data files for testing iOS data formats. It follows that the expected (verified output) are contained in the tests/data/processed/test01/ and tests/data/processed/test02/ for Android and iOS respectively. tests/data/raw/test03/ and tests/data/raw/test04/ contain data files for testing empty raw data files for android and iOS respectively. The following is a list of the sensors that testing is currently available.","title":"Test Cases"},{"location":"developers/test-cases/#messages-sms","text":"The raw message data file contains data for 2 separate days. The data for the first day contains records 5 records for every epoch . The second day's data contains 6 records for each of only 2 epoch (currently morning and evening ) The raw message data contains records for both message_types (i.e. recieved and sent ) in both days in all epochs. The number records with each message_types per epoch is randomly distributed There is at least one records with each message_types per epoch. There is one raw message data file each, as described above, for testing both iOS and Android data. There is also an additional empty data file for both android and iOS for testing empty data files","title":"Messages (SMS)"},{"location":"developers/test-cases/#calls","text":"Due to the difference in the format of the raw call data for iOS and Android the following is the expected results the calls_with_datetime_unified.csv . This would give a better idea of the use cases being tested since the calls_with_datetime_unified.csv would make both the iOS and Android data comparable. The call data would contain data for 2 days. The data for the first day contains 6 records for every epoch . The second day's data contains 6 records for each of only 2 epoch (currently morning and evening ) The call data contains records for all call_types (i.e. incoming , outgoing and missed ) in both days in all epochs. The number records with each of the call_types per epoch is randomly distributed. There is at least one records with each call_types per epoch. There is one call data file each, as described above, for testing both iOS and Android data. There is also an additional empty data file for both android and iOS for testing empty data files","title":"Calls"},{"location":"developers/test-cases/#screen","text":"Due to the difference in the format of the raw screen data for iOS and Android the following is the expected results the screen_deltas.csv . This would give a better idea of the use cases being tested since the screen_eltas.csv would make both the iOS and Android data comparable These files are used to calculate the features for the screen sensor The screen delta data file contains data for 1 day. The screen delta data contains 1 record to represent an unlock episode that falls within an epoch for every epoch . The screen delta data contains 1 record to represent an unlock episode that falls across the boundary of 2 epochs. Namely the unlock episode starts in one epoch and ends in the next, thus there is a record for unlock episodes that fall across night to morning , morning to afternoon and finally afternoon to night The testing is done for unlock episode_type. There is one screen data file each for testing both iOS and Android data formats. There is also an additional empty data file for both android and iOS for testing empty data files","title":"Screen"},{"location":"developers/test-cases/#battery","text":"Due to the difference in the format of the raw battery data for iOS and Android as well as versions of iOS the following is the expected results the battery_deltas.csv . This would give a better idea of the use cases being tested since the battery_deltas.csv would make both the iOS and Android data comparable. These files are used to calculate the features for the battery sensor. The battery delta data file contains data for 1 day. The battery delta data contains 1 record each for a charging and discharging episode that falls within an epoch for every epoch . Thus, for the daily epoch there would be multiple charging and discharging episodes Since either a charging episode or a discharging episode and not both can occur across epochs, in order to test episodes that occur across epochs alternating episodes of charging and discharging episodes that fall across night to morning , morning to afternoon and finally afternoon to night are present in the battery delta data. This starts with a discharging episode that begins in night and end in morning . There is one battery data file each, for testing both iOS and Android data formats. There is also an additional empty data file for both android and iOS for testing empty data files","title":"Battery"},{"location":"developers/test-cases/#bluetooth","text":"The raw Bluetooth data file contains data for 1 day. The raw Bluetooth data contains at least 2 records for each epoch . Each epoch has a record with a timestamp for the beginning boundary for that epoch and a record with a timestamp for the ending boundary for that epoch . (e.g. For the morning epoch there is a record with a timestamp for 6:00AM and another record with a timestamp for 11:59:59AM . These are to test edge cases) An option of 5 Bluetooth devices are randomly distributed throughout the data records. There is one raw Bluetooth data file each, for testing both iOS and Android data formats. There is also an additional empty data file for both android and iOS for testing empty data files.","title":"Bluetooth"},{"location":"developers/test-cases/#wifi","text":"There are 2 data files ( wifi_raw.csv and sensor_wifi_raw.csv ) for each fake participant for each phone platform. The raw WIFI data files contain data for 1 day. The sensor_wifi_raw.csv data contains at least 2 records for each epoch . Each epoch has a record with a timestamp for the beginning boundary for that epoch and a record with a timestamp for the ending boundary for that epoch . (e.g. For the morning epoch there is a record with a timestamp for 6:00AM and another record with a timestamp for 11:59:59AM . These are to test edge cases) The wifi_raw.csv data contains 3 records with random timestamps for each epoch to represent visible broadcasting WIFI network. This file is empty for the iOS phone testing data. An option of 10 access point devices is randomly distributed throughout the data records. 5 each for sensor_wifi_raw.csv and wifi_raw.csv . There data files for testing both iOS and Android data formats. There are also additional empty data files for both android and iOS for testing empty data files.","title":"WIFI"},{"location":"developers/test-cases/#light","text":"The raw light data file contains data for 1 day. The raw light data contains 3 or 4 rows of data for each epoch except night . The single row of data for night is for testing features for single values inputs. (Example testing the standard deviation of one input value) Since light is only available for Android there is only one file that contains data for Android. All other files (i.e. for iPhone) are empty data files.","title":"Light"},{"location":"developers/test-cases/#application-foreground","text":"The raw application foreground data file contains data for 1 day. The raw application foreground data contains 7 - 9 rows of data for each epoch . The records for each epoch contains apps that are randomly selected from a list of apps that are from the MULTIPLE_CATEGORIES and SINGLE_CATEGORIES (See testing_config.yaml ). There are also records in each epoch that have apps randomly selected from a list of apps that are from the EXCLUDED_CATEGORIES and EXCLUDED_APPS . This is to test that these apps are actually being excluded from the calculations of features. There are also records to test SINGLE_APPS calculations. Since application foreground is only available for Android there is only one file that contains data for Android. All other files (i.e. for iPhone) are empty data files.","title":"Application Foreground"},{"location":"developers/test-cases/#activity-recognition","text":"The raw Activity Recognition data file contains data for 1 day. The raw Activity Recognition data each epoch period contains rows that records 2 - 5 different activity_types . The is such that durations of activities can be tested. Additionally, there are records that mimic the duration of an activity over the time boundary of neighboring epochs. (For example, there a set of records that mimic the participant in_vehicle from afternoon into evening ) There is one file each with raw Activity Recognition data for testing both iOS and Android data formats. (plugin_google_activity_recognition_raw.csv for android and plugin_ios_activity_recognition_raw.csv for iOS) There is also an additional empty data file for both android and iOS for testing empty data files.","title":"Activity Recognition"},{"location":"developers/test-cases/#conversation","text":"The raw conversation data file contains data for 2 day. The raw conversation data contains records with a sample of both datatypes (i.e. voice/noise = 0 , and conversation = 2 ) as well as rows with for samples of each of the inference values (i.e. silence = 0 , noise = 1 , voice = 2 , and unknown = 3 ) for each epoch . The different datatype and inference records are randomly distributed throughout the epoch . Additionally there are 2 - 5 records for conversations ( datatype = 2, and inference = -1) in each epoch and for each epoch except night, there is a conversation record that has a double_convo_start timestamp that is from the previous epoch . This is to test the calculations of features across epochs . There is a raw conversation data file for both android and iOS platforms ( plugin_studentlife_audio_android_raw.csv and plugin_studentlife_audio_raw.csv respectively). Finally, there are also additional empty data files for both android and iOS for testing empty data files","title":"Conversation"},{"location":"developers/testing/","text":"Testing \u00b6 The following is a simple guide to testing RAPIDS. All files necessary for testing are stored in the /tests directory Steps for Testing \u00b6 To begin testing RAPIDS place the fake raw input data csv files in tests/data/raw/ . The fake participant files should be placed in tests/data/external/ . The expected output files of RAPIDS after processing the input data should be placed in tests/data/processesd/ . The Snakemake rule(s) that are to be tested must be placed in the tests/Snakemake file. The current tests/Snakemake is a good example of how to define them. (At the time of writing this documentation the snakefile contains rules messages (SMS), calls and screen) Edit the tests/settings/config.yaml . Add and/or remove the rules to be run for testing from the forcerun list. Edit the tests/settings/testing_config.yaml with the necessary configuration settings for running the rules to be tested. Add any additional testscripts in tests/scripts . Uncomment or comment off lines in the testing shell script tests/scripts/run_tests.sh . Run the testing shell script. tests/scripts/run_tests.sh The following is a snippet of the output you should see after running your test. test_sensors_files_exist ( test_sensor_features.TestSensorFeatures ) ... ok test_sensors_features_calculations ( test_sensor_features.TestSensorFeatures ) ... FAIL ====================================================================== FAIL: test_sensors_features_calculations ( test_sensor_features.TestSensorFeatures ) ---------------------------------------------------------------------- The results above show that the first test test_sensors_files_exist passed while test_sensors_features_calculations failed. In addition you should get the traceback of the failure (not shown here). For more information on how to implement test scripts and use unittest please see Unittest Documentation Testing of the RAPIDS sensors and features is a work-in-progress. Please see test-cases for a list of sensors and features that have testing currently available. Currently the repository is set up to test a number of sensors out of the box by simply running the tests/scripts/run_tests.sh command once the RAPIDS python environment is active.","title":"Testing"},{"location":"developers/testing/#testing","text":"The following is a simple guide to testing RAPIDS. All files necessary for testing are stored in the /tests directory","title":"Testing"},{"location":"developers/testing/#steps-for-testing","text":"To begin testing RAPIDS place the fake raw input data csv files in tests/data/raw/ . The fake participant files should be placed in tests/data/external/ . The expected output files of RAPIDS after processing the input data should be placed in tests/data/processesd/ . The Snakemake rule(s) that are to be tested must be placed in the tests/Snakemake file. The current tests/Snakemake is a good example of how to define them. (At the time of writing this documentation the snakefile contains rules messages (SMS), calls and screen) Edit the tests/settings/config.yaml . Add and/or remove the rules to be run for testing from the forcerun list. Edit the tests/settings/testing_config.yaml with the necessary configuration settings for running the rules to be tested. Add any additional testscripts in tests/scripts . Uncomment or comment off lines in the testing shell script tests/scripts/run_tests.sh . Run the testing shell script. tests/scripts/run_tests.sh The following is a snippet of the output you should see after running your test. test_sensors_files_exist ( test_sensor_features.TestSensorFeatures ) ... ok test_sensors_features_calculations ( test_sensor_features.TestSensorFeatures ) ... FAIL ====================================================================== FAIL: test_sensors_features_calculations ( test_sensor_features.TestSensorFeatures ) ---------------------------------------------------------------------- The results above show that the first test test_sensors_files_exist passed while test_sensors_features_calculations failed. In addition you should get the traceback of the failure (not shown here). For more information on how to implement test scripts and use unittest please see Unittest Documentation Testing of the RAPIDS sensors and features is a work-in-progress. Please see test-cases for a list of sensors and features that have testing currently available. Currently the repository is set up to test a number of sensors out of the box by simply running the tests/scripts/run_tests.sh command once the RAPIDS python environment is active.","title":"Steps for Testing"},{"location":"developers/virtual-environments/","text":"Virtual Environments \u00b6 Add new packages \u00b6 Try to install any new package using conda . If a package is not available in one of conda \u2018s channels you can install it with pip but make sure your virtual environment is active. Update your conda environment.yaml \u00b6 After installing a new package you can use the following command in your terminal to update your environment.yaml before publishing your pipeline. Note that we ignore the package version for libfortran to keep compatibility with Linux: conda env export --no-builds | sed 's/^.*libgfortran.*$/ - libgfortran/' > environment.yml Update and prune your conda environment from a environment.yaml file \u00b6 Execute the following command in your terminal, see these docs for more information conda env update --prefix ./env --file environment.yml --prune","title":"Virtual Environments"},{"location":"developers/virtual-environments/#virtual-environments","text":"","title":"Virtual Environments"},{"location":"developers/virtual-environments/#add-new-packages","text":"Try to install any new package using conda . If a package is not available in one of conda \u2018s channels you can install it with pip but make sure your virtual environment is active.","title":"Add new packages"},{"location":"developers/virtual-environments/#update-your-conda-environmentyaml","text":"After installing a new package you can use the following command in your terminal to update your environment.yaml before publishing your pipeline. Note that we ignore the package version for libfortran to keep compatibility with Linux: conda env export --no-builds | sed 's/^.*libgfortran.*$/ - libgfortran/' > environment.yml","title":"Update your conda environment.yaml"},{"location":"developers/virtual-environments/#update-and-prune-your-conda-environment-from-a-environmentyaml-file","text":"Execute the following command in your terminal, see these docs for more information conda env update --prefix ./env --file environment.yml --prune","title":"Update and prune your conda environment from a environment.yaml file"},{"location":"features/add-new-features/","text":"Add New Features \u00b6 Hint We recommend reading the Behavioral Features Introduction before reading this page Hint You won\u2019t have to deal with time zones, dates, times, data cleaning or preprocessing. The data that RAPIDS pipes to your feature extraction code is ready to process. New Features for Existing Sensors \u00b6 You can add new features to any existing sensors (see list below) by adding a new provider in three steps: Modify the config.yaml file Create a provider folder, script and function Implement your features extraction code As a tutorial, we will add a new provider for PHONE_ACCELEROMETER called VEGA that extracts feature1 , feature2 , feature3 in Python and that it requires a parameter from the user called MY_PARAMETER . Existing Sensors An existing sensor is any of the phone or Fitbit sensors with a configuration entry in config.yaml : Phone Accelerometer Phone Activity Recognition Phone Applications Foreground Phone Battery Phone Bluetooth Phone Calls Phone Conversation Phone Data Yield Phone Light Phone Locations Phone Messages Phone Screen Phone WiFI Connected Phone WiFI Visible Fitbit Heart Rate Summary Fitbit Heart Rate Intraday Fitbit Sleep Summary Fitbit Steps Summary Fitbit Steps Intraday Modify the config.yaml file \u00b6 In this step you need to add your provider configuration section under the relevant sensor in config.yaml . See our example for our tutorial\u2019s VEGA provider for PHONE_ACCELEROMETER : Example configuration for a new accelerometer provider VEGA PHONE_ACCELEROMETER : TABLE : accelerometer PROVIDERS : RAPIDS : COMPUTE : False ... PANDA : COMPUTE : False ... VEGA : COMPUTE : False FEATURES : [ \"feature1\" , \"feature2\" , \"feature3\" ] MY_PARAMTER : a_string SRC_FOLDER : \"vega\" SRC_LANGUAGE : \"python\" Key Description [COMPUTE] Flag to activate/deactivate your provider [FEATURES] List of features your provider supports. Your provider code should only return the features on this list [MY_PARAMTER] An arbitrary parameter that our example provider VEGA needs. This can be a boolean, integer, float, string or an array of any of such types. [SRC_LANGUAGE] The programming language of your provider script, it can be python or r , in our example python [SRC_FOLDER] The name of your provider in lower case, in our example vega (this will be the name of your folder in the next step) Create a provider folder, script and function \u00b6 In this step you need to add a folder, script and function for your provider. Create your provider folder under src/feature/DEVICE_SENSOR/YOUR_PROVIDER , in our example src/feature/phone_accelerometer/vega (same as [SRC_FOLDER] in the step above). Create your provider script inside your provider folder, it can be a Python file called main.py or an R file called main.R . Add your provider function in your provider script. The name of such function should be [providername]_features , in our example vega_features Python function def [ providername ] _features ( sensor_data_files , time_segment , provider , filter_data_by_segment , * args , ** kwargs ): R function [ providername ] _ features <- function ( sensor_data , time_segment , provider ) Implement your feature extraction code \u00b6 The provider function that you created in the step above will receive the following parameters: Parameter Description sensor_data_files Path to the CSV file containing the data of a single participant. This data has been cleaned and preprocessed. Your function will be automatically called for each participant in your study (in the [PIDS] array in config.yaml ) time_segment The label of the time segment that should be processed. provider The parameters you configured for your provider in config.yaml will be available in this variable as a dictionary in Python or a list in R. In our example this dictionary contains {MY_PARAMETER:\"a_string\"} filter_data_by_segment Python only. A function that you will use to filter your data. In R this function is already available in the environment. *args Python only. Not used for now **kwargs Python only. Not used for now The code to extract your behavioral features should be implemented in your provider function and in general terms it will have three stages: 1. Read a participant\u2019s data by loading the CSV data stored in the file pointed by sensor_data_files acc_data = pd . read_csv ( sensor_data_files [ \"sensor_data\" ]) Note that phone\u2019s battery, screen, and activity recognition data is given as episodes instead of event rows (for example, start and end timestamps of the periods the phone screen was on) 2. Filter your data to process only those rows that belong to time_segment This step is only one line of code, but to undersand why we need it, keep reading. acc_data = filter_data_by_segment ( acc_data , time_segment ) You should use the filter_data_by_segment() function to process and group those rows that belong to each of the time segments RAPIDS could be configured with . Let\u2019s understand the filter_data_by_segment() function with an example. A RAPIDS user can extract features on any arbitrary time segment . A time segment is a period of time that has a label and one or more instances. For example, the user (or you) could have requested features on a daily, weekly, and week-end basis for p01 . The labels are arbritrary and the instances depend on the days a participant was monitored for: the daily segment could be named my_days and if p01 was monitored for 14 days, it would have 14 instances the weekly segment could be named my_weeks and if p01 was monitored for 14 days, it would have 2 instances. the weekend segment could be named my_weekends and if p01 was monitored for 14 days, it would have 2 instances. For this example, RAPIDS will call your provider function three times for p01 , once where time_segment is my_days , once where time_segment is my_weeks and once where time_segment is my_weekends . In this example not every row in p01 \u2018s data needs to take part in the feature computation for either segment and the rows need to be grouped differently. Thus filter_data_by_segment() comes in handy, it will return a data frame that contains the rows that were logged during a time segment plus an extra column called local_segment . This new column will have as many unique values as time segment instances exist (14, 2, and 2 for our p01 \u2018s my_days , my_weeks , and my_weekends examples). After filtering, you should group the data frame by this column and compute any desired features , for example: acc_features [ \"maxmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . max () The reason RAPIDS does not filter the participant\u2019s data set for you is because your code might need to compute something based on a participant\u2019s complete dataset before computing their features. For example, you might want to identify the number that called a participant the most throughout the study before computing a feature with the number of calls the participant received from this number. 3. Return a data frame with your features After filtering, grouping your data, and computing your features, your provider function should return a data frame that has: One row per time segment instance (e.g. 14 our p01 \u2018s my_days example) The local_segment column added by filter_data_by_segment() One column per feature. By convention the name of your features should only contain letters or numbers ( feature1 ). RAPIDS will automatically add the right sensor and provider prefix ( phone_accelerometr_vega_ ) PHONE_ACCELEROMETER Provider Example For your reference, this a short example of our own provider ( RAPIDS ) for PHONE_ACCELEROMETER that computes five acceleration features def rapids_features ( sensor_data_files , time_segment , provider , filter_data_by_segment , * args , ** kwargs ): acc_data = pd . read_csv ( sensor_data_files [ \"sensor_data\" ]) requested_features = provider [ \"FEATURES\" ] # name of the features this function can compute base_features_names = [ \"maxmagnitude\" , \"minmagnitude\" , \"avgmagnitude\" , \"medianmagnitude\" , \"stdmagnitude\" ] # the subset of requested features this function can compute features_to_compute = list ( set ( requested_features ) & set ( base_features_names )) acc_features = pd . DataFrame ( columns = [ \"local_segment\" ] + features_to_compute ) if not acc_data . empty : acc_data = filter_data_by_segment ( acc_data , time_segment ) if not acc_data . empty : acc_features = pd . DataFrame () # get magnitude related features: magnitude = sqrt(x^2+y^2+z^2) magnitude = acc_data . apply ( lambda row : np . sqrt ( row [ \"double_values_0\" ] ** 2 + row [ \"double_values_1\" ] ** 2 + row [ \"double_values_2\" ] ** 2 ), axis = 1 ) acc_data = acc_data . assign ( magnitude = magnitude . values ) if \"maxmagnitude\" in features_to_compute : acc_features [ \"maxmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . max () if \"minmagnitude\" in features_to_compute : acc_features [ \"minmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . min () if \"avgmagnitude\" in features_to_compute : acc_features [ \"avgmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . mean () if \"medianmagnitude\" in features_to_compute : acc_features [ \"medianmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . median () if \"stdmagnitude\" in features_to_compute : acc_features [ \"stdmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . std () acc_features = acc_features . reset_index () return acc_features New Features for Non-Existing Sensors \u00b6 If you want to add features for a device or a sensor that we do not support at the moment (those that do not appear in the \"Existing Sensors\" list above), contact us or request it on Slack and we can add the necessary code so you can follow the instructions above.","title":"Add New Features"},{"location":"features/add-new-features/#add-new-features","text":"Hint We recommend reading the Behavioral Features Introduction before reading this page Hint You won\u2019t have to deal with time zones, dates, times, data cleaning or preprocessing. The data that RAPIDS pipes to your feature extraction code is ready to process.","title":"Add New Features"},{"location":"features/add-new-features/#new-features-for-existing-sensors","text":"You can add new features to any existing sensors (see list below) by adding a new provider in three steps: Modify the config.yaml file Create a provider folder, script and function Implement your features extraction code As a tutorial, we will add a new provider for PHONE_ACCELEROMETER called VEGA that extracts feature1 , feature2 , feature3 in Python and that it requires a parameter from the user called MY_PARAMETER . Existing Sensors An existing sensor is any of the phone or Fitbit sensors with a configuration entry in config.yaml : Phone Accelerometer Phone Activity Recognition Phone Applications Foreground Phone Battery Phone Bluetooth Phone Calls Phone Conversation Phone Data Yield Phone Light Phone Locations Phone Messages Phone Screen Phone WiFI Connected Phone WiFI Visible Fitbit Heart Rate Summary Fitbit Heart Rate Intraday Fitbit Sleep Summary Fitbit Steps Summary Fitbit Steps Intraday","title":"New Features for Existing Sensors"},{"location":"features/add-new-features/#modify-the-configyaml-file","text":"In this step you need to add your provider configuration section under the relevant sensor in config.yaml . See our example for our tutorial\u2019s VEGA provider for PHONE_ACCELEROMETER : Example configuration for a new accelerometer provider VEGA PHONE_ACCELEROMETER : TABLE : accelerometer PROVIDERS : RAPIDS : COMPUTE : False ... PANDA : COMPUTE : False ... VEGA : COMPUTE : False FEATURES : [ \"feature1\" , \"feature2\" , \"feature3\" ] MY_PARAMTER : a_string SRC_FOLDER : \"vega\" SRC_LANGUAGE : \"python\" Key Description [COMPUTE] Flag to activate/deactivate your provider [FEATURES] List of features your provider supports. Your provider code should only return the features on this list [MY_PARAMTER] An arbitrary parameter that our example provider VEGA needs. This can be a boolean, integer, float, string or an array of any of such types. [SRC_LANGUAGE] The programming language of your provider script, it can be python or r , in our example python [SRC_FOLDER] The name of your provider in lower case, in our example vega (this will be the name of your folder in the next step)","title":"Modify the config.yaml file"},{"location":"features/add-new-features/#create-a-provider-folder-script-and-function","text":"In this step you need to add a folder, script and function for your provider. Create your provider folder under src/feature/DEVICE_SENSOR/YOUR_PROVIDER , in our example src/feature/phone_accelerometer/vega (same as [SRC_FOLDER] in the step above). Create your provider script inside your provider folder, it can be a Python file called main.py or an R file called main.R . Add your provider function in your provider script. The name of such function should be [providername]_features , in our example vega_features Python function def [ providername ] _features ( sensor_data_files , time_segment , provider , filter_data_by_segment , * args , ** kwargs ): R function [ providername ] _ features <- function ( sensor_data , time_segment , provider )","title":"Create a provider folder, script and function"},{"location":"features/add-new-features/#implement-your-feature-extraction-code","text":"The provider function that you created in the step above will receive the following parameters: Parameter Description sensor_data_files Path to the CSV file containing the data of a single participant. This data has been cleaned and preprocessed. Your function will be automatically called for each participant in your study (in the [PIDS] array in config.yaml ) time_segment The label of the time segment that should be processed. provider The parameters you configured for your provider in config.yaml will be available in this variable as a dictionary in Python or a list in R. In our example this dictionary contains {MY_PARAMETER:\"a_string\"} filter_data_by_segment Python only. A function that you will use to filter your data. In R this function is already available in the environment. *args Python only. Not used for now **kwargs Python only. Not used for now The code to extract your behavioral features should be implemented in your provider function and in general terms it will have three stages: 1. Read a participant\u2019s data by loading the CSV data stored in the file pointed by sensor_data_files acc_data = pd . read_csv ( sensor_data_files [ \"sensor_data\" ]) Note that phone\u2019s battery, screen, and activity recognition data is given as episodes instead of event rows (for example, start and end timestamps of the periods the phone screen was on) 2. Filter your data to process only those rows that belong to time_segment This step is only one line of code, but to undersand why we need it, keep reading. acc_data = filter_data_by_segment ( acc_data , time_segment ) You should use the filter_data_by_segment() function to process and group those rows that belong to each of the time segments RAPIDS could be configured with . Let\u2019s understand the filter_data_by_segment() function with an example. A RAPIDS user can extract features on any arbitrary time segment . A time segment is a period of time that has a label and one or more instances. For example, the user (or you) could have requested features on a daily, weekly, and week-end basis for p01 . The labels are arbritrary and the instances depend on the days a participant was monitored for: the daily segment could be named my_days and if p01 was monitored for 14 days, it would have 14 instances the weekly segment could be named my_weeks and if p01 was monitored for 14 days, it would have 2 instances. the weekend segment could be named my_weekends and if p01 was monitored for 14 days, it would have 2 instances. For this example, RAPIDS will call your provider function three times for p01 , once where time_segment is my_days , once where time_segment is my_weeks and once where time_segment is my_weekends . In this example not every row in p01 \u2018s data needs to take part in the feature computation for either segment and the rows need to be grouped differently. Thus filter_data_by_segment() comes in handy, it will return a data frame that contains the rows that were logged during a time segment plus an extra column called local_segment . This new column will have as many unique values as time segment instances exist (14, 2, and 2 for our p01 \u2018s my_days , my_weeks , and my_weekends examples). After filtering, you should group the data frame by this column and compute any desired features , for example: acc_features [ \"maxmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . max () The reason RAPIDS does not filter the participant\u2019s data set for you is because your code might need to compute something based on a participant\u2019s complete dataset before computing their features. For example, you might want to identify the number that called a participant the most throughout the study before computing a feature with the number of calls the participant received from this number. 3. Return a data frame with your features After filtering, grouping your data, and computing your features, your provider function should return a data frame that has: One row per time segment instance (e.g. 14 our p01 \u2018s my_days example) The local_segment column added by filter_data_by_segment() One column per feature. By convention the name of your features should only contain letters or numbers ( feature1 ). RAPIDS will automatically add the right sensor and provider prefix ( phone_accelerometr_vega_ ) PHONE_ACCELEROMETER Provider Example For your reference, this a short example of our own provider ( RAPIDS ) for PHONE_ACCELEROMETER that computes five acceleration features def rapids_features ( sensor_data_files , time_segment , provider , filter_data_by_segment , * args , ** kwargs ): acc_data = pd . read_csv ( sensor_data_files [ \"sensor_data\" ]) requested_features = provider [ \"FEATURES\" ] # name of the features this function can compute base_features_names = [ \"maxmagnitude\" , \"minmagnitude\" , \"avgmagnitude\" , \"medianmagnitude\" , \"stdmagnitude\" ] # the subset of requested features this function can compute features_to_compute = list ( set ( requested_features ) & set ( base_features_names )) acc_features = pd . DataFrame ( columns = [ \"local_segment\" ] + features_to_compute ) if not acc_data . empty : acc_data = filter_data_by_segment ( acc_data , time_segment ) if not acc_data . empty : acc_features = pd . DataFrame () # get magnitude related features: magnitude = sqrt(x^2+y^2+z^2) magnitude = acc_data . apply ( lambda row : np . sqrt ( row [ \"double_values_0\" ] ** 2 + row [ \"double_values_1\" ] ** 2 + row [ \"double_values_2\" ] ** 2 ), axis = 1 ) acc_data = acc_data . assign ( magnitude = magnitude . values ) if \"maxmagnitude\" in features_to_compute : acc_features [ \"maxmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . max () if \"minmagnitude\" in features_to_compute : acc_features [ \"minmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . min () if \"avgmagnitude\" in features_to_compute : acc_features [ \"avgmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . mean () if \"medianmagnitude\" in features_to_compute : acc_features [ \"medianmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . median () if \"stdmagnitude\" in features_to_compute : acc_features [ \"stdmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . std () acc_features = acc_features . reset_index () return acc_features","title":"Implement your feature extraction code"},{"location":"features/add-new-features/#new-features-for-non-existing-sensors","text":"If you want to add features for a device or a sensor that we do not support at the moment (those that do not appear in the \"Existing Sensors\" list above), contact us or request it on Slack and we can add the necessary code so you can follow the instructions above.","title":"New Features for Non-Existing Sensors"},{"location":"features/feature-introduction/","text":"Behavioral Features Introduction \u00b6 Every phone or Fitbit sensor has a corresponding config section in config.yaml , these sections follow a similar structure and we\u2019ll use PHONE_ACCELEROMETER as an example to explain this structure. Hint We recommend reading this page if you are using RAPIDS for the first time All computed sensor features are stored under /data/processed/features on files per sensor, per participant and per study (all participants). Every time you change any sensor parameters, provider parameters or provider features, all the necessary files will be updated as soon as you execute RAPIDS. Config section example for PHONE_ACCELEROMETER # 1) Config section PHONE_ACCELEROMETER : # 2) Parameters for PHONE_ACCELEROMETER TABLE : accelerometer # 3) Providers for PHONE_ACCELEROMETER PROVIDERS : # 4) RAPIDS provider RAPIDS : # 4.1) Parameters of RAPIDS provider of PHONE_ACCELEROMETER COMPUTE : False # 4.2) Features of RAPIDS provider of PHONE_ACCELEROMETER FEATURES : [ \"maxmagnitude\" , \"minmagnitude\" , \"avgmagnitude\" , \"medianmagnitude\" , \"stdmagnitude\" ] SRC_FOLDER : \"rapids\" # inside src/features/phone_accelerometer SRC_LANGUAGE : \"python\" # 5) PANDA provider PANDA : # 5.1) Parameters of PANDA provider of PHONE_ACCELEROMETER COMPUTE : False VALID_SENSED_MINUTES : False # 5.2) Features of PANDA provider of PHONE_ACCELEROMETER FEATURES : exertional_activity_episode : [ \"sumduration\" , \"maxduration\" , \"minduration\" , \"avgduration\" , \"medianduration\" , \"stdduration\" ] nonexertional_activity_episode : [ \"sumduration\" , \"maxduration\" , \"minduration\" , \"avgduration\" , \"medianduration\" , \"stdduration\" ] SRC_FOLDER : \"panda\" # inside src/features/phone_accelerometer SRC_LANGUAGE : \"python\" Sensor Parameters \u00b6 Each sensor configuration section has a \u201cparameters\u201d subsection (see #2 in the example). These are parameters that affect different aspects of how the raw data is downloaded, and processed. The TABLE parameter exists for every sensor, but some sensors will have extra parameters like [PHONE_LOCATIONS] . We explain these parameters in a table at the top of each sensor documentation page. Sensor Providers \u00b6 Each sensor configuration section can have zero, one or more behavioral feature providers (see #3 in the example). A provider is a script created by the core RAPIDS team or other researchers that extracts behavioral features for that sensor. In this example, accelerometer has two providers: RAPIDS (see #4 ) and PANDA (see #5 ). Provider Parameters \u00b6 Each provider has parameters that affect the computation of the behavioral features it offers (see #4.1 or #5.1 in the example). These parameters will include at least a [COMPUTE] flag that you switch to True to extract a provider\u2019s behavioral features. We explain every provider\u2019s parameter in a table under the Parameters description heading on each provider documentation page. Provider Features \u00b6 Each provider offers a set of behavioral features (see #4.2 or #5.2 in the example). For some providers these features are grouped in an array (like those for RAPIDS provider in #4.2 ) but for others they are grouped in a collection of arrays depending on the meaning and purpose of those features (like those for PANDAS provider in #5.2 ). In either case, you can delete the features you are not interested in and they will not be included in the sensor\u2019s output feature file. We explain each behavioral feature in a table under the Features description heading on each provider documentation page.","title":"Introduction"},{"location":"features/feature-introduction/#behavioral-features-introduction","text":"Every phone or Fitbit sensor has a corresponding config section in config.yaml , these sections follow a similar structure and we\u2019ll use PHONE_ACCELEROMETER as an example to explain this structure. Hint We recommend reading this page if you are using RAPIDS for the first time All computed sensor features are stored under /data/processed/features on files per sensor, per participant and per study (all participants). Every time you change any sensor parameters, provider parameters or provider features, all the necessary files will be updated as soon as you execute RAPIDS. Config section example for PHONE_ACCELEROMETER # 1) Config section PHONE_ACCELEROMETER : # 2) Parameters for PHONE_ACCELEROMETER TABLE : accelerometer # 3) Providers for PHONE_ACCELEROMETER PROVIDERS : # 4) RAPIDS provider RAPIDS : # 4.1) Parameters of RAPIDS provider of PHONE_ACCELEROMETER COMPUTE : False # 4.2) Features of RAPIDS provider of PHONE_ACCELEROMETER FEATURES : [ \"maxmagnitude\" , \"minmagnitude\" , \"avgmagnitude\" , \"medianmagnitude\" , \"stdmagnitude\" ] SRC_FOLDER : \"rapids\" # inside src/features/phone_accelerometer SRC_LANGUAGE : \"python\" # 5) PANDA provider PANDA : # 5.1) Parameters of PANDA provider of PHONE_ACCELEROMETER COMPUTE : False VALID_SENSED_MINUTES : False # 5.2) Features of PANDA provider of PHONE_ACCELEROMETER FEATURES : exertional_activity_episode : [ \"sumduration\" , \"maxduration\" , \"minduration\" , \"avgduration\" , \"medianduration\" , \"stdduration\" ] nonexertional_activity_episode : [ \"sumduration\" , \"maxduration\" , \"minduration\" , \"avgduration\" , \"medianduration\" , \"stdduration\" ] SRC_FOLDER : \"panda\" # inside src/features/phone_accelerometer SRC_LANGUAGE : \"python\"","title":"Behavioral Features Introduction"},{"location":"features/feature-introduction/#sensor-parameters","text":"Each sensor configuration section has a \u201cparameters\u201d subsection (see #2 in the example). These are parameters that affect different aspects of how the raw data is downloaded, and processed. The TABLE parameter exists for every sensor, but some sensors will have extra parameters like [PHONE_LOCATIONS] . We explain these parameters in a table at the top of each sensor documentation page.","title":"Sensor Parameters"},{"location":"features/feature-introduction/#sensor-providers","text":"Each sensor configuration section can have zero, one or more behavioral feature providers (see #3 in the example). A provider is a script created by the core RAPIDS team or other researchers that extracts behavioral features for that sensor. In this example, accelerometer has two providers: RAPIDS (see #4 ) and PANDA (see #5 ).","title":"Sensor Providers"},{"location":"features/feature-introduction/#provider-parameters","text":"Each provider has parameters that affect the computation of the behavioral features it offers (see #4.1 or #5.1 in the example). These parameters will include at least a [COMPUTE] flag that you switch to True to extract a provider\u2019s behavioral features. We explain every provider\u2019s parameter in a table under the Parameters description heading on each provider documentation page.","title":"Provider Parameters"},{"location":"features/feature-introduction/#provider-features","text":"Each provider offers a set of behavioral features (see #4.2 or #5.2 in the example). For some providers these features are grouped in an array (like those for RAPIDS provider in #4.2 ) but for others they are grouped in a collection of arrays depending on the meaning and purpose of those features (like those for PANDAS provider in #5.2 ). In either case, you can delete the features you are not interested in and they will not be included in the sensor\u2019s output feature file. We explain each behavioral feature in a table under the Features description heading on each provider documentation page.","title":"Provider Features"},{"location":"features/fitbit-heartrate-intraday/","text":"Fitbit Heart Rate Intraday \u00b6 Sensor parameters description for [FITBIT_HEARTRATE_INTRADAY] : Key Description [TABLE] Database table name or file path where the heart rate intraday data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-07\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:1200.6102,\u201dmax\u201d:88,\u201dmin\u201d:31,\u201dminutes\u201d:1058,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:760.3020,\u201dmax\u201d:120,\u201dmin\u201d:86,\u201dminutes\u201d:366,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:15.2048,\u201dmax\u201d:146,\u201dmin\u201d:120,\u201dminutes\u201d:2,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:221,\u201dmin\u201d:148,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:72}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:68},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:67},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:67},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-08\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:1100.1120,\u201dmax\u201d:89,\u201dmin\u201d:30,\u201dminutes\u201d:921,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:660.0012,\u201dmax\u201d:118,\u201dmin\u201d:82,\u201dminutes\u201d:361,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:23.7088,\u201dmax\u201d:142,\u201dmin\u201d:108,\u201dminutes\u201d:3,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:221,\u201dmin\u201d:148,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:70}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:77},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:75},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:73},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-09\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:750.3615,\u201dmax\u201d:77,\u201dmin\u201d:30,\u201dminutes\u201d:851,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:734.1516,\u201dmax\u201d:107,\u201dmin\u201d:77,\u201dminutes\u201d:550,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:131.8579,\u201dmax\u201d:130,\u201dmin\u201d:107,\u201dminutes\u201d:29,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:220,\u201dmin\u201d:130,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:69}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:90},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:89},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:88},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} PLAIN_TEXT device_id local_date_time heartrate heartrate_zone a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:00:00 68 outofrange a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:01:00 67 outofrange a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:02:00 67 outofrange RAPIDS provider \u00b6 Available time segments Available for all time segments File Sequence - data/raw/ { pid } /fitbit_heartrate_intraday_raw.csv - data/raw/ { pid } /fitbit_heartrate_intraday_parsed.csv - data/raw/ { pid } /fitbit_heartrate_intraday_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_heartrate_intraday_features/fitbit_heartrate_intraday_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_heartrate_intraday.csv Parameters description for [FITBIT_HEARTRATE_INTRADAY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_HEARTRATE_INTRADAY features from the RAPIDS provider [FEATURES] Features to be computed from heart rate intraday data, see table below Features description for [FITBIT_HEARTRATE_INTRADAY][PROVIDERS][RAPIDS] : Feature Units Description maxhr beats/mins The maximum heart rate during a time segment. minhr beats/mins The minimum heart rate during a time segment. avghr beats/mins The average heart rate during a time segment. medianhr beats/mins The median of heart rate during a time segment. modehr beats/mins The mode of heart rate during a time segment. stdhr beats/mins The standard deviation of heart rate during a time segment. diffmaxmodehr beats/mins The difference between the maximum and mode heart rate during a time segment. diffminmodehr beats/mins The difference between the mode and minimum heart rate during a time segment. entropyhr nats Shannon\u2019s entropy measurement based on heart rate during a time segment. minutesonZONE minutes Number of minutes the user\u2019s heart rate fell within each heartrate_zone during a time segment. Assumptions/Observations There are four heart rate zones (ZONE): outofrange , fatburn , cardio , and peak . Please refer to Fitbit documentation for more information about the way they are computed.","title":"Fitbit Heart Rate Intraday"},{"location":"features/fitbit-heartrate-intraday/#fitbit-heart-rate-intraday","text":"Sensor parameters description for [FITBIT_HEARTRATE_INTRADAY] : Key Description [TABLE] Database table name or file path where the heart rate intraday data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-07\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:1200.6102,\u201dmax\u201d:88,\u201dmin\u201d:31,\u201dminutes\u201d:1058,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:760.3020,\u201dmax\u201d:120,\u201dmin\u201d:86,\u201dminutes\u201d:366,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:15.2048,\u201dmax\u201d:146,\u201dmin\u201d:120,\u201dminutes\u201d:2,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:221,\u201dmin\u201d:148,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:72}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:68},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:67},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:67},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-08\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:1100.1120,\u201dmax\u201d:89,\u201dmin\u201d:30,\u201dminutes\u201d:921,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:660.0012,\u201dmax\u201d:118,\u201dmin\u201d:82,\u201dminutes\u201d:361,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:23.7088,\u201dmax\u201d:142,\u201dmin\u201d:108,\u201dminutes\u201d:3,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:221,\u201dmin\u201d:148,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:70}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:77},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:75},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:73},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-09\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:750.3615,\u201dmax\u201d:77,\u201dmin\u201d:30,\u201dminutes\u201d:851,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:734.1516,\u201dmax\u201d:107,\u201dmin\u201d:77,\u201dminutes\u201d:550,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:131.8579,\u201dmax\u201d:130,\u201dmin\u201d:107,\u201dminutes\u201d:29,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:220,\u201dmin\u201d:130,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:69}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:90},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:89},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:88},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} PLAIN_TEXT device_id local_date_time heartrate heartrate_zone a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:00:00 68 outofrange a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:01:00 67 outofrange a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:02:00 67 outofrange","title":"Fitbit Heart Rate Intraday"},{"location":"features/fitbit-heartrate-intraday/#rapids-provider","text":"Available time segments Available for all time segments File Sequence - data/raw/ { pid } /fitbit_heartrate_intraday_raw.csv - data/raw/ { pid } /fitbit_heartrate_intraday_parsed.csv - data/raw/ { pid } /fitbit_heartrate_intraday_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_heartrate_intraday_features/fitbit_heartrate_intraday_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_heartrate_intraday.csv Parameters description for [FITBIT_HEARTRATE_INTRADAY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_HEARTRATE_INTRADAY features from the RAPIDS provider [FEATURES] Features to be computed from heart rate intraday data, see table below Features description for [FITBIT_HEARTRATE_INTRADAY][PROVIDERS][RAPIDS] : Feature Units Description maxhr beats/mins The maximum heart rate during a time segment. minhr beats/mins The minimum heart rate during a time segment. avghr beats/mins The average heart rate during a time segment. medianhr beats/mins The median of heart rate during a time segment. modehr beats/mins The mode of heart rate during a time segment. stdhr beats/mins The standard deviation of heart rate during a time segment. diffmaxmodehr beats/mins The difference between the maximum and mode heart rate during a time segment. diffminmodehr beats/mins The difference between the mode and minimum heart rate during a time segment. entropyhr nats Shannon\u2019s entropy measurement based on heart rate during a time segment. minutesonZONE minutes Number of minutes the user\u2019s heart rate fell within each heartrate_zone during a time segment. Assumptions/Observations There are four heart rate zones (ZONE): outofrange , fatburn , cardio , and peak . Please refer to Fitbit documentation for more information about the way they are computed.","title":"RAPIDS provider"},{"location":"features/fitbit-heartrate-summary/","text":"Fitbit Heart Rate Summary \u00b6 Sensor parameters description for [FITBIT_HEARTRATE_SUMMARY] : Key Description [TABLE] Database table name or file path where the heart rate summary data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-07\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:1200.6102,\u201dmax\u201d:88,\u201dmin\u201d:31,\u201dminutes\u201d:1058,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:760.3020,\u201dmax\u201d:120,\u201dmin\u201d:86,\u201dminutes\u201d:366,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:15.2048,\u201dmax\u201d:146,\u201dmin\u201d:120,\u201dminutes\u201d:2,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:221,\u201dmin\u201d:148,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:72}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:68},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:67},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:67},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-08\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:1100.1120,\u201dmax\u201d:89,\u201dmin\u201d:30,\u201dminutes\u201d:921,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:660.0012,\u201dmax\u201d:118,\u201dmin\u201d:82,\u201dminutes\u201d:361,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:23.7088,\u201dmax\u201d:142,\u201dmin\u201d:108,\u201dminutes\u201d:3,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:221,\u201dmin\u201d:148,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:70}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:77},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:75},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:73},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-09\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:750.3615,\u201dmax\u201d:77,\u201dmin\u201d:30,\u201dminutes\u201d:851,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:734.1516,\u201dmax\u201d:107,\u201dmin\u201d:77,\u201dminutes\u201d:550,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:131.8579,\u201dmax\u201d:130,\u201dmin\u201d:107,\u201dminutes\u201d:29,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:220,\u201dmin\u201d:130,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:69}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:90},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:89},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:88},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} PLAIN_TEXT device_id local_date_time heartrate_daily_restinghr heartrate_daily_caloriesoutofrange heartrate_daily_caloriesfatburn heartrate_daily_caloriescardio heartrate_daily_caloriespeak a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 72 1200.6102 760.3020 15.2048 0 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-08 70 1100.1120 660.0012 23.7088 0 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-09 69 750.3615 734.1516 131.8579 0 RAPIDS provider \u00b6 Available time segments Only available for segments that span 1 or more complete days (e.g. Jan 1 st 00:00 to Jan 3 rd 23:59) File Sequence - data/raw/ { pid } /fitbit_heartrate_summary_raw.csv - data/raw/ { pid } /fitbit_heartrate_summary_parsed.csv - data/raw/ { pid } /fitbit_heartrate_summary_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_heartrate_summary_features/fitbit_heartrate_summary_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_heartrate_summary.csv Parameters description for [FITBIT_HEARTRATE_SUMMARY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_HEARTRATE_SUMMARY features from the RAPIDS provider [FEATURES] Features to be computed from heart rate summary data, see table below Features description for [FITBIT_HEARTRATE_SUMMARY][PROVIDERS][RAPIDS] : Feature Units Description maxrestinghr beats/mins The maximum daily resting heart rate during a time segment. minrestinghr beats/mins The minimum daily resting heart rate during a time segment. avgrestinghr beats/mins The average daily resting heart rate during a time segment. medianrestinghr beats/mins The median of daily resting heart rate during a time segment. moderestinghr beats/mins The mode of daily resting heart rate during a time segment. stdrestinghr beats/mins The standard deviation of daily resting heart rate during a time segment. diffmaxmoderestinghr beats/mins The difference between the maximum and mode daily resting heart rate during a time segment. diffminmoderestinghr beats/mins The difference between the mode and minimum daily resting heart rate during a time segment. entropyrestinghr nats Shannon\u2019s entropy measurement based on daily resting heart rate during a time segment. sumcaloriesZONE cals The total daily calories burned within heartrate_zone during a time segment. maxcaloriesZONE cals The maximum daily calories burned within heartrate_zone during a time segment. mincaloriesZONE cals The minimum daily calories burned within heartrate_zone during a time segment. avgcaloriesZONE cals The average daily calories burned within heartrate_zone during a time segment. mediancaloriesZONE cals The median of daily calories burned within heartrate_zone during a time segment. stdcaloriesZONE cals The standard deviation of daily calories burned within heartrate_zone during a time segment. entropycaloriesZONE nats Shannon\u2019s entropy measurement based on daily calories burned within heartrate_zone during a time segment. Assumptions/Observations There are four heart rate zones (ZONE): outofrange , fatburn , cardio , and peak . Please refer to Fitbit documentation for more information about the way they are computed. Calories\u2019 accuracy depends on the users\u2019 Fitbit profile (weight, height, etc.).","title":"Fitbit Heart Rate Summary"},{"location":"features/fitbit-heartrate-summary/#fitbit-heart-rate-summary","text":"Sensor parameters description for [FITBIT_HEARTRATE_SUMMARY] : Key Description [TABLE] Database table name or file path where the heart rate summary data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-07\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:1200.6102,\u201dmax\u201d:88,\u201dmin\u201d:31,\u201dminutes\u201d:1058,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:760.3020,\u201dmax\u201d:120,\u201dmin\u201d:86,\u201dminutes\u201d:366,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:15.2048,\u201dmax\u201d:146,\u201dmin\u201d:120,\u201dminutes\u201d:2,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:221,\u201dmin\u201d:148,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:72}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:68},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:67},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:67},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-08\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:1100.1120,\u201dmax\u201d:89,\u201dmin\u201d:30,\u201dminutes\u201d:921,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:660.0012,\u201dmax\u201d:118,\u201dmin\u201d:82,\u201dminutes\u201d:361,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:23.7088,\u201dmax\u201d:142,\u201dmin\u201d:108,\u201dminutes\u201d:3,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:221,\u201dmin\u201d:148,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:70}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:77},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:75},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:73},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-09\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:750.3615,\u201dmax\u201d:77,\u201dmin\u201d:30,\u201dminutes\u201d:851,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:734.1516,\u201dmax\u201d:107,\u201dmin\u201d:77,\u201dminutes\u201d:550,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:131.8579,\u201dmax\u201d:130,\u201dmin\u201d:107,\u201dminutes\u201d:29,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:220,\u201dmin\u201d:130,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:69}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:90},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:89},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:88},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} PLAIN_TEXT device_id local_date_time heartrate_daily_restinghr heartrate_daily_caloriesoutofrange heartrate_daily_caloriesfatburn heartrate_daily_caloriescardio heartrate_daily_caloriespeak a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 72 1200.6102 760.3020 15.2048 0 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-08 70 1100.1120 660.0012 23.7088 0 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-09 69 750.3615 734.1516 131.8579 0","title":"Fitbit Heart Rate Summary"},{"location":"features/fitbit-heartrate-summary/#rapids-provider","text":"Available time segments Only available for segments that span 1 or more complete days (e.g. Jan 1 st 00:00 to Jan 3 rd 23:59) File Sequence - data/raw/ { pid } /fitbit_heartrate_summary_raw.csv - data/raw/ { pid } /fitbit_heartrate_summary_parsed.csv - data/raw/ { pid } /fitbit_heartrate_summary_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_heartrate_summary_features/fitbit_heartrate_summary_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_heartrate_summary.csv Parameters description for [FITBIT_HEARTRATE_SUMMARY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_HEARTRATE_SUMMARY features from the RAPIDS provider [FEATURES] Features to be computed from heart rate summary data, see table below Features description for [FITBIT_HEARTRATE_SUMMARY][PROVIDERS][RAPIDS] : Feature Units Description maxrestinghr beats/mins The maximum daily resting heart rate during a time segment. minrestinghr beats/mins The minimum daily resting heart rate during a time segment. avgrestinghr beats/mins The average daily resting heart rate during a time segment. medianrestinghr beats/mins The median of daily resting heart rate during a time segment. moderestinghr beats/mins The mode of daily resting heart rate during a time segment. stdrestinghr beats/mins The standard deviation of daily resting heart rate during a time segment. diffmaxmoderestinghr beats/mins The difference between the maximum and mode daily resting heart rate during a time segment. diffminmoderestinghr beats/mins The difference between the mode and minimum daily resting heart rate during a time segment. entropyrestinghr nats Shannon\u2019s entropy measurement based on daily resting heart rate during a time segment. sumcaloriesZONE cals The total daily calories burned within heartrate_zone during a time segment. maxcaloriesZONE cals The maximum daily calories burned within heartrate_zone during a time segment. mincaloriesZONE cals The minimum daily calories burned within heartrate_zone during a time segment. avgcaloriesZONE cals The average daily calories burned within heartrate_zone during a time segment. mediancaloriesZONE cals The median of daily calories burned within heartrate_zone during a time segment. stdcaloriesZONE cals The standard deviation of daily calories burned within heartrate_zone during a time segment. entropycaloriesZONE nats Shannon\u2019s entropy measurement based on daily calories burned within heartrate_zone during a time segment. Assumptions/Observations There are four heart rate zones (ZONE): outofrange , fatburn , cardio , and peak . Please refer to Fitbit documentation for more information about the way they are computed. Calories\u2019 accuracy depends on the users\u2019 Fitbit profile (weight, height, etc.).","title":"RAPIDS provider"},{"location":"features/fitbit-sleep-summary/","text":"Fitbit Sleep Summary \u00b6 Sensor parameters description for [FITBIT_SLEEP_SUMMARY] : Key Description [TABLE] Database table name or file path where the sleep summary data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data with Fitbit\u2019s sleep API Version 1 JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d: [{\u201cawakeCount\u201d: 2, \u201cawakeDuration\u201d: 3, \u201cawakeningsCount\u201d: 10, \u201cdateOfSleep\u201d: \u201c2020-10-07\u201d, \u201cduration\u201d: 8100000, \u201cefficiency\u201d: 91, \u201cendTime\u201d: \u201c2020-10-07T18:10:00.000\u201d, \u201cisMainSleep\u201d: true, \u201clogId\u201d: 14147921940, \u201cminuteData\u201d: [{\u201cdateTime\u201d: \u201c15:55:00\u201d, \u201cvalue\u201d: \u201c3\u201d}, {\u201cdateTime\u201d: \u201c15:56:00\u201d, \u201cvalue\u201d: \u201c3\u201d}, {\u201cdateTime\u201d: \u201c15:57:00\u201d, \u201cvalue\u201d: \u201c2\u201d},\u2026], \u201cminutesAfterWakeup\u201d: 0, \u201cminutesAsleep\u201d: 123, \u201cminutesAwake\u201d: 12, \u201cminutesToFallAsleep\u201d: 0, \u201crestlessCount\u201d: 8, \u201crestlessDuration\u201d: 9, \u201cstartTime\u201d: \u201c2020-10-07T15:55:00.000\u201d, \u201ctimeInBed\u201d: 135}, {\u201cawakeCount\u201d: 0, \u201cawakeDuration\u201d: 0, \u201cawakeningsCount\u201d: 1, \u201cdateOfSleep\u201d: \u201c2020-10-07\u201d, \u201cduration\u201d: 3780000, \u201cefficiency\u201d: 100, \u201cendTime\u201d: \u201c2020-10-07T10:52:30.000\u201d, \u201cisMainSleep\u201d: false, \u201clogId\u201d: 14144903977, \u201cminuteData\u201d: [{\u201cdateTime\u201d: \u201c09:49:00\u201d, \u201cvalue\u201d: \u201c1\u201d}, {\u201cdateTime\u201d: \u201c09:50:00\u201d, \u201cvalue\u201d: \u201c1\u201d}, {\u201cdateTime\u201d: \u201c09:51:00\u201d, \u201cvalue\u201d: \u201c1\u201d},\u2026], \u201cminutesAfterWakeup\u201d: 1, \u201cminutesAsleep\u201d: 62, \u201cminutesAwake\u201d: 0, \u201cminutesToFallAsleep\u201d: 0, \u201crestlessCount\u201d: 1, \u201crestlessDuration\u201d: 1, \u201cstartTime\u201d: \u201c2020-10-07T09:49:00.000\u201d, \u201ctimeInBed\u201d: 63}], \u201csummary\u201d: {\u201ctotalMinutesAsleep\u201d: 185, \u201ctotalSleepRecords\u201d: 2, \u201ctotalTimeInBed\u201d: 198}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d: [{\u201cawakeCount\u201d: 3, \u201cawakeDuration\u201d: 21, \u201cawakeningsCount\u201d: 16, \u201cdateOfSleep\u201d: \u201c2020-10-08\u201d, \u201cduration\u201d: 19260000, \u201cefficiency\u201d: 89, \u201cendTime\u201d: \u201c2020-10-08T06:01:30.000\u201d, \u201cisMainSleep\u201d: true, \u201clogId\u201d: 14150613895, \u201cminuteData\u201d: [{\u201cdateTime\u201d: \u201c00:40:00\u201d, \u201cvalue\u201d: \u201c3\u201d}, {\u201cdateTime\u201d: \u201c00:41:00\u201d, \u201cvalue\u201d: \u201c3\u201d}, {\u201cdateTime\u201d: \u201c00:42:00\u201d, \u201cvalue\u201d: \u201c3\u201d},\u2026], \u201cminutesAfterWakeup\u201d: 0, \u201cminutesAsleep\u201d: 275, \u201cminutesAwake\u201d: 33, \u201cminutesToFallAsleep\u201d: 0, \u201crestlessCount\u201d: 13, \u201crestlessDuration\u201d: 25, \u201cstartTime\u201d: \u201c2020-10-08T00:40:00.000\u201d, \u201ctimeInBed\u201d: 321}], \u201csummary\u201d: {\u201ctotalMinutesAsleep\u201d: 275, \u201ctotalSleepRecords\u201d: 1, \u201ctotalTimeInBed\u201d: 321}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d: [{\u201cawakeCount\u201d: 1, \u201cawakeDuration\u201d: 3, \u201cawakeningsCount\u201d: 8, \u201cdateOfSleep\u201d: \u201c2020-10-09\u201d, \u201cduration\u201d: 19320000, \u201cefficiency\u201d: 96, \u201cendTime\u201d: \u201c2020-10-09T05:57:30.000\u201d, \u201cisMainSleep\u201d: true, \u201clogId\u201d: 14161136803, \u201cminuteData\u201d: [{\u201cdateTime\u201d: \u201c00:35:30\u201d, \u201cvalue\u201d: \u201c2\u201d}, {\u201cdateTime\u201d: \u201c00:36:30\u201d, \u201cvalue\u201d: \u201c1\u201d}, {\u201cdateTime\u201d: \u201c00:37:30\u201d, \u201cvalue\u201d: \u201c1\u201d},\u2026], \u201cminutesAfterWakeup\u201d: 0, \u201cminutesAsleep\u201d: 309, \u201cminutesAwake\u201d: 13, \u201cminutesToFallAsleep\u201d: 0, \u201crestlessCount\u201d: 7, \u201crestlessDuration\u201d: 10, \u201cstartTime\u201d: \u201c2020-10-09T00:35:30.000\u201d, \u201ctimeInBed\u201d: 322}], \u201csummary\u201d: {\u201ctotalMinutesAsleep\u201d: 309, \u201ctotalSleepRecords\u201d: 1, \u201ctotalTimeInBed\u201d: 322}} PLAIN_TEXT device_id local_start_date_time local_end_date_time efficiency minutes_after_wakeup minutes_asleep minutes_awake minutes_to_fall_asleep minutes_in_bed is_main_sleep type count_awake duration_awake count_awakenings count_restless duration_restless a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 15:55:00 2020-10-07 18:10:00 91 0 123 12 0 135 1 classic 2 3 10 8 9 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 09:49:00 2020-10-07 10:52:30 100 1 62 0 0 63 0 classic 0 0 1 1 1 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-08 00:40:00 2020-10-08 06:01:30 89 0 275 33 0 321 1 classic 3 21 16 13 25 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-09 00:35:30 2020-10-09 05:57:30 96 0 309 13 0 322 1 classic 1 3 8 7 10 Example of the structure of source data with Fitbit\u2019s sleep API Version 1.2 JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d:[{\u201cdateOfSleep\u201d:\u201d2020-10-10\u201d,\u201dduration\u201d:3600000,\u201defficiency\u201d:92,\u201dendTime\u201d:\u201d2020-10-10T16:37:00.000\u201d,\u201dinfoCode\u201d:2,\u201disMainSleep\u201d:false,\u201dlevels\u201d:{\u201cdata\u201d:[{\u201cdateTime\u201d:\u201d2020-10-10T15:36:30.000\u201d,\u201dlevel\u201d:\u201drestless\u201d,\u201dseconds\u201d:60},{\u201cdateTime\u201d:\u201d2020-10-10T15:37:30.000\u201d,\u201dlevel\u201d:\u201dasleep\u201d,\u201dseconds\u201d:660},{\u201cdateTime\u201d:\u201d2020-10-10T15:48:30.000\u201d,\u201dlevel\u201d:\u201drestless\u201d,\u201dseconds\u201d:60},\u2026], \u201csummary\u201d:{\u201casleep\u201d:{\u201ccount\u201d:0,\u201dminutes\u201d:56},\u201dawake\u201d:{\u201ccount\u201d:0,\u201dminutes\u201d:0},\u201drestless\u201d:{\u201ccount\u201d:3,\u201dminutes\u201d:4}}},\u201dlogId\u201d:26315914306,\u201dminutesAfterWakeup\u201d:0,\u201dminutesAsleep\u201d:55,\u201dminutesAwake\u201d:5,\u201dminutesToFallAsleep\u201d:0,\u201dstartTime\u201d:\u201d2020-10-10T15:36:30.000\u201d,\u201dtimeInBed\u201d:60,\u201dtype\u201d:\u201dclassic\u201d},{\u201cdateOfSleep\u201d:\u201d2020-10-10\u201d,\u201dduration\u201d:22980000,\u201defficiency\u201d:88,\u201dendTime\u201d:\u201d2020-10-10T08:10:00.000\u201d,\u201dinfoCode\u201d:0,\u201disMainSleep\u201d:true,\u201dlevels\u201d:{\u201cdata\u201d:[{\u201cdateTime\u201d:\u201d2020-10-10T01:46:30.000\u201d,\u201dlevel\u201d:\u201dlight\u201d,\u201dseconds\u201d:420},{\u201cdateTime\u201d:\u201d2020-10-10T01:53:30.000\u201d,\u201dlevel\u201d:\u201ddeep\u201d,\u201dseconds\u201d:1230},{\u201cdateTime\u201d:\u201d2020-10-10T02:14:00.000\u201d,\u201dlevel\u201d:\u201dlight\u201d,\u201dseconds\u201d:360},\u2026], \u201csummary\u201d:{\u201cdeep\u201d:{\u201ccount\u201d:3,\u201dminutes\u201d:92,\u201dthirtyDayAvgMinutes\u201d:0},\u201dlight\u201d:{\u201ccount\u201d:29,\u201dminutes\u201d:193,\u201dthirtyDayAvgMinutes\u201d:0},\u201drem\u201d:{\u201ccount\u201d:4,\u201dminutes\u201d:33,\u201dthirtyDayAvgMinutes\u201d:0},\u201dwake\u201d:{\u201ccount\u201d:28,\u201dminutes\u201d:65,\u201dthirtyDayAvgMinutes\u201d:0}}},\u201dlogId\u201d:26311786557,\u201dminutesAfterWakeup\u201d:0,\u201dminutesAsleep\u201d:318,\u201dminutesAwake\u201d:65,\u201dminutesToFallAsleep\u201d:0,\u201dstartTime\u201d:\u201d2020-10-10T01:46:30.000\u201d,\u201dtimeInBed\u201d:383,\u201dtype\u201d:\u201dstages\u201d}],\u201dsummary\u201d:{\u201cstages\u201d:{\u201cdeep\u201d:92,\u201dlight\u201d:193,\u201drem\u201d:33,\u201dwake\u201d:65},\u201dtotalMinutesAsleep\u201d:373,\u201dtotalSleepRecords\u201d:2,\u201dtotalTimeInBed\u201d:443}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d:[{\u201cdateOfSleep\u201d:\u201d2020-10-11\u201d,\u201dduration\u201d:41640000,\u201defficiency\u201d:89,\u201dendTime\u201d:\u201d2020-10-11T11:47:00.000\u201d,\u201dinfoCode\u201d:0,\u201disMainSleep\u201d:true,\u201dlevels\u201d:{\u201cdata\u201d:[{\u201cdateTime\u201d:\u201d2020-10-11T00:12:30.000\u201d,\u201dlevel\u201d:\u201dwake\u201d,\u201dseconds\u201d:450},{\u201cdateTime\u201d:\u201d2020-10-11T00:20:00.000\u201d,\u201dlevel\u201d:\u201dlight\u201d,\u201dseconds\u201d:870},{\u201cdateTime\u201d:\u201d2020-10-11T00:34:30.000\u201d,\u201dlevel\u201d:\u201dwake\u201d,\u201dseconds\u201d:780},\u2026], \u201csummary\u201d:{\u201cdeep\u201d:{\u201ccount\u201d:4,\u201dminutes\u201d:52,\u201dthirtyDayAvgMinutes\u201d:62},\u201dlight\u201d:{\u201ccount\u201d:32,\u201dminutes\u201d:442,\u201dthirtyDayAvgMinutes\u201d:364},\u201drem\u201d:{\u201ccount\u201d:6,\u201dminutes\u201d:68,\u201dthirtyDayAvgMinutes\u201d:58},\u201dwake\u201d:{\u201ccount\u201d:29,\u201dminutes\u201d:132,\u201dthirtyDayAvgMinutes\u201d:94}}},\u201dlogId\u201d:26589710670,\u201dminutesAfterWakeup\u201d:1,\u201dminutesAsleep\u201d:562,\u201dminutesAwake\u201d:132,\u201dminutesToFallAsleep\u201d:0,\u201dstartTime\u201d:\u201d2020-10-11T00:12:30.000\u201d,\u201dtimeInBed\u201d:694,\u201dtype\u201d:\u201dstages\u201d}],\u201dsummary\u201d:{\u201cstages\u201d:{\u201cdeep\u201d:52,\u201dlight\u201d:442,\u201drem\u201d:68,\u201dwake\u201d:132},\u201dtotalMinutesAsleep\u201d:562,\u201dtotalSleepRecords\u201d:1,\u201dtotalTimeInBed\u201d:694}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d:[{\u201cdateOfSleep\u201d:\u201d2020-10-12\u201d,\u201dduration\u201d:28980000,\u201defficiency\u201d:93,\u201dendTime\u201d:\u201d2020-10-12T09:34:30.000\u201d,\u201dinfoCode\u201d:0,\u201disMainSleep\u201d:true,\u201dlevels\u201d:{\u201cdata\u201d:[{\u201cdateTime\u201d:\u201d2020-10-12T01:31:00.000\u201d,\u201dlevel\u201d:\u201dwake\u201d,\u201dseconds\u201d:600},{\u201cdateTime\u201d:\u201d2020-10-12T01:41:00.000\u201d,\u201dlevel\u201d:\u201dlight\u201d,\u201dseconds\u201d:60},{\u201cdateTime\u201d:\u201d2020-10-12T01:42:00.000\u201d,\u201dlevel\u201d:\u201ddeep\u201d,\u201dseconds\u201d:2340},\u2026], \u201csummary\u201d:{\u201cdeep\u201d:{\u201ccount\u201d:4,\u201dminutes\u201d:63,\u201dthirtyDayAvgMinutes\u201d:59},\u201dlight\u201d:{\u201ccount\u201d:27,\u201dminutes\u201d:257,\u201dthirtyDayAvgMinutes\u201d:364},\u201drem\u201d:{\u201ccount\u201d:5,\u201dminutes\u201d:94,\u201dthirtyDayAvgMinutes\u201d:58},\u201dwake\u201d:{\u201ccount\u201d:24,\u201dminutes\u201d:69,\u201dthirtyDayAvgMinutes\u201d:95}}},\u201dlogId\u201d:26589710673,\u201dminutesAfterWakeup\u201d:0,\u201dminutesAsleep\u201d:415,\u201dminutesAwake\u201d:68,\u201dminutesToFallAsleep\u201d:0,\u201dstartTime\u201d:\u201d2020-10-12T01:31:00.000\u201d,\u201dtimeInBed\u201d:483,\u201dtype\u201d:\u201dstages\u201d}],\u201dsummary\u201d:{\u201cstages\u201d:{\u201cdeep\u201d:63,\u201dlight\u201d:257,\u201drem\u201d:94,\u201dwake\u201d:69},\u201dtotalMinutesAsleep\u201d:415,\u201dtotalSleepRecords\u201d:1,\u201dtotalTimeInBed\u201d:483}} PLAIN_TEXT device_id local_start_date_time local_end_date_time efficiency minutes_after_wakeup minutes_asleep minutes_awake minutes_to_fall_asleep minutes_in_bed is_main_sleep type a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-10 15:36:30 2020-10-10 16:37:00 92 0 55 5 0 60 0 classic a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-10 01:46:30 2020-10-10 08:10:00 88 0 318 65 0 383 1 stages a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-11 00:12:30 2020-10-11 11:47:00 89 1 562 132 0 694 1 stages a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-12 01:31:00 2020-10-12 09:34:30 93 0 415 68 0 483 1 stages RAPIDS provider \u00b6 Available time segments Only available for segments that span 1 or more complete days (e.g. Jan 1 st 00:00 to Jan 3 rd 23:59) File Sequence - data/raw/ { pid } /fitbit_sleep_summary_raw.csv - data/raw/ { pid } /fitbit_sleep_summary_parsed.csv - data/raw/ { pid } /fitbit_sleep_summary_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_sleep_summary_features/fitbit_sleep_summary_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_sleep_summary.csv Parameters description for [FITBIT_SLEEP_SUMMARY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_SLEEP_SUMMARY features from the RAPIDS provider [SLEEP_TYPES] Types of sleep to be included in the feature extraction computation. Fitbit provides 3 types of sleep: main , nap , all . [FEATURES] Features to be computed from sleep summary data, see table below Features description for [FITBIT_SLEEP_SUMMARY][PROVIDERS][RAPIDS] : Feature Units Description countepisodeTYPE episodes Number of sleep episodes for a certain sleep type during a time segment. avgefficiencyTYPE scores Average sleep efficiency for a certain sleep type during a time segment. sumdurationafterwakeupTYPE minutes Total duration the user stayed in bed after waking up for a certain sleep type during a time segment. sumdurationasleepTYPE minutes Total sleep duration for a certain sleep type during a time segment. sumdurationawakeTYPE minutes Total duration the user stayed awake but still in bed for a certain sleep type during a time segment. sumdurationtofallasleepTYPE minutes Total duration the user spent to fall asleep for a certain sleep type during a time segment. sumdurationinbedTYPE minutes Total duration the user stayed in bed (sumdurationtofallasleep + sumdurationawake + sumdurationasleep + sumdurationafterwakeup) for a certain sleep type during a time segment. avgdurationafterwakeupTYPE minutes Average duration the user stayed in bed after waking up for a certain sleep type during a time segment. avgdurationasleepTYPE minutes Average sleep duration for a certain sleep type during a time segment. avgdurationawakeTYPE minutes Average duration the user stayed awake but still in bed for a certain sleep type during a time segment. avgdurationtofallasleepTYPE minutes Average duration the user spent to fall asleep for a certain sleep type during a time segment. avgdurationinbedTYPE minutes Average duration the user stayed in bed (sumdurationtofallasleep + sumdurationawake + sumdurationasleep + sumdurationafterwakeup) for a certain sleep type during a time segment. Assumptions/Observations There are three sleep types (TYPE): main , nap , all . The all type contains both main sleep and naps. There are two versions of Fitbit\u2019s sleep API ( version 1 and version 1.2 ), and each provides raw sleep data in a different format: Count & duration summaries . v1 contains count_awake , duration_awake , count_awakenings , count_restless , and duration_restless fields for every sleep record but v1.2 does not. API columns . Features are computed based on the values provided by Fitbit\u2019s API: efficiency , minutes_after_wakeup , minutes_asleep , minutes_awake , minutes_to_fall_asleep , minutes_in_bed , is_main_sleep and type .","title":"Fitbit Sleep Summary"},{"location":"features/fitbit-sleep-summary/#fitbit-sleep-summary","text":"Sensor parameters description for [FITBIT_SLEEP_SUMMARY] : Key Description [TABLE] Database table name or file path where the sleep summary data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data with Fitbit\u2019s sleep API Version 1 JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d: [{\u201cawakeCount\u201d: 2, \u201cawakeDuration\u201d: 3, \u201cawakeningsCount\u201d: 10, \u201cdateOfSleep\u201d: \u201c2020-10-07\u201d, \u201cduration\u201d: 8100000, \u201cefficiency\u201d: 91, \u201cendTime\u201d: \u201c2020-10-07T18:10:00.000\u201d, \u201cisMainSleep\u201d: true, \u201clogId\u201d: 14147921940, \u201cminuteData\u201d: [{\u201cdateTime\u201d: \u201c15:55:00\u201d, \u201cvalue\u201d: \u201c3\u201d}, {\u201cdateTime\u201d: \u201c15:56:00\u201d, \u201cvalue\u201d: \u201c3\u201d}, {\u201cdateTime\u201d: \u201c15:57:00\u201d, \u201cvalue\u201d: \u201c2\u201d},\u2026], \u201cminutesAfterWakeup\u201d: 0, \u201cminutesAsleep\u201d: 123, \u201cminutesAwake\u201d: 12, \u201cminutesToFallAsleep\u201d: 0, \u201crestlessCount\u201d: 8, \u201crestlessDuration\u201d: 9, \u201cstartTime\u201d: \u201c2020-10-07T15:55:00.000\u201d, \u201ctimeInBed\u201d: 135}, {\u201cawakeCount\u201d: 0, \u201cawakeDuration\u201d: 0, \u201cawakeningsCount\u201d: 1, \u201cdateOfSleep\u201d: \u201c2020-10-07\u201d, \u201cduration\u201d: 3780000, \u201cefficiency\u201d: 100, \u201cendTime\u201d: \u201c2020-10-07T10:52:30.000\u201d, \u201cisMainSleep\u201d: false, \u201clogId\u201d: 14144903977, \u201cminuteData\u201d: [{\u201cdateTime\u201d: \u201c09:49:00\u201d, \u201cvalue\u201d: \u201c1\u201d}, {\u201cdateTime\u201d: \u201c09:50:00\u201d, \u201cvalue\u201d: \u201c1\u201d}, {\u201cdateTime\u201d: \u201c09:51:00\u201d, \u201cvalue\u201d: \u201c1\u201d},\u2026], \u201cminutesAfterWakeup\u201d: 1, \u201cminutesAsleep\u201d: 62, \u201cminutesAwake\u201d: 0, \u201cminutesToFallAsleep\u201d: 0, \u201crestlessCount\u201d: 1, \u201crestlessDuration\u201d: 1, \u201cstartTime\u201d: \u201c2020-10-07T09:49:00.000\u201d, \u201ctimeInBed\u201d: 63}], \u201csummary\u201d: {\u201ctotalMinutesAsleep\u201d: 185, \u201ctotalSleepRecords\u201d: 2, \u201ctotalTimeInBed\u201d: 198}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d: [{\u201cawakeCount\u201d: 3, \u201cawakeDuration\u201d: 21, \u201cawakeningsCount\u201d: 16, \u201cdateOfSleep\u201d: \u201c2020-10-08\u201d, \u201cduration\u201d: 19260000, \u201cefficiency\u201d: 89, \u201cendTime\u201d: \u201c2020-10-08T06:01:30.000\u201d, \u201cisMainSleep\u201d: true, \u201clogId\u201d: 14150613895, \u201cminuteData\u201d: [{\u201cdateTime\u201d: \u201c00:40:00\u201d, \u201cvalue\u201d: \u201c3\u201d}, {\u201cdateTime\u201d: \u201c00:41:00\u201d, \u201cvalue\u201d: \u201c3\u201d}, {\u201cdateTime\u201d: \u201c00:42:00\u201d, \u201cvalue\u201d: \u201c3\u201d},\u2026], \u201cminutesAfterWakeup\u201d: 0, \u201cminutesAsleep\u201d: 275, \u201cminutesAwake\u201d: 33, \u201cminutesToFallAsleep\u201d: 0, \u201crestlessCount\u201d: 13, \u201crestlessDuration\u201d: 25, \u201cstartTime\u201d: \u201c2020-10-08T00:40:00.000\u201d, \u201ctimeInBed\u201d: 321}], \u201csummary\u201d: {\u201ctotalMinutesAsleep\u201d: 275, \u201ctotalSleepRecords\u201d: 1, \u201ctotalTimeInBed\u201d: 321}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d: [{\u201cawakeCount\u201d: 1, \u201cawakeDuration\u201d: 3, \u201cawakeningsCount\u201d: 8, \u201cdateOfSleep\u201d: \u201c2020-10-09\u201d, \u201cduration\u201d: 19320000, \u201cefficiency\u201d: 96, \u201cendTime\u201d: \u201c2020-10-09T05:57:30.000\u201d, \u201cisMainSleep\u201d: true, \u201clogId\u201d: 14161136803, \u201cminuteData\u201d: [{\u201cdateTime\u201d: \u201c00:35:30\u201d, \u201cvalue\u201d: \u201c2\u201d}, {\u201cdateTime\u201d: \u201c00:36:30\u201d, \u201cvalue\u201d: \u201c1\u201d}, {\u201cdateTime\u201d: \u201c00:37:30\u201d, \u201cvalue\u201d: \u201c1\u201d},\u2026], \u201cminutesAfterWakeup\u201d: 0, \u201cminutesAsleep\u201d: 309, \u201cminutesAwake\u201d: 13, \u201cminutesToFallAsleep\u201d: 0, \u201crestlessCount\u201d: 7, \u201crestlessDuration\u201d: 10, \u201cstartTime\u201d: \u201c2020-10-09T00:35:30.000\u201d, \u201ctimeInBed\u201d: 322}], \u201csummary\u201d: {\u201ctotalMinutesAsleep\u201d: 309, \u201ctotalSleepRecords\u201d: 1, \u201ctotalTimeInBed\u201d: 322}} PLAIN_TEXT device_id local_start_date_time local_end_date_time efficiency minutes_after_wakeup minutes_asleep minutes_awake minutes_to_fall_asleep minutes_in_bed is_main_sleep type count_awake duration_awake count_awakenings count_restless duration_restless a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 15:55:00 2020-10-07 18:10:00 91 0 123 12 0 135 1 classic 2 3 10 8 9 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 09:49:00 2020-10-07 10:52:30 100 1 62 0 0 63 0 classic 0 0 1 1 1 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-08 00:40:00 2020-10-08 06:01:30 89 0 275 33 0 321 1 classic 3 21 16 13 25 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-09 00:35:30 2020-10-09 05:57:30 96 0 309 13 0 322 1 classic 1 3 8 7 10 Example of the structure of source data with Fitbit\u2019s sleep API Version 1.2 JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d:[{\u201cdateOfSleep\u201d:\u201d2020-10-10\u201d,\u201dduration\u201d:3600000,\u201defficiency\u201d:92,\u201dendTime\u201d:\u201d2020-10-10T16:37:00.000\u201d,\u201dinfoCode\u201d:2,\u201disMainSleep\u201d:false,\u201dlevels\u201d:{\u201cdata\u201d:[{\u201cdateTime\u201d:\u201d2020-10-10T15:36:30.000\u201d,\u201dlevel\u201d:\u201drestless\u201d,\u201dseconds\u201d:60},{\u201cdateTime\u201d:\u201d2020-10-10T15:37:30.000\u201d,\u201dlevel\u201d:\u201dasleep\u201d,\u201dseconds\u201d:660},{\u201cdateTime\u201d:\u201d2020-10-10T15:48:30.000\u201d,\u201dlevel\u201d:\u201drestless\u201d,\u201dseconds\u201d:60},\u2026], \u201csummary\u201d:{\u201casleep\u201d:{\u201ccount\u201d:0,\u201dminutes\u201d:56},\u201dawake\u201d:{\u201ccount\u201d:0,\u201dminutes\u201d:0},\u201drestless\u201d:{\u201ccount\u201d:3,\u201dminutes\u201d:4}}},\u201dlogId\u201d:26315914306,\u201dminutesAfterWakeup\u201d:0,\u201dminutesAsleep\u201d:55,\u201dminutesAwake\u201d:5,\u201dminutesToFallAsleep\u201d:0,\u201dstartTime\u201d:\u201d2020-10-10T15:36:30.000\u201d,\u201dtimeInBed\u201d:60,\u201dtype\u201d:\u201dclassic\u201d},{\u201cdateOfSleep\u201d:\u201d2020-10-10\u201d,\u201dduration\u201d:22980000,\u201defficiency\u201d:88,\u201dendTime\u201d:\u201d2020-10-10T08:10:00.000\u201d,\u201dinfoCode\u201d:0,\u201disMainSleep\u201d:true,\u201dlevels\u201d:{\u201cdata\u201d:[{\u201cdateTime\u201d:\u201d2020-10-10T01:46:30.000\u201d,\u201dlevel\u201d:\u201dlight\u201d,\u201dseconds\u201d:420},{\u201cdateTime\u201d:\u201d2020-10-10T01:53:30.000\u201d,\u201dlevel\u201d:\u201ddeep\u201d,\u201dseconds\u201d:1230},{\u201cdateTime\u201d:\u201d2020-10-10T02:14:00.000\u201d,\u201dlevel\u201d:\u201dlight\u201d,\u201dseconds\u201d:360},\u2026], \u201csummary\u201d:{\u201cdeep\u201d:{\u201ccount\u201d:3,\u201dminutes\u201d:92,\u201dthirtyDayAvgMinutes\u201d:0},\u201dlight\u201d:{\u201ccount\u201d:29,\u201dminutes\u201d:193,\u201dthirtyDayAvgMinutes\u201d:0},\u201drem\u201d:{\u201ccount\u201d:4,\u201dminutes\u201d:33,\u201dthirtyDayAvgMinutes\u201d:0},\u201dwake\u201d:{\u201ccount\u201d:28,\u201dminutes\u201d:65,\u201dthirtyDayAvgMinutes\u201d:0}}},\u201dlogId\u201d:26311786557,\u201dminutesAfterWakeup\u201d:0,\u201dminutesAsleep\u201d:318,\u201dminutesAwake\u201d:65,\u201dminutesToFallAsleep\u201d:0,\u201dstartTime\u201d:\u201d2020-10-10T01:46:30.000\u201d,\u201dtimeInBed\u201d:383,\u201dtype\u201d:\u201dstages\u201d}],\u201dsummary\u201d:{\u201cstages\u201d:{\u201cdeep\u201d:92,\u201dlight\u201d:193,\u201drem\u201d:33,\u201dwake\u201d:65},\u201dtotalMinutesAsleep\u201d:373,\u201dtotalSleepRecords\u201d:2,\u201dtotalTimeInBed\u201d:443}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d:[{\u201cdateOfSleep\u201d:\u201d2020-10-11\u201d,\u201dduration\u201d:41640000,\u201defficiency\u201d:89,\u201dendTime\u201d:\u201d2020-10-11T11:47:00.000\u201d,\u201dinfoCode\u201d:0,\u201disMainSleep\u201d:true,\u201dlevels\u201d:{\u201cdata\u201d:[{\u201cdateTime\u201d:\u201d2020-10-11T00:12:30.000\u201d,\u201dlevel\u201d:\u201dwake\u201d,\u201dseconds\u201d:450},{\u201cdateTime\u201d:\u201d2020-10-11T00:20:00.000\u201d,\u201dlevel\u201d:\u201dlight\u201d,\u201dseconds\u201d:870},{\u201cdateTime\u201d:\u201d2020-10-11T00:34:30.000\u201d,\u201dlevel\u201d:\u201dwake\u201d,\u201dseconds\u201d:780},\u2026], \u201csummary\u201d:{\u201cdeep\u201d:{\u201ccount\u201d:4,\u201dminutes\u201d:52,\u201dthirtyDayAvgMinutes\u201d:62},\u201dlight\u201d:{\u201ccount\u201d:32,\u201dminutes\u201d:442,\u201dthirtyDayAvgMinutes\u201d:364},\u201drem\u201d:{\u201ccount\u201d:6,\u201dminutes\u201d:68,\u201dthirtyDayAvgMinutes\u201d:58},\u201dwake\u201d:{\u201ccount\u201d:29,\u201dminutes\u201d:132,\u201dthirtyDayAvgMinutes\u201d:94}}},\u201dlogId\u201d:26589710670,\u201dminutesAfterWakeup\u201d:1,\u201dminutesAsleep\u201d:562,\u201dminutesAwake\u201d:132,\u201dminutesToFallAsleep\u201d:0,\u201dstartTime\u201d:\u201d2020-10-11T00:12:30.000\u201d,\u201dtimeInBed\u201d:694,\u201dtype\u201d:\u201dstages\u201d}],\u201dsummary\u201d:{\u201cstages\u201d:{\u201cdeep\u201d:52,\u201dlight\u201d:442,\u201drem\u201d:68,\u201dwake\u201d:132},\u201dtotalMinutesAsleep\u201d:562,\u201dtotalSleepRecords\u201d:1,\u201dtotalTimeInBed\u201d:694}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d:[{\u201cdateOfSleep\u201d:\u201d2020-10-12\u201d,\u201dduration\u201d:28980000,\u201defficiency\u201d:93,\u201dendTime\u201d:\u201d2020-10-12T09:34:30.000\u201d,\u201dinfoCode\u201d:0,\u201disMainSleep\u201d:true,\u201dlevels\u201d:{\u201cdata\u201d:[{\u201cdateTime\u201d:\u201d2020-10-12T01:31:00.000\u201d,\u201dlevel\u201d:\u201dwake\u201d,\u201dseconds\u201d:600},{\u201cdateTime\u201d:\u201d2020-10-12T01:41:00.000\u201d,\u201dlevel\u201d:\u201dlight\u201d,\u201dseconds\u201d:60},{\u201cdateTime\u201d:\u201d2020-10-12T01:42:00.000\u201d,\u201dlevel\u201d:\u201ddeep\u201d,\u201dseconds\u201d:2340},\u2026], \u201csummary\u201d:{\u201cdeep\u201d:{\u201ccount\u201d:4,\u201dminutes\u201d:63,\u201dthirtyDayAvgMinutes\u201d:59},\u201dlight\u201d:{\u201ccount\u201d:27,\u201dminutes\u201d:257,\u201dthirtyDayAvgMinutes\u201d:364},\u201drem\u201d:{\u201ccount\u201d:5,\u201dminutes\u201d:94,\u201dthirtyDayAvgMinutes\u201d:58},\u201dwake\u201d:{\u201ccount\u201d:24,\u201dminutes\u201d:69,\u201dthirtyDayAvgMinutes\u201d:95}}},\u201dlogId\u201d:26589710673,\u201dminutesAfterWakeup\u201d:0,\u201dminutesAsleep\u201d:415,\u201dminutesAwake\u201d:68,\u201dminutesToFallAsleep\u201d:0,\u201dstartTime\u201d:\u201d2020-10-12T01:31:00.000\u201d,\u201dtimeInBed\u201d:483,\u201dtype\u201d:\u201dstages\u201d}],\u201dsummary\u201d:{\u201cstages\u201d:{\u201cdeep\u201d:63,\u201dlight\u201d:257,\u201drem\u201d:94,\u201dwake\u201d:69},\u201dtotalMinutesAsleep\u201d:415,\u201dtotalSleepRecords\u201d:1,\u201dtotalTimeInBed\u201d:483}} PLAIN_TEXT device_id local_start_date_time local_end_date_time efficiency minutes_after_wakeup minutes_asleep minutes_awake minutes_to_fall_asleep minutes_in_bed is_main_sleep type a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-10 15:36:30 2020-10-10 16:37:00 92 0 55 5 0 60 0 classic a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-10 01:46:30 2020-10-10 08:10:00 88 0 318 65 0 383 1 stages a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-11 00:12:30 2020-10-11 11:47:00 89 1 562 132 0 694 1 stages a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-12 01:31:00 2020-10-12 09:34:30 93 0 415 68 0 483 1 stages","title":"Fitbit Sleep Summary"},{"location":"features/fitbit-sleep-summary/#rapids-provider","text":"Available time segments Only available for segments that span 1 or more complete days (e.g. Jan 1 st 00:00 to Jan 3 rd 23:59) File Sequence - data/raw/ { pid } /fitbit_sleep_summary_raw.csv - data/raw/ { pid } /fitbit_sleep_summary_parsed.csv - data/raw/ { pid } /fitbit_sleep_summary_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_sleep_summary_features/fitbit_sleep_summary_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_sleep_summary.csv Parameters description for [FITBIT_SLEEP_SUMMARY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_SLEEP_SUMMARY features from the RAPIDS provider [SLEEP_TYPES] Types of sleep to be included in the feature extraction computation. Fitbit provides 3 types of sleep: main , nap , all . [FEATURES] Features to be computed from sleep summary data, see table below Features description for [FITBIT_SLEEP_SUMMARY][PROVIDERS][RAPIDS] : Feature Units Description countepisodeTYPE episodes Number of sleep episodes for a certain sleep type during a time segment. avgefficiencyTYPE scores Average sleep efficiency for a certain sleep type during a time segment. sumdurationafterwakeupTYPE minutes Total duration the user stayed in bed after waking up for a certain sleep type during a time segment. sumdurationasleepTYPE minutes Total sleep duration for a certain sleep type during a time segment. sumdurationawakeTYPE minutes Total duration the user stayed awake but still in bed for a certain sleep type during a time segment. sumdurationtofallasleepTYPE minutes Total duration the user spent to fall asleep for a certain sleep type during a time segment. sumdurationinbedTYPE minutes Total duration the user stayed in bed (sumdurationtofallasleep + sumdurationawake + sumdurationasleep + sumdurationafterwakeup) for a certain sleep type during a time segment. avgdurationafterwakeupTYPE minutes Average duration the user stayed in bed after waking up for a certain sleep type during a time segment. avgdurationasleepTYPE minutes Average sleep duration for a certain sleep type during a time segment. avgdurationawakeTYPE minutes Average duration the user stayed awake but still in bed for a certain sleep type during a time segment. avgdurationtofallasleepTYPE minutes Average duration the user spent to fall asleep for a certain sleep type during a time segment. avgdurationinbedTYPE minutes Average duration the user stayed in bed (sumdurationtofallasleep + sumdurationawake + sumdurationasleep + sumdurationafterwakeup) for a certain sleep type during a time segment. Assumptions/Observations There are three sleep types (TYPE): main , nap , all . The all type contains both main sleep and naps. There are two versions of Fitbit\u2019s sleep API ( version 1 and version 1.2 ), and each provides raw sleep data in a different format: Count & duration summaries . v1 contains count_awake , duration_awake , count_awakenings , count_restless , and duration_restless fields for every sleep record but v1.2 does not. API columns . Features are computed based on the values provided by Fitbit\u2019s API: efficiency , minutes_after_wakeup , minutes_asleep , minutes_awake , minutes_to_fall_asleep , minutes_in_bed , is_main_sleep and type .","title":"RAPIDS provider"},{"location":"features/fitbit-steps-intraday/","text":"Fitbit Steps Intraday \u00b6 Sensor parameters description for [FITBIT_STEPS_INTRADAY] : Key Description [TABLE] Database table name or file path where the steps intraday data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-07\u201d,\u201dvalue\u201d:\u201d1775\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:5},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:3},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:0},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-08\u201d,\u201dvalue\u201d:\u201d3201\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:14},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:11},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:10},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-09\u201d,\u201dvalue\u201d:\u201d998\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:0},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:0},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:0},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} PLAIN_TEXT device_id local_date_time steps a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:00:00 5 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:01:00 3 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:02:00 0 RAPIDS provider \u00b6 Available time segments Available for all time segments File Sequence - data/raw/ { pid } /fitbit_steps_intraday_raw.csv - data/raw/ { pid } /fitbit_steps_intraday_parsed.csv - data/raw/ { pid } /fitbit_steps_intraday_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_steps_intraday_features/fitbit_steps_intraday_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_steps_intraday.csv Parameters description for [FITBIT_STEPS_INTRADAY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_STEPS_INTRADAY features from the RAPIDS provider [FEATURES] Features to be computed from steps intraday data, see table below [THRESHOLD_ACTIVE_BOUT] Every minute with Fitbit steps data wil be labelled as sedentary if its step count is below this threshold, otherwise, active . [INCLUDE_ZERO_STEP_ROWS] Whether or not to include time segments with a 0 step count during the whole day. Features description for [FITBIT_STEPS_INTRADAY][PROVIDERS][RAPIDS] : Feature Units Description sumsteps steps The total step count during a time segment. maxsteps steps The maximum step count during a time segment. minsteps steps The minimum step count during a time segment. avgsteps steps The average step count during a time segment. stdsteps steps The standard deviation of step count during a time segment. countepisodesedentarybout bouts Number of sedentary bouts during a time segment. sumdurationsedentarybout minutes Total duration of all sedentary bouts during a time segment. maxdurationsedentarybout minutes The maximum duration of any sedentary bout during a time segment. mindurationsedentarybout minutes The minimum duration of any sedentary bout during a time segment. avgdurationsedentarybout minutes The average duration of sedentary bouts during a time segment. stddurationsedentarybout minutes The standard deviation of the duration of sedentary bouts during a time segment. countepisodeactivebout bouts Number of active bouts during a time segment. sumdurationactivebout minutes Total duration of all active bouts during a time segment. maxdurationactivebout minutes The maximum duration of any active bout during a time segment. mindurationactivebout minutes The minimum duration of any active bout during a time segment. avgdurationactivebout minutes The average duration of active bouts during a time segment. stddurationactivebout minutes The standard deviation of the duration of active bouts during a time segment. Assumptions/Observations Active and sedentary bouts . If the step count per minute is smaller than THRESHOLD_ACTIVE_BOUT (default value is 10), that minute is labelled as sedentary, otherwise, is labelled as active. Active and sedentary bouts are periods of consecutive minutes labelled as active or sedentary .","title":"Fitbit Steps Intraday"},{"location":"features/fitbit-steps-intraday/#fitbit-steps-intraday","text":"Sensor parameters description for [FITBIT_STEPS_INTRADAY] : Key Description [TABLE] Database table name or file path where the steps intraday data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-07\u201d,\u201dvalue\u201d:\u201d1775\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:5},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:3},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:0},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-08\u201d,\u201dvalue\u201d:\u201d3201\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:14},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:11},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:10},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-09\u201d,\u201dvalue\u201d:\u201d998\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:0},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:0},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:0},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} PLAIN_TEXT device_id local_date_time steps a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:00:00 5 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:01:00 3 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:02:00 0","title":"Fitbit Steps Intraday"},{"location":"features/fitbit-steps-intraday/#rapids-provider","text":"Available time segments Available for all time segments File Sequence - data/raw/ { pid } /fitbit_steps_intraday_raw.csv - data/raw/ { pid } /fitbit_steps_intraday_parsed.csv - data/raw/ { pid } /fitbit_steps_intraday_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_steps_intraday_features/fitbit_steps_intraday_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_steps_intraday.csv Parameters description for [FITBIT_STEPS_INTRADAY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_STEPS_INTRADAY features from the RAPIDS provider [FEATURES] Features to be computed from steps intraday data, see table below [THRESHOLD_ACTIVE_BOUT] Every minute with Fitbit steps data wil be labelled as sedentary if its step count is below this threshold, otherwise, active . [INCLUDE_ZERO_STEP_ROWS] Whether or not to include time segments with a 0 step count during the whole day. Features description for [FITBIT_STEPS_INTRADAY][PROVIDERS][RAPIDS] : Feature Units Description sumsteps steps The total step count during a time segment. maxsteps steps The maximum step count during a time segment. minsteps steps The minimum step count during a time segment. avgsteps steps The average step count during a time segment. stdsteps steps The standard deviation of step count during a time segment. countepisodesedentarybout bouts Number of sedentary bouts during a time segment. sumdurationsedentarybout minutes Total duration of all sedentary bouts during a time segment. maxdurationsedentarybout minutes The maximum duration of any sedentary bout during a time segment. mindurationsedentarybout minutes The minimum duration of any sedentary bout during a time segment. avgdurationsedentarybout minutes The average duration of sedentary bouts during a time segment. stddurationsedentarybout minutes The standard deviation of the duration of sedentary bouts during a time segment. countepisodeactivebout bouts Number of active bouts during a time segment. sumdurationactivebout minutes Total duration of all active bouts during a time segment. maxdurationactivebout minutes The maximum duration of any active bout during a time segment. mindurationactivebout minutes The minimum duration of any active bout during a time segment. avgdurationactivebout minutes The average duration of active bouts during a time segment. stddurationactivebout minutes The standard deviation of the duration of active bouts during a time segment. Assumptions/Observations Active and sedentary bouts . If the step count per minute is smaller than THRESHOLD_ACTIVE_BOUT (default value is 10), that minute is labelled as sedentary, otherwise, is labelled as active. Active and sedentary bouts are periods of consecutive minutes labelled as active or sedentary .","title":"RAPIDS provider"},{"location":"features/fitbit-steps-summary/","text":"Fitbit Steps Summary \u00b6 Sensor parameters description for [FITBIT_STEPS_SUMMARY] : Key Description [TABLE] Database table name or file path where the steps summary data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-07\u201d,\u201dvalue\u201d:\u201d1775\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:5},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:3},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:0},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-08\u201d,\u201dvalue\u201d:\u201d3201\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:14},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:11},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:10},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-09\u201d,\u201dvalue\u201d:\u201d998\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:0},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:0},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:0},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} PLAIN_TEXT device_id local_date_time steps a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 1775 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-08 3201 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-09 998 RAPIDS provider \u00b6 Available time segments Only available for segments that span 1 or more complete days (e.g. Jan 1 st 00:00 to Jan 3 rd 23:59) File Sequence - data/raw/ { pid } /fitbit_steps_summary_raw.csv - data/raw/ { pid } /fitbit_steps_summary_parsed.csv - data/raw/ { pid } /fitbit_steps_summary_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_steps_summary_features/fitbit_steps_summary_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_steps_summary.csv Parameters description for [FITBIT_STEPS_SUMMARY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_STEPS_SUMMARY features from the RAPIDS provider [FEATURES] Features to be computed from steps summary data, see table below Features description for [FITBIT_STEPS_SUMMARY][PROVIDERS][RAPIDS] : Feature Units Description maxsumsteps steps The maximum daily step count during a time segment. minsumsteps steps The minimum daily step count during a time segment. avgsumsteps steps The average daily step count during a time segment. mediansumsteps steps The median of daily step count during a time segment. stdsumsteps steps The standard deviation of daily step count during a time segment. Assumptions/Observations NA","title":"Fitbit Steps Summary"},{"location":"features/fitbit-steps-summary/#fitbit-steps-summary","text":"Sensor parameters description for [FITBIT_STEPS_SUMMARY] : Key Description [TABLE] Database table name or file path where the steps summary data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-07\u201d,\u201dvalue\u201d:\u201d1775\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:5},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:3},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:0},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-08\u201d,\u201dvalue\u201d:\u201d3201\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:14},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:11},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:10},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-09\u201d,\u201dvalue\u201d:\u201d998\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:0},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:0},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:0},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} PLAIN_TEXT device_id local_date_time steps a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 1775 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-08 3201 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-09 998","title":"Fitbit Steps Summary"},{"location":"features/fitbit-steps-summary/#rapids-provider","text":"Available time segments Only available for segments that span 1 or more complete days (e.g. Jan 1 st 00:00 to Jan 3 rd 23:59) File Sequence - data/raw/ { pid } /fitbit_steps_summary_raw.csv - data/raw/ { pid } /fitbit_steps_summary_parsed.csv - data/raw/ { pid } /fitbit_steps_summary_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_steps_summary_features/fitbit_steps_summary_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_steps_summary.csv Parameters description for [FITBIT_STEPS_SUMMARY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_STEPS_SUMMARY features from the RAPIDS provider [FEATURES] Features to be computed from steps summary data, see table below Features description for [FITBIT_STEPS_SUMMARY][PROVIDERS][RAPIDS] : Feature Units Description maxsumsteps steps The maximum daily step count during a time segment. minsumsteps steps The minimum daily step count during a time segment. avgsumsteps steps The average daily step count during a time segment. mediansumsteps steps The median of daily step count during a time segment. stdsumsteps steps The standard deviation of daily step count during a time segment. Assumptions/Observations NA","title":"RAPIDS provider"},{"location":"features/phone-accelerometer/","text":"Phone Accelerometer \u00b6 Sensor parameters description for [PHONE_ACCELEROMETER] : Key Description [TABLE] Database table where the accelerometer data is stored RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_accelerometer_raw.csv - data/raw/ { pid } /phone_accelerometer_with_datetime.csv - data/interim/ { pid } /phone_accelerometer_features/phone_accelerometer_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_accelerometer.csv Parameters description for [PHONE_ACCELEROMETER][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_ACCELEROMETER features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_ACCELEROMETER][PROVIDERS][RAPIDS] : Feature Units Description maxmagnitude m/s 2 The maximum magnitude of acceleration ( \\(\\|acceleration\\| = \\sqrt{x^2 + y^2 + z^2}\\) ). minmagnitude m/s 2 The minimum magnitude of acceleration. avgmagnitude m/s 2 The average magnitude of acceleration. medianmagnitude m/s 2 The median magnitude of acceleration. stdmagnitude m/s 2 The standard deviation of acceleration. Assumptions/Observations Analyzing accelerometer data is a memory intensive task. If RAPIDS crashes is likely because the accelerometer dataset for a participant is to big to fit in memory. We are considering different alternatives to overcome this problem. PANDA provider \u00b6 These features are based on the work by Panda et al . Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_accelerometer_raw.csv - data/raw/ { pid } /phone_accelerometer_with_datetime.csv - data/interim/ { pid } /phone_accelerometer_features/phone_accelerometer_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_accelerometer.csv Parameters description for [PHONE_ACCELEROMETER][PROVIDERS][PANDA] : Key Description [COMPUTE] Set to True to extract PHONE_ACCELEROMETER features from the PANDA provider [FEATURES] Features to be computed for exertional and non-exertional activity episodes, see table below Features description for [PHONE_ACCELEROMETER][PROVIDERS][PANDA] : Feature Units Description sumduration minutes Total duration of all exertional or non-exertional activity episodes. maxduration minutes Longest duration of any exertional or non-exertional activity episode. minduration minutes Shortest duration of any exertional or non-exertional activity episode. avgduration minutes Average duration of any exertional or non-exertional activity episode. medianduration minutes Median duration of any exertional or non-exertional activity episode. stdduration minutes Standard deviation of the duration of all exertional or non-exertional activity episodes. Assumptions/Observations Analyzing accelerometer data is a memory intensive task. If RAPIDS crashes is likely because the accelerometer dataset for a participant is to big to fit in memory. We are considering different alternatives to overcome this problem. See Panda et al for a definition of exertional and non-exertional activity episodes","title":"Phone Accelerometer"},{"location":"features/phone-accelerometer/#phone-accelerometer","text":"Sensor parameters description for [PHONE_ACCELEROMETER] : Key Description [TABLE] Database table where the accelerometer data is stored","title":"Phone Accelerometer"},{"location":"features/phone-accelerometer/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_accelerometer_raw.csv - data/raw/ { pid } /phone_accelerometer_with_datetime.csv - data/interim/ { pid } /phone_accelerometer_features/phone_accelerometer_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_accelerometer.csv Parameters description for [PHONE_ACCELEROMETER][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_ACCELEROMETER features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_ACCELEROMETER][PROVIDERS][RAPIDS] : Feature Units Description maxmagnitude m/s 2 The maximum magnitude of acceleration ( \\(\\|acceleration\\| = \\sqrt{x^2 + y^2 + z^2}\\) ). minmagnitude m/s 2 The minimum magnitude of acceleration. avgmagnitude m/s 2 The average magnitude of acceleration. medianmagnitude m/s 2 The median magnitude of acceleration. stdmagnitude m/s 2 The standard deviation of acceleration. Assumptions/Observations Analyzing accelerometer data is a memory intensive task. If RAPIDS crashes is likely because the accelerometer dataset for a participant is to big to fit in memory. We are considering different alternatives to overcome this problem.","title":"RAPIDS provider"},{"location":"features/phone-accelerometer/#panda-provider","text":"These features are based on the work by Panda et al . Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_accelerometer_raw.csv - data/raw/ { pid } /phone_accelerometer_with_datetime.csv - data/interim/ { pid } /phone_accelerometer_features/phone_accelerometer_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_accelerometer.csv Parameters description for [PHONE_ACCELEROMETER][PROVIDERS][PANDA] : Key Description [COMPUTE] Set to True to extract PHONE_ACCELEROMETER features from the PANDA provider [FEATURES] Features to be computed for exertional and non-exertional activity episodes, see table below Features description for [PHONE_ACCELEROMETER][PROVIDERS][PANDA] : Feature Units Description sumduration minutes Total duration of all exertional or non-exertional activity episodes. maxduration minutes Longest duration of any exertional or non-exertional activity episode. minduration minutes Shortest duration of any exertional or non-exertional activity episode. avgduration minutes Average duration of any exertional or non-exertional activity episode. medianduration minutes Median duration of any exertional or non-exertional activity episode. stdduration minutes Standard deviation of the duration of all exertional or non-exertional activity episodes. Assumptions/Observations Analyzing accelerometer data is a memory intensive task. If RAPIDS crashes is likely because the accelerometer dataset for a participant is to big to fit in memory. We are considering different alternatives to overcome this problem. See Panda et al for a definition of exertional and non-exertional activity episodes","title":"PANDA provider"},{"location":"features/phone-activity-recognition/","text":"Phone Activity Recognition \u00b6 Sensor parameters description for [PHONE_ACTIVITY_RECOGNITION] : Key Description [TABLE][ANDROID] Database table where the activity data from Android devices is stored (the AWARE client saves this data on different tables for Android and iOS) [TABLE][IOS] Database table where the activity data from iOS devices is stored (the AWARE client saves this data on different tables for Android and iOS) [EPISODE_THRESHOLD_BETWEEN_ROWS] Difference in minutes between any two rows for them to be considered part of the same activity episode RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_activity_recognition_raw.csv - data/raw/ { pid } /phone_activity_recognition_with_datetime.csv - data/raw/ { pid } /phone_activity_recognition_with_datetime_unified.csv - data/interim/ { pid } /phone_activity_recognition_episodes.csv - data/interim/ { pid } /phone_activity_recognition_episodes_resampled.csv - data/interim/ { pid } /phone_activity_recognition_episodes_resampled_with_datetime.csv - data/interim/ { pid } /phone_activity_recognition_features/phone_activity_recognition_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_activity_recognition.csv Parameters description for [PHONE_ACTIVITY_RECOGNITION][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_ACTIVITY_RECOGNITION features from the RAPIDS provider [FEATURES] Features to be computed, see table below [ACTIVITY_CLASSES][STATIONARY] An array of the activity labels to be considered in the STATIONARY category choose any of still , tilting [ACTIVITY_CLASSES][MOBILE] An array of the activity labels to be considered in the MOBILE category choose any of on_foot , walking , running , on_bicycle [ACTIVITY_CLASSES][VEHICLE] An array of the activity labels to be considered in the VEHICLE category choose any of in_vehicule Features description for [PHONE_ACTIVITY_RECOGNITION][PROVIDERS][RAPIDS] : Feature Units Description count rows Number of episodes. mostcommonactivity activity type The most common activity type (e.g. still , on_foot , etc.). If there is a tie, the first one is chosen. countuniqueactivities activity type Number of unique activities. durationstationary minutes The total duration of [ACTIVITY_CLASSES][STATIONARY] episodes durationmobile minutes The total duration of [ACTIVITY_CLASSES][MOBILE] episodes of on foot, running, and on bicycle activities durationvehicle minutes The total duration of [ACTIVITY_CLASSES][VEHICLE] episodes of on vehicle activity Assumptions/Observations iOS Activity Recognition names and types are unified with Android labels: iOS Activity Name Android Activity Name Android Activity Type walking walking 7 running running 8 cycling on_bicycle 1 automotive in_vehicle 0 stationary still 3 unknown unknown 4 In AWARE, Activity Recognition data for Android and iOS are stored in two different database tables, RAPIDS automatically infers what platform each participant belongs to based on their participant file .","title":"Phone Activity Recognition"},{"location":"features/phone-activity-recognition/#phone-activity-recognition","text":"Sensor parameters description for [PHONE_ACTIVITY_RECOGNITION] : Key Description [TABLE][ANDROID] Database table where the activity data from Android devices is stored (the AWARE client saves this data on different tables for Android and iOS) [TABLE][IOS] Database table where the activity data from iOS devices is stored (the AWARE client saves this data on different tables for Android and iOS) [EPISODE_THRESHOLD_BETWEEN_ROWS] Difference in minutes between any two rows for them to be considered part of the same activity episode","title":"Phone Activity Recognition"},{"location":"features/phone-activity-recognition/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_activity_recognition_raw.csv - data/raw/ { pid } /phone_activity_recognition_with_datetime.csv - data/raw/ { pid } /phone_activity_recognition_with_datetime_unified.csv - data/interim/ { pid } /phone_activity_recognition_episodes.csv - data/interim/ { pid } /phone_activity_recognition_episodes_resampled.csv - data/interim/ { pid } /phone_activity_recognition_episodes_resampled_with_datetime.csv - data/interim/ { pid } /phone_activity_recognition_features/phone_activity_recognition_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_activity_recognition.csv Parameters description for [PHONE_ACTIVITY_RECOGNITION][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_ACTIVITY_RECOGNITION features from the RAPIDS provider [FEATURES] Features to be computed, see table below [ACTIVITY_CLASSES][STATIONARY] An array of the activity labels to be considered in the STATIONARY category choose any of still , tilting [ACTIVITY_CLASSES][MOBILE] An array of the activity labels to be considered in the MOBILE category choose any of on_foot , walking , running , on_bicycle [ACTIVITY_CLASSES][VEHICLE] An array of the activity labels to be considered in the VEHICLE category choose any of in_vehicule Features description for [PHONE_ACTIVITY_RECOGNITION][PROVIDERS][RAPIDS] : Feature Units Description count rows Number of episodes. mostcommonactivity activity type The most common activity type (e.g. still , on_foot , etc.). If there is a tie, the first one is chosen. countuniqueactivities activity type Number of unique activities. durationstationary minutes The total duration of [ACTIVITY_CLASSES][STATIONARY] episodes durationmobile minutes The total duration of [ACTIVITY_CLASSES][MOBILE] episodes of on foot, running, and on bicycle activities durationvehicle minutes The total duration of [ACTIVITY_CLASSES][VEHICLE] episodes of on vehicle activity Assumptions/Observations iOS Activity Recognition names and types are unified with Android labels: iOS Activity Name Android Activity Name Android Activity Type walking walking 7 running running 8 cycling on_bicycle 1 automotive in_vehicle 0 stationary still 3 unknown unknown 4 In AWARE, Activity Recognition data for Android and iOS are stored in two different database tables, RAPIDS automatically infers what platform each participant belongs to based on their participant file .","title":"RAPIDS provider"},{"location":"features/phone-applications-foreground/","text":"Phone Applications Foreground \u00b6 Sensor parameters description for [PHONE_APPLICATIONS_FOREGROUND] (these parameters are used by the only provider available at the moment, RAPIDS): Key Description [TABLE] Database table where the applications foreground data is stored [APPLICATION_CATEGORIES][CATALOGUE_SOURCE] FILE or GOOGLE . If FILE , app categories (genres) are read from [CATALOGUE_FILE] . If [GOOGLE] , app categories (genres) are scrapped from the Play Store [APPLICATION_CATEGORIES][CATALOGUE_FILE] CSV file with a package_name and genre column. By default we provide the catalogue created by Stachl et al in data/external/stachl_application_genre_catalogue.csv [APPLICATION_CATEGORIES][UPDATE_CATALOGUE_FILE] if [CATALOGUE_SOURCE] is equal to FILE , this flag signals whether or not to update [CATALOGUE_FILE] , if [CATALOGUE_SOURCE] is equal to GOOGLE all scraped genres will be saved to [CATALOGUE_FILE] [APPLICATION_CATEGORIES][SCRAPE_MISSING_CATEGORIES] This flag signals whether or not to scrape categories (genres) missing from the [CATALOGUE_FILE] . If [CATALOGUE_SOURCE] is equal to GOOGLE , all genres are scraped anyway (this flag is ignored) RAPIDS provider \u00b6 The app category (genre) catalogue used in these features was originally created by Stachl et al . Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_applications_foreground_raw.csv - data/raw/ { pid } /phone_applications_foreground_with_datetime.csv - data/raw/ { pid } /phone_applications_foreground_with_datetime_with_categories.csv - data/interim/ { pid } /phone_applications_foreground_features/phone_applications_foreground_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_applications_foreground.csv Parameters description for [PHONE_APPLICATIONS_FOREGROUND][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_APPLICATIONS_FOREGROUND features from the RAPIDS provider [FEATURES] Features to be computed, see table below [SINGLE_CATEGORIES] An array of app categories to be included in the feature extraction computation. The special keyword all represents a category with all the apps from each participant. By default we use the category catalogue pointed by [APPLICATION_CATEGORIES][CATALOGUE_FILE] (see the Sensor parameters description table above) [MULTIPLE_CATEGORIES] An array of collections representing meta-categories (a group of categories). They key of each element is the name of the meta-category and the value is an array of member app categories. By default we use the category catalogue pointed by [APPLICATION_CATEGORIES][CATALOGUE_FILE] (see the Sensor parameters description table above) [SINGLE_APPS] An array of apps to be included in the feature extraction computation. Use their package name (e.g. com.google.android.youtube ) or the reserved keyword top1global (the most used app by a participant over the whole monitoring study) [EXCLUDED_CATEGORIES] An array of app categories to be excluded from the feature extraction computation. By default we use the category catalogue pointed by [APPLICATION_CATEGORIES][CATALOGUE_FILE] (see the Sensor parameters description table above) [EXCLUDED_APPS] An array of apps to be excluded from the feature extraction computation. Use their package name, for example: com.google.android.youtube Features description for [PHONE_APPLICATIONS_FOREGROUND][PROVIDERS][RAPIDS] : Feature Units Description count apps Number of times a single app or apps within a category were used (i.e. they were brought to the foreground either by tapping their icon or switching to it from another app) timeoffirstuse minutes The time in minutes between 12:00am (midnight) and the first use of a single app or apps within a category during a time_segment timeoflastuse minutes The time in minutes between 12:00am (midnight) and the last use of a single app or apps within a category during a time_segment frequencyentropy nats The entropy of the used apps within a category during a time_segment (each app is seen as a unique event, the more apps were used, the higher the entropy). This is especially relevant when computed over all apps. Entropy cannot be obtained for a single app Assumptions/Observations Features can be computed by app, by apps grouped under a single category (genre) and by multiple categories grouped together (meta-categories). For example, we can get features for Facebook (single app), for Social Network apps (a category including Facebook and other social media apps) or for Social (a meta-category formed by Social Network and Social Media Tools categories). Apps installed by default like YouTube are considered systems apps on some phones. We do an exact match to exclude apps where \u201cgenre\u201d == EXCLUDED_CATEGORIES or \u201cpackage_name\u201d == EXCLUDED_APPS . We provide three ways of classifying and app within a category (genre): a) by automatically scraping its official category from the Google Play Store, b) by using the catalogue created by Stachl et al. which we provide in RAPIDS ( data/external/stachl_application_genre_catalogue.csv ), or c) by manually creating a personalized catalogue. You can choose a, b or c by modifying [APPLICATION_GENRES] keys and values (see the Sensor parameters description table above).","title":"Phone Applications Foreground"},{"location":"features/phone-applications-foreground/#phone-applications-foreground","text":"Sensor parameters description for [PHONE_APPLICATIONS_FOREGROUND] (these parameters are used by the only provider available at the moment, RAPIDS): Key Description [TABLE] Database table where the applications foreground data is stored [APPLICATION_CATEGORIES][CATALOGUE_SOURCE] FILE or GOOGLE . If FILE , app categories (genres) are read from [CATALOGUE_FILE] . If [GOOGLE] , app categories (genres) are scrapped from the Play Store [APPLICATION_CATEGORIES][CATALOGUE_FILE] CSV file with a package_name and genre column. By default we provide the catalogue created by Stachl et al in data/external/stachl_application_genre_catalogue.csv [APPLICATION_CATEGORIES][UPDATE_CATALOGUE_FILE] if [CATALOGUE_SOURCE] is equal to FILE , this flag signals whether or not to update [CATALOGUE_FILE] , if [CATALOGUE_SOURCE] is equal to GOOGLE all scraped genres will be saved to [CATALOGUE_FILE] [APPLICATION_CATEGORIES][SCRAPE_MISSING_CATEGORIES] This flag signals whether or not to scrape categories (genres) missing from the [CATALOGUE_FILE] . If [CATALOGUE_SOURCE] is equal to GOOGLE , all genres are scraped anyway (this flag is ignored)","title":"Phone Applications Foreground"},{"location":"features/phone-applications-foreground/#rapids-provider","text":"The app category (genre) catalogue used in these features was originally created by Stachl et al . Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_applications_foreground_raw.csv - data/raw/ { pid } /phone_applications_foreground_with_datetime.csv - data/raw/ { pid } /phone_applications_foreground_with_datetime_with_categories.csv - data/interim/ { pid } /phone_applications_foreground_features/phone_applications_foreground_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_applications_foreground.csv Parameters description for [PHONE_APPLICATIONS_FOREGROUND][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_APPLICATIONS_FOREGROUND features from the RAPIDS provider [FEATURES] Features to be computed, see table below [SINGLE_CATEGORIES] An array of app categories to be included in the feature extraction computation. The special keyword all represents a category with all the apps from each participant. By default we use the category catalogue pointed by [APPLICATION_CATEGORIES][CATALOGUE_FILE] (see the Sensor parameters description table above) [MULTIPLE_CATEGORIES] An array of collections representing meta-categories (a group of categories). They key of each element is the name of the meta-category and the value is an array of member app categories. By default we use the category catalogue pointed by [APPLICATION_CATEGORIES][CATALOGUE_FILE] (see the Sensor parameters description table above) [SINGLE_APPS] An array of apps to be included in the feature extraction computation. Use their package name (e.g. com.google.android.youtube ) or the reserved keyword top1global (the most used app by a participant over the whole monitoring study) [EXCLUDED_CATEGORIES] An array of app categories to be excluded from the feature extraction computation. By default we use the category catalogue pointed by [APPLICATION_CATEGORIES][CATALOGUE_FILE] (see the Sensor parameters description table above) [EXCLUDED_APPS] An array of apps to be excluded from the feature extraction computation. Use their package name, for example: com.google.android.youtube Features description for [PHONE_APPLICATIONS_FOREGROUND][PROVIDERS][RAPIDS] : Feature Units Description count apps Number of times a single app or apps within a category were used (i.e. they were brought to the foreground either by tapping their icon or switching to it from another app) timeoffirstuse minutes The time in minutes between 12:00am (midnight) and the first use of a single app or apps within a category during a time_segment timeoflastuse minutes The time in minutes between 12:00am (midnight) and the last use of a single app or apps within a category during a time_segment frequencyentropy nats The entropy of the used apps within a category during a time_segment (each app is seen as a unique event, the more apps were used, the higher the entropy). This is especially relevant when computed over all apps. Entropy cannot be obtained for a single app Assumptions/Observations Features can be computed by app, by apps grouped under a single category (genre) and by multiple categories grouped together (meta-categories). For example, we can get features for Facebook (single app), for Social Network apps (a category including Facebook and other social media apps) or for Social (a meta-category formed by Social Network and Social Media Tools categories). Apps installed by default like YouTube are considered systems apps on some phones. We do an exact match to exclude apps where \u201cgenre\u201d == EXCLUDED_CATEGORIES or \u201cpackage_name\u201d == EXCLUDED_APPS . We provide three ways of classifying and app within a category (genre): a) by automatically scraping its official category from the Google Play Store, b) by using the catalogue created by Stachl et al. which we provide in RAPIDS ( data/external/stachl_application_genre_catalogue.csv ), or c) by manually creating a personalized catalogue. You can choose a, b or c by modifying [APPLICATION_GENRES] keys and values (see the Sensor parameters description table above).","title":"RAPIDS provider"},{"location":"features/phone-battery/","text":"Phone Battery \u00b6 Sensor parameters description for [PHONE_BATTERY] : Key Description [TABLE] Database table where the battery data is stored [EPISODE_THRESHOLD_BETWEEN_ROWS] Difference in minutes between any two rows for them to be considered part of the same battery charge or discharge episode RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_battery_raw.csv - data/interim/ { pid } /phone_battery_episodes.csv - data/interim/ { pid } /phone_battery_episodes_resampled.csv - data/interim/ { pid } /phone_battery_episodes_resampled_with_datetime.csv - data/interim/ { pid } /phone_battery_features/phone_battery_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_battery.csv Parameters description for [PHONE_BATTERY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_BATTERY features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_BATTERY][PROVIDERS][RAPIDS] : Feature Units Description countdischarge episodes Number of discharging episodes. sumdurationdischarge minutes The total duration of all discharging episodes. countcharge episodes Number of battery charging episodes. sumdurationcharge minutes The total duration of all charging episodes. avgconsumptionrate episodes/minutes The average of all episodes\u2019 consumption rates. An episode\u2019s consumption rate is defined as the ratio between its battery delta and duration maxconsumptionrate episodes/minutes The highest of all episodes\u2019 consumption rates. An episode\u2019s consumption rate is defined as the ratio between its battery delta and duration Assumptions/Observations We convert battery data collected with iOS client v1 (autodetected because battery status 4 do not exist) to match Android battery format: we swap status 3 for 5 and 1 for 3 We group battery data into discharge or charge episodes considering any contiguous rows with consecutive reductions or increases of the battery level if they are logged within [EPISODE_THRESHOLD_BETWEEN_ROWS] minutes from each other.","title":"Phone Battery"},{"location":"features/phone-battery/#phone-battery","text":"Sensor parameters description for [PHONE_BATTERY] : Key Description [TABLE] Database table where the battery data is stored [EPISODE_THRESHOLD_BETWEEN_ROWS] Difference in minutes between any two rows for them to be considered part of the same battery charge or discharge episode","title":"Phone Battery"},{"location":"features/phone-battery/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_battery_raw.csv - data/interim/ { pid } /phone_battery_episodes.csv - data/interim/ { pid } /phone_battery_episodes_resampled.csv - data/interim/ { pid } /phone_battery_episodes_resampled_with_datetime.csv - data/interim/ { pid } /phone_battery_features/phone_battery_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_battery.csv Parameters description for [PHONE_BATTERY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_BATTERY features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_BATTERY][PROVIDERS][RAPIDS] : Feature Units Description countdischarge episodes Number of discharging episodes. sumdurationdischarge minutes The total duration of all discharging episodes. countcharge episodes Number of battery charging episodes. sumdurationcharge minutes The total duration of all charging episodes. avgconsumptionrate episodes/minutes The average of all episodes\u2019 consumption rates. An episode\u2019s consumption rate is defined as the ratio between its battery delta and duration maxconsumptionrate episodes/minutes The highest of all episodes\u2019 consumption rates. An episode\u2019s consumption rate is defined as the ratio between its battery delta and duration Assumptions/Observations We convert battery data collected with iOS client v1 (autodetected because battery status 4 do not exist) to match Android battery format: we swap status 3 for 5 and 1 for 3 We group battery data into discharge or charge episodes considering any contiguous rows with consecutive reductions or increases of the battery level if they are logged within [EPISODE_THRESHOLD_BETWEEN_ROWS] minutes from each other.","title":"RAPIDS provider"},{"location":"features/phone-bluetooth/","text":"Phone Bluetooth \u00b6 Sensor parameters description for [PHONE_BLUETOOTH] : Key Description [TABLE] Database table where the bluetooth data is stored RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_bluetooth_raw.csv - data/raw/ { pid } /phone_bluetooth_with_datetime.csv - data/interim/ { pid } /phone_bluetooth_features/phone_bluetooth_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_bluetooth.csv \" Parameters description for [PHONE_BLUETOOTH][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_BLUETOOTH features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_BLUETOOTH][PROVIDERS][RAPIDS] : Feature Units Description countscans devices Number of scanned devices during a time_segment , a device can be detected multiple times over time and these appearances are counted separately uniquedevices devices Number of unique devices during a time_segment as identified by their hardware ( bt_address ) address countscansmostuniquedevice scans Number of scans of the most scanned device during a time_segment across the whole monitoring period Assumptions/Observations NA","title":"Phone Bluetooth"},{"location":"features/phone-bluetooth/#phone-bluetooth","text":"Sensor parameters description for [PHONE_BLUETOOTH] : Key Description [TABLE] Database table where the bluetooth data is stored","title":"Phone Bluetooth"},{"location":"features/phone-bluetooth/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_bluetooth_raw.csv - data/raw/ { pid } /phone_bluetooth_with_datetime.csv - data/interim/ { pid } /phone_bluetooth_features/phone_bluetooth_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_bluetooth.csv \" Parameters description for [PHONE_BLUETOOTH][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_BLUETOOTH features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_BLUETOOTH][PROVIDERS][RAPIDS] : Feature Units Description countscans devices Number of scanned devices during a time_segment , a device can be detected multiple times over time and these appearances are counted separately uniquedevices devices Number of unique devices during a time_segment as identified by their hardware ( bt_address ) address countscansmostuniquedevice scans Number of scans of the most scanned device during a time_segment across the whole monitoring period Assumptions/Observations NA","title":"RAPIDS provider"},{"location":"features/phone-calls/","text":"Phone Calls \u00b6 Sensor parameters description for [PHONE_CALLS] : Key Description [TABLE] Database table where the calls data is stored RAPIDS Provider \u00b6 Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_calls_raw.csv - data/raw/ { pid } /phone_calls_with_datetime.csv - data/raw/ { pid } /phone_calls_with_datetime_unified.csv - data/interim/ { pid } /phone_calls_features/phone_calls_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_calls.csv Parameters description for [PHONE_CALLS][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_CALLS features from the RAPIDS provider [CALL_TYPES] The particular call_type that will be analyzed. The options for this parameter are incoming, outgoing or missed. [FEATURES] Features to be computed for outgoing , incoming , and missed calls. Note that the same features are available for both incoming and outgoing calls, while missed calls has its own set of features. See the tables below. Features description for [PHONE_CALLS][PROVIDERS][RAPIDS] incoming and outgoing calls: Feature Units Description count calls Number of calls of a particular call_type occurred during a particular time_segment . distinctcontacts contacts Number of distinct contacts that are associated with a particular call_type for a particular time_segment meanduration seconds The mean duration of all calls of a particular call_type during a particular time_segment . sumduration seconds The sum of the duration of all calls of a particular call_type during a particular time_segment . minduration seconds The duration of the shortest call of a particular call_type during a particular time_segment . maxduration seconds The duration of the longest call of a particular call_type during a particular time_segment . stdduration seconds The standard deviation of the duration of all the calls of a particular call_type during a particular time_segment . modeduration seconds The mode of the duration of all the calls of a particular call_type during a particular time_segment . entropyduration nats The estimate of the Shannon entropy for the the duration of all the calls of a particular call_type during a particular time_segment . timefirstcall minutes The time in minutes between 12:00am (midnight) and the first call of call_type . timelastcall minutes The time in minutes between 12:00am (midnight) and the last call of call_type . countmostfrequentcontact calls The number of calls of a particular call_type during a particular time_segment of the most frequent contact throughout the monitored period. Features description for [PHONE_CALLS][PROVIDERS][RAPIDS] missed calls: Feature Units Description count calls Number of missed calls that occurred during a particular time_segment . distinctcontacts contacts Number of distinct contacts that are associated with missed calls for a particular time_segment timefirstcall minutes The time in hours from 12:00am (Midnight) that the first missed call occurred. timelastcall minutes The time in hours from 12:00am (Midnight) that the last missed call occurred. countmostfrequentcontact calls The number of missed calls during a particular time_segment of the most frequent contact throughout the monitored period. Assumptions/Observations Traces for iOS calls are unique even for the same contact calling a participant more than once which renders countmostfrequentcontact meaningless and distinctcontacts equal to the total number of traces. [CALL_TYPES] and [FEATURES] keys in config.yaml need to match. For example, [CALL_TYPES] outgoing matches the [FEATURES] key outgoing iOS calls data is transformed to match Android calls data format. See our algorithm","title":"Phone Calls"},{"location":"features/phone-calls/#phone-calls","text":"Sensor parameters description for [PHONE_CALLS] : Key Description [TABLE] Database table where the calls data is stored","title":"Phone Calls"},{"location":"features/phone-calls/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_calls_raw.csv - data/raw/ { pid } /phone_calls_with_datetime.csv - data/raw/ { pid } /phone_calls_with_datetime_unified.csv - data/interim/ { pid } /phone_calls_features/phone_calls_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_calls.csv Parameters description for [PHONE_CALLS][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_CALLS features from the RAPIDS provider [CALL_TYPES] The particular call_type that will be analyzed. The options for this parameter are incoming, outgoing or missed. [FEATURES] Features to be computed for outgoing , incoming , and missed calls. Note that the same features are available for both incoming and outgoing calls, while missed calls has its own set of features. See the tables below. Features description for [PHONE_CALLS][PROVIDERS][RAPIDS] incoming and outgoing calls: Feature Units Description count calls Number of calls of a particular call_type occurred during a particular time_segment . distinctcontacts contacts Number of distinct contacts that are associated with a particular call_type for a particular time_segment meanduration seconds The mean duration of all calls of a particular call_type during a particular time_segment . sumduration seconds The sum of the duration of all calls of a particular call_type during a particular time_segment . minduration seconds The duration of the shortest call of a particular call_type during a particular time_segment . maxduration seconds The duration of the longest call of a particular call_type during a particular time_segment . stdduration seconds The standard deviation of the duration of all the calls of a particular call_type during a particular time_segment . modeduration seconds The mode of the duration of all the calls of a particular call_type during a particular time_segment . entropyduration nats The estimate of the Shannon entropy for the the duration of all the calls of a particular call_type during a particular time_segment . timefirstcall minutes The time in minutes between 12:00am (midnight) and the first call of call_type . timelastcall minutes The time in minutes between 12:00am (midnight) and the last call of call_type . countmostfrequentcontact calls The number of calls of a particular call_type during a particular time_segment of the most frequent contact throughout the monitored period. Features description for [PHONE_CALLS][PROVIDERS][RAPIDS] missed calls: Feature Units Description count calls Number of missed calls that occurred during a particular time_segment . distinctcontacts contacts Number of distinct contacts that are associated with missed calls for a particular time_segment timefirstcall minutes The time in hours from 12:00am (Midnight) that the first missed call occurred. timelastcall minutes The time in hours from 12:00am (Midnight) that the last missed call occurred. countmostfrequentcontact calls The number of missed calls during a particular time_segment of the most frequent contact throughout the monitored period. Assumptions/Observations Traces for iOS calls are unique even for the same contact calling a participant more than once which renders countmostfrequentcontact meaningless and distinctcontacts equal to the total number of traces. [CALL_TYPES] and [FEATURES] keys in config.yaml need to match. For example, [CALL_TYPES] outgoing matches the [FEATURES] key outgoing iOS calls data is transformed to match Android calls data format. See our algorithm","title":"RAPIDS Provider"},{"location":"features/phone-conversation/","text":"Phone Conversation \u00b6 Sensor parameters description for [PHONE_CONVERSATION] : Key Description [TABLE][ANDROID] Database table where the conversation data from Android devices is stored (the AWARE client saves this data on different tables for Android and iOS) [TABLE][IOS] Database table where the conversation data from iOS devices is stored (the AWARE client saves this data on different tables for Android and iOS) RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_conversation_raw.csv - data/raw/ { pid } /phone_conversation_with_datetime.csv - data/raw/ { pid } /phone_conversation_with_datetime_unified.csv - data/interim/ { pid } /phone_conversation_features/phone_conversation_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_conversation.csv Parameters description for [PHONE_CONVERSATION][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_CONVERSATION features from the RAPIDS provider [FEATURES] Features to be computed, see table below [RECORDING_MINUTES] Minutes the plugin was recording audio (default 1 min) [PAUSED_MINUTES] Minutes the plugin was NOT recording audio (default 3 min) Features description for [PHONE_CONVERSATION][PROVIDERS][RAPIDS] : Feature Units Description minutessilence minutes Minutes labeled as silence minutesnoise minutes Minutes labeled as noise minutesvoice minutes Minutes labeled as voice minutesunknown minutes Minutes labeled as unknown sumconversationduration minutes Total duration of all conversations maxconversationduration minutes Longest duration of all conversations minconversationduration minutes Shortest duration of all conversations avgconversationduration minutes Average duration of all conversations sdconversationduration minutes Standard Deviation of the duration of all conversations timefirstconversation minutes Minutes since midnight when the first conversation for a time segment was detected timelastconversation minutes Minutes since midnight when the last conversation for a time segment was detected noisesumenergy L2-norm Sum of all energy values when inference is noise noiseavgenergy L2-norm Average of all energy values when inference is noise noisesdenergy L2-norm Standard Deviation of all energy values when inference is noise noiseminenergy L2-norm Minimum of all energy values when inference is noise noisemaxenergy L2-norm Maximum of all energy values when inference is noise voicesumenergy L2-norm Sum of all energy values when inference is voice voiceavgenergy L2-norm Average of all energy values when inference is voice voicesdenergy L2-norm Standard Deviation of all energy values when inference is voice voiceminenergy L2-norm Minimum of all energy values when inference is voice voicemaxenergy L2-norm Maximum of all energy values when inference is voice silencesensedfraction - Ratio between minutessilence and the sum of (minutessilence, minutesnoise, minutesvoice, minutesunknown) noisesensedfraction - Ratio between minutesnoise and the sum of (minutessilence, minutesnoise, minutesvoice, minutesunknown) voicesensedfraction - Ratio between minutesvoice and the sum of (minutessilence, minutesnoise, minutesvoice, minutesunknown) unknownsensedfraction - Ratio between minutesunknown and the sum of (minutessilence, minutesnoise, minutesvoice, minutesunknown) silenceexpectedfraction - Ration between minutessilence and the number of minutes that in theory should have been sensed based on the record and pause cycle of the plugin (1440 / recordingMinutes+pausedMinutes) noiseexpectedfraction - Ration between minutesnoise and the number of minutes that in theory should have been sensed based on the record and pause cycle of the plugin (1440 / recordingMinutes+pausedMinutes) voiceexpectedfraction - Ration between minutesvoice and the number of minutes that in theory should have been sensed based on the record and pause cycle of the plugin (1440 / recordingMinutes+pausedMinutes) unknownexpectedfraction - Ration between minutesunknown and the number of minutes that in theory should have been sensed based on the record and pause cycle of the plugin (1440 / recordingMinutes+pausedMinutes) Assumptions/Observations The timestamp of conversation rows in iOS is in seconds so we convert it to milliseconds to match Android\u2019s format","title":"Phone Conversation"},{"location":"features/phone-conversation/#phone-conversation","text":"Sensor parameters description for [PHONE_CONVERSATION] : Key Description [TABLE][ANDROID] Database table where the conversation data from Android devices is stored (the AWARE client saves this data on different tables for Android and iOS) [TABLE][IOS] Database table where the conversation data from iOS devices is stored (the AWARE client saves this data on different tables for Android and iOS)","title":"Phone Conversation"},{"location":"features/phone-conversation/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_conversation_raw.csv - data/raw/ { pid } /phone_conversation_with_datetime.csv - data/raw/ { pid } /phone_conversation_with_datetime_unified.csv - data/interim/ { pid } /phone_conversation_features/phone_conversation_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_conversation.csv Parameters description for [PHONE_CONVERSATION][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_CONVERSATION features from the RAPIDS provider [FEATURES] Features to be computed, see table below [RECORDING_MINUTES] Minutes the plugin was recording audio (default 1 min) [PAUSED_MINUTES] Minutes the plugin was NOT recording audio (default 3 min) Features description for [PHONE_CONVERSATION][PROVIDERS][RAPIDS] : Feature Units Description minutessilence minutes Minutes labeled as silence minutesnoise minutes Minutes labeled as noise minutesvoice minutes Minutes labeled as voice minutesunknown minutes Minutes labeled as unknown sumconversationduration minutes Total duration of all conversations maxconversationduration minutes Longest duration of all conversations minconversationduration minutes Shortest duration of all conversations avgconversationduration minutes Average duration of all conversations sdconversationduration minutes Standard Deviation of the duration of all conversations timefirstconversation minutes Minutes since midnight when the first conversation for a time segment was detected timelastconversation minutes Minutes since midnight when the last conversation for a time segment was detected noisesumenergy L2-norm Sum of all energy values when inference is noise noiseavgenergy L2-norm Average of all energy values when inference is noise noisesdenergy L2-norm Standard Deviation of all energy values when inference is noise noiseminenergy L2-norm Minimum of all energy values when inference is noise noisemaxenergy L2-norm Maximum of all energy values when inference is noise voicesumenergy L2-norm Sum of all energy values when inference is voice voiceavgenergy L2-norm Average of all energy values when inference is voice voicesdenergy L2-norm Standard Deviation of all energy values when inference is voice voiceminenergy L2-norm Minimum of all energy values when inference is voice voicemaxenergy L2-norm Maximum of all energy values when inference is voice silencesensedfraction - Ratio between minutessilence and the sum of (minutessilence, minutesnoise, minutesvoice, minutesunknown) noisesensedfraction - Ratio between minutesnoise and the sum of (minutessilence, minutesnoise, minutesvoice, minutesunknown) voicesensedfraction - Ratio between minutesvoice and the sum of (minutessilence, minutesnoise, minutesvoice, minutesunknown) unknownsensedfraction - Ratio between minutesunknown and the sum of (minutessilence, minutesnoise, minutesvoice, minutesunknown) silenceexpectedfraction - Ration between minutessilence and the number of minutes that in theory should have been sensed based on the record and pause cycle of the plugin (1440 / recordingMinutes+pausedMinutes) noiseexpectedfraction - Ration between minutesnoise and the number of minutes that in theory should have been sensed based on the record and pause cycle of the plugin (1440 / recordingMinutes+pausedMinutes) voiceexpectedfraction - Ration between minutesvoice and the number of minutes that in theory should have been sensed based on the record and pause cycle of the plugin (1440 / recordingMinutes+pausedMinutes) unknownexpectedfraction - Ration between minutesunknown and the number of minutes that in theory should have been sensed based on the record and pause cycle of the plugin (1440 / recordingMinutes+pausedMinutes) Assumptions/Observations The timestamp of conversation rows in iOS is in seconds so we convert it to milliseconds to match Android\u2019s format","title":"RAPIDS provider"},{"location":"features/phone-data-yield/","text":"Phone Data Yield \u00b6 This is a combinatorial sensor which means that we use the data from multiple sensors to extract data yield features. Data yield features can be used to remove rows ( time segments ) that do not contain enough data. You should decide what is your \u201cenough\u201d threshold depending on the type of sensors you collected (frequency vs event based, e.g. acceleroemter vs calls), the length of your study, and the rates of missing data that your analysis could handle. Why is data yield important? Imagine that you want to extract PHONE_CALL features on daily segments ( 00:00 to 23:59 ). Let\u2019s say that on day 1 the phone logged 10 calls and 23 hours of data from other sensors and on day 2 the phone logged 10 calls and only 2 hours of data from other sensors. It\u2019s more likely that other calls were placed on the 22 hours of data that you didn\u2019t log on day 2 than on the 1 hour of data you didn\u2019t log on day 1, and so including day 2 in your analysis could bias your results. Sensor parameters description for [PHONE_DATA_YIELD] : Key Description [SENSORS] One or more phone sensor config keys (e.g. PHONE_MESSAGE ). The more keys you include the more accurately RAPIDS can approximate the time an smartphone was sensing data. The supported phone sensors you can include in this list are outlined below ( do NOT include Fitbit sensors ). Supported phone sensors for [PHONE_DATA_YIELD][SENSORS] PHONE_ACCELEROMETER PHONE_ACTIVITY_RECOGNITION PHONE_APPLICATIONS_FOREGROUND PHONE_BATTERY PHONE_BLUETOOTH PHONE_CALLS PHONE_CONVERSATION PHONE_MESSAGES PHONE_LIGHT PHONE_LOCATIONS PHONE_SCREEN PHONE_WIFI_VISIBLE PHONE_WIFI_CONNECTED RAPIDS provider \u00b6 Before explaining the data yield features, let\u2019s define the following relevant concepts: A valid minute is any 60 second window when any phone sensor logged at least 1 row of data A valid hour is any 60 minute window with at least X valid minutes. The X or threshold is given by [MINUTE_RATIO_THRESHOLD_FOR_VALID_YIELDED_HOURS] The timestamps of all sensors are concatenated and then grouped per time segment. Minute and hour windows are created from the beginning of each time segment instance and these windows are marked as valid based on the definitions above. The duration of each time segment is taken into account to compute the features described below. Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } / { sensor } _raw.csv # one for every [PHONE_DATA_YIELD][SENSORS] - data/interim/ { pid } /phone_yielded_timestamps.csv - data/interim/ { pid } /phone_yielded_timestamps_with_datetime.csv - data/interim/ { pid } /phone_data_yield_features/phone_data_yield_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_data_yield.csv Parameters description for [PHONE_DATA_YIELD][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_DATA_YIELD features from the RAPIDS provider [FEATURES] Features to be computed, see table below [MINUTE_RATIO_THRESHOLD_FOR_VALID_YIELDED_HOURS] The proportion [0.0 ,1.0] of valid minutes in a 60-minute window necessary to flag that window as valid. Features description for [PHONE_DATA_YIELD][PROVIDERS][RAPIDS] : Feature Units Description ratiovalidyieldedminutes rows The ratio between the number of valid minutes and the duration in minutes of a time segment. ratiovalidyieldedhours lux The ratio between the number of valid hours and the duration in hours of a time segment. If the time segment is shorter than 1 hour this feature will always be 1. Assumptions/Observations We recommend using ratiovalidyieldedminutes on time segments that are shorter than two or three hours and ratiovalidyieldedhours for longer segments. This is because relying on yielded minutes only can be misleading when a big chunk of those missing minutes are clustered together. For example, let\u2019s assume we are working with a 24-hour time segment that is missing 12 hours of data. Two extreme cases can occur: the 12 missing hours are from the beginning of the segment or 30 minutes could be missing from every hour (24 * 30 minutes = 12 hours). ratiovalidyieldedminutes would be 0.5 for both a and b (hinting the missing circumstances are similar). However, ratiovalidyieldedhours would be 0.5 for a and 1.0 for b if [MINUTE_RATIO_THRESHOLD_FOR_VALID_YIELDED_HOURS] is between [0.0 and 0.49] (hinting that the missing circumstances might be more favorable for b . In other words, sensed data for b is more evenly spread compared to a .","title":"Phone Data Yield"},{"location":"features/phone-data-yield/#phone-data-yield","text":"This is a combinatorial sensor which means that we use the data from multiple sensors to extract data yield features. Data yield features can be used to remove rows ( time segments ) that do not contain enough data. You should decide what is your \u201cenough\u201d threshold depending on the type of sensors you collected (frequency vs event based, e.g. acceleroemter vs calls), the length of your study, and the rates of missing data that your analysis could handle. Why is data yield important? Imagine that you want to extract PHONE_CALL features on daily segments ( 00:00 to 23:59 ). Let\u2019s say that on day 1 the phone logged 10 calls and 23 hours of data from other sensors and on day 2 the phone logged 10 calls and only 2 hours of data from other sensors. It\u2019s more likely that other calls were placed on the 22 hours of data that you didn\u2019t log on day 2 than on the 1 hour of data you didn\u2019t log on day 1, and so including day 2 in your analysis could bias your results. Sensor parameters description for [PHONE_DATA_YIELD] : Key Description [SENSORS] One or more phone sensor config keys (e.g. PHONE_MESSAGE ). The more keys you include the more accurately RAPIDS can approximate the time an smartphone was sensing data. The supported phone sensors you can include in this list are outlined below ( do NOT include Fitbit sensors ). Supported phone sensors for [PHONE_DATA_YIELD][SENSORS] PHONE_ACCELEROMETER PHONE_ACTIVITY_RECOGNITION PHONE_APPLICATIONS_FOREGROUND PHONE_BATTERY PHONE_BLUETOOTH PHONE_CALLS PHONE_CONVERSATION PHONE_MESSAGES PHONE_LIGHT PHONE_LOCATIONS PHONE_SCREEN PHONE_WIFI_VISIBLE PHONE_WIFI_CONNECTED","title":"Phone Data Yield"},{"location":"features/phone-data-yield/#rapids-provider","text":"Before explaining the data yield features, let\u2019s define the following relevant concepts: A valid minute is any 60 second window when any phone sensor logged at least 1 row of data A valid hour is any 60 minute window with at least X valid minutes. The X or threshold is given by [MINUTE_RATIO_THRESHOLD_FOR_VALID_YIELDED_HOURS] The timestamps of all sensors are concatenated and then grouped per time segment. Minute and hour windows are created from the beginning of each time segment instance and these windows are marked as valid based on the definitions above. The duration of each time segment is taken into account to compute the features described below. Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } / { sensor } _raw.csv # one for every [PHONE_DATA_YIELD][SENSORS] - data/interim/ { pid } /phone_yielded_timestamps.csv - data/interim/ { pid } /phone_yielded_timestamps_with_datetime.csv - data/interim/ { pid } /phone_data_yield_features/phone_data_yield_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_data_yield.csv Parameters description for [PHONE_DATA_YIELD][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_DATA_YIELD features from the RAPIDS provider [FEATURES] Features to be computed, see table below [MINUTE_RATIO_THRESHOLD_FOR_VALID_YIELDED_HOURS] The proportion [0.0 ,1.0] of valid minutes in a 60-minute window necessary to flag that window as valid. Features description for [PHONE_DATA_YIELD][PROVIDERS][RAPIDS] : Feature Units Description ratiovalidyieldedminutes rows The ratio between the number of valid minutes and the duration in minutes of a time segment. ratiovalidyieldedhours lux The ratio between the number of valid hours and the duration in hours of a time segment. If the time segment is shorter than 1 hour this feature will always be 1. Assumptions/Observations We recommend using ratiovalidyieldedminutes on time segments that are shorter than two or three hours and ratiovalidyieldedhours for longer segments. This is because relying on yielded minutes only can be misleading when a big chunk of those missing minutes are clustered together. For example, let\u2019s assume we are working with a 24-hour time segment that is missing 12 hours of data. Two extreme cases can occur: the 12 missing hours are from the beginning of the segment or 30 minutes could be missing from every hour (24 * 30 minutes = 12 hours). ratiovalidyieldedminutes would be 0.5 for both a and b (hinting the missing circumstances are similar). However, ratiovalidyieldedhours would be 0.5 for a and 1.0 for b if [MINUTE_RATIO_THRESHOLD_FOR_VALID_YIELDED_HOURS] is between [0.0 and 0.49] (hinting that the missing circumstances might be more favorable for b . In other words, sensed data for b is more evenly spread compared to a .","title":"RAPIDS provider"},{"location":"features/phone-light/","text":"Phone Light \u00b6 Sensor parameters description for [PHONE_LIGHT] : Key Description [TABLE] Database table where the light data is stored RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_light_raw.csv - data/raw/ { pid } /phone_light_with_datetime.csv - data/interim/ { pid } /phone_light_features/phone_light_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_light.csv Parameters description for [PHONE_LIGHT][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_LIGHT features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_LIGHT][PROVIDERS][RAPIDS] : Feature Units Description count rows Number light sensor rows recorded. maxlux lux The maximum ambient luminance. minlux lux The minimum ambient luminance. avglux lux The average ambient luminance. medianlux lux The median ambient luminance. stdlux lux The standard deviation of ambient luminance. Assumptions/Observations NA","title":"Phone Light"},{"location":"features/phone-light/#phone-light","text":"Sensor parameters description for [PHONE_LIGHT] : Key Description [TABLE] Database table where the light data is stored","title":"Phone Light"},{"location":"features/phone-light/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_light_raw.csv - data/raw/ { pid } /phone_light_with_datetime.csv - data/interim/ { pid } /phone_light_features/phone_light_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_light.csv Parameters description for [PHONE_LIGHT][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_LIGHT features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_LIGHT][PROVIDERS][RAPIDS] : Feature Units Description count rows Number light sensor rows recorded. maxlux lux The maximum ambient luminance. minlux lux The minimum ambient luminance. avglux lux The average ambient luminance. medianlux lux The median ambient luminance. stdlux lux The standard deviation of ambient luminance. Assumptions/Observations NA","title":"RAPIDS provider"},{"location":"features/phone-locations/","text":"Phone Locations \u00b6 Sensor parameters description for [PHONE_LOCATIONS] : Key Description [TABLE] Database table where the location data is stored [LOCATIONS_TO_USE] Type of location data to use, one of ALL , GPS or FUSED_RESAMPLED . This filter is based on the provider column of the AWARE locations table, ALL includes every row, GPS only includes rows where provider is gps, and FUSED_RESAMPLED only includes rows where provider is fused after being resampled. [FUSED_RESAMPLED_CONSECUTIVE_THRESHOLD] if FUSED_RESAMPLED is used, the original fused data has to be resampled, a location row will be resampled to the next valid timestamp (see the Assumptions/Observations below) only if the time difference between them is less or equal than this threshold (in minutes). [FUSED_RESAMPLED_TIME_SINCE_VALID_LOCATION] if FUSED_RESAMPLED is used, the original fused data has to be resampled, a location row will be resampled at most for this long (in minutes) Assumptions/Observations Types of location data to use AWARE Android and iOS clients can collect location coordinates through the phone's GPS, the network cellular towers around the phone or Google's fused location API. If you want to use only the GPS provider set [LOCATIONS_TO_USE] to GPS , if you want to use all providers (not recommended due to the difference in accuracy) set [LOCATIONS_TO_USE] to ALL , if your AWARE client was configured to use fused location only or want to focus only on this provider, set [LOCATIONS_TO_USE] to RESAMPLE_FUSED . RESAMPLE_FUSED takes the original fused location coordinates and replicates each pair forward in time as long as the phone was sensing data as indicated by the joined timestamps of [PHONE_DATA_YIELD][SENSORS] , this is done because Google's API only logs a new location coordinate pair when it is sufficiently different in time or space from the previous one. There are two parameters associated with resampling fused location. FUSED_RESAMPLED_CONSECUTIVE_THRESHOLD (in minutes, default 30) controls the maximum gap between any two coordinate pairs to replicate the last known pair (for example, participant A's phone did not collect data between 10.30am and 10:50am and between 11:05am and 11:40am, the last known coordinate pair will be replicated during the first period but not the second, in other words, we assume that we cannot longer guarantee the participant stayed at the last known location if the phone did not sense data for more than 30 minutes). FUSED_RESAMPLED_TIME_SINCE_VALID_LOCATION (in minutes, default 720 or 12 hours) stops the last known fused location from being replicated longer that this threshold even if the phone was sensing data continuously (for example, participant A went home at 9pm and their phone was sensing data without gaps until 11am the next morning, the last known location will only be replicated until 9am). If you have suggestions to modify or improve this resampling, let us know. BARNETT provider \u00b6 These features are based on the original open-source implementation by Barnett et al and some features created by Canzian et al . Available time segments and platforms Available only for segments that start at 00:00:00 and end at 23:59:59 of the same day (daily segments) Available for Android and iOS File Sequence - data/raw/ { pid } /phone_locations_raw.csv - data/interim/ { pid } /phone_locations_processed.csv - data/interim/ { pid } /phone_locations_processed_with_datetime.csv - data/interim/ { pid } /phone_locations_features/phone_locations_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_locations.csv Parameters description for [PHONE_LOCATIONS][PROVIDERS][BARNETT] : Key Description [COMPUTE] Set to True to extract PHONE_LOCATIONS features from the BARNETT provider [FEATURES] Features to be computed, see table below [ACCURACY_LIMIT] An integer in meters, any location rows with an accuracy higher than this will be dropped. This number means there\u2019s a 68% probability the true location is within this radius [TIMEZONE] Timezone where the location data was collected. By default points to the one defined in the Configuration [MINUTES_DATA_USED] Set to True to include an extra column in the final location feature file containing the number of minutes used to compute the features on each time segment. Use this for quality control purposes, the more data minutes exist for a period, the more reliable its features should be. For fused location, a single minute can contain more than one coordinate pair if the participant is moving fast enough. Features description for [PHONE_LOCATIONS][PROVIDERS][BARNETT] adapted from Beiwe Summary Statistics : Feature Units Description hometime minutes Time at home. Time spent at home in minutes. Home is the most visited significant location between 8 pm and 8 am including any pauses within a 200-meter radius. disttravelled meters Total distance travelled over a day (flights). rog meters The Radius of Gyration (rog) is a measure in meters of the area covered by a person over a day. A centroid is calculated for all the places (pauses) visited during a day and a weighted distance between all the places and that centroid is computed. The weights are proportional to the time spent in each place. maxdiam meters The maximum diameter is the largest distance between any two pauses. maxhomedist meters The maximum distance from home in meters. siglocsvisited locations The number of significant locations visited during the day. Significant locations are computed using k-means clustering over pauses found in the whole monitoring period. The number of clusters is found iterating k from 1 to 200 stopping until the centroids of two significant locations are within 400 meters of one another. avgflightlen meters Mean length of all flights. stdflightlen meters Standard deviation of the length of all flights. avgflightdur seconds Mean duration of all flights. stdflightdur seconds The standard deviation of the duration of all flights. probpause - The fraction of a day spent in a pause (as opposed to a flight) siglocentropy nats Shannon\u2019s entropy measurement based on the proportion of time spent at each significant location visited during a day. circdnrtn - A continuous metric quantifying a person\u2019s circadian routine that can take any value between 0 and 1, where 0 represents a daily routine completely different from any other sensed days and 1 a routine the same as every other sensed day. wkenddayrtn - Same as circdnrtn but computed separately for weekends and weekdays. Assumptions/Observations Barnett's et al features These features are based on a Pause-Flight model. A pause is defined as a mobiity trace (location pings) within a certain duration and distance (by default 300 seconds and 60 meters). A flight is any mobility trace between two pauses. Data is resampled and imputed before the features are computed. See Barnett et al for more information. In RAPIDS we only expose two parameters for these features (timezone and accuracy limit). You can change other parameters in src/features/phone_locations/barnett/library/MobilityFeatures.R . Significant Locations Significant locations are determined using K-means clustering on pauses longer than 10 minutes. The number of clusters (K) is increased until no two clusters are within 400 meters from each other. After this, pauses within a certain range of a cluster (200 meters by default) will count as a visit to that significant location. This description was adapted from the Supplementary Materials of Barnett et al . The Circadian Calculation For a detailed description of how this is calculated, see Canzian et al . DORYAB provider \u00b6 These features are based on the original implementation by Doryab et al. . Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_locations_raw.csv - data/interim/ { pid } /phone_locations_processed.csv - data/interim/ { pid } /phone_locations_processed_with_datetime.csv - data/interim/ { pid } /phone_locations_features/phone_locations_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_locations.csv Parameters description for [PHONE_LOCATIONS][PROVIDERS][BARNETT] : Key Description [COMPUTE] Set to True to extract PHONE_LOCATIONS features from the BARNETT provider [FEATURES] Features to be computed, see table below [DBSCAN_EPS] The maximum distance in meters between two samples for one to be considered as in the neighborhood of the other. This is not a maximum bound on the distances of points within a cluster. This is the most important DBSCAN parameter to choose appropriately for your data set and distance function. [DBSCAN_MINSAMPLES] The number of samples (or total weight) in a neighborhood for a point to be considered as a core point of a cluster. This includes the point itself. [THRESHOLD_STATIC] It is the threshold value in km/hr which labels a row as Static or Moving. [MAXIMUM_GAP_ALLOWED] The maximum gap (in seconds) allowed between any two consecutive rows for them to be considered part of the same displacement. If this threshold is too high, it can throw speed and distance calculations off for periods when the the phone was not sensing. [MINUTES_DATA_USED] Set to True to include an extra column in the final location feature file containing the number of minutes used to compute the features on each time segment. Use this for quality control purposes, the more data minutes exist for a period, the more reliable its features should be. For fused location, a single minute can contain more than one coordinate pair if the participant is moving fast enough. [SAMPLING_FREQUENCY] Expected time difference between any two location rows in minutes. If set to 0 , the sampling frequency will be inferred automatically as the median of all the differences between any two consecutive row timestamps (recommended if you are using FUSED_RESAMPLED data). This parameter impacts all the time calculations. Features description for [PHONE_LOCATIONS][PROVIDERS][BARNETT] : Feature Units Description locationvariance \\(meters^2\\) The sum of the variances of the latitude and longitude columns. loglocationvariance - Log of the sum of the variances of the latitude and longitude columns. totaldistance meters Total distance travelled in a time segment using the haversine formula. averagespeed km/hr Average speed in a time segment considering only the instances labeled as Moving. varspeed km/hr Speed variance in a time segment considering only the instances labeled as Moving. circadianmovement - \"It encodes the extent to which a person\u2019s location patterns follow a 24-hour circadian cycle.\" Doryab et al. . numberofsignificantplaces places Number of significant locations visited. It is calculated using the DBSCAN clustering algorithm which takes in EPS and MIN_SAMPLES as parameters to identify clusters. Each cluster is a significant place. numberlocationtransitions transitions Number of movements between any two clusters in a time segment. radiusgyration meters Quantifies the area covered by a participant timeattop1location minutes Time spent at the most significant location. timeattop2location minutes Time spent at the 2 nd most significant location. timeattop3location minutes Time spent at the 3 rd most significant location. movingtostaticratio - Ratio between the number of rows labeled Moving versus Static outlierstimepercent - Ratio between the number of rows that belong to non-significant clusters divided by the total number of rows in a time segment. maxlengthstayatclusters minutes Maximum time spent in a cluster (significant location). minlengthstayatclusters minutes Minimum time spent in a cluster (significant location). meanlengthstayatclusters minutes Average time spent in a cluster (significant location). stdlengthstayatclusters minutes Standard deviation of time spent in a cluster (significant location). locationentropy nats Shannon Entropy computed over the row count of each cluster (significant location), it will be higher the more rows belong to a cluster (i.e. the more time a participant spent at a significant location). normalizedlocationentropy nats Shannon Entropy computed over the row count of each cluster (significant location) divided by the number of clusters, it will be higher the more rows belong to a cluster (i.e. the more time a participant spent at a significant location). Assumptions/Observations Significant Locations Identified Significant locations are determined using DBSCAN clustering on locations that a patient visit over the course of the period of data collection. The Circadian Calculation For a detailed description of how this is calculated, see Canzian et al .","title":"Phone Locations"},{"location":"features/phone-locations/#phone-locations","text":"Sensor parameters description for [PHONE_LOCATIONS] : Key Description [TABLE] Database table where the location data is stored [LOCATIONS_TO_USE] Type of location data to use, one of ALL , GPS or FUSED_RESAMPLED . This filter is based on the provider column of the AWARE locations table, ALL includes every row, GPS only includes rows where provider is gps, and FUSED_RESAMPLED only includes rows where provider is fused after being resampled. [FUSED_RESAMPLED_CONSECUTIVE_THRESHOLD] if FUSED_RESAMPLED is used, the original fused data has to be resampled, a location row will be resampled to the next valid timestamp (see the Assumptions/Observations below) only if the time difference between them is less or equal than this threshold (in minutes). [FUSED_RESAMPLED_TIME_SINCE_VALID_LOCATION] if FUSED_RESAMPLED is used, the original fused data has to be resampled, a location row will be resampled at most for this long (in minutes) Assumptions/Observations Types of location data to use AWARE Android and iOS clients can collect location coordinates through the phone's GPS, the network cellular towers around the phone or Google's fused location API. If you want to use only the GPS provider set [LOCATIONS_TO_USE] to GPS , if you want to use all providers (not recommended due to the difference in accuracy) set [LOCATIONS_TO_USE] to ALL , if your AWARE client was configured to use fused location only or want to focus only on this provider, set [LOCATIONS_TO_USE] to RESAMPLE_FUSED . RESAMPLE_FUSED takes the original fused location coordinates and replicates each pair forward in time as long as the phone was sensing data as indicated by the joined timestamps of [PHONE_DATA_YIELD][SENSORS] , this is done because Google's API only logs a new location coordinate pair when it is sufficiently different in time or space from the previous one. There are two parameters associated with resampling fused location. FUSED_RESAMPLED_CONSECUTIVE_THRESHOLD (in minutes, default 30) controls the maximum gap between any two coordinate pairs to replicate the last known pair (for example, participant A's phone did not collect data between 10.30am and 10:50am and between 11:05am and 11:40am, the last known coordinate pair will be replicated during the first period but not the second, in other words, we assume that we cannot longer guarantee the participant stayed at the last known location if the phone did not sense data for more than 30 minutes). FUSED_RESAMPLED_TIME_SINCE_VALID_LOCATION (in minutes, default 720 or 12 hours) stops the last known fused location from being replicated longer that this threshold even if the phone was sensing data continuously (for example, participant A went home at 9pm and their phone was sensing data without gaps until 11am the next morning, the last known location will only be replicated until 9am). If you have suggestions to modify or improve this resampling, let us know.","title":"Phone Locations"},{"location":"features/phone-locations/#barnett-provider","text":"These features are based on the original open-source implementation by Barnett et al and some features created by Canzian et al . Available time segments and platforms Available only for segments that start at 00:00:00 and end at 23:59:59 of the same day (daily segments) Available for Android and iOS File Sequence - data/raw/ { pid } /phone_locations_raw.csv - data/interim/ { pid } /phone_locations_processed.csv - data/interim/ { pid } /phone_locations_processed_with_datetime.csv - data/interim/ { pid } /phone_locations_features/phone_locations_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_locations.csv Parameters description for [PHONE_LOCATIONS][PROVIDERS][BARNETT] : Key Description [COMPUTE] Set to True to extract PHONE_LOCATIONS features from the BARNETT provider [FEATURES] Features to be computed, see table below [ACCURACY_LIMIT] An integer in meters, any location rows with an accuracy higher than this will be dropped. This number means there\u2019s a 68% probability the true location is within this radius [TIMEZONE] Timezone where the location data was collected. By default points to the one defined in the Configuration [MINUTES_DATA_USED] Set to True to include an extra column in the final location feature file containing the number of minutes used to compute the features on each time segment. Use this for quality control purposes, the more data minutes exist for a period, the more reliable its features should be. For fused location, a single minute can contain more than one coordinate pair if the participant is moving fast enough. Features description for [PHONE_LOCATIONS][PROVIDERS][BARNETT] adapted from Beiwe Summary Statistics : Feature Units Description hometime minutes Time at home. Time spent at home in minutes. Home is the most visited significant location between 8 pm and 8 am including any pauses within a 200-meter radius. disttravelled meters Total distance travelled over a day (flights). rog meters The Radius of Gyration (rog) is a measure in meters of the area covered by a person over a day. A centroid is calculated for all the places (pauses) visited during a day and a weighted distance between all the places and that centroid is computed. The weights are proportional to the time spent in each place. maxdiam meters The maximum diameter is the largest distance between any two pauses. maxhomedist meters The maximum distance from home in meters. siglocsvisited locations The number of significant locations visited during the day. Significant locations are computed using k-means clustering over pauses found in the whole monitoring period. The number of clusters is found iterating k from 1 to 200 stopping until the centroids of two significant locations are within 400 meters of one another. avgflightlen meters Mean length of all flights. stdflightlen meters Standard deviation of the length of all flights. avgflightdur seconds Mean duration of all flights. stdflightdur seconds The standard deviation of the duration of all flights. probpause - The fraction of a day spent in a pause (as opposed to a flight) siglocentropy nats Shannon\u2019s entropy measurement based on the proportion of time spent at each significant location visited during a day. circdnrtn - A continuous metric quantifying a person\u2019s circadian routine that can take any value between 0 and 1, where 0 represents a daily routine completely different from any other sensed days and 1 a routine the same as every other sensed day. wkenddayrtn - Same as circdnrtn but computed separately for weekends and weekdays. Assumptions/Observations Barnett's et al features These features are based on a Pause-Flight model. A pause is defined as a mobiity trace (location pings) within a certain duration and distance (by default 300 seconds and 60 meters). A flight is any mobility trace between two pauses. Data is resampled and imputed before the features are computed. See Barnett et al for more information. In RAPIDS we only expose two parameters for these features (timezone and accuracy limit). You can change other parameters in src/features/phone_locations/barnett/library/MobilityFeatures.R . Significant Locations Significant locations are determined using K-means clustering on pauses longer than 10 minutes. The number of clusters (K) is increased until no two clusters are within 400 meters from each other. After this, pauses within a certain range of a cluster (200 meters by default) will count as a visit to that significant location. This description was adapted from the Supplementary Materials of Barnett et al . The Circadian Calculation For a detailed description of how this is calculated, see Canzian et al .","title":"BARNETT provider"},{"location":"features/phone-locations/#doryab-provider","text":"These features are based on the original implementation by Doryab et al. . Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_locations_raw.csv - data/interim/ { pid } /phone_locations_processed.csv - data/interim/ { pid } /phone_locations_processed_with_datetime.csv - data/interim/ { pid } /phone_locations_features/phone_locations_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_locations.csv Parameters description for [PHONE_LOCATIONS][PROVIDERS][BARNETT] : Key Description [COMPUTE] Set to True to extract PHONE_LOCATIONS features from the BARNETT provider [FEATURES] Features to be computed, see table below [DBSCAN_EPS] The maximum distance in meters between two samples for one to be considered as in the neighborhood of the other. This is not a maximum bound on the distances of points within a cluster. This is the most important DBSCAN parameter to choose appropriately for your data set and distance function. [DBSCAN_MINSAMPLES] The number of samples (or total weight) in a neighborhood for a point to be considered as a core point of a cluster. This includes the point itself. [THRESHOLD_STATIC] It is the threshold value in km/hr which labels a row as Static or Moving. [MAXIMUM_GAP_ALLOWED] The maximum gap (in seconds) allowed between any two consecutive rows for them to be considered part of the same displacement. If this threshold is too high, it can throw speed and distance calculations off for periods when the the phone was not sensing. [MINUTES_DATA_USED] Set to True to include an extra column in the final location feature file containing the number of minutes used to compute the features on each time segment. Use this for quality control purposes, the more data minutes exist for a period, the more reliable its features should be. For fused location, a single minute can contain more than one coordinate pair if the participant is moving fast enough. [SAMPLING_FREQUENCY] Expected time difference between any two location rows in minutes. If set to 0 , the sampling frequency will be inferred automatically as the median of all the differences between any two consecutive row timestamps (recommended if you are using FUSED_RESAMPLED data). This parameter impacts all the time calculations. Features description for [PHONE_LOCATIONS][PROVIDERS][BARNETT] : Feature Units Description locationvariance \\(meters^2\\) The sum of the variances of the latitude and longitude columns. loglocationvariance - Log of the sum of the variances of the latitude and longitude columns. totaldistance meters Total distance travelled in a time segment using the haversine formula. averagespeed km/hr Average speed in a time segment considering only the instances labeled as Moving. varspeed km/hr Speed variance in a time segment considering only the instances labeled as Moving. circadianmovement - \"It encodes the extent to which a person\u2019s location patterns follow a 24-hour circadian cycle.\" Doryab et al. . numberofsignificantplaces places Number of significant locations visited. It is calculated using the DBSCAN clustering algorithm which takes in EPS and MIN_SAMPLES as parameters to identify clusters. Each cluster is a significant place. numberlocationtransitions transitions Number of movements between any two clusters in a time segment. radiusgyration meters Quantifies the area covered by a participant timeattop1location minutes Time spent at the most significant location. timeattop2location minutes Time spent at the 2 nd most significant location. timeattop3location minutes Time spent at the 3 rd most significant location. movingtostaticratio - Ratio between the number of rows labeled Moving versus Static outlierstimepercent - Ratio between the number of rows that belong to non-significant clusters divided by the total number of rows in a time segment. maxlengthstayatclusters minutes Maximum time spent in a cluster (significant location). minlengthstayatclusters minutes Minimum time spent in a cluster (significant location). meanlengthstayatclusters minutes Average time spent in a cluster (significant location). stdlengthstayatclusters minutes Standard deviation of time spent in a cluster (significant location). locationentropy nats Shannon Entropy computed over the row count of each cluster (significant location), it will be higher the more rows belong to a cluster (i.e. the more time a participant spent at a significant location). normalizedlocationentropy nats Shannon Entropy computed over the row count of each cluster (significant location) divided by the number of clusters, it will be higher the more rows belong to a cluster (i.e. the more time a participant spent at a significant location). Assumptions/Observations Significant Locations Identified Significant locations are determined using DBSCAN clustering on locations that a patient visit over the course of the period of data collection. The Circadian Calculation For a detailed description of how this is calculated, see Canzian et al .","title":"DORYAB provider"},{"location":"features/phone-messages/","text":"Phone Messages \u00b6 Sensor parameters description for [PHONE_MESSAGES] : Key Description [TABLE] Database table where the messages data is stored RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_messages_raw.csv - data/raw/ { pid } /phone_messages_with_datetime.csv - data/interim/ { pid } /phone_messages_features/phone_messages_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_messages.csv Parameters description for [PHONE_MESSAGES][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_MESSAGES features from the RAPIDS provider [MESSAGES_TYPES] The messages_type that will be analyzed. The options for this parameter are received or sent . [FEATURES] Features to be computed, see table below for [MESSAGES_TYPES] received and sent Features description for [PHONE_MESSAGES][PROVIDERS][RAPIDS] : Feature Units Description count messages Number of messages of type messages_type that occurred during a particular time_segment . distinctcontacts contacts Number of distinct contacts that are associated with a particular messages_type during a particular time_segment . timefirstmessages minutes Number of minutes between 12:00am (midnight) and the first message of a particular messages_type during a particular time_segment . timelastmessages minutes Number of minutes between 12:00am (midnight) and the last message of a particular messages_type during a particular time_segment . countmostfrequentcontact messages Number of messages from the contact with the most messages of messages_type during a time_segment throughout the whole dataset of each participant. Assumptions/Observations [MESSAGES_TYPES] and [FEATURES] keys in config.yaml need to match. For example, [MESSAGES_TYPES] sent matches the [FEATURES] key sent","title":"Phone Messages"},{"location":"features/phone-messages/#phone-messages","text":"Sensor parameters description for [PHONE_MESSAGES] : Key Description [TABLE] Database table where the messages data is stored","title":"Phone Messages"},{"location":"features/phone-messages/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_messages_raw.csv - data/raw/ { pid } /phone_messages_with_datetime.csv - data/interim/ { pid } /phone_messages_features/phone_messages_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_messages.csv Parameters description for [PHONE_MESSAGES][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_MESSAGES features from the RAPIDS provider [MESSAGES_TYPES] The messages_type that will be analyzed. The options for this parameter are received or sent . [FEATURES] Features to be computed, see table below for [MESSAGES_TYPES] received and sent Features description for [PHONE_MESSAGES][PROVIDERS][RAPIDS] : Feature Units Description count messages Number of messages of type messages_type that occurred during a particular time_segment . distinctcontacts contacts Number of distinct contacts that are associated with a particular messages_type during a particular time_segment . timefirstmessages minutes Number of minutes between 12:00am (midnight) and the first message of a particular messages_type during a particular time_segment . timelastmessages minutes Number of minutes between 12:00am (midnight) and the last message of a particular messages_type during a particular time_segment . countmostfrequentcontact messages Number of messages from the contact with the most messages of messages_type during a time_segment throughout the whole dataset of each participant. Assumptions/Observations [MESSAGES_TYPES] and [FEATURES] keys in config.yaml need to match. For example, [MESSAGES_TYPES] sent matches the [FEATURES] key sent","title":"RAPIDS provider"},{"location":"features/phone-screen/","text":"Phone Screen \u00b6 Sensor parameters description for [PHONE_SCREEN] : Key Description [TABLE] Database table where the screen data is stored RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_screen_raw.csv - data/raw/ { pid } /phone_screen_with_datetime.csv - data/raw/ { pid } /phone_screen_with_datetime_unified.csv - data/interim/ { pid } /phone_screen_episodes.csv - data/interim/ { pid } /phone_screen_episodes_resampled.csv - data/interim/ { pid } /phone_screen_episodes_resampled_with_datetime.csv - data/interim/ { pid } /phone_screen_features/phone_screen_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_screen.csv Parameters description for [PHONE_SCREEN][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_SCREEN features from the RAPIDS provider [FEATURES] Features to be computed, see table below [REFERENCE_HOUR_FIRST_USE] The reference point from which firstuseafter is to be computed, default is midnight [IGNORE_EPISODES_SHORTER_THAN] Ignore episodes that are shorter than this threshold (minutes). Set to 0 to disable this filter. [IGNORE_EPISODES_LONGER_THAN] Ignore episodes that are longer than this threshold (minutes). Set to 0 to disable this filter. [EPISODE_TYPES] Currently we only support unlock episodes (from when the phone is unlocked until the screen is off) Features description for [PHONE_SCREEN][PROVIDERS][RAPIDS] : Feature Units Description sumduration minutes Total duration of all unlock episodes. maxduration minutes Longest duration of any unlock episode. minduration minutes Shortest duration of any unlock episode. avgduration minutes Average duration of all unlock episodes. stdduration minutes Standard deviation duration of all unlock episodes. countepisode episodes Number of all unlock episodes |firstuseafter |minutes |Minutes until the first unlock episode. Assumptions/Observations In Android, lock events can happen right after an off event, after a few seconds of an off event, or never happen depending on the phone's settings, therefore, an unlock episode is defined as the time between an unlock and a off event. In iOS, on and off events do not exist, so an unlock episode is defined as the time between an unlock and a lock event. Events in iOS are recorded reliably albeit some duplicated lock events within milliseconds from each other, so we only keep consecutive unlock/lock pairs. In Android you cand find multiple consecutive unlock or lock events, so we only keep consecutive unlock/off pairs. In our experiments these cases are less than 10% of the screen events collected and this happens because ACTION_SCREEN_OFF and ACTION_SCREEN_ON are sent when the device becomes non-interactive which may have nothing to do with the screen turning off . In addition to unlock/off episodes, in Android it is possible to measure the time spent on the lock screen before an unlock event as well as the total screen time (i.e. ON to OFF ) but these are not implemented at the moment. We transform iOS screen events to match Android\u2019s format, we replace lock episodes with off episodes (2 with 0) in iOS. However, as mentioned above this is still computing unlock to lock episodes.","title":"Phone Screen"},{"location":"features/phone-screen/#phone-screen","text":"Sensor parameters description for [PHONE_SCREEN] : Key Description [TABLE] Database table where the screen data is stored","title":"Phone Screen"},{"location":"features/phone-screen/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_screen_raw.csv - data/raw/ { pid } /phone_screen_with_datetime.csv - data/raw/ { pid } /phone_screen_with_datetime_unified.csv - data/interim/ { pid } /phone_screen_episodes.csv - data/interim/ { pid } /phone_screen_episodes_resampled.csv - data/interim/ { pid } /phone_screen_episodes_resampled_with_datetime.csv - data/interim/ { pid } /phone_screen_features/phone_screen_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_screen.csv Parameters description for [PHONE_SCREEN][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_SCREEN features from the RAPIDS provider [FEATURES] Features to be computed, see table below [REFERENCE_HOUR_FIRST_USE] The reference point from which firstuseafter is to be computed, default is midnight [IGNORE_EPISODES_SHORTER_THAN] Ignore episodes that are shorter than this threshold (minutes). Set to 0 to disable this filter. [IGNORE_EPISODES_LONGER_THAN] Ignore episodes that are longer than this threshold (minutes). Set to 0 to disable this filter. [EPISODE_TYPES] Currently we only support unlock episodes (from when the phone is unlocked until the screen is off) Features description for [PHONE_SCREEN][PROVIDERS][RAPIDS] : Feature Units Description sumduration minutes Total duration of all unlock episodes. maxduration minutes Longest duration of any unlock episode. minduration minutes Shortest duration of any unlock episode. avgduration minutes Average duration of all unlock episodes. stdduration minutes Standard deviation duration of all unlock episodes. countepisode episodes Number of all unlock episodes |firstuseafter |minutes |Minutes until the first unlock episode. Assumptions/Observations In Android, lock events can happen right after an off event, after a few seconds of an off event, or never happen depending on the phone's settings, therefore, an unlock episode is defined as the time between an unlock and a off event. In iOS, on and off events do not exist, so an unlock episode is defined as the time between an unlock and a lock event. Events in iOS are recorded reliably albeit some duplicated lock events within milliseconds from each other, so we only keep consecutive unlock/lock pairs. In Android you cand find multiple consecutive unlock or lock events, so we only keep consecutive unlock/off pairs. In our experiments these cases are less than 10% of the screen events collected and this happens because ACTION_SCREEN_OFF and ACTION_SCREEN_ON are sent when the device becomes non-interactive which may have nothing to do with the screen turning off . In addition to unlock/off episodes, in Android it is possible to measure the time spent on the lock screen before an unlock event as well as the total screen time (i.e. ON to OFF ) but these are not implemented at the moment. We transform iOS screen events to match Android\u2019s format, we replace lock episodes with off episodes (2 with 0) in iOS. However, as mentioned above this is still computing unlock to lock episodes.","title":"RAPIDS provider"},{"location":"features/phone-wifi-connected/","text":"Phone WiFi Connected \u00b6 Sensor parameters description for [PHONE_WIFI_CONNECTED] : Key Description [TABLE] Database table where the wifi (connected) data is stored RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_wifi_connected_raw.csv - data/raw/ { pid } /phone_wifi_connected_with_datetime.csv - data/interim/ { pid } /phone_wifi_connected_features/phone_wifi_connected_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_wifi_connected.csv Parameters description for [PHONE_WIFI_CONNECTED][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_WIFI_CONNECTED features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_WIFI_CONNECTED][PROVIDERS][RAPIDS] : Feature Units Description countscans devices Number of scanned WiFi access points connected during a time_segment, an access point can be detected multiple times over time and these appearances are counted separately uniquedevices devices Number of unique access point during a time_segment as identified by their hardware address countscansmostuniquedevice scans Number of scans of the most scanned access point during a time_segment across the whole monitoring period Assumptions/Observations A connected WiFI access point is one that a phone was connected to. By default AWARE stores this data in the sensor_wifi table.","title":"Phone WiFI Connected"},{"location":"features/phone-wifi-connected/#phone-wifi-connected","text":"Sensor parameters description for [PHONE_WIFI_CONNECTED] : Key Description [TABLE] Database table where the wifi (connected) data is stored","title":"Phone WiFi Connected"},{"location":"features/phone-wifi-connected/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_wifi_connected_raw.csv - data/raw/ { pid } /phone_wifi_connected_with_datetime.csv - data/interim/ { pid } /phone_wifi_connected_features/phone_wifi_connected_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_wifi_connected.csv Parameters description for [PHONE_WIFI_CONNECTED][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_WIFI_CONNECTED features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_WIFI_CONNECTED][PROVIDERS][RAPIDS] : Feature Units Description countscans devices Number of scanned WiFi access points connected during a time_segment, an access point can be detected multiple times over time and these appearances are counted separately uniquedevices devices Number of unique access point during a time_segment as identified by their hardware address countscansmostuniquedevice scans Number of scans of the most scanned access point during a time_segment across the whole monitoring period Assumptions/Observations A connected WiFI access point is one that a phone was connected to. By default AWARE stores this data in the sensor_wifi table.","title":"RAPIDS provider"},{"location":"features/phone-wifi-visible/","text":"Phone WiFi Visible \u00b6 Sensor parameters description for [PHONE_WIFI_VISIBLE] : Key Description [TABLE] Database table where the wifi (visible) data is stored RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_wifi_visible_raw.csv - data/raw/ { pid } /phone_wifi_visible_with_datetime.csv - data/interim/ { pid } /phone_wifi_visible_features/phone_wifi_visible_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_wifi_visible.csv Parameters description for [PHONE_WIFI_VISIBLE][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_WIFI_VISIBLE features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_WIFI_VISIBLE][PROVIDERS][RAPIDS] : Feature Units Description countscans devices Number of scanned WiFi access points visible during a time_segment, an access point can be detected multiple times over time and these appearances are counted separately uniquedevices devices Number of unique access point during a time_segment as identified by their hardware address countscansmostuniquedevice scans Number of scans of the most scanned access point during a time_segment across the whole monitoring period Assumptions/Observations A visible WiFI access point is one that a phone sensed around itself but that it was not connected to. Due to API restrictions, this sensor is not available on iOS. By default AWARE stores this data in the wifi table.","title":"Phone WiFI Visible"},{"location":"features/phone-wifi-visible/#phone-wifi-visible","text":"Sensor parameters description for [PHONE_WIFI_VISIBLE] : Key Description [TABLE] Database table where the wifi (visible) data is stored","title":"Phone WiFi Visible"},{"location":"features/phone-wifi-visible/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_wifi_visible_raw.csv - data/raw/ { pid } /phone_wifi_visible_with_datetime.csv - data/interim/ { pid } /phone_wifi_visible_features/phone_wifi_visible_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_wifi_visible.csv Parameters description for [PHONE_WIFI_VISIBLE][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_WIFI_VISIBLE features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_WIFI_VISIBLE][PROVIDERS][RAPIDS] : Feature Units Description countscans devices Number of scanned WiFi access points visible during a time_segment, an access point can be detected multiple times over time and these appearances are counted separately uniquedevices devices Number of unique access point during a time_segment as identified by their hardware address countscansmostuniquedevice scans Number of scans of the most scanned access point during a time_segment across the whole monitoring period Assumptions/Observations A visible WiFI access point is one that a phone sensed around itself but that it was not connected to. Due to API restrictions, this sensor is not available on iOS. By default AWARE stores this data in the wifi table.","title":"RAPIDS provider"},{"location":"setup/configuration/","text":"Configuration \u00b6 You need to follow these steps to configure your RAPIDS deployment before you can extract behavioral features Add your database credentials Choose the timezone of your study Create your participants files Select what time segments you want to extract features on Modify your device data source configuration Select what sensors and features you want to process When you are done with this configuration, go to executing RAPIDS . Hint Every time you see config[\"KEY\"] or [KEY] in these docs we are referring to the corresponding key in the config.yaml file. Database credentials \u00b6 Create an empty file called .env in your RAPIDS root directory Add the following lines and replace your database-specific credentials (user, password, host, and database): [ MY_GROUP ] user=MY_USER password=MY_PASSWORD host=MY_HOST port=3306 database=MY_DATABASE Warning The label MY_GROUP is arbitrary but it has to match the following config.yaml key: DATABASE_GROUP : &database_group MY_GROUP Note You can ignore this step if you are only processing Fitbit data in CSV files. Timezone of your study \u00b6 Single timezone \u00b6 If your study only happened in a single time zone, select the appropriate code form this list and change the following config key. Double check your timezone code pick, for example US Eastern Time is America/New_York not EST TIMEZONE : &timezone America/New_York Multiple timezones \u00b6 Support coming soon. Participant files \u00b6 Participant files link together multiple devices (smartphones and wearables) to specific participants and identify them throughout RAPIDS. You can create these files manually or automatically . Participant files are stored in data/external/participant_files/pxx.yaml and follow a unified structure . Note The list PIDS in config.yaml needs to have the participant file names of the people you want to process. For example, if you created p01.yaml , p02.yaml and p03.yaml files in /data/external/participant_files/ , then PIDS should be: PIDS : [ p01 , p02 , p03 ] Tip Attribute values of the [PHONE] and [FITBIT] sections in every participant file are optional which allows you to analyze data from participants that only carried smartphones, only Fitbit devices, or both. Optional: Migrating participants files with the old format If you were using the pre-release version of RAPIDS with participant files in plain text (as opposed to yaml), you can run the following command and your old files will be converted into yaml files stored in data/external/participant_files/ python tools/update_format_participant_files.py Structure of participants files \u00b6 Example of the structure of a participant file In this example, the participant used an android phone, an ios phone, and a fitbit device throughout the study between Apr 23 rd 2020 and Oct 28 th 2020 PHONE : DEVICE_IDS : [ a748ee1a-1d0b-4ae9-9074-279a2b6ba524 , dsadas-2324-fgsf-sdwr-gdfgs4rfsdf43 ] PLATFORMS : [ android , ios ] LABEL : test01 START_DATE : 2020-04-23 END_DATE : 2020-10-28 FITBIT : DEVICE_IDS : [ fitbit1 ] LABEL : test01 START_DATE : 2020-04-23 END_DATE : 2020-10-28 For [PHONE] Key Description [DEVICE_IDS] An array of the strings that uniquely identify each smartphone, you can have more than one for when participants changed phones in the middle of the study, in this case, data from all their devices will be joined and relabeled with the last 1 on this list. [PLATFORMS] An array that specifies the OS of each smartphone in [DEVICE_IDS] , use a combination of android or ios (we support participants that changed platforms in the middle of your study!). If you have an aware_device table in your database you can set [PLATFORMS]: [multiple] and RAPIDS will infer them automatically. [LABEL] A string that is used in reports and visualizations. [START_DATE] A string with format YYY-MM-DD . Only data collected after this date will be included in the analysis [END_DATE] A string with format YYY-MM-DD . Only data collected before this date will be included in the analysis For [FITBIT] Key Description [DEVICE_IDS] An array of the strings that uniquely identify each Fitbit, you can have more than one in case the participant changed devices in the middle of the study, in this case, data from all devices will be joined and relabeled with the last device_id on this list. [LABEL] A string that is used in reports and visualizations. [START_DATE] A string with format YYY-MM-DD . Only data collected after this date will be included in the analysis [END_DATE] A string with format YYY-MM-DD . Only data collected before this date will be included in the analysis Automatic creation of participant files \u00b6 You have two options a) use the aware_device table in your database or b) use a CSV file. In either case, in your config.yaml , set [PHONE_SECTION][ADD] or [FITBIT_SECTION][ADD] to TRUE depending on what devices you used in your study. Set [DEVICE_ID_COLUMN] to the name of the column that uniquely identifies each device and include any device ids you want to ignore in [IGNORED_DEVICE_IDS] . aware_device table Set the following keys in your config.yaml CREATE_PARTICIPANT_FILES : SOURCE : TYPE : AWARE_DEVICE_TABLE DATABASE_GROUP : *database_group CSV_FILE_PATH : \"\" TIMEZONE : *timezone PHONE_SECTION : ADD : TRUE # or FALSE DEVICE_ID_COLUMN : device_id # column name IGNORED_DEVICE_IDS : [] FITBIT_SECTION : ADD : TRUE # or FALSE DEVICE_ID_COLUMN : fitbit_id # column name IGNORED_DEVICE_IDS : [] Then run snakemake -j1 create_participants_files CSV file Set the following keys in your config.yaml . CREATE_PARTICIPANT_FILES : SOURCE : TYPE : CSV_FILE DATABASE_GROUP : \"\" CSV_FILE_PATH : \"your_path/to_your.csv\" TIMEZONE : *timezone PHONE_SECTION : ADD : TRUE # or FALSE DEVICE_ID_COLUMN : device_id # column name IGNORED_DEVICE_IDS : [] FITBIT_SECTION : ADD : TRUE # or FALSE DEVICE_ID_COLUMN : fitbit_id # column name IGNORED_DEVICE_IDS : [] Your CSV file ( [SOURCE][CSV_FILE_PATH] ) should have the following columns but you can omit any values you don\u2019t have on each column: Column Description phone device id The name of this column has to match [PHONE_SECTION][DEVICE_ID_COLUMN] . Separate multiple ids with ; fitbit device id The name of this column has to match [FITBIT_SECTION][DEVICE_ID_COLUMN] . Separate multiple ids with ; pid Unique identifiers with the format pXXX (your participant files will be named with this string platform Use android , ios or multiple as explained above, separate values with ; label A human readable string that is used in reports and visualizations. start_date A string with format YYY-MM-DD . end_date A string with format YYY-MM-DD . Example device_id,pid,label,platform,start_date,end_date,fitbit_id a748ee1a-1d0b-4ae9-9074-279a2b6ba524;dsadas-2324-fgsf-sdwr-gdfgs4rfsdf43,p01,julio,android;ios,2020-01-01,2021-01-01,fitbit1 4c4cf7a1-0340-44bc-be0f-d5053bf7390c,p02,meng,ios,2021-01-01,2022-01-01,fitbit2 Then run snakemake -j1 create_participants_files Time Segments \u00b6 Time segments (or epochs) are the time windows on which you want to extract behavioral features. For example, you might want to process data on every day, every morning, or only during weekends. RAPIDS offers three categories of time segments that are flexible enough to cover most use cases: frequency (short time windows every day), periodic (arbitrary time windows on any day), and event (arbitrary time windows around events of interest). See also our examples . Frequency Segments These segments are computed on every day and all have the same duration (for example 30 minutes). Set the following keys in your config.yaml TIME_SEGMENTS : &time_segments TYPE : FREQUENCY FILE : \"data/external/your_frequency_segments.csv\" INCLUDE_PAST_PERIODIC_SEGMENTS : FALSE The file pointed by [TIME_SEGMENTS][FILE] should have the following format and can only have 1 row. Column Description label A string that is used as a prefix in the name of your time segments length An integer representing the duration of your time segments in minutes Example label,length thirtyminutes,30 This configuration will compute 48 time segments for every day when any data from any participant was sensed. For example: start_time,length,label 00:00,30,thirtyminutes0000 00:30,30,thirtyminutes0001 01:00,30,thirtyminutes0002 01:30,30,thirtyminutes0003 ... Periodic Segments These segments can be computed every day, or on specific days of the week, month, quarter, and year. Their minimum duration is 1 minute but they can be as long as you want. Set the following keys in your config.yaml . TIME_SEGMENTS : &time_segments TYPE : PERIODIC FILE : \"data/external/your_periodic_segments.csv\" INCLUDE_PAST_PERIODIC_SEGMENTS : FALSE # or TRUE If [INCLUDE_PAST_PERIODIC_SEGMENTS] is set to TRUE , RAPIDS will consider instances of your segments back enough in the past as to include the first row of data of each participant. For example, if the first row of data from a participant happened on Saturday March 7 th 2020 and the requested segment duration is 7 days starting on every Sunday, the first segment to be considered would start on Sunday March 1 st if [INCLUDE_PAST_PERIODIC_SEGMENTS] is TRUE or on Sunday March 8 th if FALSE . The file pointed by [TIME_SEGMENTS][FILE] should have the following format and can have multiple rows. Column Description label A string that is used as a prefix in the name of your time segments. It has to be unique between rows start_time A string with format HH:MM:SS representing the starting time of this segment on any day length A string representing the length of this segment.It can have one or more of the following strings XXD XXH XXM XXS to represent days, hours, minutes and seconds. For example 7D 23H 59M 59S repeats_on One of the follow options every_day , wday , qday , mday , and yday . The last four represent a week, quarter, month and year day repeats_value An integer complementing repeats_on . If you set repeats_on to every_day set this to 0 , otherwise 1-7 represent a wday starting from Mondays, 1-31 represent a mday , 1-91 represent a qday , and 1-366 represent a yday Example label,start_time,length,repeats_on,repeats_value daily,00:00:00,23H 59M 59S,every_day,0 morning,06:00:00,5H 59M 59S,every_day,0 afternoon,12:00:00,5H 59M 59S,every_day,0 evening,18:00:00,5H 59M 59S,every_day,0 night,00:00:00,5H 59M 59S,every_day,0 This configuration will create five segments instances ( daily , morning , afternoon , evening , night ) on any given day ( every_day set to 0). The daily segment will start at midnight and will last 23:59:59 , the other four segments will start at 6am, 12pm, 6pm, and 12am respectively and last for 05:59:59 . Event segments These segments can be computed before or after an event of interest (defined as any UNIX timestamp). Their minimum duration is 1 minute but they can be as long as you want. The start of each segment can be shifted backwards or forwards from the specified timestamp. Set the following keys in your config.yaml . TIME_SEGMENTS : &time_segments TYPE : EVENT FILE : \"data/external/your_event_segments.csv\" INCLUDE_PAST_PERIODIC_SEGMENTS : FALSE # or TRUE The file pointed by [TIME_SEGMENTS][FILE] should have the following format and can have multiple rows. Column Description label A string that is used as a prefix in the name of your time segments. If labels are unique, every segment is independent; if two or more segments have the same label, their data will be grouped when computing auxiliary data for features like the most frequent contact for calls (the most frequent contact will be computed across all these segments). There cannot be two overlaping event segments with the same label (RAPIDS will throw an error) event_timestamp A UNIX timestamp that represents the moment an event of interest happened (clinical relapse, survey, readmission, etc.). The corresponding time segment will be computed around this moment using length , shift , and shift_direction length A string representing the length of this segment. It can have one or more of the following keys XXD XXH XXM XXS to represent a number of days, hours, minutes, and seconds. For example 7D 23H 59M 59S shift A string representing the time shift from event_timestamp . It can have one or more of the following keys XXD XXH XXM XXS to represent a number of days, hours, minutes and seconds. For example 7D 23H 59M 59S . Use this value to change the start of a segment with respect to its event_timestamp . For example, set this variable to 1H to create a segment that starts 1 hour from an event of interest ( shift_direction determines if it\u2019s before or after). shift_direction An integer representing whether the shift is before ( -1 ) or after ( 1 ) an event_timestamp device_id The device id (smartphone or fitbit) to whom this segment belongs to. You have to create a line in this event segment file for each event of a participant that you want to analyse. If you have participants with multiple device ids you can choose any of them Example label,event_timestamp,length,shift,shift_direction,device_id stress1,1587661220000,1H,5M,1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 stress2,1587747620000,4H,4H,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 stress3,1587906020000,3H,5M,1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 stress4,1584291600000,7H,4H,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 stress5,1588172420000,9H,5M,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 mood,1587661220000,1H,0,0,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 mood,1587747620000,1D,0,0,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 mood,1587906020000,7D,0,0,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 This example will create eight segments for a single participant ( a748ee1a... ), five independent stressX segments with various lengths (1,4,3,7, and 9 hours). Segments stress1 , stress3 , and stress5 are shifted forwards by 5 minutes and stress2 and stress4 are shifted backwards by 4 hours (that is, if the stress4 event happened on March 15 th at 1pm EST ( 1584291600000 ), the time segment will start on that day at 9am and end at 4pm). The three mood segments are 1 hour, 1 day and 7 days long and have no shift. In addition, these mood segments are grouped together, meaning that although RAPIDS will compute features on each one of them, some necessary information to compute a few of such features will be extracted from all three segments, for example the phone contact that called a participant the most or the location clusters visited by a participant. Segment Examples \u00b6 5-minutes Use the following Frequency segment file to create 288 (12 * 60 * 24) 5-minute segments starting from midnight of every day in your study label,length fiveminutes,5 Daily Use the following Periodic segment file to create daily segments starting from midnight of every day in your study label,start_time,length,repeats_on,repeats_value daily,00:00:00,23H 59M 59S,every_day,0 Morning Use the following Periodic segment file to create morning segments starting at 06:00:00 and ending at 11:59:59 of every day in your study label,start_time,length,repeats_on,repeats_value morning,06:00:00,5H 59M 59S,every_day,0 Overnight Use the following Periodic segment file to create overnight segments starting at 20:00:00 and ending at 07:59:59 (next day) of every day in your study label,start_time,length,repeats_on,repeats_value morning,20:00:00,11H 59M 59S,every_day,0 Weekly Use the following Periodic segment file to create non-overlapping weekly segments starting at midnight of every Monday in your study label,start_time,length,repeats_on,repeats_value weekly,00:00:00,6D 23H 59M 59S,wday,1 Use the following Periodic segment file to create overlapping weekly segments starting at midnight of every day in your study label,start_time,length,repeats_on,repeats_value weekly,00:00:00,6D 23H 59M 59S,every_day,0 Week-ends Use the following Periodic segment file to create week-end segments starting at midnight of every Saturday in your study label,start_time,length,repeats_on,repeats_value weekend,00:00:00,1D 23H 59M 59S,wday,6 Around surveys Use the following Event segment file to create two 2-hour segments that start 1 hour before surveys answered by 3 participants label,event_timestamp,length,shift,shift_direction,device_id survey1,1587661220000,2H,1H,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 survey2,1587747620000,2H,1H,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 survey1,1587906020000,2H,1H,-1,rqtertsd-43ff-34fr-3eeg-efe4fergregr survey2,1584291600000,2H,1H,-1,rqtertsd-43ff-34fr-3eeg-efe4fergregr survey1,1588172420000,2H,1H,-1,klj34oi2-8frk-2343-21kk-324ljklewlr3 survey2,1584291600000,2H,1H,-1,klj34oi2-8frk-2343-21kk-324ljklewlr3 Device Data Source Configuration \u00b6 You might need to modify the following config keys in your config.yaml depending on what devices your participants used and where you are storing your data. You can ignore [PHONE_DATA_CONFIGURATION] or [FITBIT_DATA_CONFIGURATION] if you are not working with either devices. Phone The relevant config.yaml section looks like this by default: PHONE_DATA_CONFIGURATION : SOURCE : TYPE : DATABASE DATABASE_GROUP : *database_group DEVICE_ID_COLUMN : device_id # column name TIMEZONE : TYPE : SINGLE # SINGLE (MULTIPLE support coming soon) VALUE : *timezone Parameters for [PHONE_DATA_CONFIGURATION] Key Description [SOURCE] [TYPE] Only DATABASE is supported (phone data will be pulled from a database) [SOURCE] [DATABASE_GROUP] *database_group points to the value defined before in Database credentials [SOURCE] [DEVICE_ID_COLUMN] A column that contains strings that uniquely identify smartphones. For data collected with AWARE this is usually device_id [TIMEZONE] [TYPE] Only SINGLE is supported for now [TIMEZONE] [VALUE] *timezone points to the value defined before in Timezone of your study Fitbit The relevant config.yaml section looks like this by default: FITBIT_DATA_CONFIGURATION : SOURCE : TYPE : DATABASE # DATABASE or FILES (set each [FITBIT_SENSOR][TABLE] attribute with a table name or a file path accordingly) COLUMN_FORMAT : JSON # JSON or PLAIN_TEXT DATABASE_GROUP : *database_group DEVICE_ID_COLUMN : device_id # column name TIMEZONE : TYPE : SINGLE # Fitbit devices don't support time zones so we read this data in the timezone indicated by VALUE VALUE : *timezone Parameters for For [FITBIT_DATA_CONFIGURATION] Key Description [SOURCE] [TYPE] DATABASE or FILES (set each [FITBIT_SENSOR] [TABLE] attribute accordingly with a table name or a file path) [SOURCE] [COLUMN_FORMAT] JSON or PLAIN_TEXT . Column format of the source data. If you pulled your data directly from the Fitbit API the column containing the sensor data will be in JSON format [SOURCE] [DATABASE_GROUP] *database_group points to the value defined before in Database credentials . Only used if [TYPE] is DATABASE . [SOURCE] [DEVICE_ID_COLUMN] A column that contains strings that uniquely identify Fitbit devices. [TIMEZONE] [TYPE] Only SINGLE is supported (Fitbit devices always store data in local time). [TIMEZONE] [VALUE] *timezone points to the value defined before in Timezone of your study Sensor and Features to Process \u00b6 Finally, you need to modify the config.yaml section of the sensors you want to extract behavioral features from. All sensors follow the same naming nomenclature ( DEVICE_SENSOR ) and parameter structure which we explain in the Behavioral Features Introduction . Done Head over to Execution to learn how to execute RAPIDS.","title":"Configuration"},{"location":"setup/configuration/#configuration","text":"You need to follow these steps to configure your RAPIDS deployment before you can extract behavioral features Add your database credentials Choose the timezone of your study Create your participants files Select what time segments you want to extract features on Modify your device data source configuration Select what sensors and features you want to process When you are done with this configuration, go to executing RAPIDS . Hint Every time you see config[\"KEY\"] or [KEY] in these docs we are referring to the corresponding key in the config.yaml file.","title":"Configuration"},{"location":"setup/configuration/#database-credentials","text":"Create an empty file called .env in your RAPIDS root directory Add the following lines and replace your database-specific credentials (user, password, host, and database): [ MY_GROUP ] user=MY_USER password=MY_PASSWORD host=MY_HOST port=3306 database=MY_DATABASE Warning The label MY_GROUP is arbitrary but it has to match the following config.yaml key: DATABASE_GROUP : &database_group MY_GROUP Note You can ignore this step if you are only processing Fitbit data in CSV files.","title":"Database credentials"},{"location":"setup/configuration/#timezone-of-your-study","text":"","title":"Timezone of your study"},{"location":"setup/configuration/#single-timezone","text":"If your study only happened in a single time zone, select the appropriate code form this list and change the following config key. Double check your timezone code pick, for example US Eastern Time is America/New_York not EST TIMEZONE : &timezone America/New_York","title":"Single timezone"},{"location":"setup/configuration/#multiple-timezones","text":"Support coming soon.","title":"Multiple timezones"},{"location":"setup/configuration/#participant-files","text":"Participant files link together multiple devices (smartphones and wearables) to specific participants and identify them throughout RAPIDS. You can create these files manually or automatically . Participant files are stored in data/external/participant_files/pxx.yaml and follow a unified structure . Note The list PIDS in config.yaml needs to have the participant file names of the people you want to process. For example, if you created p01.yaml , p02.yaml and p03.yaml files in /data/external/participant_files/ , then PIDS should be: PIDS : [ p01 , p02 , p03 ] Tip Attribute values of the [PHONE] and [FITBIT] sections in every participant file are optional which allows you to analyze data from participants that only carried smartphones, only Fitbit devices, or both. Optional: Migrating participants files with the old format If you were using the pre-release version of RAPIDS with participant files in plain text (as opposed to yaml), you can run the following command and your old files will be converted into yaml files stored in data/external/participant_files/ python tools/update_format_participant_files.py","title":"Participant files"},{"location":"setup/configuration/#structure-of-participants-files","text":"Example of the structure of a participant file In this example, the participant used an android phone, an ios phone, and a fitbit device throughout the study between Apr 23 rd 2020 and Oct 28 th 2020 PHONE : DEVICE_IDS : [ a748ee1a-1d0b-4ae9-9074-279a2b6ba524 , dsadas-2324-fgsf-sdwr-gdfgs4rfsdf43 ] PLATFORMS : [ android , ios ] LABEL : test01 START_DATE : 2020-04-23 END_DATE : 2020-10-28 FITBIT : DEVICE_IDS : [ fitbit1 ] LABEL : test01 START_DATE : 2020-04-23 END_DATE : 2020-10-28 For [PHONE] Key Description [DEVICE_IDS] An array of the strings that uniquely identify each smartphone, you can have more than one for when participants changed phones in the middle of the study, in this case, data from all their devices will be joined and relabeled with the last 1 on this list. [PLATFORMS] An array that specifies the OS of each smartphone in [DEVICE_IDS] , use a combination of android or ios (we support participants that changed platforms in the middle of your study!). If you have an aware_device table in your database you can set [PLATFORMS]: [multiple] and RAPIDS will infer them automatically. [LABEL] A string that is used in reports and visualizations. [START_DATE] A string with format YYY-MM-DD . Only data collected after this date will be included in the analysis [END_DATE] A string with format YYY-MM-DD . Only data collected before this date will be included in the analysis For [FITBIT] Key Description [DEVICE_IDS] An array of the strings that uniquely identify each Fitbit, you can have more than one in case the participant changed devices in the middle of the study, in this case, data from all devices will be joined and relabeled with the last device_id on this list. [LABEL] A string that is used in reports and visualizations. [START_DATE] A string with format YYY-MM-DD . Only data collected after this date will be included in the analysis [END_DATE] A string with format YYY-MM-DD . Only data collected before this date will be included in the analysis","title":"Structure of participants files"},{"location":"setup/configuration/#automatic-creation-of-participant-files","text":"You have two options a) use the aware_device table in your database or b) use a CSV file. In either case, in your config.yaml , set [PHONE_SECTION][ADD] or [FITBIT_SECTION][ADD] to TRUE depending on what devices you used in your study. Set [DEVICE_ID_COLUMN] to the name of the column that uniquely identifies each device and include any device ids you want to ignore in [IGNORED_DEVICE_IDS] . aware_device table Set the following keys in your config.yaml CREATE_PARTICIPANT_FILES : SOURCE : TYPE : AWARE_DEVICE_TABLE DATABASE_GROUP : *database_group CSV_FILE_PATH : \"\" TIMEZONE : *timezone PHONE_SECTION : ADD : TRUE # or FALSE DEVICE_ID_COLUMN : device_id # column name IGNORED_DEVICE_IDS : [] FITBIT_SECTION : ADD : TRUE # or FALSE DEVICE_ID_COLUMN : fitbit_id # column name IGNORED_DEVICE_IDS : [] Then run snakemake -j1 create_participants_files CSV file Set the following keys in your config.yaml . CREATE_PARTICIPANT_FILES : SOURCE : TYPE : CSV_FILE DATABASE_GROUP : \"\" CSV_FILE_PATH : \"your_path/to_your.csv\" TIMEZONE : *timezone PHONE_SECTION : ADD : TRUE # or FALSE DEVICE_ID_COLUMN : device_id # column name IGNORED_DEVICE_IDS : [] FITBIT_SECTION : ADD : TRUE # or FALSE DEVICE_ID_COLUMN : fitbit_id # column name IGNORED_DEVICE_IDS : [] Your CSV file ( [SOURCE][CSV_FILE_PATH] ) should have the following columns but you can omit any values you don\u2019t have on each column: Column Description phone device id The name of this column has to match [PHONE_SECTION][DEVICE_ID_COLUMN] . Separate multiple ids with ; fitbit device id The name of this column has to match [FITBIT_SECTION][DEVICE_ID_COLUMN] . Separate multiple ids with ; pid Unique identifiers with the format pXXX (your participant files will be named with this string platform Use android , ios or multiple as explained above, separate values with ; label A human readable string that is used in reports and visualizations. start_date A string with format YYY-MM-DD . end_date A string with format YYY-MM-DD . Example device_id,pid,label,platform,start_date,end_date,fitbit_id a748ee1a-1d0b-4ae9-9074-279a2b6ba524;dsadas-2324-fgsf-sdwr-gdfgs4rfsdf43,p01,julio,android;ios,2020-01-01,2021-01-01,fitbit1 4c4cf7a1-0340-44bc-be0f-d5053bf7390c,p02,meng,ios,2021-01-01,2022-01-01,fitbit2 Then run snakemake -j1 create_participants_files","title":"Automatic creation of participant files"},{"location":"setup/configuration/#time-segments","text":"Time segments (or epochs) are the time windows on which you want to extract behavioral features. For example, you might want to process data on every day, every morning, or only during weekends. RAPIDS offers three categories of time segments that are flexible enough to cover most use cases: frequency (short time windows every day), periodic (arbitrary time windows on any day), and event (arbitrary time windows around events of interest). See also our examples . Frequency Segments These segments are computed on every day and all have the same duration (for example 30 minutes). Set the following keys in your config.yaml TIME_SEGMENTS : &time_segments TYPE : FREQUENCY FILE : \"data/external/your_frequency_segments.csv\" INCLUDE_PAST_PERIODIC_SEGMENTS : FALSE The file pointed by [TIME_SEGMENTS][FILE] should have the following format and can only have 1 row. Column Description label A string that is used as a prefix in the name of your time segments length An integer representing the duration of your time segments in minutes Example label,length thirtyminutes,30 This configuration will compute 48 time segments for every day when any data from any participant was sensed. For example: start_time,length,label 00:00,30,thirtyminutes0000 00:30,30,thirtyminutes0001 01:00,30,thirtyminutes0002 01:30,30,thirtyminutes0003 ... Periodic Segments These segments can be computed every day, or on specific days of the week, month, quarter, and year. Their minimum duration is 1 minute but they can be as long as you want. Set the following keys in your config.yaml . TIME_SEGMENTS : &time_segments TYPE : PERIODIC FILE : \"data/external/your_periodic_segments.csv\" INCLUDE_PAST_PERIODIC_SEGMENTS : FALSE # or TRUE If [INCLUDE_PAST_PERIODIC_SEGMENTS] is set to TRUE , RAPIDS will consider instances of your segments back enough in the past as to include the first row of data of each participant. For example, if the first row of data from a participant happened on Saturday March 7 th 2020 and the requested segment duration is 7 days starting on every Sunday, the first segment to be considered would start on Sunday March 1 st if [INCLUDE_PAST_PERIODIC_SEGMENTS] is TRUE or on Sunday March 8 th if FALSE . The file pointed by [TIME_SEGMENTS][FILE] should have the following format and can have multiple rows. Column Description label A string that is used as a prefix in the name of your time segments. It has to be unique between rows start_time A string with format HH:MM:SS representing the starting time of this segment on any day length A string representing the length of this segment.It can have one or more of the following strings XXD XXH XXM XXS to represent days, hours, minutes and seconds. For example 7D 23H 59M 59S repeats_on One of the follow options every_day , wday , qday , mday , and yday . The last four represent a week, quarter, month and year day repeats_value An integer complementing repeats_on . If you set repeats_on to every_day set this to 0 , otherwise 1-7 represent a wday starting from Mondays, 1-31 represent a mday , 1-91 represent a qday , and 1-366 represent a yday Example label,start_time,length,repeats_on,repeats_value daily,00:00:00,23H 59M 59S,every_day,0 morning,06:00:00,5H 59M 59S,every_day,0 afternoon,12:00:00,5H 59M 59S,every_day,0 evening,18:00:00,5H 59M 59S,every_day,0 night,00:00:00,5H 59M 59S,every_day,0 This configuration will create five segments instances ( daily , morning , afternoon , evening , night ) on any given day ( every_day set to 0). The daily segment will start at midnight and will last 23:59:59 , the other four segments will start at 6am, 12pm, 6pm, and 12am respectively and last for 05:59:59 . Event segments These segments can be computed before or after an event of interest (defined as any UNIX timestamp). Their minimum duration is 1 minute but they can be as long as you want. The start of each segment can be shifted backwards or forwards from the specified timestamp. Set the following keys in your config.yaml . TIME_SEGMENTS : &time_segments TYPE : EVENT FILE : \"data/external/your_event_segments.csv\" INCLUDE_PAST_PERIODIC_SEGMENTS : FALSE # or TRUE The file pointed by [TIME_SEGMENTS][FILE] should have the following format and can have multiple rows. Column Description label A string that is used as a prefix in the name of your time segments. If labels are unique, every segment is independent; if two or more segments have the same label, their data will be grouped when computing auxiliary data for features like the most frequent contact for calls (the most frequent contact will be computed across all these segments). There cannot be two overlaping event segments with the same label (RAPIDS will throw an error) event_timestamp A UNIX timestamp that represents the moment an event of interest happened (clinical relapse, survey, readmission, etc.). The corresponding time segment will be computed around this moment using length , shift , and shift_direction length A string representing the length of this segment. It can have one or more of the following keys XXD XXH XXM XXS to represent a number of days, hours, minutes, and seconds. For example 7D 23H 59M 59S shift A string representing the time shift from event_timestamp . It can have one or more of the following keys XXD XXH XXM XXS to represent a number of days, hours, minutes and seconds. For example 7D 23H 59M 59S . Use this value to change the start of a segment with respect to its event_timestamp . For example, set this variable to 1H to create a segment that starts 1 hour from an event of interest ( shift_direction determines if it\u2019s before or after). shift_direction An integer representing whether the shift is before ( -1 ) or after ( 1 ) an event_timestamp device_id The device id (smartphone or fitbit) to whom this segment belongs to. You have to create a line in this event segment file for each event of a participant that you want to analyse. If you have participants with multiple device ids you can choose any of them Example label,event_timestamp,length,shift,shift_direction,device_id stress1,1587661220000,1H,5M,1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 stress2,1587747620000,4H,4H,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 stress3,1587906020000,3H,5M,1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 stress4,1584291600000,7H,4H,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 stress5,1588172420000,9H,5M,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 mood,1587661220000,1H,0,0,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 mood,1587747620000,1D,0,0,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 mood,1587906020000,7D,0,0,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 This example will create eight segments for a single participant ( a748ee1a... ), five independent stressX segments with various lengths (1,4,3,7, and 9 hours). Segments stress1 , stress3 , and stress5 are shifted forwards by 5 minutes and stress2 and stress4 are shifted backwards by 4 hours (that is, if the stress4 event happened on March 15 th at 1pm EST ( 1584291600000 ), the time segment will start on that day at 9am and end at 4pm). The three mood segments are 1 hour, 1 day and 7 days long and have no shift. In addition, these mood segments are grouped together, meaning that although RAPIDS will compute features on each one of them, some necessary information to compute a few of such features will be extracted from all three segments, for example the phone contact that called a participant the most or the location clusters visited by a participant.","title":"Time Segments"},{"location":"setup/configuration/#segment-examples","text":"5-minutes Use the following Frequency segment file to create 288 (12 * 60 * 24) 5-minute segments starting from midnight of every day in your study label,length fiveminutes,5 Daily Use the following Periodic segment file to create daily segments starting from midnight of every day in your study label,start_time,length,repeats_on,repeats_value daily,00:00:00,23H 59M 59S,every_day,0 Morning Use the following Periodic segment file to create morning segments starting at 06:00:00 and ending at 11:59:59 of every day in your study label,start_time,length,repeats_on,repeats_value morning,06:00:00,5H 59M 59S,every_day,0 Overnight Use the following Periodic segment file to create overnight segments starting at 20:00:00 and ending at 07:59:59 (next day) of every day in your study label,start_time,length,repeats_on,repeats_value morning,20:00:00,11H 59M 59S,every_day,0 Weekly Use the following Periodic segment file to create non-overlapping weekly segments starting at midnight of every Monday in your study label,start_time,length,repeats_on,repeats_value weekly,00:00:00,6D 23H 59M 59S,wday,1 Use the following Periodic segment file to create overlapping weekly segments starting at midnight of every day in your study label,start_time,length,repeats_on,repeats_value weekly,00:00:00,6D 23H 59M 59S,every_day,0 Week-ends Use the following Periodic segment file to create week-end segments starting at midnight of every Saturday in your study label,start_time,length,repeats_on,repeats_value weekend,00:00:00,1D 23H 59M 59S,wday,6 Around surveys Use the following Event segment file to create two 2-hour segments that start 1 hour before surveys answered by 3 participants label,event_timestamp,length,shift,shift_direction,device_id survey1,1587661220000,2H,1H,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 survey2,1587747620000,2H,1H,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 survey1,1587906020000,2H,1H,-1,rqtertsd-43ff-34fr-3eeg-efe4fergregr survey2,1584291600000,2H,1H,-1,rqtertsd-43ff-34fr-3eeg-efe4fergregr survey1,1588172420000,2H,1H,-1,klj34oi2-8frk-2343-21kk-324ljklewlr3 survey2,1584291600000,2H,1H,-1,klj34oi2-8frk-2343-21kk-324ljklewlr3","title":"Segment Examples"},{"location":"setup/configuration/#device-data-source-configuration","text":"You might need to modify the following config keys in your config.yaml depending on what devices your participants used and where you are storing your data. You can ignore [PHONE_DATA_CONFIGURATION] or [FITBIT_DATA_CONFIGURATION] if you are not working with either devices. Phone The relevant config.yaml section looks like this by default: PHONE_DATA_CONFIGURATION : SOURCE : TYPE : DATABASE DATABASE_GROUP : *database_group DEVICE_ID_COLUMN : device_id # column name TIMEZONE : TYPE : SINGLE # SINGLE (MULTIPLE support coming soon) VALUE : *timezone Parameters for [PHONE_DATA_CONFIGURATION] Key Description [SOURCE] [TYPE] Only DATABASE is supported (phone data will be pulled from a database) [SOURCE] [DATABASE_GROUP] *database_group points to the value defined before in Database credentials [SOURCE] [DEVICE_ID_COLUMN] A column that contains strings that uniquely identify smartphones. For data collected with AWARE this is usually device_id [TIMEZONE] [TYPE] Only SINGLE is supported for now [TIMEZONE] [VALUE] *timezone points to the value defined before in Timezone of your study Fitbit The relevant config.yaml section looks like this by default: FITBIT_DATA_CONFIGURATION : SOURCE : TYPE : DATABASE # DATABASE or FILES (set each [FITBIT_SENSOR][TABLE] attribute with a table name or a file path accordingly) COLUMN_FORMAT : JSON # JSON or PLAIN_TEXT DATABASE_GROUP : *database_group DEVICE_ID_COLUMN : device_id # column name TIMEZONE : TYPE : SINGLE # Fitbit devices don't support time zones so we read this data in the timezone indicated by VALUE VALUE : *timezone Parameters for For [FITBIT_DATA_CONFIGURATION] Key Description [SOURCE] [TYPE] DATABASE or FILES (set each [FITBIT_SENSOR] [TABLE] attribute accordingly with a table name or a file path) [SOURCE] [COLUMN_FORMAT] JSON or PLAIN_TEXT . Column format of the source data. If you pulled your data directly from the Fitbit API the column containing the sensor data will be in JSON format [SOURCE] [DATABASE_GROUP] *database_group points to the value defined before in Database credentials . Only used if [TYPE] is DATABASE . [SOURCE] [DEVICE_ID_COLUMN] A column that contains strings that uniquely identify Fitbit devices. [TIMEZONE] [TYPE] Only SINGLE is supported (Fitbit devices always store data in local time). [TIMEZONE] [VALUE] *timezone points to the value defined before in Timezone of your study","title":"Device Data Source Configuration"},{"location":"setup/configuration/#sensor-and-features-to-process","text":"Finally, you need to modify the config.yaml section of the sensors you want to extract behavioral features from. All sensors follow the same naming nomenclature ( DEVICE_SENSOR ) and parameter structure which we explain in the Behavioral Features Introduction . Done Head over to Execution to learn how to execute RAPIDS.","title":"Sensor and Features to Process"},{"location":"setup/execution/","text":"Execution \u00b6 After you have installed and configured RAPIDS, use the following command to execute it. ./rapids -j1 Ready to extract behavioral features If you are ready to extract features head over to the Behavioral Features Introduction Info The script ./rapids is a wrapper around Snakemake so you can pass any parameters that Snakemake accepts (e.g. -j1 ). Updating RAPIDS output after modifying config.yaml Any changes to the config.yaml file will be applied automatically and only the relevant files will be updated. This means that after modifying the features list for PHONE_MESSAGE for example, RAPIDS will update the output file with the correct features. Multi-core You can run RAPIDS over multiple cores by modifying the -j argument (e.g. use -j8 to use 8 cores). However , take into account that this means multiple sensor datasets for different participants will be load in memory at the same time. If RAPIDS crashes because it ran out of memory reduce the number of cores and try again. As reference, we have run RAPIDS over 12 cores and 32 Gb of RAM without problems for a study with 200 participants with 14 days of low-frequency smartphone data (no accelerometer, gyroscope, or magnetometer). Forcing a complete rerun If you want to update your data from your database or rerun the whole pipeline from scratch run one or both of the following commands depending on the devices you are using: ./rapids -j1 -R download_phone_data ./rapids -j1 -R download_fitbit_data Deleting RAPIDS output If you want to delete all the output files RAPIDS produces you can execute the following command: ./rapids -j1 --delete-all-output","title":"Execution"},{"location":"setup/execution/#execution","text":"After you have installed and configured RAPIDS, use the following command to execute it. ./rapids -j1 Ready to extract behavioral features If you are ready to extract features head over to the Behavioral Features Introduction Info The script ./rapids is a wrapper around Snakemake so you can pass any parameters that Snakemake accepts (e.g. -j1 ). Updating RAPIDS output after modifying config.yaml Any changes to the config.yaml file will be applied automatically and only the relevant files will be updated. This means that after modifying the features list for PHONE_MESSAGE for example, RAPIDS will update the output file with the correct features. Multi-core You can run RAPIDS over multiple cores by modifying the -j argument (e.g. use -j8 to use 8 cores). However , take into account that this means multiple sensor datasets for different participants will be load in memory at the same time. If RAPIDS crashes because it ran out of memory reduce the number of cores and try again. As reference, we have run RAPIDS over 12 cores and 32 Gb of RAM without problems for a study with 200 participants with 14 days of low-frequency smartphone data (no accelerometer, gyroscope, or magnetometer). Forcing a complete rerun If you want to update your data from your database or rerun the whole pipeline from scratch run one or both of the following commands depending on the devices you are using: ./rapids -j1 -R download_phone_data ./rapids -j1 -R download_fitbit_data Deleting RAPIDS output If you want to delete all the output files RAPIDS produces you can execute the following command: ./rapids -j1 --delete-all-output","title":"Execution"},{"location":"setup/installation/","text":"Installation \u00b6 You can install RAPIDS using Docker (the fastest), or native instructions for MacOS and Ubuntu Docker Install Docker Pull our RAPIDS container docker pull agamk/rapids:latest ` Run RAPIDS' container (after this step is done you should see a prompt in the main RAPIDS folder with its python environment active) docker run -it agamk/rapids:latest Pull the latest version of RAPIDS git pull Make RAPIDS script executable chmod +x rapids Check that RAPIDS is working ./rapids -j1 Optional . You can edit RAPIDS files with vim but we recommend using Visual Studio Code and its Remote Containers extension How to configure Remote Containers extension Make sure RAPIDS container is running Install the Remote - Containers extension Go to the Remote Explorer panel on the left hand sidebar On the top right dropdown menu choose Containers Double click on the agamk/rapids container in the CONTAINERS tree A new VS Code session should open on RAPIDS main folder insidethe container. MacOS We tested these instructions in Catalina Install brew Install MySQL brew install mysql brew services start mysql Install R 4.0, pandoc and rmarkdown. If you have other instances of R, we recommend uninstalling them brew install r brew install pandoc Rscript --vanilla -e 'install.packages(\"rmarkdown\", repos=\"http://cran.us.r-project.org\")' Install miniconda (restart your terminal afterwards) brew cask install miniconda conda init zsh # (or conda init bash) Clone our repo git clone https://github.com/carissalow/rapids Create a python virtual environment cd rapids conda env create -f environment.yml -n rapids conda activate rapids Install R packages and virtual environment: snakemake -j1 renv_install snakemake -j1 renv_restore Note This step could take several minutes to complete, especially if you have less than 3Gb of RAM or packages need to be compiled from source. Please be patient and let it run until completion. Make RAPIDS script executable chmod +x rapids Check that RAPIDS is working ./rapids -j1 Ubuntu We tested on Ubuntu 18.04 & 20.04 Install dependencies sudo apt install libcurl4-openssl-dev sudo apt install libssl-dev sudo apt install libxml2-dev Install MySQL sudo apt install libmysqlclient-dev sudo apt install mysql-server Add key for R\u2019s repository. sudo apt-key adv --keyserver keyserver.ubuntu.com --recv-keys E298A3A825C0D65DFD57CBB651716619E084DAB9 Add R\u2019s repository For 18.04 sudo add-apt-repository 'deb https://cloud.r-project.org/bin/linux/ubuntu bionic-cran40/' For 20.04 sudo add-apt-repository 'deb https://cloud.r-project.org/bin/linux/ubuntu focal-cran40/' Install R 4.0. If you have other instances of R, we recommend uninstalling them sudo apt update sudo apt install r-base Install Pandoc and rmarkdown sudo apt install pandoc Rscript --vanilla -e 'install.packages(\"rmarkdown\", repos=\"http://cran.us.r-project.org\")' Install git sudo apt install git Install miniconda Restart your current shell Clone our repo: git clone https://github.com/carissalow/rapids Create a python virtual environment: cd rapids conda env create -f environment.yml -n MY_ENV_NAME conda activate MY_ENV_NAME Install R packages and virtual environment: snakemake -j1 renv_install snakemake -j1 renv_restore Note This step could take several minutes to complete, especially if you have less than 3Gb of RAM or packages need to be compiled from source. Please be patient and let it run until completion. Make RAPIDS script executable chmod +x rapids Check that RAPIDS is working ./rapids -j1","title":"Installation"},{"location":"setup/installation/#installation","text":"You can install RAPIDS using Docker (the fastest), or native instructions for MacOS and Ubuntu Docker Install Docker Pull our RAPIDS container docker pull agamk/rapids:latest ` Run RAPIDS' container (after this step is done you should see a prompt in the main RAPIDS folder with its python environment active) docker run -it agamk/rapids:latest Pull the latest version of RAPIDS git pull Make RAPIDS script executable chmod +x rapids Check that RAPIDS is working ./rapids -j1 Optional . You can edit RAPIDS files with vim but we recommend using Visual Studio Code and its Remote Containers extension How to configure Remote Containers extension Make sure RAPIDS container is running Install the Remote - Containers extension Go to the Remote Explorer panel on the left hand sidebar On the top right dropdown menu choose Containers Double click on the agamk/rapids container in the CONTAINERS tree A new VS Code session should open on RAPIDS main folder insidethe container. MacOS We tested these instructions in Catalina Install brew Install MySQL brew install mysql brew services start mysql Install R 4.0, pandoc and rmarkdown. If you have other instances of R, we recommend uninstalling them brew install r brew install pandoc Rscript --vanilla -e 'install.packages(\"rmarkdown\", repos=\"http://cran.us.r-project.org\")' Install miniconda (restart your terminal afterwards) brew cask install miniconda conda init zsh # (or conda init bash) Clone our repo git clone https://github.com/carissalow/rapids Create a python virtual environment cd rapids conda env create -f environment.yml -n rapids conda activate rapids Install R packages and virtual environment: snakemake -j1 renv_install snakemake -j1 renv_restore Note This step could take several minutes to complete, especially if you have less than 3Gb of RAM or packages need to be compiled from source. Please be patient and let it run until completion. Make RAPIDS script executable chmod +x rapids Check that RAPIDS is working ./rapids -j1 Ubuntu We tested on Ubuntu 18.04 & 20.04 Install dependencies sudo apt install libcurl4-openssl-dev sudo apt install libssl-dev sudo apt install libxml2-dev Install MySQL sudo apt install libmysqlclient-dev sudo apt install mysql-server Add key for R\u2019s repository. sudo apt-key adv --keyserver keyserver.ubuntu.com --recv-keys E298A3A825C0D65DFD57CBB651716619E084DAB9 Add R\u2019s repository For 18.04 sudo add-apt-repository 'deb https://cloud.r-project.org/bin/linux/ubuntu bionic-cran40/' For 20.04 sudo add-apt-repository 'deb https://cloud.r-project.org/bin/linux/ubuntu focal-cran40/' Install R 4.0. If you have other instances of R, we recommend uninstalling them sudo apt update sudo apt install r-base Install Pandoc and rmarkdown sudo apt install pandoc Rscript --vanilla -e 'install.packages(\"rmarkdown\", repos=\"http://cran.us.r-project.org\")' Install git sudo apt install git Install miniconda Restart your current shell Clone our repo: git clone https://github.com/carissalow/rapids Create a python virtual environment: cd rapids conda env create -f environment.yml -n MY_ENV_NAME conda activate MY_ENV_NAME Install R packages and virtual environment: snakemake -j1 renv_install snakemake -j1 renv_restore Note This step could take several minutes to complete, especially if you have less than 3Gb of RAM or packages need to be compiled from source. Please be patient and let it run until completion. Make RAPIDS script executable chmod +x rapids Check that RAPIDS is working ./rapids -j1","title":"Installation"},{"location":"workflow-examples/minimal/","text":"Minimal Working Example \u00b6 This is a quick guide for creating and running a simple pipeline to extract missing, outgoing, and incoming call features for daily and night epochs of one participant monitored on the US East coast. Install RAPIDS and make sure your conda environment is active (see Installation ) Make the changes listed below for the corresponding Configuration step (we provide an example of what the relevant sections in your config.yml will look like after you are done) Things to change on each configuration step 1. Setup your database connection credentials in .env . We assume your credentials group is called MY_GROUP . 2. America/New_York should be the default timezone 3. Create a participant file p01.yaml based on one of your participants and add p01 to [PIDS] in config.yaml . The following would be the content of your p01.yaml participant file: PHONE : DEVICE_IDS : [ aaaaaaaa-1111-bbbb-2222-cccccccccccc ] # your participant's AWARE device id PLATFORMS : [ android ] # or ios LABEL : MyTestP01 # any string START_DATE : 2020-01-01 # this can also be empty END_DATE : 2021-01-01 # this can also be empty 4. [TIME_SEGMENTS][TYPE] should be the default PERIODIC . Change [TIME_SEGMENTS][FILE] with the path of a file containing the following lines: label,start_time,length,repeats_on,repeats_value daily,00:00:00,23H 59M 59S,every_day,0 night,00:00:00,5H 59M 59S,every_day,0 5. If you collected data with AWARE you won\u2019t need to modify the attributes of [DEVICE_DATA][PHONE] 6. Set [PHONE_CALLS][PROVIDERS][RAPIDS][COMPUTE] to True Example of the config.yaml sections after the changes outlined above PIDS: [p01] TIMEZONE: &timezone America/New_York DATABASE_GROUP: &database_group MY_GROUP # ... other irrelevant sections TIME_SEGMENTS: &time_segments TYPE: PERIODIC FILE: \"data/external/timesegments_periodic.csv\" # make sure the three lines specified above are in the file INCLUDE_PAST_PERIODIC_SEGMENTS: FALSE # No need to change this if you collected AWARE data on a database and your credentials are grouped under `MY_GROUP` in `.env` DEVICE_DATA: PHONE: SOURCE: TYPE: DATABASE DATABASE_GROUP: *database_group DEVICE_ID_COLUMN: device_id # column name TIMEZONE: TYPE: SINGLE # SINGLE or MULTIPLE VALUE: *timezone ############## PHONE ########################################################### ################################################################################ # ... other irrelevant sections # Communication call features config, TYPES and FEATURES keys need to match PHONE_CALLS: TABLE: calls # change if your calls table has a different name PROVIDERS: RAPIDS: COMPUTE: True # set this to True! CALL_TYPES: ... Run RAPIDS ./rapids -j1 The call features for daily and morning time segments will be in /data/processed/features/p01/phone_calls.csv","title":"Minimal"},{"location":"workflow-examples/minimal/#minimal-working-example","text":"This is a quick guide for creating and running a simple pipeline to extract missing, outgoing, and incoming call features for daily and night epochs of one participant monitored on the US East coast. Install RAPIDS and make sure your conda environment is active (see Installation ) Make the changes listed below for the corresponding Configuration step (we provide an example of what the relevant sections in your config.yml will look like after you are done) Things to change on each configuration step 1. Setup your database connection credentials in .env . We assume your credentials group is called MY_GROUP . 2. America/New_York should be the default timezone 3. Create a participant file p01.yaml based on one of your participants and add p01 to [PIDS] in config.yaml . The following would be the content of your p01.yaml participant file: PHONE : DEVICE_IDS : [ aaaaaaaa-1111-bbbb-2222-cccccccccccc ] # your participant's AWARE device id PLATFORMS : [ android ] # or ios LABEL : MyTestP01 # any string START_DATE : 2020-01-01 # this can also be empty END_DATE : 2021-01-01 # this can also be empty 4. [TIME_SEGMENTS][TYPE] should be the default PERIODIC . Change [TIME_SEGMENTS][FILE] with the path of a file containing the following lines: label,start_time,length,repeats_on,repeats_value daily,00:00:00,23H 59M 59S,every_day,0 night,00:00:00,5H 59M 59S,every_day,0 5. If you collected data with AWARE you won\u2019t need to modify the attributes of [DEVICE_DATA][PHONE] 6. Set [PHONE_CALLS][PROVIDERS][RAPIDS][COMPUTE] to True Example of the config.yaml sections after the changes outlined above PIDS: [p01] TIMEZONE: &timezone America/New_York DATABASE_GROUP: &database_group MY_GROUP # ... other irrelevant sections TIME_SEGMENTS: &time_segments TYPE: PERIODIC FILE: \"data/external/timesegments_periodic.csv\" # make sure the three lines specified above are in the file INCLUDE_PAST_PERIODIC_SEGMENTS: FALSE # No need to change this if you collected AWARE data on a database and your credentials are grouped under `MY_GROUP` in `.env` DEVICE_DATA: PHONE: SOURCE: TYPE: DATABASE DATABASE_GROUP: *database_group DEVICE_ID_COLUMN: device_id # column name TIMEZONE: TYPE: SINGLE # SINGLE or MULTIPLE VALUE: *timezone ############## PHONE ########################################################### ################################################################################ # ... other irrelevant sections # Communication call features config, TYPES and FEATURES keys need to match PHONE_CALLS: TABLE: calls # change if your calls table has a different name PROVIDERS: RAPIDS: COMPUTE: True # set this to True! CALL_TYPES: ... Run RAPIDS ./rapids -j1 The call features for daily and morning time segments will be in /data/processed/features/p01/phone_calls.csv","title":"Minimal Working Example"}]}
\ No newline at end of file
diff --git a/0.1/sitemap.xml b/0.1/sitemap.xml
index 7e02bf4e..d018a09e 100644
--- a/0.1/sitemap.xml
+++ b/0.1/sitemap.xml
@@ -1,147 +1,147 @@
 <?xml version="1.0" encoding="UTF-8"?>
 <urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9"><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url>
 </urlset>
\ No newline at end of file
diff --git a/0.1/sitemap.xml.gz b/0.1/sitemap.xml.gz
index 40efebc501f25b75b836f0ba54feec55cb08c729..4c7772715a3212064978fd3da90d2c06afb06091 100644
GIT binary patch
delta 195
zcmcc0c#V-=zMF$%srbo>?6UPy;-*vb`uwUGoHnX14B(F1yZ6GCBib`;#2!4a|Cv_t
zTS~X;m8HacsTjewiDjqCkA3Cc7w7j>Zu_DK{WsZU1GRQvc%XRfrN_%nIw~_(xLdM5
zdOnZMbcy$ZXb(xtgo){{jh`M}(=PnxS3JxAen6YWFTX2m(o#dG=S-isJ!ka_N8?>_
t_tR>{=bC&w`_g|?nZ;S2h=fK~F=XOzcj%hvU%B1<8@V%%&*f)e008$VSkV9g

delta 194
zcmcb{c$JY|zMF$XT<gR{cG>z<OWH1-z39pBps`IwH^gD}c5xryJxwQ*8SDN{&vz?1
zU+Kgzyy)w?qr8a{lFMuJ?_PLx`e(%Z0(I>|{$LZ%r<=NU<$|O5rmTDIwa7Ou%v;5d
z$K;h^ywY8#Qzzt38)<8YOq;Q1n)>^m1y8^5hWc!NubMvRD(`y9^XH?tg??Fa$0%Xl
s!8<AI-+U4JqW-glMVBWcp^;S#nfTisTD$Gf?P>Yv^TR6U@-r|10C(nA`2YX_

diff --git a/latest/file-structure/index.html b/latest/file-structure/index.html
index 66b2f900..fb682026 100644
--- a/latest/file-structure/index.html
+++ b/latest/file-structure/index.html
@@ -895,13 +895,15 @@
                 <h1 id="file-structure">File Structure<a class="headerlink" href="#file-structure" title="Permanent link">&para;</a></h1>
 <div class="admonition tip">
 <p class="admonition-title">Tip</p>
-<p>Read this page if you want to learn more about how RAPIDS is structured. If you want to start using it go to <a href="../setup/installation/">Installation</a>, then to <a href="../setup/configuration/">Configuration</a>, and then to <a href="../setup/execution/">Execution</a></p>
+<ul>
+<li>Read this page if you want to learn more about how RAPIDS is structured. If you want to start using it go to <a href="../setup/installation/">Installation</a>, then to <a href="../setup/configuration/">Configuration</a>, and then to <a href="../setup/execution/">Execution</a></li>
+<li>All paths mentioned in this page are relative to RAPIDS&rsquo; root folder.</li>
+</ul>
 </div>
-<p>All paths mentioned in this page are relative to RAPIDS&rsquo; root folder.</p>
-<p>If you want to extract the behavioral features that RAPIDS offers, you will only have to create or modify the <a href="../setup/configuration/#database-credentials"><code>.env</code> file</a>, <a href="../setup/configuration/#participant-files">participants files</a>, <a href="../setup/configuration/#time-segments">time segment files</a>, and the <code>config.yaml</code> file. The <code>config.yaml</code> file is the heart of RAPIDS and includes parameters to manage participants, data sources, sensor data, visualizations and more.</p>
+<p>If you want to extract the behavioral features that RAPIDS offers, you will only have to create or modify the <a href="../setup/configuration/#database-credentials"><code>.env</code> file</a>, <a href="../setup/configuration/#participant-files">participants files</a>, <a href="../setup/configuration/#time-segments">time segment files</a>, and the <code>config.yaml</code> file as instructed in the <a href="../setup/configuration">Configuration page</a>. The <code>config.yaml</code> file is the heart of RAPIDS and includes parameters to manage participants, data sources, sensor data, visualizations and more.</p>
 <p>All data is saved in <code>data/</code>. The <code>data/external/</code> folder stores any data imported or created by the user, <code>data/raw/</code> stores sensor data as imported from your database, <code>data/interim/</code> has intermediate files necessary to compute behavioral features from raw data, and <code>data/processed/</code> has all the final files with the behavioral features in folders per participant and sensor.</p>
-<p>All the source code is saved in <code>src/</code>. The <code>src/data/</code> folder stores scripts to download, clean and pre-process sensor data, <code>src/features</code> has scripts to extract behavioral features organized in their respective subfolders , <code>src/models/</code> can host any script to create models or statistical analyses with the behavioral features you extract, and <code>src/visualization/</code> has scripts to create plots of the raw and processed data.</p>
-<p>There are other important files and folders but only relevant if you are interested in extending RAPIDS (e.g. virtual env files, docs, tests, Dockerfile, the Snakefile, etc.). In the figure below, we represent the interactions between users and files. After a user modifies <code>config.yaml</code> and <code>.env</code> the <code>Snakefile</code> file will decide what Snakemake rules have to be executed to produce the required output files (behavioral features) and what scripts are in charge of producing such files. In addition, users can add or modifiy files in the <code>data</code> folder (for example to configure the <a href="../setup/configuration/#participant-files">participants files</a> or the <a href="../setup/configuration/#time-segments">time segment files</a>).</p>
+<p>RAPIDS source code is saved in <code>src/</code>. The <code>src/data/</code> folder stores scripts to download, clean and pre-process sensor data, <code>src/features</code> has scripts to extract behavioral features organized in their respective sensor subfolders , <code>src/models/</code> can host any script to create models or statistical analyses with the behavioral features you extract, and <code>src/visualization/</code> has scripts to create plots of the raw and processed data. There are other files and folders but only relevant if you are interested in extending RAPIDS (e.g. virtual env files, docs, tests, Dockerfile, the Snakefile, etc.). </p>
+<p>In the figure below, we represent the interactions between users and files. After a user modifies the configuration files mentioned above, the <code>Snakefile</code> file will search for and execute the Snakemake rules that contain the Python or R scripts necessary to generate or update the required output files (behavioral features, plots, etc.).</p>
 <figure>
   <img src="../img/files.png" width="600" />
   <figcaption>Interaction diagram between the user, and important files in RAPIDS</figcaption>
diff --git a/latest/img/files.png b/latest/img/files.png
index 58ee5dec797ec1e0d3a9a7ffbcd2ea53c6b92d5d..1fef7eebd110c8f0a3c69025c16892469c9da3c8 100644
GIT binary patch
literal 321183
zcmeEvcR1Gl`?o|H8IcG@nHh;fwh}_2?2(zhSIFwBJBo}uQORr|dsA7NNk+<+Eqm|%
zoF9$5>*D+S-QVMQj_04Jj=Q7lzQ%i;?{mLj=lQv(C@)P!Kuv&!g+(ND;rtaWEW8CQ
zEP`&lJ@C$n6HJouKRsp=5{fbs67-7pHpXU_Mp#%E?u9(stE{9!eRI-n*dGf&<^p~J
zegi%|(;Sv;>NBzj;^*+WkFvIeyxUJpasGVTJ|o5R?>_AhN~b@V#L73en}?NkRPUYl
z2T?uzQ<)~tbsGZwi$a^Lle!bH7b7-ziR1UkG9F6f!FtkbX?Ff5|EUwAsC%o2gKnOn
z$9B>y9jIa)OG_iY`F?Iq&DD!ioP3ru5np$6W39-SO_#oJ7Z%g8@Q;tCdG=dNSFlH(
z4<o`FIr)rVJB;&5cRSwamltl~k;vPQ9PoP5-7h1>uF!T~a*5s&2P^X9fUMFk`gZyo
z#>89{(-OoIEm9vT@9o#Qowiru{Q$%LGTh8lr!tvcsPs>j>0o2s2{_5VtRSki*(;{e
z6IFdlWlH&K@U2CPD#k+{9h(jo8@bnv-^Up~roYYcaE}!IsokwbZo?Yy2oAE;kwh5p
z_iFHbaA&4?$n0B{J6Y_Lx_Khiq?*TcJT=FUc?Fax@UfVO;qnx{Hw!hEWHoqs!K3B#
z$B^2xkT_}fxpTXud&Cr|D1>SRo;_av%pn)JFEV}l^3cl}Jjo4gyPJy_wY?<ApJa$N
zviGt!852IorpR_MnW4PfdP>yD&bMO;Tes-YQ`0*IjO|+c<sR&hO{2Vq!#bA{exbb0
zvfs8giAZTopPHZ5G7QBvwfpG%K>TMYk+j`8#kha%4&qb~6}fz0A(T;#O_sv=RO@?x
z@{<Avk@J?Z5&nnmXP74@zx?e&!|lp-wBpsFtE!y#54#wgV_l>kX%mRc?ZToaF?N)4
z!;1d4yT0RHu$)7z<Ip{^eKV@j*n;)vJ4w8Y_o*Lk71+h!edii2&bhM(?(X%&BEdV$
z9CYiY9<Dkef%yIwf}Gm~X_Q}Xu6L83!PYFsx80+SrB+O2yH^;eKo2V)hoM`9>NY*z
z-plmVFYt%YRs=C!+nY{L@}7hMbu8M26<_vFSrDz}u4EtIXtH?hgj*vqQ89h>hla?$
z5PkJoi9R|+G<Qhq-L=QWqs7cqw2yH!d-To_=-|bb$aM>!q3hiDsI+nFY~CT!-MXmB
z;({qTmjk$pj0bkV49qx}t|%I>sEo~q%Osh6j^fz|#m0CkWts$XC&ETQYAKI%hM!KK
zCbznCU1IOE=bwb<$mZDQm>qF%h`l-FLF_^J)#u1ruTsU6ZB%mi*}ZSyD}7-QYCv$(
z=&|5O?vJh?4}Zk^CUVfvNbgd5ZHu_2Fdxxd+PC;`Y1;p?s=RKvn*RI@tvtEOJ%`FW
zy?J?-+Pkb-Cg>+lPcU0!E$&w+8SNRL+E~KLrz#@W^%lD+cBkl`R{$LYEBSr=l)n_Q
zqk`mG&c!oj?N1@!gJ*c2;@R;}5mb^F8B&-+$!YEl-V3~IBg6P$SJj;Z@??KE1$~h5
zk)f3Fc%g7jB?GtBC;UQaFn{pk-|bb4RW`>D_~MuQJrJ*owT%6uRC^^-QIze|!=#Wi
zvhFV%WkY0ZWcfpMF5@d3vORrsH%yxI`H`18am4K84=+9Rc=-B5;zRb3K=wt}z=V2b
zOO<mfJSu#zXcb?-;EDHD;=MetG?4f$u_-=MWjrp%)cZ6Ybzn}Us+^DlO|;=~tRu0A
zzs`BF^OMg4<6k8R#Fxi&DkUlly=tl!=e|fqE49avun({AL|^8c070#*(S?@_F0aR~
z?Df~6)}hAsKj+WvFGL;1`y|XWY&fhuj591SyI%94W{;*^_JV0dGj*h7)Bv9!f1+*E
z-x?%Y`c{PFc;f?YylraP(oT*}*p7^XLQb_K`$l;j3miW?$~h?v6Ayoy6P`PN<K>OG
zd7YDuk1dV&WQMy22s}$ori<rlPHmWMpgVPv_b#8hN&U5{mLSvjJQuk+c*joCayOg`
zGs!B6e>a5Ue)rgj<8-HfhuPY@vM-EH*KW)BRYkw@K3;jcF6-Ug--pUSezv@)k=<(L
zkR(to{M9{onTW8Z<kLCR%Vu_8$#NVs4rol|j79oC8RyXwetYA3AVsUMRfu7uUZY7(
z8gaUg$ybAQ^MFs`e+iU32oAbu+GSOxy~=hqO*E&u_O>>sX2Cz=&f&V0hKU@J^3E>L
zY>&isi@kDtQKd(G+yYKlx(3kD3egq`3JA$WP)B%&`-caII}0qP4JDmyVs4V8yziG=
zZT^V-M3i=fcIMT<yp0~?fzXNT#a6Z{!`kmU+Fq|u?3vJBWLwlBiIsj(8F-{1awSr`
z&b%&8$6LoeRWNnX(&0MwD*b9mu7B>A)yIo_>C~b^M8^wRyRf>>b)67XDKaf0C}P>z
zv+-(!XoF!_z^>}uZM(Tq?@;9^5!%DVfl76U@a>5;4>_KAOiO!6d;i+$vSa6|8~8{f
zDNoz9=S$D}5Bd$}{&h=HekiBDQa@H-uEdk5(%a{bnzxu=n!oz~Plsguv}s-GXFMJ{
ze+iBE`DFZQW+;Hl{G!1{kB>g;T`dn<I`ahV*!xn-EXw_Qwa9Jpx$&LwpA)VfwA4Kr
zH1t?K$i+EkxWtH%i?c!VVfs+N!HB^`?{k^QFIrygmeI|N=`f!#<zaDEo58E|$-TY!
zV6vQRfCRq-m+98v@rmP1$8W!yc-4Nz=dZ{To}RS#(#l6~L>zpa!#^%@!*r!^Zu)-9
z{n(4{##g@Cr6-)Z>V86w-$Fw*x9rLl`r)g+Z?3IA*l(7WCBLWY;0Mjj;S0|lRjipe
z<vSns{jKz5*3Rr>!}Dt8(0h-r)?~l``gm2iPHZx6U}zwJ;4u4~TGG`|xep35%qg<i
zTvaN~OB<IH_cq}qwHAfu@p+7AJMoRmHWg^my)ER_W%0V?x#_&Ci}wI4#Rb9}e0f$a
zt0F6IpBAsujPp-)EF}l&y1afL8=o{(&D<GVE!MK3xKX<F?P1MTt?-<&R^P8|oi}n@
zGF`mx$;#-+Z05MOhsiy8dFSP14p*n=tB*03lf=Em%K@D;K{HJeQbM^CYgdjY<tYtc
zZOxK%Yp%FTcYW5mtD`SF&MvjK^I1WZ*MW7HRJQ1Z*KIG}B<;}(R+>%Xi_Mv3X;PSX
z(sYt^(p|i_I6o2C;+4|*x^Uy>0*aH!_IAGhkY4XQ8XtF(Y(7n1RbK8eU*g-PGo@Uf
z@+->|e|Iw+me!D_@#5YvYUXZBFENf6$aQ@o+0xKX-PPtHJmez%RI5E*H`(i}^SRCQ
z12WmcTJrl92BN6NBAjnzU2PC`XP&?4RoNThW71dKDN5;?GU`6H>_@rA6+V8=IkIrb
zbGnSFS+0^R^F*$;-J1sY#_5rCi-6_;Cjlp3UFX8i8>ItMT8kYMO8g==k*^2bMoX+}
z$Jw2~bqyENcr5BD7${7NY!tj(J+!DR_TKg6f^7}M$LA%L9|Ef*O~qbq)U4&k>cn*h
zdlY)8MM;TzwA77yEx#RG<l4Pl$6EIxa#SRhE{jH(yI;()ulyF}ZZ0OSN^ioAn3=$<
zsULOLXMCFJ)TDc*_UOpz1{Sr>sm_iUIulO{N|j1Q<=453EyPSc|5mvkr5+_HR_W;}
znlnc@A-AkHn`+f{t4M^pj+X9`?aPiTRjfPv&M@KL<X;WiICBK6jR@;5VeHDN3@Ww|
zpBaCRC355)OF90{up?N{LU)te6WxrUxFF}Z6nCr(S6})3<Q_dYxALV&@2`xF$GZP5
z@^@4J^wN^nDJ<J^EO*W$baaV_=r%QPGp>@}T<CaJ=y~oT3AUn<C3)`wd);pAjT1^D
zn^ER#tNWOFY_YteHt{n(Xaera>t%t@>uaPgV|?in78|_A!@}7`jfD%ZcEMlKT{OR5
zOYLIC+P(EYHWrqT85Yiu@5sZy$e$qii~Q#Mzq=oJW9^0iBZI##(bzw~jkgfJ`{!$d
zZukuAtdfL`4E(EPXm4a>?O<x-C@#TY0B_*iUeIvB!aB%={M{vUg>eeTZ!uF=cT~T0
zQP9xF>V*C^8v~;gE>^b4I9S3ig7DJH$Wfo(#mds!LC{6y*w!}$;WhF$*D?C7uQ*zW
z98<rfNH1YyZ$!^`;^c{w$3zL}>FI^-uNezoIWP6&=kR|b$4ng^Z3VfwoSmIdIG;LU
zV{gL6Eg&Gkb&`jRhldls!Rg>??Wpg<Y3;!9eUP8yoHueXv^TSLG_$d$N5<7RuyJw}
zId%-0=+~d`>ojsP`)4L=hacO54RRr$aB-hF$@Ob&_^B}RuArisi;<<qc{3}Z8O$Nd
z!+++q@YWBuee}<izx`DGx1aI|p#Ai>k1&6#;$UPiVPgf8I*R^tU_a*l{lgzW6y`$q
z{u?d66ME}iptLA~FxRhB6D9adls^{^lGg0Jf-?LIY=->7MUI#C`@irS`&j>8pGFBR
zEO9KE^JkS^c8z^?&QocO;Gd#XQPwNlgKfaI|A0jq;~om$?(@>u$Rx_jblx3|xEU`q
zw&ywP2qUxlUEGM4XZvYnIT)I{>kVw8&!*icl@xdB@`zlV>Z~rHJYzdiI5@W6TDN;J
zJiA%PVqJ7}W7(t0I;VeAr{|dXO{`tmIC!M=|I>@;XE<T_3(8+S)%X4HU%rWjx^+`8
zPz>jP9Tu6If|b5S3H8)K=6_qm);e+A)`<S+rT*AXy!ftnkpvwF<o?wL>0A12{tX>&
zisKUCEhu-#s+0X2I_!FP#B2Az+83Lmcv>7bfBA#|<siN<5^rIN=6^h|pL<b6Vc!y=
z$U7c-{^vXY(<zImRd)RgW`Of0I6yk~s`yem<G(uMTOu2K{?)#eqi^ad;9C1K?f*Bs
zBOROO|DU$=Ys=BtxHxWXwn`lL{>|=yb&LMh;r?OWEeii(U4&cyAG5A<V)0%&chb~?
z1x@$0=}K<riH?#YuEM2bO_~<Trp+l=d%A9nCR1wJ6gW9fbiCHA&a&P?RA1rYExahP
zBt!owNCefIX;S$YK6&qSC0!t;j!O*bkW7g%yBpKlD8}y0cI&t0qN`)%ZsXZ%mTB+y
zu&?HM?x5s&xaeDgmpN7G(S?aMs`%rRQ+-WZHva84MIt(`10m`e*JN7v*fb?+%#F34
z+CxlB<#YJDy5n%eOG`tVyW3?}oEK5TeJL8J(1*UD(Y>)c8F=}*RQb|$pW$?Gxxc8}
z?6;cd3Q?lm*WL%6nJV5*;~`4Eg8_*@59UI)gTeyCTu{&l;kazf#fg=dQ|fA;iHLg@
z`Y^oLVWHKevS@R)NSbWPZU<fU-VgMc&UMWQbh1R_pze{gRLHPI9PnCg?~-*%uzNd^
z^;zETt5dUNs!jp5^RegT+cjT2DPqc6@v1RC3Zb8>Vu>oN^V*oD61m~Xp^@=|EL7Qo
zs3wfp&oDrM`9#57gSzuqCyTGMw|5|76Aq4BM$Yq*qv$#`k#%bBvqEJtYPVs<{X>ao
z&fN%f8gJvZE?Qq$s&y_5?tEd4K_g_7K^d^)58bg0=!1nY2ZeuhXf=9N{(3Ny#_u3!
zLe>CpUSuezRs}Alw&EFmR)MNY$zO{4*9&?Juv;4Km*>%s3Ppx<kVBnRY`IFIH$Z5D
zwdd7`ksCuXJ{6C-4WBuUHicxLOvKm@vbeM`7%dTh$PS&)6_c@nIs7)(S1SzzMO2fS
zIIb=hu1?S_b~q?xS+!p`d&-4=ep|~+xC)lK&AEOBA8l>LY<TV)%Y&cd6rx<F%PG}*
zhCjGpYrg92KiyX;y}B^Y?toI=VWZIq-;up@WI?A}<B6$8HC@Rf&-LDWERpdGS4NV&
zCJPn=?D{IosV$N&4@T}m-^<p|8=1kCvJx^bKwq{VFU~<O9a^!q9+JM4w-cKFm)TZ7
z?b^MkEW)|-P|Z{R)Ui97htVYo;WQKlXon=M_g3gL1u>x<Y6VzMycv4N5u$9vS?WU)
zNXV6+*6Bu#PLVB^Q${2qd+ebT85;AV8V|Vi|1JH@rmHahTGfL_+E+7li2MuB)k~yr
zZmg<p7$gVnz_YvIAc=}inbE1GLAEqiN@7@f*!g~-=nAWx5oc~F5r4-?=1XA*Xsoh6
z*{{Bq`NdklUPNC(yaFeaBtP_t9$xMak|4h4Ht{;{K7(qdqRdy2AvGtCCbk{+$H73~
zlDY406gtI$efui>{dN<dp|>zfb`&8IUDBHyZMJ#2I(p&P{J&mAUja#EGh#`?kkZq)
zir1EBpGyZGnf~_FhOcC(k-Pt(mgRn>*o*i0tlE?-ZzVrKUpm5(#hc{d7{#Ge2#1Hm
zwoyf`+$=64w1+?~CF^en>55lzk#FoH=68^0#iNMZn6W?f?5{;`z4*YHruXhvi~DSi
z;kS=Z29CTLsTgf|DVIOrr1j~I(`bU$TiosXt6xn#jityf)gFzH-sct1H|x?>(s{X8
zBx*0ucrL!}Dz~dVoI67@w9{eTgTo4Q*?my~eMQuAdhJ%Z{%X1&+3DV?Z|rhgjh`;9
zc5Ui7jppc~O}NF<YhXC~ohMb~(MCCR&;>--aPg~io$Z}O)i#GZ5z4N$Z~SWG`U*Q}
z_X`@ZAtT$*=pUg@<Ln_hLffq4W<{-}(n!k^<!-O-P{XT|ZS2Q{aT<7}?^8i8JQsBa
z&=@D^GD_0#pk$nUL~mc9Sgq-uM4cwYjAt6x_#WicOj@CjWZX6seb56KOSV21L-U#8
zGp0TrX-rhp{H*d-n4Cp^E-WLVotF?pcyRFgm0&+nt10Ga^m?CG%=Ac`%p}U=x;MLa
zLBol{rCusN%a#$MUNf{Wei!WHh%0f^k^udqpx9#1)p5!|kp=ZICDlgl)rq$@!Yf11
z{{vCtU?ZZ7R~ylZqK~>h-d3LPI?JQoGZNC2W7Ymx$a&(wTSsW|{X?e=`uzFjE#4$V
z46ZMB1*Tkmt2(~I8NN+~=;?s_ms?C!&~%o#JWfR*9bHvotadf!1s#<p(DVb&_K&vx
z{H8cLA>P92={=7l(LZ~r)cs~O#V=Sv^mVv|v3-(7Y5||T;P8u!O|h)#WP>wCjR8Tg
znQ*C`Lt|B(Udo~+$#AcYWkywc+fLWP@>I|HM2b24r|5Gbnn3(BaQo@xqE>XhC9Z*6
zd#3Mx-pmK~YC+?Jdxh5LKKJ>ur@S{XGvBTY-Uq>%c=IV8M>`Yo?iRy9f5f8D!!kZ{
zSC&nBZLa>e%RoFm!^!~8(Z?FJ9CmNUbJ{c1!@{;liad<qNygMD6QCQZGre~idO<Vp
zNykQ&4Oh_EXrJZEMc?ZdQBEhr7Cx^$jX)>SLlwREK}YRsc$*(OedHkKbD8R<;x$)3
zD5pe@J`X}lZ&1-ZSx;m!WDJoGZTl)Fpu$PZmbAR3UO%<^Y7C3HC9#*->03T$Z5U%1
zdLxRZ`BbmNVw=l#TZbVRUMo17#1R_HyqzQV;R8pQo(jR>hzamns7~r(-KD@|lwV(6
z)ctR%(3WNPslaV9Bj8E-yIA=MYEY<G|0^o!dd#aN6)dz(gGMhk6^{lT1}>29`O&6i
znxU8{=w#pGOe0`!<qA=CTy#paJ}q8ox1j?Acf8$2gC-tKAGhmFe^SWHBRh}-xci!|
zQV4o#WBK5jfrm;4U*az>OR3AxOL*_QmzY|3LqMcsJ8#n~9s)@ZNheXov~$e1;B855
zZmd`H=``(mGZ^XC+kKm)R#OAhTL!6kk0dSj@T;IFu!3w*3S#T0`C9aDFRhHG&JtZ#
znCH~Yx(kt&v=8y&k=KbkBqPL);E0u*Jk^OX=;9}ixPmc0N7$;>HK<>J+r(|%k;f78
z;@}Q+0qt|m#UVrrO>rz!<bizkrj-&`bBXT`VL7pTFR^8Jit)DJ@4^mY0yk1)Ri77K
z<fA#Tbv>7j9LHL?ztu))g-&Ojyzut=xB<9r>EFUEjALkd(LCa<WA7IS+p_m*(s{Ib
z-p3Vpzq{SnW|3NBGCi*x;c}mvHz87J2hqn<js@x(P57x`Slh;fE^tH!)ttHa?+{X^
zET}h%iYKZj_xkg9-nIVvM#b%T6NamRL@5m5ZPQxiTtU+k^j_G~bRLc!wxvYN?Cwu|
z0|iIoA)uP$N%lmu0N?pI-V(<6&`Fnw5$Yr|O@S_|xZZ9>NeDqovR%S*PgAm1WtQ1z
z13%83_-{N(7zzleMehR)DA3`^gyuGh7vj{4EOz8gl@QLVIQpB{2@Hnkb>AX1&svg2
zJJhWMiWUUjlwcT@gy9S(WfVWs^*r?Tt#e>*m#14uWqi|K{&vgE$AV7gNw2?!HEP~+
zLYwycfvLlRRX$C1L_e^=VhIwF1AA$0#UadO<d@{PXyQ2HKJ{*>J=ezCT$L4_UlEO&
zgM^paQ3)6tlZYVP_1lx_r(J8Y?^ZS5{{2PBpzQ9>0!|CU&~KU<g1+bFgX(#<`tIN^
z`a>{neB_y0s?KzOu!0nvK96o9Hrml{Q7Z#fZ}8Uwl|$%@jAxkYbek+c;x@LQ&VA}^
zgqw!hP{(Tr;nlIsPsv)jNe^ETqtO8Ib5P-k1L@coatZAmmMI<!yW02kN7jk1aya0P
zCS}U^yoemn?Gkm!KE6GH0SpHYM!N$AI{8s9Bd_8NHrE#emuCk}q@G`T$~XO<A+Rjc
zZQ{w3umc!sA7N)3U`6-Iq7k$`(^q1XGhe*xwVLbYS0}caKSI%#YeNUlt?n3P|5bGT
zu(w-SyaV7~`;v*G4RlkS7#9hlJXRO{H8PD=tmY#GoqUlH6k>1nY+m&<=mgy&&<Y&;
zh|(u?<%46-I0Xrx@}oYqWU2WxvW+TVGfnF9z#BB2t>I67ldW}fhoLCo{F5%c_dq{|
zpxr1SPbEY$2a8s}PA0^*fvYSk6+nB)sx9kP7_ZiLUc^En5ti(y1`KEIns#@s(B%Dh
zxA#_#wPv#CG!XH%(5HZCEcK9hsi#|HokX8{%a9L{l8znsdB%-SL~sy9!H=R6b$98R
zi!Aq}RJ&*dp><&Y!$`jqZ&(E#hZMO-Ge~zB&mXkyXZ4H<j2J=X#8iVE*H+pDLP?7m
z*_HWGN=j|pvtt?6CoUBpBt)o<LA0Py(3P4E-!Mc@g>7o4nu}7t>h)r$8$%7b`vdYD
z#G*^xCBtWhCi5LaMl|>^20}Ppg_3kk!%0X1T?L}F@jpK1X0U!e5K48*_~OMV(SlFr
zUDj{Mdm#c3JbQbOB;+1xtd8tPXYH-=;~jxrWxI}!VrWhZ)~OHdsb$XtMX3YDHe7_~
zzbJ9rbmpt*7`5kE8P<s`-uGB&WdX;aXs9)FkS%cF><(nqAR`^CYOqeYgHDvwsN!dm
ze!bwGy5+BoB;N5tGNzD!WI;?A-kLl3Dd~_hl>|d}41YSQXk#TEg4s&Gc1xwDTrikY
z)uEhE^#)>h2)h(p5$B;mOjUZ@LF{leO0f5E>--FK2}57(PxHy91u3SC#mW(M9^3iL
z_H3ZtKKrZfaE9AXRon=C1LcJc^&QeZ+rKOiI>G+U^oMN$<Ii>ew!Hsj713RYAB1}U
z3iwo>U#E4)PcaxfgcH1x)>T8NfAeJl3gCj_3U-WWC@G;B!pGjP&RzabPMMro=QpVh
zUmDA->;0@C#_3)gB_@WX6$vR%aI9IhW*B9eH?nEwT3=lPUVf`?65$lOG1`)Dh$Pad
zQso|<NT05tk)FAakvBr9A|44eo5^1_$8htj_g}ar6wIP<V(lrPm1aKt{*Kq?Ql(hv
ztg@>$Q%vL~zE;MZ4*TjB^Tgz5;Ps}L5(^y=4OR@=X^tbH%x2Q6%`teAo%F&vVXpkC
zk_IoXg1Ia1GgU0?E^O=VNIJNNze~7qxyGhq;_al)QtoW65T*PsSub8=j4==z*XIFW
z`;S_ndjRx`e`9}+QwW5#22Z5t%5X`MbEj+EvF6u2ewvWwfCTEoT?hdZr?{3Ucy-tr
z;z+hL)Juf>R){s|&_80mEpC6Tjv-K(tgkXK5Hdo-qH9y7Joh!9NMzY`i7vpgA#Mh(
zZSk&8FXbM6(2mD&2DgS&N8Qw`Be>at!L5yKC&4b{f#xIJ)B$lv?i#W~Gc-r#bsC-O
z&+etz7>5jeB5$bA4wCueILKCP&g<V89XVtMLPVELSBh<LPvp<mx~$JP)41l0(0D?A
zAwYsyfYEdAIUSPFPe{dL$5;<g_<able@c#4IvPg@Np9V`0G|<}BH_;Es2>xb?4_sH
zZ_H>R#m#NoT@2dK<oQQ^#T~R?B9ev_yb+0EL@u$O)GzrfZ=ubCFttrH%k(mN%8Qja
z`CB5Dv<??(uea!bijnXc;ktl}wvsLF^#sgZRT&@KEi|n&b8KpPFsc&L?Y&4UWB7C2
z;LEWuBbJ|IFYTi`Ngp7x&@u!Lr}=EycCY8AxCb!S{3}yT_0N0~ZW@quL>~l7fv&UP
zR9Z$dUWfM|hyw8}m`X?iH`*lg!VZV7oCTDWv<WOn3l79Pabqt%J;)$+|1R8r@g6*(
z{n$Bx`BjlpoV;Cw3`;p=)Tu^Oi#&sq-o4nq@nyK$t{N;%3`5&Eo3`Fef7P^~?%J`5
zg2hgRi|(ZI(N$}~C79t<_dC09PfQ4w`VFg9wk0o3?LqZ3cVS~%>`mQn@W=dgH)bAj
zUmsIPMQ+wzx!l#3V?~1mtx@Y^Ch46wMpeGFad@LG1z|>dEg)rUn?fU+Di#kRK#Vi^
zH2H{eb_c#x(UdF<{R(OfQg`a%b`^)xE2u5{6I~k7%{n4MFOC7T6J*L@F#9^TuKTp)
zeJbu_p6jE!(lcLGJg}cy<!I=W?A=GD=;4=Zu#F#4z6SvwFf^Wi2JNHw3B?bX<a%@3
zt^I)Qr0g{V6IXC=EKVGLF6HkF&T`?3-{oECssWKUNhDWL#;1;h_EEo#9^@_TI_xUT
zPt^$r$7MKNwtxu0U}>ty4sPdQojcsN=sAJvRF$sFL_hXuoGq79lJe?-M^~?FaA@bH
z{l(&x%!fLBz2&GiXzLc&!H7hw;U>9lmR|8J90Wzyp-^=1yp_TQE^d04`T_%+b-hI1
zN+u+Qii0ab$g&s?dh-+bB$|tB7?uuMR2mPA$Nu~UhW1pKM1d!%U+P`1qIQ)+E|f!~
z{IalW4Yu24fpZ6BSU;s3R+^tEZQHKH!XN;T!>_-D&X&uhF3SUt4C|xMa(k@IxgSXS
zum@MrX+#-s5s49q7VI#tA`mt!&VZ#=`!1n}J~$ufZObXt%Iaez40}rM7*W-bYLH&g
zsdO5C`KaQFN#v6|_Fr~zhX5d?K=3{mjRL4OJPw`wGx9>?Sx6$z(^u6@1{;wDxU7(Z
zn>x80@+N2uh5Q<|hD3=rbt)Ip&c>hB_QUZwy!q45Y*fAFOGMy;z)AJP<-KS;v}Lgw
z;P<D=lJwC?2N4>?&aUB3O&ZJW+U1FiCk@K*QU^4d1W3?S4I-E`pnHR(3RI7yjeYuA
zab}$;P1pK_vmqR}{~5oSOvd^QqiQ6`Lt<79HaSjoHbgdH1ls2~Lu?G15Yk6sBPlk=
z$xe|@xO;KOj=vFto6}<8w-w5RL?;S+??qex7A<y;dtWfxB1e;^UwS31=ClTRUjG9*
zgQ5Bu4R$$^9oB<HJG-A9MPE-5w!>hpKWLST!6>gtU7V2AP;t29L@ta!!Mlyrm8IK<
zYK42Ls?Nv+bjxt%KpxoAzAs7&m+YjJj4snJ3sLi2yRc*avDh3`LDHT`Y4p8{*F}o-
zd2I}OiIb{xqWULI8(zpH@_O%J&@))zs0u_>_M;uu3&tsMiz;HDT8A}Q##X&42K`C?
z)G|Y+sahGsRRN9@#|JTB(i=w4O5m6iAo(M^xlhu~mNOB>Y=rlZhzJHiCP~G!^Y*_c
z2HDT#<?$Nhuq48Lhj<uM3Rgyw9=#IX&iQ0WP+Qy(f^HqY92ja&)kQ8+d7rwoq<z3f
z!gmPTlV@g+Z1?uwM}i<OEGSE&EiveUICw9;aAz-P#`pQ!RXxZa1R~j%f{XLp_#f{f
zJ$UHYp($uXh)2hY2ObeI82R$*$2yQi)N!TQr2lRmWMz-RW0uW)!aRx%iRp-F^(KA|
z5A>-(3y^`|EsOIRBg6J1t5JOv7Dy~<h-NGis%1dz9h|(bjGk_N0f@xlW7mme$;PVQ
zF|+@TPG3@b#b-dXdwOrsLtGT{?&T5_0XUTpF{b-4&uiy|7z$b9m9pA*%8&<axyH23
zEXyJaVgE&E^x+hdcV(A2d|*z21O6E4Yj81<OhC?p5_$?Z2w9?Ui@;;q9dfb=Yu`gy
z`_?e}?I?~fyeo+xwR7SFh|`LM-c6nf=x)jtSfb2+p}k7@3;lY*A=~b~SucZOI)Cyj
zhAboRvi4-s!khz&r}v-<DxBYcaD0#@9yb`&V}tarWd<4gn6N}9NZUQcixom2juqZL
zz~mk(0dvy&R-hRp<eS&zAQ!5&`idHj_!nS_ERtii7nxduZ(mY9hCU{}_<w+ONP-vm
z?5{d|P+<=xK2dGsU`wz=*DS<TB+%^GaaiJuyAiivK`5S9CA&?4A0p`4V~;1$hl@tu
zeK@5-40GDEQg1WnL7eb`<#?Pqx>`q;cp%;3WQd*uF`vdg^fCVk8QQ@BasCJyx?}f8
z$j~I_|34v1f%NI)Qzr7hr^48(v5^p)*15w@<HThvG{S6g&CblSO$Be6Y=G9_xHTy9
zZTH?WO*0Y@?Cc6`v;AADPn;SIhZ*bBZ5k4tv%fF@KbS`}9r{a0&=Y|&J^ip|n`z!@
zi~qr7Y^NIXPLMZPaq17>b{go{^gmzh?5kj!tRLUL0EUo-$jD)5T4B(7XZwxurJrkj
z0UlVecZR^uwEDT&A4K2TreS>P=NcOzK#`GXrQMlU7{|D?GllV`pKCM&OZ_6oa{DFV
zPSpFu&OeVD4(1O#|L!pUVdvjO_m6h|sa_C;^+!AZu1o)D=ie;RA7|rti~h&i_}#Jk
z<81tKHh$-a-z4S#Q)lCS(AL!=+|2qxin%OYz+np+dPL+#S<zZ`5l8Y2lvDX{ZcJ_l
zl#+-EYvx$K&hjuSa)+Bj1YitEjE((tUTb!iS)Zg3>_r(|22cZVLaSBZd|nIihWBas
zQ_{XcsTcy@4eg8)^YYAJo2i}#2&?2^ZNzi9(D_v8vHEh+0|R`8oK5`!h%zt5Sj+$J
z7J~J^RZOYtz8~-<s*L)N2Cgzl2izxG5eeV;*$aTLC0g|n<45}ve3tWM+Gay-aKq3C
zP$|oyg@y&F`irugcs*EwAnzaZ+jKT8^gaUQYx#+{V`Oh9-dJ?NRneWGi)0h60M78e
zs=DL~1yTnpyZ{%%ps|{vhG2K&BCh*F@z%+O4u?7;o95M3C<(g<fxD=6UXS-<UJKQj
zI=Jg6h<V;$dDZzV1%N+h8r)>GDgzF06>y4rtT2jQ{f+&8MrC2=A*CwQatY&-L{+DI
zI4&xnv%sygx7_d1YCYV~Qy`bg$)Z(Ng2<Iia?Xk;qaoKMW4KAAEAt*A>H{(tVqKEq
zj_VWL=`VF{u6G$IMbyPb-e4=7jhOsmrkkU`GB<1oB?t%Pmo*l2Jc^$3+Z=(<&9xmr
zNg}}Hrr{aDrC07I5hg8KAGhiITl&Z=LuY_$RP2|g-Tz4EhS@$EzNFe&nNol>(AX5M
z=LEZrG=AEau>Mbhw-M<Qz)yRNHb#pKprBwr(pX71jpfNgHCNI7#V)TRTb@f~f62&q
zYAk5#q?s&?wcb&)@_>7yJC)br_ByZEmz_ZWIip^1KHKQKy>$2CYP<qUdlfB)Lhu(D
z-@Uv|EXayL3rjMgUgUZYPDH<xdevw~Bw*r@n=``Hfdt;A&Gl)28+QgkioU)eKXY8C
z%1l%y+5~E}%3}qiWYSn(2MEu+7hE*wyH6)33QQjrcBv5k`eMrwyKo5-0KDep=yyD*
z89c_+62XbQzL9-RifM7il?Y~GI#G8$!C0whb>^1H?mheNzMU+5ThE2xa54N$dZ&`*
zgJy^M(Ppj((el%pm(xt(N^UyY#v_qbR5r5JWASTWsFw9T&fJcSGv|9w;gc{eA{U#Q
z<5htO7&HFz^UxfYxPy(t8m7bk)VkgzbUJt}Ofhi-SGkJT?0Xt@t4I5e{O&q_cdAhp
za1&w1q0xtai-M9OyC7c<)x>fUpVXLJg7dE=7OENY9vG9s=Hb<y3<-x`S5#l>=aEZ5
zVxck8D6b9Im}=NoRn^(IV;OQ=z`;Tqn|!;APk3H-)+;s~9exzvI}2o_SfSjMlP8ib
zk$O9xqZKn8qNZ~X>XQ!YoV|_|y#j7qX-so4(`|9W4lb`Kbaq{D(lC)ElV{VZsydI<
z0Q-u0pfC3OIY5mmi<3P$`bVfn@}-f))MtPTcG{!yM6WL6Y(y_T5w067yztCY_iewg
zy<rw9$Mcyj%p^o~AAEL@7Z`Y05;1`5_MrIEjCoq`Jb*MSp+s7DO=EHpkVb~x#VDr<
zz@Nc2ZlU~Y$sNMy0FwDTzrwQBqLhfABfrAp^Cf=H4p|+w>wuJ$w)a(>>?|N;*##BB
zPOX(|+p$gbL1B=cGXKJiQEf4>AA!s;#(m*%OTL&E7xUSZnqYx3bzD}bOC|pS`E?za
zo;#*UCBds&lbN<8f4(->2a1;@-bMCPs8`{O$aM9O0Q%{uqLN9P5%mLdCjiuvBIU$R
zEE;tH+BO`|Zcb^8T)QCODI-6vu^@ZyoX?RPLm4%mJ8*eC6{O~TTL}m;R5d$GkZ~G|
zt4QO(BV8Q}{VFqI(8$`>d4t8fN$&4G;u8H%ukI56*8|c=T>BgBf}7o#QkONKSOj0<
zy9e;HRQYDm=IK-BsYW*LaIz=z%@d8?0N4g~^yYla7<Xz2!$$W)T9eIMjsyKh9FIen
zzDBPOa2Q%HOpO`rBY=H7dN;FCRg2RtacY&zYgz6n>BGOAa=JWM1wA}hK1U(t&*s@I
zal4W~&I4$;v3)$yYt3*U_38N90>I?4Z>%&IHJQ^vjX15-m&@60I{{A=&A>*5<S;a$
z-#_7~(j!C>Xf%duixg+cWX)_}as^>-Z{ow6F7v7d5!9vtTjcBmfT3mgL;`OVXdpW%
zrfhc2gQELs0Khbx7XjR{?=C|?eB-wrGPo#Nzy-X@I-HJC;**ArI(=KDN*gg{h^Zw?
zNdU%2F4!x*)+)Wu=URWI9#hMohQq3Wo(C#x8cS=Pi-nY8OYca`vqM%GTnkjms5BSa
zoVz>>*OsQU!*Jcq?UuIZV-;gy=M-<aLoonD)MOV<wFD>i0Ep!tBkd`#jWzF3Wm5(D
zDS?^y41{stpb#egUEwK{+WP=`XnGSBo)sLP_nE5sXTf7~%zr<j5KPzy__VvI-FqvN
zG%~3*n;I6N`jMhU1Wa{XLr$1(?8PT!0EPCbi7l~6e~@;`P#e&iG3FQ1eG7;r|3lWC
z^|`nbj@yF)FQDXfX4E=|b-{eog{HJ;qu>PDAs_|>BKBoO(5|-(iu^>(vptYni=vJB
z)a<qo^Ac%s+7Bu)BnLcz3xHW>v9g=RFddBtOcV4J>{<~->e22@p0nZojoLDKao<$B
z?ukSQIS2GpT=loZUpVcZz`gC(7H8+=ZuEU@|NUdj02X`>U_qwj8Kmh0Vm6`FCUms}
ztPADd!;bY0qI8~1`iy5IjZ%S!*og)DgN&Z4jVvlDh){HGEaZ+RpG`A)iRXl&sbDYv
zsdrE{x#@kqI>2the+TE~S78N}Q1hcRxz0%H2#T_Ns;ATtl-Rsz5h#R8s7olE>jdNx
zg++3<DnOfD%(P~VMw4>`UMa;Pz;cW1)S=2H=v2|=vFN{5R1Z}?f?sjnMuJR#i*SCV
z&*I5IZLzhoR%2WO^Wth|n^VY8#eL%iS9a*rbLpB`!7eed2SbR-Tfikec%J85Mssp*
zC-v<;M81_r+&I8x${ZViSrn{Tr{FmgjMU<X=1>%6S$`EQ1X<IJ8y*8xEVo(13xh{5
zqc)eLHV?6c+m<wRvCqDAp6o>GZ4P2fiLDPSj0zuXQSv#zdz&>D4?%EyRZ<$v=pV%&
z2dJ12g1$EAD@Tfox>h?E&9gaIpU3Fc0ZQ8dRJl%$o3`ijfE^gia%e$9hxl!y9jUEB
z;f!;Abem@?0uF0szMk>`2Pa#n$O({MW#BW6%#D6|r6dgP8<6@P!Nn1arE1*`vaZdI
zMZ!3I#N|q#cr)5G%>3UjI0#3#(`$1*?2;WA#WL_8H4?2_>7cp8)o?&rSMy1e`L^oa
zCK}121cg;VBDR*+*=*P7&WcZCmEEYy4Pf)In!j;;`PxjMzqCSXbAajRuOjzj9RMsO
z>f{og1+`7F@-advcS>W{Nw-|I%48-irQhi)Zz`N;<z?x^Dw$On7?7=G5d|Qcl<c$&
z&Z0Zvdga(|W9ia}FQ$rkz9%=<2F04q6PNpX{kQ^pN`0CwHCGB^nq4Svi^x<W4nG3j
zeOY&UZXo(LvG3|+k=O8eKVvkMFlN;z7G8%2B^OEr=C5*U<=o%TAa&k8LjeGJ7ET)Z
zE(iqkupNNfDxijjaMAkBaBTb}uXa@pqxeIdd*(Vt06Xu!f2uaVwI2h1fv`Br7t#KC
zoayxF`YX)<)&C`;e=fo(6|hT^U4<SNyeur^;Q*_&18HeAPhB<y(3)Kp)1xnzZ0@Ot
z#xwwh-Ur}h03F@<GOU&v<F&bA2S)3J74;cgy$-0$lk=i1VzdzHzPa01Y;@fc!2NS0
zjqHW<$u^QmKm!QW0C2odLJ91fEa*WoNe3r6)zDS#`HBn2sm&~|9K5^NmTE=Y^%S1+
zL-2Yet$fOV20%X?(h%|s@=42|0zH-E9PHm*nh{LtpI53H%{~I$xP=gn+FUb<v9bS9
z1Gr9=!PnVMn(S|!Cw1`ZkQ#w$I4mjq2KgF2g_(MQ&C(!s^hjHrS}GZdnGeZ~OTppw
zO8F1y<)J#%SIokLdDqjGCZ)vB_MS<ZwSs?#^L@#-Xubh{%5a-yW)0F)AjL_{^{P4<
zv3X;HO7eBF@5@9JP-Uc>0Mod4Hs)3ir~`o95{$(ia!b{jzw@9F$p8SPMBetkeU3x*
zN0Er6#rUb!2v`@<@%CJ<@QL7X>@Q%&dJU(c(vG4;sw%Ub9iPmfRv=?3vv%+nE}=p3
z&0W*rUVV9;`^+0U9F##waK)zhxGNvG;J3$yNOgCLb9{fJ5tNDhGk)pcuB+)45ztad
z9(xmd9WajSIeGzsLEX)HT?YJ0j!E;cZ=8CR!^!2(#MnNQmxvo5gJ|!L+I6PXT3Ma9
zA==fEElLiNwGl#<Nc=oISQqHjd^MM}_ITvJhjT(ukI)Ax0pI=uF#}g+uvj$OLC-#F
zPHrl=Hqk$?1J%=r_vr#4aHwY7&kd#%37#WXd1)ZkaHwmkg1>idwl45CiD=$8&aNF8
z==5u#daw-&7lZ05lzG64gmDdwsb3Q&W##8zuP?M44n{bi=#bY5Q2_@9$AwsAuKvvM
zi(rG?uh$b#o#$==H2%{T)93!GBY#Erhv=y3xEhT?38+(rmQE?8KWO8kJa0hdmWS(C
zg4B6xK#C)EPxA|_??^mJy}+LwDQH$SBqI$wseeu)H+u&U+7vNG%=-N3kXKS@?Sz!s
zka#UgTnsyX9J&)km_iAcWlpQ{^!lV%5oyXRSA$we&O8Iz{5xDZINx^!3LNkEH);`G
zMY@_GWe;GQ$l<K~`j?^VI%SH>b-oD{iy_qBq*pONJ&#}tp*z%F=-?psp!r(z$UF#1
z1&Gn03s}}(*xvpAvKa7{UfBvN+9xRTJo+z6Xwa4DPrEA!n@wov+1@w<{j9d6=>2Ut
z$0OG@{C6A*!>Fqu%G*iH_}#zxvMW-MQF<bA8yxYsOM{<D!RQf)EE24%IUfIf=eH|3
zG--z|(35}wxwaVbqDPyUU;93Sas=$J(dx+#eO|UxqyH@Ugd}g1q|kNy4hsfDkboG1
zH=LQ<yS8j!2}&4zi+=q_joXtn+Xq*4g8ixF8XZJOi0@4EpuKR|!ebd2y^M-C^(S?@
zIcY{WQDP_b*<?R3^)oTeKmTImYc2cwn)ZyGzZRR!J!5;8LyzpLID_v!>>zu#G8)&x
z!*liTifp=LY%%8vUf-EnOj>~K%izb?!Y1J53>4^copMVph{}!}T)+BlKHN!kRb$6x
z#fJmc2V`S+Kvi%5{v}`(vvVQ-|B6xU5#`DLgm?!b$6!_*ywh-;ari2H|LQnD0Gv@d
z<(lBXA$mOyqE3`ssdi9|+qjAi5t`QzhIgnK-o_~6<RAxOibjSQF7tMx!8?D5{0~|F
z5cwA`q5cCWf8ewow<7WOD7^kTEh4<IoV_&GM$5Wo(Q+qV(@TTmL_LyJ41WuS5AfcL
zq>2tZC`f?l9*M*LwT$mCf)YUyq=a9`D6fZM>-9i}_E2`!qsx#u7f8U37;}sUWr#Ar
z1ORO_gGPUh{w#AyCt9Dp>lkG|$P}zFg;Yqp9>x@t0AasjGlusM`~2Ky4Wf#qBK$Er
zQyf70)KO*cM$Z)>JN%*zlYhhI6GkV9B2wq~sW)JIrM03Otm%zvQ{K-hzb`^O3bwt1
zGnux1P1%T@xhg}1k#w7bE@g2=8v)z%bc%E^#mbk43m8)zhbd$)?}^x^CrRU>6jg!9
z*A#;y(HBA48YOraZqsRa%Ft3Q?nC2ej1W)p9867iQ1TJ_)QC}MgVI!`s*K^&=p4d|
z^c8cpcy6+N<BgCUf2CS57mDr^Jk)`y=a?GKqLX|pHbWx#7m9DtYsYXBpiodLM0S`3
zeQKbCcq_2N2c4tqm?fSF&DGyIV)S1CIz(pyO<r>P2cYu;${YkkIn(;ab}m#D1)9t>
zo)1K)iTF#Hn(eA+1V-gG(1G=DpvgVo7<P1;L=(f*PVvvUw;#tt#M13kso#rE6Gfzb
z3^m1PMf9l=daghf{%1AMy%@94BOBMf<d4B@$i~HB_H{dFWsE7bV2ZZTrsC~W5I|Ao
zajK3U^iRaU!oEjz^!qSKj_g|&cEM^^iIJ=$+7Ci`A%-Xzoez*T4Z@m4SgvECuc@8`
z)@0n+Zi+7V$j@(pG5##1<%V$);E$-5Ypq1+HMGdWqxT|Hh-b@TOmPjSU{>pD**?V#
zj8&!dxe$HZSSV=O&A46k=z{{`utftF(CaaA@UlQ!!wJwiV9TYQ0diDTe$v4lA1Y`9
zl~CJgbW>aucCdV=jSz$T>p{Hi*tYW$>3eDLg6X8)FpdF76ULFpMW@4-_qvQYw|%CV
z%}2$d8m02pcK=qK9#K&4wBs19FGx+n1v^85u`guR{Yb@P;PV2^#qMZw5Jlh)#GA3p
zqQqEiG?Y*EDgQYHgv5Ui;TLEAk%WIl;*UD|MT<Y`=pTLipKSsIuAAuG#3P^F)WQSo
z82a{RaUM>nSL=(ZUP@6LZ7cW3;)g*ok0?qS{vtB{$@|4d%^mz7RiA=Zs&_*PO?q&I
zJ1mP7hy^kHGynJR`|odY8&&g}XW{z#c$%6Ol>??ztEB56BWcc@wsE8=zV+@tpY4Va
z3l$4a+Ue|>3eJ{3+V_|A+U)rIbGU?l<!B%0729kqy)Sb%Ao*v=-^>v!;wIz!MiY8~
z2_7bN6CoL)>6uSTyo~M(V%S;n=q@mv>nS(;7QrCK>CW&9r+~NZXMSPFNq$jg`)q72
zN_B6jDYlJojqN9D^Dgbf3|D-6+WmiaF3yQF@p^N5s{XkG$ss?zI#MzkYH)==2c4tj
zzW6Tk*u%ry%ZT)zVW+C!xN^oQivN-$f7@?qmipZ&hPpyd?dvt`5(^wHr~WGQ_6Ea$
z`bE9uKBcrcJfq5AONSTY*|>aes~P&q&4n-8Zr8g(0Kj0|b$fGPrN?5x(RGTkL+1?s
zxr0MBz(@~c4+vgC615KwJR|=4SQoHIbIr~AE)_o&b$od7Tq{3uyUA>(z6OM?(ex-I
z-3(K8`)Nr6!LU+bA4inX=8XDM+2v*;ebRq5MJ;8Qj!(~_pG%3h$NpyWTYgCZIso<W
zGk5;ys0biLJdV|RlnrS*`N}@zZKkK4^2Ex}j}BG!!n<ZOzcqH;{q7i|-oe@57Rmp`
zWAuu+utPRw89G1k6OUhc`>k@u?|B$y%hY#9gXsNnl*&JOsD2J`eNM%H{fEU^kp>x9
z*L>oUE1Z`fR3>Uy$}hhjeg0#O;<Gp(>%!C=7=F08dte9q7N&fE$%8lr$Y6pZlLyBY
z80f-y7+p12fVqxd=?{mTjYZE>^`CrIP$o=M8-4wsP&OJ1c7Ppp_}LZMTL_!lYoSp}
z>4B8M&L3;6AwMZ*p;P1kBbQ4BJ2=%+`kQW00TRAC7c!TA0R}pjZQ;pLbSA+SMHKjb
zjgZ$`*7zrRC|(QG2;6S_rv(@a8+Je-e*La2lCw+D9+*hdr44y2dgI3$Pb(C!CAsqQ
z|7i5GJ(r=3gi8b}4VA|r7V-U(vYfuw8~;xf|Az^1{xHEGCiuey;{O+!U{0bNfEf4)
zEU6U0XW{G40KhTwQh3V8iL@|!?KlijN#en92l>3AXMQEnPL~`(PKnMwsr$<rpu?}n
zy5u3}kEF{yEG(}5GD!2#LGYE0l|Kgpg{`@isE@POYf5zh@us%yrpJMkVxtU&66B^q
z#Yj6-`7=F#*+S1&A3$Mu?PiFA@@I9Mg84l<aiYr~)C$RB6{3!G&c*WIQSca1>H)Yo
zkzG5WQf6#d$8&vTf|*}VXbNHDP(urq&o__J{Awm4J`OtBT<yy4Fz7B_hqH^<OTL&Z
zATo36{tv^WxCaa)kMdbMywlR%fF2>DvCh-y>!vK6y3>Q&dj!vqBjY{kO@MO?gaaf`
zADWl-K&v`JE|+(N-tNrScaO#ZNltwX<7H0*bW{&qq7p-+Qd6X((AHyKvf)7rRweKR
z0G^RTZ;_jDpLzNmX!-p{d>4LtrKm8l7hjU1y-(?b;lU?wJF5k2Y$MzFnGJd<0fvLt
zsh$w~(*C=-AKKti2AMt9h0EW1+yL|w6jRymm<-)*cDDn}WNr#j5{}(YgTY+xJ%sx3
zw3+H#VsN3Ym2*zPeG__~5(1#p)aC1_4yC;CbQy}7VceoyA`+aGGcb8YNXN}zsUum&
zsZ=i!^1{VTu{&O3vmf(r@Gib7@@}7Wb-eJ@IeEDuZG@oI>#XT}A1xZ4g<i?hH#Y9A
zYn?d};!Hqvw#%{RjtyJ^371(4ex&ll81y`w!Zm4*qj>NvVNE?BPt|>VGXss=McjMI
zqo^8^p_46F5<t0o;DJ0JJN**ZkQ;RgQD?Z4_U3I;XFz(I>O*%VeLyTph@NWlATl!r
zAITrVD=WVy`jwK6{tMg3W439kZ*;RQH*#EbUPb?TcIa^X^x?<iN5F2ovbISxcT-Bs
z|9rNdxDAf=%6e(fM*(Mlx4~I!58D>LVBOhX{oKH;>C(|ustt<)CaIOXi+H4_X~n1N
zJZG%|m`ucNNlpj2NN06GS@}@BhDOt+hfKovcS5%+AGkX|&XHD(Eg_m)eo*UL?y&FV
zSGdX7paOzLemxoLLDXE!hm8{kNu<vRQGWW*;sUt6;qH~sQwjUDvsfOz<5+zlSvtJB
z;jC^N6VrUeUf^Z~DeDE`A^IU^TB+Y}5~Jy`6*fn^^j8^Dh!!`-qt?&3-xb4(%9(sx
zQbNet@~BS;dioAf<^h<$q`xXSGXUw)+XMIwxrsTXA6n;1Q?BuN*L;fe9XjZ$S+&{_
zgoV0+zt%N*ggAH8Y{-6%!GFevMC?%E{PVXy?$@fyqbPGUpdE3$bo*3oGZ#gurorEz
zM8EysX;=_vCEGl(@rIypx7qj9a-qGG4uBu$>uJ6VK(r1v<y9w}$MX?FD?IUEMYCid
zXDnoR2;|%*qgNzCK@|zdSkdN&hRGtpHA!{jgf{>K;p3d&cen>?I6j8aL2m@4cH#Kj
z25&4BH9^PqW(NI7Twm*R|1NNURbKL{&O3rqvp&~jRTXFNPmv;xe<3cWI-ahnz@$JC
z5edo8I(NOW{N<*%8~aZ@b>7Uw-&h><I+9cKZBpI3Wt_oE+iRvlzmY=sOoMd(PoLBD
zp?J~O7|-QT-5$R!o^R9$kINxd)i4T4Yco$In(GF1KE9f^og~uyv;=_uBdGClmh>7x
zca%Wq?r&27C@w5;{Zh0!^Qh=A=uvqb0fFg1F$%@EaVe!Kk!4RlZT_~FWNl-yh{Wzr
zBmz0iAGfk0Ouf;VSR&N5Tst)zwJE&X!sR*bXX$40bYT|o)%0cQ>mXA^Pi+ca-5gv1
zub0{GJPtKIJ@7!6jpButItn#yZsAGSI+~UyPo5%=?5@cTp(98o3K60ywyJnk)%EX;
zUDpd{zBIBiR6HuSF&((2b&+zJ%(~Rv&Kvsa<p&9ozHFrli78>rMD8u**f<3HTQ(m)
zE#<P=a64K#BKlrOt3c3l>f6h={BFNpK@?Nr`08X=nu{2+`UWaX0Wb5sWJ`G-xH+o}
zn7SH560sFWyS?LQ8pycR&x9{8SLruABXqKyfA}+6h^9+iP*1LOm8RE4QF-2r38MP=
z8h~D<2p)9|3m}%~kXD1GbQ@DXQ-H=fNm!c|h&)!xr-rYc!|hNhhtNcx;S#`{n>~fJ
zZQtSQLmA5v7m<$OD0oI$4`6Bb0S0|Kb1_7hHNP8|3y(x(PI2LrRZc6j`*O7|F_&wV
zoVSHw8s`EC|6<L;{!&f_(dBD**~b8mDcS?wEe%{&hT}@F$Z=92MFnZyw>Dop>F%p9
zTQqz;2w<$*PCp^{0e0Odcl>&1YujDQn@dHIt{+J?rmNH`<6v$MAwWW;_ujQ#&)^vl
z!h{QDb%3bK=y@P_;t*NAS)c^~WP5ZsSJYhKMCguqtxb<j4uoo*K-w7ULv@Tkv~?^!
z+yENp25Wt)id$0C=hA0wyuY*z1CFmYas>WoUQ+X3KrSTR#%_xY&b%!>PT482ut&ht
z-p5~!K&!yy4(+0;+7H8yx`B6~0FmbryG|;}*EqfAr@5~=0Icy2l<BxfJJ?Z#`_Orw
z$sZJ%#OKccyo}hvGk_$`q_FQ@Em|Ak66zw3TKN)JqE@uhc(%QbD-z&)X`khV?%25Z
z(P%^u-rP-$hd?e5jx8fFDx^E#6g0InV&5!l*DbHFvklDKtaJZ%N*{%L@V#|rHatY{
z4!|tM2pP(|FR$*GIER#hts3#gLK|$svXeCc`Q$yBLuKHrAG>t7Ph`12xD@GgY2}wl
zRbC0e<0#j%!Sp(A^;2zyy(U6*+#ksUDI_k?Ex(s8IlK4p0abXeo-Gvm_*mtB<y#$k
zcwDLE4y8^@D+jGz+5R4fs8#KQih`5*vl-pW+T%kKbSw46RwYuJPfY7;^C6k-jPgus
zGt7?cs5bTa<r)zkU^Q<%k~n{Qsy|^l2*B++-Cug<Es(cfs9ppw2Jw-ik7G-wAU}-_
zSE*<vLDj05_j2Ze*&}b#aT~Xkd<?rd_4aV1zR%%MC(b<%0b(0VV@)dD7s=E*@AW)_
z_RPkUP=mG51H?Pba~~R9@>^z9F}M1jvBQn9@fg)Q>iw4>-6S{=C4e(4wN%p1X7P3w
zqyvklR&@^*MzwM;k;tj@T%w0i+V!VYBlmhDdyYipj{5G3gW3(p_m#{Fk-}-3m-n@o
z)z{btep;YdEf`ICYn(VzmL7Gz%}gULt{|0)`NpuEZdiePwN@t&Bgee2TK=&Pzyr1>
zzk$Bt?bZi-N5rNNYTHpnQI+|!FS|f3U%vP!M6~0UsD=J<;zV$L)e&Bu7j_O4h2CVb
zQM(xK7j5YPlfQ1c{6wZrC5d0&(hI{<(Ye^uO^S9DVM|M6CSo)xZ$=Wva<maueDKV@
z#E%I6wRP3b2vNfPJuiagKVXY<f3fdI?Ip%>BL=vtrPGtK(qu7%scQh#Bt*1dCMVJ)
z^dR2I+C}%y)<7z3-^3+HQ3Ds&BkZEyLbZs&Oly#3H=3;MR>&u6w#}XMO&l*9x5JH*
zJgC7)W08MYeXtMdpU$`7LX(u2>hd&$DOCO0EJN#T`WSmBwIwuY|0qhgN~bn5>|PeF
zyvnDE6iporH!s=*g>;R5UwJcoN}j=W!gRzr9nY26Cjs&RhjBWs1~jiPC96HcLI-qp
zrGG&u;#0*_)etM^Q5V4+2n9n`jdVt1lBoz|T_))L`G6g9x#+uyhUjx-sMkWdJbhJt
z;w^{Jd&Il6oo`e#Nvz0U$WE=vDOYhvsvO2#_bru8$>e){0sHpE%B{V<lh0NzUhc$e
zMoXx7Y+YEO<9=^H^=_{Tm*EuigvBRGN~<D~DQ9i{K~TxI3oE>-u9ptdn-?9DkV2mE
zkRy0WyOG&Dus1nRkj@S_%Ec{V>8=OV^ARoHO~fu2zvy1pes6t9T8w%o9SRpWi8g1&
zHiNCaW}kM2KiRijpGhOtt~O>e=A2)yz9y8CN3J$*CWiEzwOj>Y{6kz9#8*x$A-HX`
z_pojkJe5o%7(o1jnKroq-n`)g3q`!PPo4cXZ~;xP_$5I^b#z$5?(i?TMRXH(!IrYj
z7ImT)IRnb_k=NWi%GZb@luGqc_?ZxX_MO^+v-ibqa1m^B+LtSGcMy}=is~IXt-IRR
zTmqGGK?NH?2U0}0j4fB|EM%lX?av)590aPHVFg^7CzA5v6ci0u8a)Nqk;9E^W4~I>
zz1#=plVUj*sl=zJVw!?$vFa0b<`bscyEgdTSW=S6C8W~1rtP4-j0DFagD6?7*RwA{
zY<j=VVWhw*yYtYR+P=~<Est{P{A4vZ-iTU-@(@4cy2uEtur30L>6%Xq^A-Gs+}d92
z^R1q}oVC3lB*!b^;gX~Pn10LgI1?Vp<g>ZBxuI*^?pd_H$Rav@4>3<!y2k7qS!ySI
z+4q!7XDx!_D#DiB(s>(AEuVhH5+ld!6A=zms+%p3KL<Cw2k*0fC-kB8eVsq~b>IG>
z0fhL6O00IHv~hdKQ{VvtnJAAGz;m~b2#q(|7Opg=dMg09cK2ju``NEhG8N-7nqUJZ
zfV<fj%T#O)hGJ*!;7OL=hfWz^d)mwS0(q*_)HBM0_*sVCb$#X{c8!rsL)tQfP?kZ6
z$ZX~~l!Go5kBGN=H_>tH9Viu_4%L`Oo+NsQw^=)LexkdmcMwW811$Sez1Bz5pj4D(
zqc}6QN!`$Y%4>7ct91c+Fa=WZEFntHe?Kb~454rtMNlupn{vG?`Pph6M{MelF23Gi
z6J7P8cKD+;cg|dcoJrL9MjycA>sDqT6|EU{Cj(>hUxY`4bZ4{R;LYhqH*YDzP2Q&6
zzaqsO1ak(6kbjxg=1$mz<n$SGDGJuDx9)51q5GNX-bXgnxJAdZF<ew6-WWuzM>};A
z*P*;wXK513w->6@mcKpqnXMD^>YfKriQm>QA<qs%X%Ym}ys!aY=%Zx4-~!Y2=@GRe
zLBywMO*H!)OO$l$^YW7=;XOKtlZ#Su1x+7^@+`ku3yyCP@4d4xpacJlx2`QvJX8vZ
z^^9@EEjUO|JriTXkJJtv?fk|r+RK>&1tckBu7h-TAu&dw8qDok5IXr#QTxWcn5Z;e
z_065}1^pT5vYJdabFa|O3slQBz5wJokc4}kCf}~kgAu0U`|9|43PQ+hHX`uYueyr{
zY42}iL6Pm;mL7Z<S9{|ZctKGN^?%s=%cv~7t$iF;1St`fQc@A5L_ujJ6ahgAQMyGb
zC8VUo01Hr3Ktcg2=@cnJK|!QDB&17P`ajooU3lL+p1q&(+xz|SjQ9THa3G%Nxz?J!
z<~)ufj2+zgw*RQJqGZO1FlpWJ(fSD_lc$*+Ben%KmI%ASPw!4!;0Y6b?7H<S`1(NX
z=}Q#)5|DfLvz~O4LsLEjH9x2Cl#0TniJ%MP*25!#2Vbv1DE`voJ<wjDfLP?`CJ@wx
z^)PdKI}8lI|Gr@fkqr0l-Wk`q;u*J-u6?_h4z@1@Mz6D=5rZRXiSv*PxEmmYMH!-T
z?U7g8telP;o45VIRN27MG*4SVT)$EGe(P6bXbB~3Wd1GNqMmk0le;&b?^Cbn0K$W;
zsd?=q>?Ohw$#9C@e(?I2<A~_@L%nk$`PVmGE9iRnfy;jRz%5I#>X+<DYW20igY=Kj
za*^ZVS0BYGXB6!O@$Udpj9?C9^A^v@J{=c^;=su9GjwNTLXlLO;5%Avm;Pn;eRe@I
z`P6fY8&8-R9BL_Xy4%#v?BY+_=RURE?Br_s=*#<&!s=0r`^Prk?k0C+WBOsv+GuGm
zINf!qY$~;L=FYT)<$DO>v$f?R9a&$~opDzGHKybdZlMLkKUrOFpJ-4KsvXrnrnpjb
z8YPK1x_(YF@#^s1s1CA*`IQ{O2L}&4<{~3o)_~_%+z8bTDdsPYxx7c{NPtl0HWmPU
zn0}48EqHqGgNcWxc^gZRoj#kgt^)}4yVNSCQe3(~@dGW!b;LOQ>``7&(4$$zpjEWk
z&iv$fV^8-@a1!kGPcRX6)m@R_jboBV_;XsAh+6mtt*9~-Q$pNbcg<iGQJ=weyNRcs
z``HM_!?Z4jOKw7#qB={hHoJe<QDCL{5OWWcK_tfdRp))$p|mH#P8^M#XUvqW8G|`1
zMrE>Fi%}@h?^*)W?@&)7j;2}i?kbIl56nY`sor5OFf}#1q>AtivO{66au2T#1rzUe
z;8NINcj?_6gyCiK+0uh?kNgkXIKf7oD^gX}`M|b2q|?7(t{&wq1M=r)skOn}>1Vfi
zhyb>L1f4p=V>f^Ql(a(5x5yaJf0F>8tCPf**52g)v9H0K&a&NI_{R*&2m8?!fn>co
zKo(<iL=(WvkE!bg^LAu+bY}|tC!VW1WLQ=GfypJRuV{<??&?X*mt@2kzTT0C4I9w&
zQx(Qz?4^fsDZKM*hjH?0py5@7a__o~B7p;+bzy8wfXH_E>z5_ET`Zs-JuyZfwPil3
zg6=^8Oa_&MjhJO4<sqo>K^VjH?Hnn0P>&*4nm{%TY2J4vO)E(L3iG(n-G%}6e^0vO
zC$8O7M|-EadS7#<ZBj7x<nI!Nd8cxyR0b3;XN49Y?}lh_SANE0H$S(>JL6ETt;3`h
zupO)c58lTgh$}R$N6JZ0e$A_rDB0SWA;<Ubo%OqOAJGwLW{hcnUse0n6un7FtzqCb
z*{V4uskmn+M00mZPY7lBF9kw4Es+l%mmJ@3fcx_R$=M1qL_6ZG>(IeEnOViS>&Tmf
zC!B#sfgZtHA5jorctuV&$>sfez@qpJbgYF+Mq&0)C$|MxW$~1hz(>@hKcNOt!sKJw
z&6{4bOw}JH8+wT`zr{IVjPLC;sNp_%+%l6kj5(E6-OE+DHoS4>XH`gKlWIB)=x<Oz
z<m&#|C$GbLEX{^u*YT6i4NrC#(yM!G8a|-PYO5$_vLbCcr1Pl%1x!y5qRLD^3@smQ
zuOk%y32W83T>%TALnh2?c|cqoIH<kBQVMtF!C1<D<jL=;EeZor10lg<#I6KPd!qNz
zs5I&P$H7lKR7CA^fE)IP5zsqkmm|*ZTlNu4JAqob%^Hd)H%<=9OWj^SZT$6fx!jbh
z)in)v@4Ev=fQ><XTcSd&GAsSyurTCnyThw8t6Clm$e+2*|LUOU<COiMT3kh}R;bhm
zHLpEl&FXsksjg^D!HTwB)-?<2Wgy=LN9+hAce{NiMO})|+Au17^AN)}@2N7am*e{}
z@TjGH8%egZQCk7Q<=PUfwI-$Orxz}Y$NF}ei^UQ<m7aj|qkYo#vxA>)G2!}EJ|CRy
zo1f%qz{E&vS3d5z$h^<$eeT$s2;<@o7(prrQU$(56Q}_ljCthWZ`nkpUZ4p@N5b3=
zh=D2qLP5;*L4@PQ<gL~8EfN^Gv(~r~tXfX){`20X!xdg3$_n1wR1pO#-_ShXHB&b6
z8HlNG6odWU13{34h@eRwWa)L*P%QF+9fCk~cW*SGnTkVAT0Ttn?dlo`6#91X^&@}B
zCus$*mG|_bVWf_{U$T1^bF4h5PKG;_2>yI!68-dsJ|fA}a9eE2j=B8aVvkg6(OmVB
z5*jOoysxKf?t{7K>X4V6s?)jsiK>6f%EM)8&OY5m7gHnHIA5U=dT`%k#fDR@mEg*4
zNPc9Z5W8AayLkJiniA0%z5O+QWiLrhT3&Af%&14*4m4A~3=Zlsb(((9H%ubEkaE`#
z$1>5ZYQT(9hBT80w}TZ7AStFRaBD&A@d!BNtnSnFF0L`CK71!aY5(rxEttwR=34EB
zQPmE})shZojqM%&C<wO^t+%8jxzdYHtxaCkUv2ye19$f!5E0U*V%u$d?U#wWKN>!E
zoit*iBl`R?_<`}amtm=UDZ4&uYv!&kLy7MU>2T3<@eD)O`vQ?VV4)1R@LzkFvWbQP
z?}^i>b3A?(3vNpQrO290gl;j5*N(3}X7c5lhOm%-9Rd!zoQU{PJrTMqI>M%3SP4(g
zGTalNZPnpzrchUT-2bLPM5#+;cjSP{Kx#cQ1L<YwF;S1x&6Mp%X0O-iZkb9+6M3pn
zt)ClK5uvL7mOP`FKSubzG)~}Ln({TTX3UQUUXRmy8A7e+4A2$Ss)ar>41CTwo2Lzh
zS6K}0>OkrnHC_9bTSHNfHA$IgHbJ7ik)`XHDjzTxh%|grF!v2r8BitX0W41Ny>z?v
z+SB~fLKuF`YPmNsa;+?J{x*<MUN;Ocu!weBI%Ib-^IP4gk=Ut}`T0zJ-fqe&MyA_C
zuLVnHeLIrRMn>P1h=ySHVcu-Hg)7kKdVIz^$9svYLVUKgU;u`Xpv?W^Gg46w54Xi_
z26ScAnsB+Xj=`h|3B-u7Xx;cp5T`BL5%Kazm}*P+#cSJz8m~)-Zy*g7aw6I{notJO
zY&q2O7za{|uI)1cs*XvMJM<Aya3+ew{7+iS@B6kdw}hdg=f^|gN?l(QOk}jbrOF)2
z{8u7Ws70SW!nm?)H-}8I6(LDm{L1%8%)@Oh55m<5$w#UOD7&zRi?|19=9u2D{_Ks(
zH{$ybE1&fW!koy5a}%d<jxYvMUy6NXFl;l@MQ3{SuyaZX{3!SM%^AYrLiAu~eL7iO
zY7oxN6NR48G->;3l$erxpOt@$lW5A5lO;j`%EGE0KO$JVqJT<@`gWo8{yV(FNuJ(8
zYiM3L&vP7?hm_&N{mwB@L9aJP7%`&Q(VBxuO@g^~w?w0|&?{H<1;uc$9Ovd~#eDS4
zeAy^bkb*%<Y90xuZk+ahn${0&%};>0@IhjM2@Jsp62X0sUPs|u34EffItFJj+B@*6
zb^BgHk{sUe>?u;I>z9XKR!KBwl#9CZRfV50e09W#eCFb9K3pa39Uk^GYsDm<zLadh
zLlll0#$h+IDlDOd-G84#kx1L}-3^2wR1UcijX=lsz}N48IKow(q`_Ov%XhNfJF-32
zhv*^>)Z3V!YDDt&g}B2KH{}zYQZb0xkD=tqUR+6Fk?NoYX8ZD&WMpFA;z2HLAq32r
z6{w60$SSWgkaGc6$Ok}?lJHiy^&}p?`C*<A%!b`NN4Gt=g6sEYs5&u{Z(bKiIDCIz
zu~*_au#^q9NBAE|ECI=0yxHEi|EWEC-9qlAy|@VsK4pVA%!!v#ls_X5lI*)vx~5xz
zisg_pQ0vWyx7{b8)`-C`ZiF7{@#8$57)Nir=<ACsw)60UX??V!kh5r=vUFy%KDwML
z!ID87;6bkIGai@@OUznQMVqgZE8p3ci4pDqYNdQ!-%uyXS1Zg0<o(!<o3`0=ln1kQ
zhjxEG_B*q^1eX9?Z0DAf#RT=8&;M%L(tT!s3nEC(0+Uw{sK-5I7d<6+f{YOyZrj)w
zn~Dlph0!}JaaS?GL0y2^vR3mk8mwZ}lX|USeofD`YkR^X2KAh8q~Cz}d#c4b%4t}C
z7eg2FpPD9-RQFwWUpSML2^;DAMEgg*3?`PBRx8ik*6KQ9J51qh<klHnPg3|%o^G^V
zW{MlW+jkSsn4i%?_1%Y|{8Y{HK%386IZi-fR1PsuM3+Wq%934LYtuX1{wPJVx(R4O
zx-v3OEtH4-QoJX9%G-5zLwpe=$8|iC0=Psgmu3`mXOesJ-Og=UJa!yj_70NiasVt6
z?jKaQCnOjDWuO66+O`w!I=Lpbwf-dPQF#iZI;jcsI{+rk{uktDW<ybmlyx`J1mOFC
zB`e7a%tmdtR6O6#gD=Mvn=84xca~U@ke<vja2?;qdx1MIO!i`Sb?o<b>odeF{*iia
z@6t!TOU|<-{j-EKjG0nUx={dv^walUWUkH`2ZP^9)+r8HEXHJuL2o4saS%CtYKPN0
zDVVOb>g8qb-n}ih+VBO)D7rBo6x)vZ-@@@31=Fs-)5CY!&1ATlp4a2hb-X*f?`}!E
zkO%9`RaKCp<X|qdQc;W%kb4rRjWe=?_yfsevxd2F#odd$;`Z9>RbXBxlI~&F-dB6^
z&Y!NAJj;|p=3gB%>S(MpP@4Si0}Smy5il6h^3!-N@hhzGubiqWcy1QyofPk#hez0a
z<er?k>ZLdFyJo$8Q{@SRCDSD{vl-nNX-G-A3nY7d%w2JZyc3TCn(gFvP2S01TXF{E
zD^jSw6k0235$AFuYDYQJD?^r_R9yy)QkLH;4;haB6W|Bj)}Rp#GW0MmXY&jb$U*Av
z3PDV)-s$yT%k9TeCSriS3Aa`XeG=M&8Z#`X<h$kz;Dn*lp2OP6yRC+QdfM-A#2*IN
zb(-nLWr2j<F1<g!1@8mx;mIQ4jNR+>z(stu!QUI8U{niCg2J8NBOLga#6Z1m0o~S}
zKc0s92}YI8K2s1cVN5>o;C?O_yYqb`TJ&Sw;=nsKrnUB}a*l+T%-!?!k4&i0C!6-$
z{T&<ScGdOwJLZ6D^~?8WM&f-J;BRvIO}`7cm<g5NzxVqu=4gN}`!4NXug8c|vB<OV
zAGxl#yF9VpYz101ewdgk;S*hg5j&Fs@1?)}_rF^woC-fQ<tE7;A-gB30FJ|4<wC`F
zmF^$DANyTikKkJ_b=NZjGlmj0&q!Q!*$#gHr~mQ$>tl>z|M|HR`~QdE+&*8)LJ+UQ
z$Pk!efdDA5G)e5=9hfwzFQ=AJ&oO1hz#2~7n(=!V|M_1G*5OkI1)J{f@alz4Tlm%*
zwQXGfag#7ViMa;<gNp0L^npZ#;!sO^3taHOTnLW?d`qVG3d>HsVmR@!;>>^O&HlS5
zNJ_&YaH_EPu+;B^|LaZJ$@6c%%D~Qx`JNaAzGM=Rs>;j+Dg8Kxs=z)dw|X~K=SQ@l
z1y-l6eCfCBC_gL%m`sM@+6eW}ne;y-@F3#Ra;$Udz4woe&7P8}_>cE2AyUr%XHt8U
zg7rHHv&-;EAu16%;tC)VCtTDQ{Ut$A2haZ&oWuv7OrB3p4nh@!@*5<=8Bh<R_OSlr
z-sksk9<E2Tc8niFZsqfKXM^H~zEv-jGDw7e2W1NF-@D5RL^n;*uZfozs(LPCHl)lN
zFXShaVtjtr@&51H`Yzi8c?;NP@lQ<45VVn_8>(4Chx2%Y&^vpGFvS127zLp@P><aG
z(__)aM=|Z%UWz;CUn%-ti_$JHJ!m4kUGTvqbUBV5{(>$`lZaz<)1kb9_$f4UoyPmN
z-a9vyu<~WW^1YSB;<f}y>f*06r<4X$PLHLR#$un%ZXn)VI$w5Rb0hoCb(}TB&FpGT
zyD{QgM24p0dim8rr(P|b2w;g($arKWp|CwRT)$(dZO$f*W<kOhe?6w;Zkp%0qU<rT
zHFVNL7_OY7$E?`se=}nprh$#sImneCc+zd`Vm}~&-5Od=tyn~sCo%I03$dv%|Lj!4
z%fnMxKBs2oGWt|{X{u=sS%w7+8)!HOeDqc!Y{&Ks;n;BPNouxC*~Yab&{B4jUo^cf
z7c8J0dn)Fs0`OFN-YIP7bn^vx>X}cJ(;Q0N3-^)J31>ABzv|}<PsMbMVdcN<Jc#|V
z*XkR2&_Z&n9qMDr68nxO`huFB<IrW1ldO3MTQRiVQ694!Z`gE*^R>0oyYW|o=n?gM
z;PB>}I}U6PB|U&{F|5@?bl@t@<1NFe;{nzDC-~W8lis4Ij_JTtC(crQcOT!Gtcwu~
zFA~xhvBBw%<zhoAj~G!@BUw!F)u&oj|KhLE6h`N5#Jq#38^<wce67RCPxu%rcP%o+
zJXeJ_5w=%X(YY-q!pANWwOV2%Rg!?<YeTgrUw@QMt2AaZJSGZnSZXLGg>}6~okDEe
zcs`W|gMD1S_Se@2StbUA)M~t^mHi?jwO6L4kvmZD-2O2?+WFzlHq4&IoI`uS9b)}^
z3|RRj?14>d6YRm;Cwjlj-w^vAl1p%!VE{}(L%2&a?=AB!-~&@e9B=)cWoK%Yg6mpR
zRha|UJa8U1@Kc*+CSDHOZVX@C$HT;F#lYRK-`hEQ2xp-UP&Z!x)$rme)Qo~|p0oEq
zEwMsR#fUCdJ$x(lniFAUUHKY_?g5&F7XJRIxwZ@BXkTJF(I%ut!zSo56>m?q+L2s6
z(w+^3+}?oYTAJPE>#_+&7*T}r`oMRMRR7nxc4%oK6^+`p-xS<X4r}FX{u7qLGR|MI
zB^y_H3Mw$m_Bucu_pL#o-QXS%K&2h<jr)~`ZbEdNhJqi^3wgVIJ2U?^1ArE(Y|ap3
z^<3e=g-CM#FNxoca_@VTa`pYf*=4UxV)&rbIMvT2OMDak_(()SfZ}t11TgG6Z5z!s
z_smcFFwqjrOMG<rROpmQK02^tSI+F!EF8|{DYe8M$n3Er&qnFdW>o_f`nSYj$#K_2
zH6p5mPg&Qwlv#9fw~VwX<@dc3!}KAF-L~D(=K&xArPE$;)+B{3zkjOmLV%~5HOMXJ
z_KdJ3F}})(>0Pl;gf0VP%-vV+ZkgF(23pP8o{2C$zW57uL*>pNtn)-rx*FvkXAH!*
z8BQ^tjGv@C-9V{X7`Oy0MkKaY%q7Qe41qsAApgSoIZMBC_$A4*{ODtz3D_RM#2CqO
zdWDPxi4ReZkBOJB{`BC}MSW{tDx@#S<3kgeO7~^l_i!%G*}gW#IiQuvQl|M|-=dgo
zJ|8iJR@#~FAHs0^2RoPXMvn!o#y;0Ij{f^i7&Ae%?_ePIH69<LG&K67cq4novCvPI
zEC%9y$KDu9G^gtm!E)Md+PCpxM&E;Of4|7{X79pH#_t{Ppv?r`wAxwDy_B3r!WoB-
zC^sC?<<!5s%cqAzgn?vH9&~A}WxR?$H9NmtERr?~Yqjk_q3qtffAcxsAFU_Iz3qvM
zL8Zom-)<Z@{0wpF#tbvd;c2DFm7WOrwRFX~7PoHI{XDW4)31)xTL1>kUUgl2)m1N1
ztxD@U0Gd3M2mN<`sZlt0G`OFKiPpchukOt^pWQR88^NsWO;dktA8N>@NnVokluU-@
z-F=D7Z0fO%q6-6EN@Mo8DDvvrGJ5V}`6SyMRE$0Y1!~k`HAXmN3fa`&JxW+xbQe4Z
z`qpt*_Kk;9+FIoviI2AFNxj@r9CWewHu1iLa0O{16nr`F-Gx&m^+Z<7Ys_HMe3j|3
zw`vfMzKD&d;j#Clz9$AWu2+OMe+@oxX<W(09{D?B5QHnR7}OqR-T#2bn=Q(bd83bk
z8?wK?eG{YG+1lZHeGux+RZ)?hc=tf!cDp<s3x2|4bd7dNDv3tUyGL_C2F~ez6Sf^X
zNC&M^64vOD!N#arz}BOG5u;|$LSbEnKh6(JLHX0>Vd<|gT-}&EG=!?`#-sJy?O)P>
zg_YlT0Nd_|<lzNY9^x?aXJ?B&VgSzk+R(@qDIz<uXX#@s>2h3~B<?*v=sg4HN2Cy*
zv?H{FM{5_bwVq1r9sFZSXiR&vK$LSQUIiK=nmIL~Mk6Ql<IrsFBON5>`^}26NDH(l
z4g5mfNiJRYqqh3gva)Y_tobLlTu;emARaUwS><X9Fj|@gBLI%5G;=|=24wku!CVg1
z53%-1auycxDFU@3&4XhPE3I|?e*5|L6290H-i9YEB^!L4ndBI6=v7GqM;7Y-+BRqU
zDCx4@6Gv>n<6{s#`8(fkirYcI@B`F-_=9S?14_#k=q^c|=eN+~2Y@pTXy;>=jFCk<
zmx0~6AGe=_y=&aSoxD9Y5_(HUcnj>?@3%Ogg!Ndy9C3vvnbphf0HSG2y6g&7JaW|S
zz1sI%*7qIE2j^Pwg%6Yx6%L6DaIdk)yyj*MD`x45DW1Yw&Nt@6W9^qE3lg+H!lfsQ
z0<@9BelLRK0ajn?X(w>bjiN0LHc(S11vD{@Dr&4Apg_bKkTsev>aHD=G$?{M4N|!*
z&EsC?gn8M-_RC_?%T(uI!pqM6>1CqmWnFq-eV|v24i)8|sTYi(mOCMUk!vDM!hR}*
z5oIqxv@0(S0Y5Ay#EpTTlURe%OCvXsA3Z=2P&BPQ4XT{mf=vh)w2_$OIn^~3x-fS)
zaSvd-B1e6Z@ZC%wmU8lOLEzrjANR|~kzd7%CNjXhpukTbeer?PIy7Lj?d+5ez|T<&
z)YD=4;$mim3|>0U;6)>Yw_5xHmL&7XlAe{KH<+Cqk-~J8oqQv<IvP91FlqV@a87-W
z`k%=%dXj*ZJlH`eQa<5!Tj;;?0eIFS1#^$1M5qmm4BG6biK2h1H6a)BZ3hZnv=p3i
z2lyIyfKSk#vbylgPE0?yR?2^f4QdYSqdl$DtMdW0(&+oXg1tgW9H9lK92zBq#SQ)t
zM;S^}n+&$ak;nVb77M>FgS$+6lxpzUk2mj^7^xkG5;se;_##s_;7R*HYX~cSrnrK$
z({mR0H4eSQUnWY1AYh1||Av8ki8+M!!+v^*KJ~qNxmXXg9@YV@<A3#O<bLvkd}AR&
zf()SRzHoPE174js>fUs#|HH0)g>+)%*-EEltRP;wSn=)D8a)D+1Ds>ngHAt*@eCH*
zo`D7_i9z~^X^$Ai^ceLl_Uiq|QKa-5&90*+5^I_uz0k+eDB9Iv0~hE-RrI!({w%$7
zkkIU({4#ikc1HZZOC?w9%Q{k6Fz&>ev98E%`-K>-)Y}hQ**nBAEer)-Zq+W|NsL_a
zDVDji7ypp{<}<f<c@o-@F#W-+eK?JV<30#p?r8hzA93ua_pFu9)>DZ-g$VisilA@u
zbBr^~_0n7b*Ihh}-4CzV$b7~>LCUH8hj2li7S6WxH3%29!&J!7Sy0Yty)%K1Zc-HV
zi+Ey16O}^lfvEv!5V?<_bl)_`!6G~#3P>*kuzpWkkL8PK_aJh8ZCHJwJOXw!P3H$3
zUg)0=59htV`atu!{`LMSxb0;cW0*002Lg&p+V;0*w~L~ZhoLej3*yxMB_+~YNiSfr
z-5-|7fKm@w;wtif@on;0O;hHFXm;T-l^Dow(w}5k0rs($J8!(rb@VY;8WHMUJ*U?n
zRp-8x-4FM1DkAs)SYa5Tju<E%+Ej=2Nl{fd5JG4tQUIY{t9he8GTcFWCvnaw80}_C
zi@&he9jgZ}h0;aX8G%6>h#5*X-%Cr3NFn6Mc0hCnqY8MBVu+1#>eb5ca@1j%+l0mS
z=Zwiw^hZTys97d}Wo5NR(SJ*vK8+kYJY(G9B@=<4P*2teSz-K>1+Qd-pi4aco6@q2
zIAw*{?V%{>qb|tS{!*8q5J8Rl^t$z=w!IS4d;kbq*z&Yi3QKN?E&*LEc#mCO9d>oN
zXbPrk(dwG(V0Gz#TAdME-Fww#wi+HL=#5Ru3U-+zS%%xV6Gy(K!qp)z!K;K=86<Nv
z`%OOgeHZ5m?WJPSf~%jgcF$cOx!Z+KGpxI<J^^B)_sSia)f0c1)q@e9EA{O!z1Oa4
z4ojgH@NDZk=yKxI5v(e<00hpx2obrU%WS)lW(u__baq=zaCSzjkH-QbCd<ak=8XbO
zOty%N$$~VIAs4xU%&gv@G@%gQI4U#0c0O!a3NhA0qC$p51vL)Bg#xsWzxe*sV!NcV
zZxrIq{gY!Ntn%|J1qpod>^Vo_{JOEuDrg8RoXR-iR71j99dyI5RSd5`%5r}?Ch#l!
z4B6JvwWkY~9T@k4`r>{<a}z66jP0l$aU;Zse*yrZzd@a?iPTCIq=4ziil?8$>N0Sv
zgWO}8-~jj$iIX_X;)JAyE5r=mM)Ie(twY{CY<0R*^u|kaX!$9vR-1N$Hdfh*%@@5}
z%?aXEFxzz=VtTWceb`W@-VTDMSx8`#0|VKVEEkea-<j^V9tu{czIM3#3X!eZUS>#x
zNwB`8+Tl6NjI}?gWHcR4Y9FEKhBU+iz9ZmFg_wKa6m%~kFFy{+LXci<);Aw0chR)!
zdu$*t4Sj9bf5CTgb_cpAo8oY<_<+pBWc@=F&SCKX*mKDJMUsU2p3?k={K8h@!ppki
zN8l^uuwS9F#rx#@vB$5%^v*y6jfv0E3m+rnE_iZ~KGlWqPsZpK?7x&kWnui+`Byj7
z-W?-dUWYLurU`4A^=>FHA1ndSNxb|t@rMFdiDJeeCIPbRE%LZco#!@h9q9U_^3E4;
zQA7169eW%}TQ3$*wNtzw22;q73mt-_k;EOi5gBXmf8kFOr%~o=no;Gb7wje;bf;j1
zBKp^P1SJce!#3d8aoiaI?i=HWs2~b)@Tx&Z_IXeoYfc=o1=CcBvd&iXfK`9Z^m}5N
zE>8M~k!H=G#|qsuX3(tHOVI)lnImZ9x1h3Pgo>uZ0Eul4*v!Xr=sQ1>7fC7HLfN*L
zIm#2;PTGrK{m7rPl$ZwG1sT206Ay~irZE>&ZKa#`{ZRrLzY&-jFnhs}Q1K%A&880T
z@UEvn&$fxkmL+}A21Ssz!x%Ca@b#}x1pHu_MxSOYjr}wpJ<x~Q$70JX&BC3wE{W5p
zC4Q>qJ2+#mail;@y~Cf{UxY+F@uu3;Z5_CifN&eSe6(Gf?r)+-0Y*iboav7u<}k-W
zTs8wQI9Su6BE^k&%tIO#xwlhog@T=Lfs0Cv{e3c2xEL}88}E!JK)h4vK2FcQ@o>oe
zkVl``*7|$}jB8Z4@D^iB5QCIp^4K)0s;-Pk!B`UsOm<$VigMmsfZ(4Tjqg|zc0!*k
zSSPrFe$gd9wGB3=Y>z!RTOl|%n28nDv2Ri3@S$&%DKH_D;}!tp8`OZch|L~fp_uzI
zZ58rVUz~&AH~<10suUR;&y?xPZSwWSe&`F)Eb$xLj}eQ5kn+?EWfY2cF09m7i7#cM
zu4?@Ls5FWS><&xDWMV~?&11ZdYW7Y)OnKV95U|jyHuTBaKYi$&$_g2BH@{EmG#C&;
zoa^Sp$ZI3W&BD=;^mqfM74z4!D8l4%hiG9oB3Jtb`2!x9CCUJ&ohP@xJ)~s{NP%F4
ztG9N5g67b?!zQ3AeRTCQP=))9HH~qbMj`GWZE_c;5^P$etT36r9t!jpZ>`dER}(MA
zMN44pholpV^KKf7Ul$K@U)r_@e?I8f43m$lD5vF~r+d(WKG0&F8PbQ<{WWck`O($5
zN*)X#fhB5dpszid)mX=f@!<dnI4m{R!ozfmNnXNbJJ%V9n)cE5KpVY;O`V2A@OY|e
zxrX#R=<%z6s2le2AGQmH|BsJHUr#+Pmn7-dg~H<2-_p?NuTc_m_m6_77;?uGrhdaT
znqwyZn0Q6`7@Wj{`3v$e^7v>mti9cdXAhTo`zPgu)311>9X!!a;eWC_Q00Mh9{emZ
z<+m1KM2f)WKieGMQn=%Uz*C-eybYc1n?IlKTrTA#hrYrBMj6IYAgCv6%UAk&d!xNh
zp<0A`h3jXicuY3nZrl=MvEt(txoZ0fhRfQ?`=f%)n^Sxt=o7;ReR3OB$3a(ba<pG5
z2_MDiLbDVyu71O&?$3j_#pK2T$e+`_SDvjXuXe=ZQa)u_tVYg@fJQdl70WzXF2+HX
zq=e2^>u=9iB_qL=zu_n{`iVX$Ct|2=SZC(i{?uxW`y!h^0!NT7yRC4jbOF9$#2ysR
zh<NGaCed=-)p$k=$hXEvu_?#6gz=c{as7*bU?|2gVsacl+4(1Lx1Yrx2nnAv|5qOn
zf1I#_eHMY-2{@5owt;Rv<!`50V}Y{`8_!2)`|97E?JOE8MJ;x#d*z$wP_GPIq+Mfy
z%XL2i%Tr_+6}*qMLHc&Bd<NV0GGXz5J`qRQGJV*xmLvy^I=uVseK6{9()Vu!swLKs
zg5eX)C=BtqtuEZ2sie2)zD|omL`h;W@QqK<sE-c>{P8m}=edE}><en>@qaHnEH!wn
z33v6H2#poUpo~SXzW+J9+vVE#`z|V+S9=7&bmd|rW#nKmR_Y1M?+XH?FXNZ19{HWi
zFnN6!gZm>H3@?RFg7sBP@u28<F<c?O7OqH;!mM5Kpouy#`DlXz^ihQ$qmQ*1eT15B
z*4x{^rRX38j~3UX-xN*IXeuP-+(tDlxb+&Asz~Cqida#eH@EzV&BdWh8zT=Ustq8@
zvLMRxs_96SYf*}pVn{(FXRdKo3SBgK$wTE%wb1pQeJ?$?pgab|$%};K1-WX5sn}@N
ziZZ$K6i!D>F|nXhwcK~um&pXszue>&q<?xy|5`F0Jz?QLo_qvzPZAGT9Gto(-6B?g
zA$0K!n4iJ5hYzPl4x?R%Ya(Hy!*-Abl||$|<7U*MsX(=WE&f`I8JIz3E)-Rnw||rl
zprvkkh7puT5ZK@o`j(M`vNgVqewd%#g$<xOu0>vWOT2m4)%V*qb7&PowzP5iG>9t~
zh-<ES4ieWuB(D6o2AJixk)xE&ARv$rJ(>S4B6di|4C8E+&w_~~p-v^#6=3>(9Ry?h
zfa-~kY-|neryH`LIxCp-!sHjU0=J*Xc}w%NV01SJbT_^xRC}bh!6r5btIsi@Tm(;A
z@12MsU)<4h0$V=zs)Qm9(yoOYL`xstp}2T8y&I#sgGh79A_^|3RJ^7=fSm7DEx0{v
z9I8p3A2xk=xf5fAt}2(gCf=2vI2&5)<Lqf3iW7lBOv<dHxSE9zviksJSECmrySGtr
z<gK8?l4N+9>=BAkIl1AAXKZe$)yX!en;jcQaKZ)U)eo;1V`_$_{($X_xsCJeZegI2
zTcDAu$4!t%7NS&Dt>V~;faZ)uZ%kP}7-|eDb}Kd$MZ#K^K@OOoQ9($PY|!9be`Ed?
zV@xN`IwB+o6lVsO-1-nDv>S<Q2u57tBmWEHs$XfKMDGLQ%3p&KSH}N}xSqV3!R99o
z;`(|BBd(<X6>$v}Zl|a^3gX&ggAvzz{}pkqVVrzqbO6Nl))q!w%aD2bkHu9L#Pv2Y
zh--c+64!O)CjVpG5=}d)39L!wXhh%U^A7<SksyXzqm|9tH*X-Uv=%~L^uYE5V@y%t
zA(Zgm{8x79$~`fTDo$<4se0AUXRv)}`>*Vf!rrCx=?ABUaiP7+ube|rc`?C%xl%mD
z{AK-mcy5n$z&5~c%sN=Nq4dl9#rR~o&UNr&tlL|#ZZ}ddWp12e;%x0|KtBVU?Tqus
z9SY=2=d;lr0NiW6`Du3gF4W42gyvmpqytvI@ZEg>grzn)zX3T4$rW_ztRKtlZ0e_W
zKg$T?4K=|cS8Ih`Te~5C^=*DE&M__GBDrh=sM)A|i^v2?JoH4R%|wS~49{%zB8?oa
zmSY9xg0B&>?BG1U2M%B9Q<W_V(WR|d0K`|pZx6eT`>iNP_c7ogP7;H$?v$7tW0!lM
z#`_x7X#^FDgOQ`TAf;b;p|QT4P5>FXAT=;@Iq@7hM#H8LY(x)Yes&3r+;xJnD|f#7
z`~mL`tthCjq@rkwu^;os$3YM)UpFrvqLC|<q+M;44)CV#%|Lb*@TP}N<GB@HF`Sm=
z3hMfR+1=}~yR2Lnq6>}0o7o(FSmE>GW#)iLhkVC5QjA#^Gy?rTO&y<BS8$<`ps_ym
zL;zpTPj5`(5l_V<%Vl9kGW9}O>s9w&Our~A1zeA7o?UHbn13byK@^uuD|2CO9EBD(
zS3GnP%w{ILy<4Spz=>zxC3er_UOf=Jg<J=FD!2}_i&spmzBZ5yIq+b976z{43rS<b
z^b7Ju;$B0X>(Iivj(TkPSF6zp^?vE3iPI+n?7a9E;-9dPaTgh)(zT=+oWHzw_&HHy
zKQ;FJdl--W1+pA@wP)h?Rj!FjKlg6W9{_wb^2qn8G0x7avQY;0A?QYfdn1i|byAnb
z%(R~jQ%scS0?*OeNZHA2n{)0R1Ljgr{up?Uth)<IqQg>dckgJR&xie2!+DMzOjPA`
zIs}}N(dF^gr%zaP50kk;VO)yTsvZ5VvKBbx#*jDleqpgx)l&wTpPd4S96%MP)Wgd(
zPEk~WyT3zC)W{(N<KF)jQ5}WWZB{g&%As(vZTNRQyvA=-`IHB&-R=A)0|biw+|An!
ztdOB5Z9e4c+c-xtKR}FGPb#170Wl`(G{kIe5QuU7@r(VDU?g64@5IRlEyhSR|IA@*
zd$z8_dm=5$LnDLDs~v0EX8kq)p93JXZn^LL1`_IaUJEObp`Y7{bJJTdTK=D%e`Vx?
z?ye<UX_^bs`alp$iE~kZeI_zV|Laid*VbeY;rfsB8(Xvc<L-HSE-)7#b+$V{-mqVJ
zpFgw0`5$o=Oa;4yC4arTkly%QZ*PB}r>JC~=}1EgTd}QOP}+~^<?**>5w*h+lg3%;
zXBh9i|8guaY~)8o<aLR#voSOHM3YA{5@z<UUhsOC_Rw>|F{CVto~leCfZmk9`+lG%
z8CNj#5zifXL_*w>bhTsST*v2g8c1SUj^OW;+_968z?58A!;JNM*pf1FBFiKAy~FU!
zZfueU<X&QKcj{$HzS>LU$5mzHFa1c|VHxm4GNu36%dh4dNtDDp;SptKJ3CDk)%6X-
z5`5JTvJzuo#V+ac`GY*yk(wMIc}Z3|%&YMCq26N`qRFxM_rz5U(zuupi$$?!u*niw
z2eJ@qkqP!+SF4bZzkvAwMbdg1+rm|nnllzT<?M1qm{*<M*(tSGO^aCbjo4K@PB~_@
zs>7a=z~R*{1uZ2E3(xUvb@%Tan?3F&Cd}y>!6IC5UO>#2O{>N!eQpOl!dlByAiSwX
zU1QiRmP$tM9OhMfN$MGG0CwZ^si{{7l;sYfSHUl@DteYN;<>C&$M{pen0QQd!QX^<
z^2_C&-7H_c)G8E&!dfy^I1Zo>@NB}%X?U4dQhrK_E|o|A0QObNo=eZ}1*@G4F(VK5
zln=nXis*1Xuu3fLYj3(k=oVL|qkdk}`OHNFB|nnBV{&_QXf4aXy!8&2e~$Tpm((%7
z!*+{QAx@p94=xAn!@TMqUXBm_+nfVgms}F@uLhh)tCHN|RYGw%!>25&M0E5h!Ehln
zUgMx9o{P=Z9fEQ;F)h{<ipKnJuRgwywyQ5^hXr3kN_xiECK;}$0rxSlqNj`z`(Y3t
z{Om@j6l+WX3wjm&@<QYeFERY3g`Tra96W7Q`=2q^9cT}X#Ve2t*p=!XVSOsaFkXcP
zvt0=Urm@1hZVzrm31}Gxv0z^nvt#NAAy?}C<4ug5JA+yFqix+M`IpH4C9?mwMK<M;
zoWwR3z+V>kFN^z^#r-Qt_$x^GD@gdS$F)DmCZXd~_bcGJ^gaiU>z*5xzI1I30$Kn;
zs3aievv<nzy#t-KZi8Z?g!o9|Ypumg5KON2T@U%s;<LHkOLi2G%c3mJm%o=3q$R3F
zQb61%{$Ggu4B6BE3R3^dtNzNX{{KF&@@j~cjR2DAS6oAk{N_Oon<9~@i{~`Zmn#b1
zd@tTcW$|eB8bKOYvT7}=J&-FJ|0n7EKK9Eo=2O3thZ4>^2Pp3ED`X}}AY|I#9=Izq
zQ1d@Xy?5fvQzmMOmYA<_JdcXIf?tq!;ltPmS@>efk^jgH@9;W6klgB*uH`Nt$2LZ}
zGmT5d+WF^xsBF{=hYkG)RT#;b^IplhgDD}=2gc+?q&+7dQ@&($J#V12*Y2+^{%ec>
zlH$J%1*RwnmAbzS1-4}Q|5ICBH1!K+sap~C$AghuKR!Pn21xN4p&kIV>)68}k~28f
z9utnIqkc6-j|d9TOin;L{nf@m-4V*VRB7~C-l(iw9||?-)vQjV2XB0-j~ebo6<kRs
z;(EsM*lI-%&Z$?F^x~MhG+GcR`?Guo{oL0-wH;f`uN$oN7hW@Pm$p&>7tm0|rVd^p
zgR2PY<|OKSHKkC^cJ!iGkwwmjdxQ~Z@9kcU*JL0SGc3oq1%xFBog6^hkJx`y#;%Bg
z4pTKgtjy|FG<9TJQ#Fa2Fp6ecldSt0B~GGtcv+JCu&BCo+Fx_&lf5)G>z&qqNL%cs
zQQGp^#byXx=KFF~S+*W}#*Zy+Q-1D}nYehIgaK;YY%=9|n0hlI!CanC^?3RBv*lXX
zohI8ehXD*}tl)T#kH|5jU6@}%rRII`CBM!&=ov7tQ|lczCI}eF|4DN7++kAuC48Q&
zl3L6gRoi_Rg~HxlR{DXd{l|FV&*e3&%03KuL;O=pEqre@=1!x48SXTK!;_80&B#@=
z^*&SKisa4uF(;-R-AU8iGOMlnTauNa@ZK!j8_1<3F;tfJ^Vpe<tNqnSZel9xZXD0@
zeOp5{C`FEyk!$iEpvG*N5caktMH#LNR1;5qCfP2(2TSU9&f+gMPh_xWBxo(lp-oDx
z`NA*T&=I#|r)2=8=Y<2<N<9lvz05^P!^_5S855SIu^&@sFNeDVP95RB*L|0s?Fv3m
zy<L9~!Lv83qG-R8UJ+OpW!}ODA;oy?3AQ$RX_G)a`~o?Z9n5mrlYDkql8$@l1L-iR
z?%#NGO&W6`B&79>$<{+c&-mds=IQ~>?K=o9lIC~@B?G16y|4wzs?jKD^Dlg<V@nyS
zK(Y@mICef{@xtEN+}&Z3gHK6)18z7%!X8s$K8*9Yq)TFX@StQX+(;?QUB|2egwLEq
zIn?2dm)N4n8Dikfc0)H1;^Q_pwU_M#I3(echv0AkK6b~5K?e>fYXoU}3TPQ~s$o7X
zF=NNnD-w0TU7E3CKO|Y7W7YuMlNi6fxj$L{+T8y$n+sD@KJNg6gzJjH&<M;hjtfcP
zFWFsE8zUnE;aJ?aU8e9=2*>MYF;n@u8>zw$uX7;5RMj1K9yQ2Jyyr|PL*|hi?mk<x
zwBD<+cS`BT8p~o;3H~9TU`|hcU;C9Y^C=-)%;U#*gEaTmx}2)X<Pu1F1@~&dFF<1s
z;0;#!3sM}rlz3k6^L-b*d%MX_0m%MqmqTn-*hk{y7no3$^;uNmb!U&icuMx$7;2IH
zC{8<^k)Y?Z%S-GJ)1m37n<+;|0GjywK6LxrK!2`E#+d8=Qun-n-0twtdra$P2uzpZ
z)`#Rje#f5j^P090(ivK`Em)nIi~4Fa`B=f6T@oldw=adUdXBwF#pvemKa_WQ6_q{!
zGjXq)oYJc-YKII#FVejOj;kDmv=Mv+j;l#>ALzDCE#_+PpZt;d4DJxAlAkgfKPZc0
z9AKLaFN*@z$C%Xr`Q?}oss||jr3GzYw$RYBsOzD6++YRO=O$?|C;G4d@FIab;VZAI
zM-A2BzWIoIc^dJgoTTA?oMru&FW!J*W2Mp-M;j;jKMrFpDtIo=taa`<>-kTXh>H=B
zQTc1Jf1T{#3)g=p#!v!2bjC5`5TB{?1Nn1myP0{mBd?bqfA&|@KHPs)q}PwEk$`AY
zZn^);zf(sftE=Y@TNx<n_bGqQ0|m=rt<J^s5_{L)j%l*^m+}pU)EOyO`s$>E8POz%
z?8~w!8yz{Sf31wze}$R!$=k_=oT1Q}P@QxD{<zJm`oH<ymj@1KR3(ik>lhPHawN#Z
zYpa5<{%^F=S{^z$5|UGF!oYC6<b*C6dUX})f3raJv1Ndv;_@;%t@;M^(%2;5|7($d
zm^k>`|EZJonjNf)Af`Ab6l`)zo52h6moQ*1%M>}3ET;&*QBGQA&;LdUQ4E0-{W!iq
z9PPjU8~=Xr?Z>F(c()z?ztfX(R&YJV<AIHTZ$ti>gGN7VKJQr;6@HcF)88D>zgSRW
zD7fRrWF`4~za9Pm^la=mC{utgn5#Ue-1YB19=AxK;)9eWMVy1|ADwNzARaIjy*wo9
z@HWgmIqB<xi^mPsw9Mj5I7A8lte9Toqy^Ts1Lqi1^qUrI`U#m_4eV}>eg_18i{Z#S
zdfC8;6NXUUje(7dl*%!tsvfA;3}yW2&7$0vG_wH?Hyn_oxP!z`SMF=Cxctvn*!QyR
z1+URpAl5V%2dD+(UVqVPA=|Q@gnT1Tf4ese@?{ThQq#<}r>su2ZTQg%enTkQrv<yM
z@6Db$S;+$srC}%s?z!{5=MwKHO5?*-^?-$J`b~W8mhJSrk#JxZ<zLDLPLBu(3z>mQ
zM;`?kOgn>^{&7pKl??!RQ_%m++y(ZL!lYjSwH^Tg?k93y9{QV8jp@QWp_5(ktje-Y
zyL5U6pe{~HdGFyB>~%Ccb{EK^KI+)k@*4x3%IIk4dP;i9saRQm0U%M38C0GB@Te>*
zrgE#7qJ3V1wD+GyVnWO@eUUKE$`aC5#pKa_ym+B8bLjZr%2**sUpfC1Qm0Ad$i;~g
zJ@=Ix*#N$OP*>@u765_O0ye)1wZ7|ZTL6&~ylX8ESVsrz;G=HvT~}|tIS=GeU+Dp3
zN=9rkCmdW?$dL@TjtBAoOrNNzHEv~o`ugz^g^{o;3@U#Gx)yrvZ`eA(_NK%{Z0|>_
zro{9F>_5-=yRoE%D04~=V9^tVdDk4kbaD!(2ju2xfa<lsq+;!V*E35mf(fYNhWqZ3
z3jpwX!e#QaOdjjEZJCzHyu{R)*X0eA+V2T(zva)qgAo5jK_6Z=qMfrT1W4vEpkU$3
zdHQ$9M1%V*y=GmBsd-?{5IidL)RcM!$be20(pyW~_Lj(rs4aBBc)96<rnR}f1uzUo
z(}_KeEsFw*g4*K+eRpNNr2qYqElBnO_acActf0-0&y|69HM3);fY-r%9x!OafcCTk
z5%?f|sVNkmI0CgHk{@E<7miR#%U~DfwMDWgZUcLyZ`VEuJu|v>SRfr9aI|0;_;+p{
zn+Mv9Fd+O%7iMzJhC$zOL^V|K8BJpwhnv0m0$&c~AfFKh%BuyUN#fglCPb6J6R^LP
z^cWgEI*Xj=hs2z7fIal)SC~iacyc^L2`Wt#JHH9KsWebhVE%^}9)YO@<wH4@M|i~|
zm5O4ESJC~DaTJ=wKSZ!{Nqpr;?TrtDTflomTfCC!6NQ+6jRA{=2e4S4HaIxFaa-!B
zLQHM16o~)LH+A75A<;BXihnxd1X&+l^v02T6TFvJt8dFWPGF500a3j>`;O-<Dg@3>
zPg9{{Hw?^HH;1j>gc+ZjDcy3nEoq@6_{E-KlAQF$JXikBP=z<yX;~&WJ5yaemvi*B
zhwxm&DT~hjS&neBHJVojJXJ<%fFNFJIhL8wwK_@ewRxb=wMl<b!3-Pju%Gi>lG^_`
z?)kC$6r)L`)c~4nAM3*yQ3~5wQ?%51(;;u64T^Cu5b#|N)4zM?n=)VmFh>B$!zYMw
zhebl&=X!;Q{R`GE)Dz<6Fs;S}xafS;V>ANN3%@X6usnxHdZHl435?shn(CLAt;EK2
z*@X7DQ#Akj?t<sCY4SnzpJ<HN3F3926y?A9Hq(0rIvT=VpzwmpAla2Y=M}tOusp!{
zB*|#l<k7&#(}uh*ZcoL4G0E33^Bz8o+CK~vx_!<1a?%W#{_o)k|09R|-!0L*jLxda
za5ow4bv<D8sh>uY$pZ$SfXKRp15Og_wMI|x>?C(n>lb6Ur1Ik>piYHjc-43kpwWiG
zY7ldC*#et|AFmI+%>`2${H6Os;ivJJ_&6N@)>Gg-W?mT_jX5w&U6}{q#g7ibf*B(K
z*>f6+*9y3fn+W3XW1~fGA5f|0!>G`uqh*OLdHQaPs)LooV9L!)fJvLKXSBIA=}maD
zYF6#gc&E;aGdPA?==-oN*_g75fVPm<OQ&o3YVCK=4c{Xd=kb@=EO45y6|4oW(3`_A
zZNJH^k{-|ST<L5p5qK6S+S@u`BqJQ{I6xe0a}prmje$)};Pf{kqtmIS@Vwz(ml=NW
zuWl7tD`zVOdpKtRFVZB+bxv-6eZEd`F5S7Q6tQa_vh8%{5A}a{YF%P$-PRW91;UoU
zdwhQj@+YIaC7S>!H=jn$t`XpkA8h*`J$|(2YURosx+>?8tjxVw6_O@#g@p`9xlHo{
zJ?aBUZmf@+<<2a)tphyN3qbO(cndUp!@#O&e5?2jDPT@;&ksB~DX{4__l3<kMAMiT
z%0KC09-AwKIpo1+3!@Fbd1KVtw+xK|i`{ss`Q|7HM)PRcbs#1_g^=d5Q`KeG&We6>
zscrna;PE|CEGaL!QwGo*%yqOAr&6QT-bA~vj*c_bj(jmX>CZrw_H?l7lYqG_)yDT=
zdxbPTyRBJ&4?aPCpmd@2n5ZRSC>dKg$ayp);pqf5@4~uBS=|nbe++dX=7=Xn#<3T;
zPq>wqKL%{Nfo)p$-gD@{+6ZCg$GN3`A0}r6shtm`CVWkufFW-*!OH#9>aL3{qf?yV
zch`CT{1#X1MiHR7i5&S(Z<S^=e+Hrrr-Ufy<dN?m_gXf(dVo7~0c3G<F^PTbc~igx
zHjXevdl47=@M>hqyY;!6VxR@6in#We<JuUP|LLzh_eQ?QEF$tW#lUID4a6+i3n5V8
zXk5|f_N^nIL;+8%7^q@$p6g2#B*bpm#TJ1<eKA7#(rJ@i_0b#7Crru%kxs*5y?`{{
zd3iH8tENil&?d6H#T(1tzc+oqxADFZ$myig^1E#=4hrQ(6Y95VTZt*zhX_zPl{B~)
z-hP`9>{EwTJ$9)w^6N{$(qmu+cxhIyY-=C-U-kSt+tsV&K1X2%9_yyNN!QGEiDw_X
zKqhT-eb)~kshbI|hyZ*Xqp{9~V47UBI#@u+)U>kkmgpcM(}U``g%gFNuZNQna8GuT
zgFfe0d!|wG>n`$L?~Kiif0bG>y8kk?0v^4E^|FKzV|P#gJAi=b?EX{9Za!q*X}&$h
zNS1(O;@9m=-_ERSN0~-D7ZItf?#;CibD2h8JYF2hQT(~8Q)SQ?w0uxSYg1V#^UP@X
zGZW6iLg|&nYI{%SColeK(9f>ghk(N-y(5dC_fEIW1iuN(hwQ`=!EF_zKrbZfu9=pC
zd!1-6&`XYU{9@zitK7Y0S{$e-@}-Ha)?uZQOg7bujbLIAllYlgAAeIV1d%T*p=`=j
z=v53TOnJ4TJ)!QNm+O7Eeq`5PdzC&|={vvP>ve8r*5%aPx}Nj7np^IZDR|LYmwMRK
zZ!H+9@m!FX;8`iI;=7xTteu7?<;y%kcXf8Jlgjh}Yw~_X;z%cZ)I^m|!{#dQS&2C0
z_@dksv#aj8<$8%F2QoOcA6a^*TUWALjPJO4rSg%F%H<Kd2*-}9!XEN;Y)oe-C27(6
z{HUxm-K~tXZ+mF>Oun0|<|+n!_N=Kpb8|H(XNvm>o34j7E2@ZcawvbE>CYM}QPZ~I
zubYX_8ai?H7{L%ru$%>=8!XLC{VsZ;#x2ppu<KA{$?kOZoY<na<Gz<<Hs@Ug795~u
z?tHe|%7P2=;(F8-W{w9Y^FJ~}tNd*AD`y97==gPF$814hZcPt%F`A%=*_6%6@0Bv%
z<|XK^kk-$~pVx?=*n#JAm%qa0pZPHre9F%Q<k?N@2|M<=+^a7cm+R;gSt^#eA9!bD
zvCReQzmF}9@>_1EltncuN%S-|rF^iDouO)vWz)XX>on557?4Nt;(KIi{^sh$ot!Cv
zj*0Hj`oOdKb7$=WRl6OqsX0UFmwE@plh5*D-nd2OjAGB)XsrLK8Lpbg?)h7<dPCQb
z)E9xdE?%+AkkFAD{9@{1de1m2RTl~tmO*VyC2u>PJgbe6V2KzZC)m$?koVg8Ylk2A
z99wk;TjG?)WGlVK<j`&yCGk{kQiG_IBcqCGwH&74hE&=}EfGy8+mbtf{k+1@9qelw
z<kL@<v4<GM5i`T*#s)cDH+Shx3xhhKI%&In_|e^KjE;R?&68Y?>SA`gOKV!*h|PYa
z5c6P<krA~S`1q(7af{cBIBX!WKwh`p!KUiCDcBGXfmF5hrB$i)NWFA+ThL88&hCpR
zi+^zEHmkgHP0y<4>_w!iKh^x@=Pr(alFeH^BQo^GDBYhS_!U^`Ap!Foy{_66AJ)Z?
zL+hanJWuc@?@r)<pE!F+0ZvqHh0fGKsiBSggR9z<2qek|Q$HF1HW0-GVliBJkeFQ?
z-JNThrVnWo_i}2KiB^wievLN4(jLY$1ygyYK%7P^Wz~_hKsmcyvMe?O)YHNF5Dg1@
z=VwI|=u_Q_Wn$)gdd2$P-CNC~&)hvP6gH&U+Xr6l3CWLW9$MjRy3};ZBR*Jdu35p=
zPKBoyP}iGvPQHkSojhH*>N%_`5K)5XqZKSA0I`F!t*%AWiHrp8p_YAyd3&5y`@TRL
z_Hxt)pfbH%$t-Ey(&yf}ehuzVXR3Ow*x<An6mRt9gnFHpvxvS&o5MlDHO-%k;4>B9
zX7`mV)(L*Fs5#SmZ>?Vp>!xmyKBeVPUJ;{PKS5TaZh(N)D93AgemQlzEOrM9;0HMi
zJhk*_QB`N9*8U*UV9%fKxim^Sb@r6mdlYjuM=5H_)?9TE9Ws;H^CSGzbXy5$pD-=U
z)`b36=Ni0rlCxHw>_bE|X@K{|D=ue*uqQTDiaoIm#rYET7{vvvvs1GUSE7s(Rl|2d
zVQ;XI^+jE>F5$j*ink2CNB>F6ZYLenL^lKsT55wyR~Mh2c3=PXk*+Nj&xMcA(|huA
z+bnP&o}C=N@zeN?_w`0!4}vfF32f1>jTdP{<hmerXs+c8CQPM!*?gl2kMizX=FHSj
zsSk_yX29}#I6vXe)3EIVan8T7&o|En>|jxtad^_Zk{*A{g__ida!~@h#SjjCN&1pZ
z__D7l#>`8shcn%|nK<O^%wmphikuSBO`^0>0g^tEp`a__TMKq;$UPqPEB|DacHG0>
ze`}#hNpN%b;&r=bUZM1a`^%udGBaWyR)rwKrEpQyf|cD(VJ(?ia5bXfnAgZ(>%-Pn
zy)20n8B%xnBAd`)e2YqmZt{rC_g4^$KD+}IQjU;b<Rt8v?^||QDDLc(YXT=pL;Cje
zNr}{B1ipk_{tgo3a=ao*weg1Dd~uBkqBC3~cna{#EfyS4Lt5_Bs!AU8fkSD{@NnDl
z4{ltJvSRb?HWR8+!EY2&$7Q5m?M^u3G&z(RTNG-)UeGB+);c2DJ3==3#dc3_wMlPT
zR92eyE%6jb21x_|R0h3&R+%k4_H0DPiN{v1LNAwOlGt(?s0}_vx#kLfT^y_&^p{}N
z5nI><YTTQn!-YMH=~onmZyYsc&pB%#BghztXX(9{fmVZSQ|+wGhYesk=AI2^zKD42
z>jfOvsMIu$WEq!qKt_5Zs08msrvk}l=E?=q$uSux`#tzR{MjoiJJdRZns54hn>s?|
z*(5NpBQjjrpKjP6;GL%|W_r*Ub}T<W!Mi<c@rFJkn)JS1<GW5i7iCp_rnPCV#$?-r
z^?{r6cT4Y_QO~ug0~}Kg0ZR7hR&Ae>nm}jCb)asg?+SGezZ8LDye?y~v<h8u4H3S0
z-nWGBrjs<b`Y$@D&?y#|y#+eEUSNj*Q4B6{DYo)v?~)YHb8unf3Vbi_<VL=(__TXD
zb~$zY29!Je(rCl$0;?<;MXPTK`Z0xeCcYw}FS`W$K6+xFUP{2X$V&_e^RCW<g=2JS
z7Ft+^nvn}!N1C3m8n1k$X#DuR!g#M@O&#E=R~9^nXuVV_o+TxX*cB=bQ6-<L$iytW
zh1#-|{{Dum@HkIB-H@DaCd@rv<_&bNYiN3JRkC8N{QxnC+6C$gc2i~80{CKHkD5})
zZmar<4aDGSBf5Big1++N+P70l@7>6ZD0@P^)s#IOBoju-$!N1Kec&>K-BPi-P<;g)
zq-F5(BetlA9f^Cp6o6mT;g)DHTWSJR#2eBoo&5D10eq8Q-VKhBXdc%bm;l_g2n|B3
zqFI@NRR_b~c?Hd?!de<!fU%$=S;{Zd^(0(f+246dE@JPQpJg8{a*N#){T1cB4o!Kl
z3j_P+wYB<v@21|wEfvnif6ovJo6BRmyY|_(_O*d6!3n~NH@en-OHRNw{va>2i=0m2
z_>1`VHIV+{%-B75ARLVMrA_(xlJqeNlKzh4BgB(qmG&2@(svO}X3F~RN-rGC>@bdC
zbr4PooWGS_+d8`1?w6V<QE8xFuQdWSAKRF?>kX_sr3IN>e|~s1od}@sH$xvw{=L#T
zA4XeW;Y04<Q}B(pJ$7-@*K*Anpk|F#5bwgZE<`#OmL~>?-bP)By^}j1Vbw`Z<WKal
zy{LDcuchg14Y?xNPbe}~y0w{^gJ@Xu$)*ASSb??+RB%kqX>sA*Vn#Y5MLrgC$oR#P
zyeUY@IvliV{N7ovkaCR?=0g#!AayWKF7VoBa>XMGwl|g$%dS^iGo!*a#j#BvF&FV@
z#ar_p+)Z!O_i*c?Po%gT*UKKP*vJz`a>jqt_#IHrzL{oAnmXh7v#;4=p&uYbXt~-D
z>q%ahwEE{RccM3eBjM7eL%S!WX87aG-plP=hXBCNRGXlc(1k-@{-_e;d#6eEx*idC
z9kbYDEk;o{qzRUO2z%dhpNAUFU}5~@ww?G(?Xhpl(`jXGfNFPd<il9Dg}!wl+8q#E
zYB!3iyu0T_n@eQb6US32<vQ0>N%6)R<QqCgPOwGX%4)wYW1vLC>7|Rw-4}vAEkh7<
zVsEgzUdO>*I=Wusu&)*#XD&)LwzoSEtN`PK#me=ZLKa|yk0^#rw2n^9+0H21%xvxu
z*1JX$?OBjnfor*IZLylWTo8zSdmj8{;p5E(_lvC?GFFnDkvV#f0nWcJ&j@L5O3#~)
zWaBI4q{uEFwrC<PvCR0eJ836%VJKpC6PVU!or={SQgD8KwfNKZFi3yHG&%%3O=I&S
z3a!7|*XV1=d2At^b8T6&tyfVAcdH_MC4TkYOX@l{H3NKJay>FkA@*<cdw999)roq7
zodlIKrB@ckBJ3Y8nHn~QLnciT_^W5eeWb%Mi_*2}ga+%Fi_k?+zt&8r%=E&oma(Ax
zjeXrpQ8z=Nj(9Z4V`Fc^<<nW0Xm=~L(bH;8K0W>IXY=6o_Z)CkZj*_7dW75q#)}?X
zG&|UbFV4TlbD0Ax<?n>&Qh2*3z;3f)<$6q8QC_!`9*k4RwtlkEa~@+o=~y9YR6OA?
zB{9w(U=QU);>hkPnr5gM4=Dh4{iEo3U#jn)UT1R25-bqBiO^lrxaiE9QPoh*+J!OB
z!4%3g9{`3%(YF2Z=yF=r!M0B&cipoTQBo=<cAE$ONuabYmL?Hpq(^&3H~G>N(kRdO
zny&)D=Zx8&g|}OqfF7MIU%ghkwN~0}w^wvBBcy8%+ef!Z5MUW@yz;6iWD=@CX6~UK
zGjHGS6j+VQ>S}L$>m^3(@Ws+dKYLpJ!#fAMwL%D=yO=-e(P){zYF~ADUzF|~ZWGhQ
zaHqTBl)iJ~rQ+ox2>cml-e?c0l8auJAkz)u`|6gnukblt53mtC1Y%8J@B)iljzVX(
z9mFDjEgv9IY%*W0oGTiM$nslk{yKLAD{i_6>cw80zJC)~#z=N37`z<b<poL2#<(MP
z3x_pR#-N%s_f2@@<;PdH!EG=W4=CLZZ*@o<#_Xy=`KA+^Zc-BrgR3<tzD_1dKAj)9
z965a1Y<Zx)!M#Qtsa*Bzhe?TLQD>^2)p|sMz1BgVCT73GSTy|$PWC?Xq#Gds<310h
zocuO|5KYiYIJZE7{!DFmFOYy#Ze>=vLjAq=#*!}+{`K{Ta-v<XZs66PdhG0eIYNdP
zNkk9j_ev(`ny>Jcl~x9aN}uoV>L#D$RoA&KeM^`Wmqj0r@a%Mjv@->voS(Mn`_97)
zh2~IUs8#qf{bJ~<^vLS?Wpp_L-rR&Z;z_2-QwO&mUULWbBZ1DKO^V4%ra;jJC%a0I
z!fFKdQ8{CHc=b7|_J&Wjwfj6LsLq!-_I~!{CUf+<i35~xPnlEve~i6#RFrGmH@pSG
zK*Ru4kT8%oNCgH^1Vl<YhLUcSlo~`-R0Nb1kS@s~hh{(&q+v)I38h=Q;XN+hd;8qq
zdw=hG{@Kg57YsAkoaYh0IF6!7WJ7(ER2ISD2J*DU=l26!#e!6x_sF-a81o?!J`XF5
zT=0=;-gn^avCUqpoNJh$a^_luCS;;??^-lduhI6vqabxEZaBsrubEDt&<=CLV*>e2
z0Xn0!Df@In_9Dh*N-uD)P2Id7_Nff6qwr6|oP>t=P8-tdJ^HR7NRfrbeTuv3(}dLN
z+KlfPe6_D4JClaX2AF8M)|Gy2*@pEnbyu%@x8!=Pi}6H26)T7nDK>z0#x@11PZ~RI
zw}zSaPU}#+>#rDPN9}&x*j>Zhfs^c%eVz@B_<4G!s~x%yy3O0OriGMfdDm7!le5!V
zi(l-2cfY0FKLM_#^P+y!X1i0xxCF)9v#{i5$ZrBK@duDj&)@yF1(96L9X9{^57xXc
z83~eoEw*^)XJVUsp~)UBt$i9MX%^IHg)s>H`ml~6F(ajA+(vBw$=T&d$yqDBON+sY
z+W7^-`c3k96Ui3GNa@+KyLADJDxC|%cj?tkE*sMEn!aA}EuvuQn%d+mP&U5Qdd)bv
z&GWm&(Hoyy1oS#EW^>(6vMgBAZMkb|s^W^wThew*v9+plk<=!1I<1v!_&CEx{-Yu%
zNl1_13n}!RYQeue;di!sK{^mcdfXI^d{A%9sp74<vWg*<FYw(&j~~GEJ#QuileoRQ
z4+}<L6fP|@%;-OIb<1PFo+dweVB}J^9ns%SNPBQvX7~=#DFHjBD=u;FHY-Yt&v{1E
z?ir2xo&IrPnno+&Rok15mfSWu`?6vCwh0Oq8eK^LLo?wwrS{=zm3Zvq6(3&1L4Gaw
zmY0~K3HQrZfhZYWd$ay|O2Kc(8uKgh!IwM*rSV=}ehzAv#Mo+<d|V7&Uyo^_)+aEX
zQFfIPkt?=g8WK+=m(pA*Rr+j|?fTknc9`P&Y-p`o+59;e9Qv0}h3zHEG+l$2#uzAC
zqs4Mq3y+f9SF&$vDmJq*{T`>$Ch6_HTAGk)`6bnycxUN?bU>48P?T+HJsL)U8Cpz~
zp=+Y0I$8)uflR$8iP;*cRY$kx1igYp6Fr;4E*`Bh-Lwu7OEI<U&LM#bfk?T^u_vCI
z;P$IvDtGBKQWV6Z)14Wr?&++4Nb8*jg!Tlz{mmNt!pLpbI73!9r}@k)kCE-~1FJ;4
z`3=RRCR57AU80%+;z5u1E}G2rUdUmUd1dmuCNDVkX#KmdghbY+B02x%kVJGPq-4E}
z%hLP>w;yOfs3tQS>}sdq592MBFR*>N+;fk*)FO_FJs4A-H?8Zj@VQ#|*#PHhm>|6S
z@JA)hyN>##@1--MW)UW|iZ&*fejWccGV(=<vb$n$qI@s!Uo~fQ+S)PG51r?b$53Gg
zX7alWZnOO}3!s&;F}2z5r<%lzm<JyZapw;<FK(hp+~{gW--%W9T#FE3AL^Y~x{8bT
z{ENeen9c-WNqDf8SR}P;W-0@#$1frGV{t*UR|i{ux_^CH=--(w1eA-ipFLFBZl&`y
z)`63lPKa*f^6#$3=i2<lywSzZd7}=WE;1>^S{8-f7%j=38qK}nmlwd))AxZCby?@g
z%;c(*Y-l!AsAndQ>l0Q}-`n4v_>^^xJ2>VAu51rlnu?0GfwF61$+jVEx4P#!S}{$W
zB*}*+iu3&`wt)Y%_;&F%Ld@JV+J}$2ZwHm96$U#cGXe|3b4*^w0uUTF{BElzlCNmV
z*J))tzJkuC-aPQlY8kN-3|%7w5NvxUDpDgJ?@6Ifen46F<YST5yblU{x1bpj>)d+N
zuYfx0#6#xOyFz{77nClJ0<qAB;+5tsx0wZB{XoyP?A%(q(2Q-UmopYf<&@FYn6(4?
zzDLgdVE^m7kW*Rt?!Dn_`lBj>MQC~Hvkz}uN5LYoqOPfUNU<}MIJ_S`WT#Buqeo1C
z=aY$e@#EEJrmnH{cIVHdSDt<t2=BtY!ZGZ8H?0_t^%6~L5A;TXIqiOviu&$PiE~ls
z&D>~uJ0^1zdBQ!ni+iOQzT^%W8YmWF(Ie<k1S}-CF9<K5g{JVN(8E@Mdm>g-{)$bC
zx)32?%Cg`7;8|GH8*ilMF97Or@w`L7vdJLDZM;g?l402-jLi4*q?#_%tTAAbr%U0S
zWOuOPg<b0I780k|jZ*4X`C$brj`Gr4-JJlN)E%U(mC<_5twMC})-tn6T0qu{3Uq2(
zlm0CzzW@RQHv>gN@Cy^MEa5jk(u(TShb%o?i?)h@um0@ZHM=;quN3tBb&I86&ig$=
z&`>U~DJ`Y7@h2u|So$uqvKwrY4lhx(qr2N~90vr>@H3~spWSnDz)59n4i<(ZB!cGg
zOY)H;-1zOq_|;CpIObuAw_2`KVH^{w`v|SIUEdlq7dt`eIE6br!TQ`pd*|Wxhb8`z
z<C+hy<*-f`Pj)-mW04N3-XSEsycxf@Gv9kvr(cL!_A*_4&pqZz3w}_bXtw{JHx#Gp
z4X^j9<Cw~-p=ZfxSYrVm*%=X2%8ny`)j-uJGCmGUNs2F_)il^QUosvet6fndr>OmP
z*$0a>jt<dDf^=)w*#p5mEw7EjarKfdU@<Ehz2m=Q@dFLO374Ukny21wDV)awuk|f7
zo9VFQU|Q28_%hbQbnNRtl@oa;R$9Gw9oXnA*s#yLwP97gk*ivnT+LmP#V;2IAXAS(
zz4vec$|44BMG5J0pkd;9lA2XfIjO{5r}g%W^SLq3?|=T?GxSrlozIaLb38zNv|8a7
zcvNBPp()JOKbn=X+6Aw3T!(wP5+0{xY%?9X{))IkjEs#4+1v+>S#=dGiF+PE11?1Y
zu|fq(=q?9>{V7uSQqTUNk?suSU<O$x)-@uRuvZ2xR(18<JxTAx03i<N@)Nx(pCivJ
z{tBdOwAYdTaWGZOM^7~cTAM2cSrrpKpC1yE;?9)?cve778>~*r-Jx>Ld1q)Ev=|c8
zEX^N;M~4L5umdjC?>cP{>)jKBQeakn+~HT;^o1)dLp7s>bqbaj(3%kjQOK#4P9ulZ
z4xnP_Ehy;)GCSl?_Gr+<nA!zx3>2x4@>@mSx%P1KIo;Y;&j_<>t~3(`uvo3ADYY5v
z84bL1IcN@wY_QMbKW^~;y4YSh{$@5;K%T;e{!#usvk>a>f*97&FXAnNl1$StpBesL
zC8LhrIXu(uV#ON#q-Az3*HewEwcWs;5^o|_gq<3_VRIK6dLF)6IL`FM@VN>nA<Cp3
zgSAjrbE%5cS{Js`tCoOX?Y$JW;RIBed6E<GtVQ20{q9YNk>1oY5trF^R1{8hYE%!m
zTG`QwnVE0Z-c{)vWS!BGC?%qX6KT^^aT>A!(ank81qiD&R<A^UaqOyE2B6b<;Tem8
zZ6r&1+xfGfC#*6IqLZRdXkl)*Q#rOTv=+941}cYpHri$23}-A31%<Xsks5O#clh&F
zgI~U$58>hj7iHFp_K}i(He?x$#`l*tqUSwQH+0SMCs)!3jD@0?FTB823_YO`>gU5%
z?!1+(zGcyYR1Vj_g)UwL(6Scub5hc^?)91B=BqVFD5Ibt?-2f3MOwjK^o@Xqx~0Mn
zvlKM|2=AN?w7<U)EO<k%X^O(BTE)<fb_|9x$u|gfcLdOg{_+ip$Dv8Gp707CI<IIM
zj8J*;=h>JgsUDFL^WFCPtgz3z{aSkGzYjT2+2r<*K)7)%=0Wc|ueB#!zEZhvnApti
zzVfKU@@{$!)9z2-EOk4s`R?4rXs0CviHX#{!Qe4rXb||S$A6mEeV2VRvdCxZWfnZ!
z*GX#>wSi1bTlQ#>4!Ol%vwtl9qzX}t{d)8y)PBE1*`4E>=JC%A2UAxQQ1jCur#-n=
z3=~@JYA8#N^vh;I%yMmzAo6MF_Nl45TyrfxIqALv*exVbgb=G8>Z82B&iVneAQJa%
zK=qnf>K#b?F3sI~^K3C)Z#!WyHZm3#P?kf7UQe4kJN-c^k!})EtqcIKMs(hB=0vn)
zR!M8H{4$|G-9H8nxoOlV``cbklg_SorIRVOfebuq!sCdN2^~|;`Rg&`4~s<_-xv&W
z${a@p^d9}KlsQkPO=ea=H-NGjM95#G*>W(D-6)s>CCFK{G)e^*HA6LBz50_QBa>ju
zi*2fk;hOa|AYanSpAf_uM9jMT&jl6Q0+z(7Sbd{Uy+S2L-gU7z%Sd`-ZVV~I`LjfY
zH)cSo^OC{JEhoN%9ZK-Ex8p8jf477XG*Mp>mdXvJW~-OWf8J7L74eskP9}Py<F)iw
z_P;6Bser~n1Qnf`YSRf2mA=PYlci3l&g2dN_J|XvYe=}((J->ipyx%EB<3N|%gbpK
ztq&(dJe9x1WQJ=RzT}F+=|~GqUDC9mawJeP``2f)mV@r^>uI}f<b#^$wb){f<^cg+
zTQ&-w(l`1veVS>)pJ!)wLi28hpp(gmVI8sBNjq6DD<QpnIl4vpl@HI`5iJE<4R8fP
zCLL`;t_sER```W6=)|>>E$>(nJyxB7*|R&MAAgQ(akqoanUKU9HG`S4>aGIOK?Ez#
zrtChIH6yaysu$O{H(NAlt8u7eO6pmuZMV?oLf#8!PxAv!zab_aX7-%a;23iZKofP9
zXb}D6%Z+Y+ZN<<Moj<)e4yYbnMMI)#|2%YqOhkd%+DRk@pa(Ht2<FU<m-ZoR|M`AZ
zRK;;UW5_MKyh!i{KUe6SenHfL*Pf^1MxkpF7K`6?Gw*<gU`^$TV!(8aLIi>;2<`;;
zC6!{MOzv6+<0u4L-7DFrHCvSxy8r&L*52yXXRzgqcT;pT7(N`Awk9fIOT$m@H2<un
zw>y$$)DuYVD(w<U-JAdMFz0Ozzh*3G-g0*)i*O35<+Hv+m_?oZ-Z+nGYLBi5Yg(5m
z9ZZXYz4}5}1N()XL*{K6#j*>OI;Qf2TWyQKKOcK&7<j6%P&gXkA!jC;3pEA8EoBe;
zl|N5ECXl#Oizl!u@r}izjwN7`8N(NPS%q2>vc5QDfhGxa=~R+-l{w|$tUtx$0=4n#
z9O!GD>pzS3$XI~hkgEUDi`|L%I|9Ve-2PNgUHFx9SYO2zX8XXqG+!8C;PP`4&-wN|
zK}Hl1F)1bSAbMy^y;^AVhX^$3<tAZTDG@QHCt7&Z&O+y7F)U4kuAInlqY+wa;#Dbg
zXw07<5>U#BlOSvr#GmV|a3KgdM?C*z&OdqRvSgD2H2U(y55CXS+j1X*qQO1d5&a=b
z`(H^%NB2LHj&Qz~6x;M@x2ZJlwKP1L<Rk`ATDugxrR*r{&8P-@m6kEzDn`y~&fiGS
z#(q60GkRU@yP0U0W^##&F})(`#BV@Tv8#8h&vqc|;F_I$Dd|mvzyD&6;-p9rxMxYn
z4lBDixVPmMe}W@NrO{E0S!h!GtF4*-0V>9`y4(+US;aPFtkM13Ej-bnl*_ddgk48m
zQEo7BM%S81Mytu}gr;n^XguZFRk(AA_5Cq=!rentn5a3UNm1|q>=*ZoLvR)KX^u{V
z7aEhc?g}uuMcbB6o*;E8QyVb9a?2GnQ?G1W7Ql%h`oV4vO>{cR=Q^I$1q~EUpBJ0o
zNNF<aXQVIxyZz)u&LCY3YhzLXZNgwZT{oYQKfT~(TntY5$BPRI83N;l@>dTM=ggp-
zu*ptTxadeFU(vBYmA^-a7bmvDM)(`89I&F@skLrA>$%e&h@qy4$=m9*8?SzBzoRfX
zcfSN6eMltszY~e&w0j&DftzNocQ|RM0w?)pPNJekI`E;cFfOX0(l+gfllXUUlpBdz
zxUTa;=QHhs6WYzwfg^C}tTe=$Tb-V$vncX+sYa~Ddve~Bs0c}tKU`G|`@QM=3)rt-
zh%}=I?mr_`Z+U5=oWze2cE*L3dBD*irW}{&6=L!)=nU3G1R(T+cM_&?^e3-HDuphA
zB6T(8o#87}QJSWNE0W2f)T%34cAkwy=>O<-%hKG7hVY2wcnf|op{!nHP$qi%w5qCA
z2D`=SS8?I1G|?e0q1=<}>Mq^Yvp(}0Af-@5Z52d1%4NhSG2wzpY#ncPCaow4O(uP+
zm3!djhOJ68Fp`2>P$tvh?uhRH)<E|frb1+BoBRU5{v!FhpV4c3L6iR0zC{!Hio-KX
zuB-3{G_|^(-*r3Dr1yT(VI`pc*W|~0V&2@i-t(&!SD}4<&UkvW3ybN}Gy`-#U>{pA
zs~ajR*$Tr_@clV`Ew1Y-iB^hHAW8t2#Rx?hOaJw*=G6k38cE5<GtnA>BBdK%(BiuT
znd)}o7{GO5hi$_7C)l~KtmY5Nw^*b^)T+3r^cO(Q>btnq5xdLxFxWOk14Syu)U6yP
zu}`~{DuoHN{4;XaQ>TJ=`nxEKD_JkHE74_HX8F^{{j{a_O-?oXTrz9aD{<OkGHd8W
zFVZO+luZK!$0be0#w)+wzszlv$iYPJ7;#ST=7`&n*m`v4{2HK<S8%h!^W`VCTWxuP
z$*mJIZJQ4&_3E4|$^>x>LFH!yFev4r>e^{<Z5<z_1F>)ehlh?^I1KAdfzVHrV-==K
zz7!UCVNi_E_&aB~eN<;XK-VQfb^A6`dLvJPXN-vRbnjae!9XUBz7N$P%$*N@-bNTq
zWo(-<ZU7!?c@1i*-ocC?$NJ=dQ-N`9RgC_#3>6k}^c+)C8{P*Pf5w3HPf?MdW#vcW
z#1+p0Tjk|shLdeE>np~=wN{;yD~MF(bm^UaCsvn+GHj0((p{R)sX4k76tjNN77Mvu
ztpcc5mJ}wPo=T1IhAcO6CB07ByFc}n_opiiR69MHw;9PSbzlUal4YgPp_%r3%C9^M
zN86@d@LRg`pHJx@Q43DEAN(Zl36X4Y5YBM?tZ)SAITKP@CmhT7cO0qIP!yc<9|yqt
zuDrJY(XQn4XdhzZCY9H^&#Bp?+Ws#msO~rT23RnJ-&Ne`SYNDOp4m!@deabRTb#DW
zQKm=8AaM_=bc?;aS>DA%e=O;H315<Q#!66_nZMp8*cy0sSmIZ$w!iN%TNP?k_x(y;
zHSQ&A<}sUA^Fe_}sQW8z=L?e)U7Lld!%ryRb0_#mAZmw#SfrY2-kFdV|Ahy^p!c(m
zZ8w{jJb!MP0G+(H95KiKb(D@oSO}3tHMh>Xm8KG=7+?46QB^lVRLZsKGVuyyPa6+~
z@<zBZ#owJy09)Myx}ytD@ByN$XNeU(l)FjdeVx%M9;eIe(>$jK!sl`eBKev=B2ssr
z9{*l-m9L`=Vi*s$E3@0$p4nRtQyW_Q(AB;f38VJ2?>LG2brQ-^4?vgmn(%pS<$^N7
z^A1^I@-HNpY1+bd?01>Ezk$}A=6(EppR<Iq<+85<z0BaO*G8khwyD<a=+NeB1*5aZ
z;|jM;CEvqSB2}2~DYfY1y|`qln}=5KBo2}XV|=^c+^=5IHs$+y6DVuumr~Lv+U<=?
zOqJ-$W^rh=pLt)kE9&A~lLiOfu4W@?lL-)9Dbw@IR^ZUmcwyER+hn5&z&fmoTql-*
zQIH=xAhv#U=6cUy&SH-($I>q{?-|+Z#1E94%5BykNx>s9^DOkyLP~Ht$wh7XSg+j@
zv)FHite18s;`N>cr3?oMUtYMS=*`qVL~VnI)~5F8*)pcB^a15Cs1?WDHa{xaaUFj5
zlDxXtW;uOsB2VFP=ONpSwLY^li&)Rqmo}&5Q^5moelh(?jJD(i3o47YzbLwumrJ`~
zW-9rET2gxLk7?sq&hh=}Xv-A=<3<C0EUZoYq|4%TBgJM82;JUIH^Y{co~ac<{D!>I
z1ou`6-JMNzTX-I3h7jxuAu5ATU^D8KvkM9Y>3nVnlyj7nNu@0&##F;nTMy2>RAc{}
z;h4C?oL0wJn9&&NcBVIx$EKiND%aW1w(mohU8=#={Fnf<-)&y%W83Wm*!$>;o_n3`
zGJ<r$m~R?%v$h7)(mqL^>=n+{*)piP-hyT0hkcH49tvhWwR>i(=RcYYZCXPkF4I)h
z@0Fp^FE+VcYwft#zA+)MFpCED;GK2z$9|0akC0nf==MLBk7~G#ozpPgX#WsuL9kzW
zm&8OGbg3txglC85QMhBEe2SZ;{+d5pz7N-xLp}9*yq-1Iy3k_Sw>HVm@3J2#Cq%o^
zrthf!MvR__XAB+j6q{O8+}Ta^9q<Weyr`krQ9XaJFmVNq?n}C_=tC@U$3yGmI@QK(
zgDMpBiIy_I_-g2?YP(I>nC*(NnexB^`-+5Ra{K0l*{*#)LOL(PE;Mc*=FEKK<{7`t
zKiX_07|1Z1KQmlcD3C5k&NFGp!ksOCjS6HBaw)nasOKTY*?2TVE3!P2IEQbV4&%@}
zKz*V&(eL{NS!n`C$0Z_4yC9G6h&o7l7Rh@w9^5L^(`o^gGoms<7Utoi5Z!J2aNzEv
zF84rP!AUhiNr0&Fy&ZR^cmyVEgDwl@E4Y%ncdop5lOlwt$B5i@TV$~^2y}b<GE8_@
zR!e_IM)0<(Gy=c^H_@;QU|GI$Qg&+g+eFvS@-P=3GW-Y)S1@1Nw73&D!?#%5(*=>#
z#}dgUA9z-1%uBUTK0BpnexxU5SmCB*yUmlOSGT$cy-TAh6iOey?L@uDb;%!(dOjuQ
z7A#TY$9}yz`=p1Bi-d|ROwG_dSuXus>gKR-<;3aj-NN}U()$B4VlN8Z(&Kx)QP8p}
zaA7H*YP3}{&`EgjwKE-9ip*j71yk~qC9Jv4ti^eD)m&YTzt-=5FOX5}XY7P5U2UIj
zqQ(kaR(t`CYJyCS(ex%>m#V2WJi%QD8pvB#%65%(0vww;S`WNdcR)ryZ!tkar1zhr
zGC#PP<bDab(sXyWK6MEci+t%uDOiFoVgmd460nD972NHGEX1_JSo%i_wz@@E5?{_e
zOLIvy$2QXCRj3KG6wl{1X3)1qAK9>N3|~zjNC~l%|D|f&o3-<@Izo?K)s2jQ-qYqu
zjCSrzyVmraYvpH3Uye+LG<k<Iy>d+&gG1A`J}v<#`^sE;mthcF{rE?0MUg~i);q6R
zsKmec3}dVhETa+LbJB&AOt&D)(My_0bVOIAd!{Jy^Res*m=$%{IpQvOKOC;*7cyx3
zsk*wPjx7z)G>tGR2eB)z(SCSI;g{-Irr`~y#;PmhZU)a(-zap53U{+}Ghc`}FZC*=
zGy;NT0|?SL0j6ybn_D}czJ>t4JLvZLAC}=(Ns)l~)=(Us7I<TH2k)C9$i>hzn?3C`
z8)wy!6u@h(<X35%JE;!5?3*gal}jf?TqlyEIxhq~YqQE=%TYHvn=ZU-J{kl5rs|96
zN<}(ZrHu-xy|ir;AIGboTAP~VwE5x<$it<Pc+W|yi_?K8;20-7lc95Cmo*4nv?$A=
z=+3AVA*3Yfxf`zoV~}bSw#6IG7v!czkgQZVtaV9p5w?<BH*YErDG`!#JAG)~@NKs!
zShD(bQnKGA5+`a6`Yvvnw`vGbN=m2rYcZY8`v{D=lS-1B;_ZgOS|<O1vNRjtqYRN@
z{k3L@C)tqfcd}L{qYjCtev>ti*{Zo`N%EGeE`Wh(?sEjBIg9mPrtS>AiK_|(Dj~~!
zE5fld9A&VRdD~vd>%CtpG##3@inz(t*h<&kp0&qgP8DhMGQw%61h*L8u5^hSB>D#A
z*&6lauYJ&7X|U>%b<>6ZivFwm&pU*1i$?15V1a&6(!>uWp_2>Yait8jDH5g2T4&@H
zP;yD%skBGW73);1J73>$o;)CSCBHmT%D*9G&EQ5~g~qX0DWzu}BsZq#)1q~k9nM(Z
zEmb_DD=0KLb-yihJgPQeXRp*Xljynl>bK>kXg`2A6W{le245!(79IZ3+?V0oko4FS
z^*+{-NsfVo?Pnc%585XQS3GCK&RnB#I3RE=tG}F)_KT1c0dx8bzih@f8=Ia}f<a1w
z{p6Zbp&e0PawL~n7to*zIoOiEP?w=-QwM)4l|jNHmdw4D;mft8qilrL;N=Ln$994=
zPW^^!D$sgk*&Rgh&DDD?epy$%aHgWKe(%Cn*sc2{THa$N_6t7-E1KK7&BXlLD=2__
zEjEm2_t8}I=&nEMu(Tzen097KW*0p^Ltywo!6(sG-y$AN+|va-mtm}hH&gNb?bN(m
zvD`UTB#3E>7b7lX?)soMaFGvm$0iSyhAYJMN*6AmjT#dvYG0o*+GZu(@(b0#OUJk6
zIrCzcGjBlmEs;_G9Y9(NZoigwuX~7x<tX<mzRDlumgZWHweym->o<AYKiAGr)~)Mu
z)rFnPtztUdXkD?=otsqfZSQ;_ci`opi4^`!C4HJRgBP!2!s}nwsS>VrW-DylhWYPF
zGxa5lS%<YRycO6G+<QHUaxtgb8`BsrUP|<rXiVfC+-PxWf)bTZJF|3&H2AfcwXQj$
zDX?$X0en0D^RB%3u$akdZQAIC*e5){>_WLypXOsk{ul$zxrfC1ZIC9g_b`YJpD*}w
zyx>+u4-LM%iU~y3Qcdx1VgeN`Pp&KSJ)_;KSmr0*j;A`$yZZ(=2ED@}?iJDj|Guec
zIZLKZ@7-MaZShg5n65cBVSO9k!uH>$gQ+V+gGn|XVw0k;nZ8*r_fM=nMejYKihByG
zz^kSD=Xm=a7VQWE-?n3(HG!~zZf%<UjmK2V?wbQrgG7+Z1SoBZuRYz?OQPS~K2ZL8
z?`h70FS*^w&9!&idVX|krP}>Ivi)DZ#JEb!OIt_9SS%UcXRZw&2s*Pj<~Xe8s^Pk<
zX!6t7>gsNdiND5PHQTV*dUvsO%51Fr+>Jyd!Rs4hclhVN?+@?#?%BRu!8bYL(cx!Q
z5iT%8k3gAm`2C2BcUa7WhND1?<K+TVgJps1cFY4hy!@1CGoV2R+cBq`!Qu3Eo$?_F
zcdu?X^R<qw>De%XsIJzo023%?3c^_H0{u`7A!$o<)4z9KE-rTf*<V$loz@7MB_#AX
zKA&Xxz?IW`@uEvp8CW{657|-->orbic9whx?{53&zusP+>Q&Ox5y9kj-lPf0<=V*H
z6q3{}Qe70Pl42P3>{4OFkUig(Ok@uxs8gNWedcK{?9?;nb?uY7;;U~{#^#(TnzQ}k
z&G}60Cy24SZ#q%=4Ckw|gO`OY2ub%!W-2c!ynk2VX04@!T1C7ZVS!wz)%Rf4$(=<w
zCQZ>`+9rVEp?#2d)(TEgUEwR^$6=>7uQyW9R4OP%WL?^mQZT>6IcqlPxn@)7W>gY$
zwP+(3J9VqPKS?NWny4Xp(JTTQt)gBWW|*TfMNb(sd-A!(p*ADMe`Z=SzmXTO82LKC
zGn7#(jFWg_EK@J)l%4vX@2Cr*bDGgTgmq>A8HO;4(W9%iAHJXZ*V5w6v)?hcZf{@O
z*WFTpbnR!QYMQMk4v1ByMTEIMH~n|en9i{~Q%;|$W*(lK9?7jEIV+6H{>ynVoP?C|
zRioFBBG>QfQ43R|@-_4(5!<YK<%qZAT0rC;pMB$;H`Vo_$8DkoOq6?F^&rjKVGnyw
zkfA6zs5fM#yR#H)YFmb;Bu%~|TITP$lBh_u&7c8QrjkpC^WZ8!WUW2bN7lk(Qvxaj
z{Fjsa!#$e`DD{+2=**<=`7@PIu!!x$k9#*?bDw$j3eoA*hs#CQf-V@o2bscE6D^2Z
z{N84tMp*XFFa4e~==b<Wq;llkVv}1!U3R#oiqsyU%fLtNe9I%?cL;ai=YR6-4kK;}
zFwqQ`t=Xa?`A=SS)%iEO>Gq1dlB#s0aGuSZM%_RvoFsP7TjX?QD~Gcd-mt+er{9-~
z_+2T33WaC?He+a%O{8;8irH8E99at}zx?@jQhW2QKMMU;J%AFamF!^4ge%SUsC@wp
zW*Gsmg}JWARWXu{-L5lir2C8TGUD<JAHj_kJ)=X;JRAKQu5R`>A_Hpat@+s>T`0wb
zRWT-q^Q}5`Goj$2E%@GfH7X|i#@DU)W!M&|_)Aj<9`?!8De{(PJ;67;S1n4fo51Xx
zJE!bIe^&6?Y#r9%Jf^2HW-!_Dfzm`p?Y!vtdxvtwa397rdPUZt_02g60I(_Hg$rNU
zL?ySV9rSvE>&1bd^w9?o8E{TsOv^d)>#7fuZJ#?^4w{(HDsM7(XThS(?bDkTDrPM;
zQ`U>K>!e-sEtR(pO!H3>NRK&MG`&^x;H$9_cK6=t=9QVv!yzp^EQO6bb6)w@bm%&!
z6M|t>QB3&hq2A3uX7yLcUJM8$Cd=dXZo6-c{c-+cd>-c}-yZyps)DWo_gD296N(-|
z&w%L7Zf=&0?5f8k?wzFm6&tqt`~^yHJJ5C&G(_K3CusLw%A=BosFITnUMI0nT8x4C
zPeze<5Cr5hWiPooIdY^?O}%cpSgj`_zwmur$Ew9XZrrizY~NW?Fav6AJ+#0gY81m5
zG^#)QuV{mxT`-Q~T3k5_;f=v2L($3<jJn3ueIA}8h!Bs-Ua&GyRK9XkYf9%C9;y(-
zEnP4o_Lag8nXnf$zfR4j+z>dlm#N<w;`e4V!X_=|)Iq@1Pfe_~wbJjp5coU^wf(b(
z-g|E9tNm~MtouUpYILW98vXf3WA`fs8|}ZO)g~REoSxJqkaslEnTNb%qxeV3?@ddB
zkOJ1%?@>1Lx5=O=`*2Or?}@aH^!{t7nRo>cQTKlKsKRw;ORX&G^dBUohi8DFM@OE@
zTK%Dh3Z=e*0xmJv(_CN6m*U_U3;R833x6gt(PHhToi!l+dn@zFS1*zscM{GSM^?Wg
zC$bLsGDBiLdt;!0-{n$sYodr#SLV~#cAf63KiQq!MY3Tl`Ac@4N8g_{d$e4uI~(J+
zQ_nkj;QGp{(VCe)|Ldsd%=t@5IA}%Iw-W<E$5>b+@cR-iOAP@{6zCaF#61?3jRNm9
zHjLQEN6<5iw78@wrdb%N2ep4~TS+@#r;$;a%&8ffnr4nKZ1avQT^e-_8ADcN_j&&v
zo*+qa7Wn>LT%AQ~v_M?H2#l0yzF*u)+qh`(?XmNh+J7(ZO%lYsY-n?#pV#nKT=ZXc
z3c}|^rQ;rX?~9J6$vM8GWB|TZ4!-ay8J1G2<_wJp1y=uv=#XyUsKVZ<;>lOpzDTZ|
zE>zCB+qYbOfa#HnJinTC8OX0uHu)b#5$~qamin0>l5{?~aPGl#uwncm<6QZYf*qKT
zEK}kO_-@z!BOrP^5DTK$wsCFP92AiD^>0)9_<lyuWhJ|*wA6q?(oWiY%(Adld_2}2
z#L_Z9aJX5*lv+&-&a%fcc8c2^+^33i=#2_+Uuc`k6Vp@Uk~}j`#R8R<H;S`h`T#AP
z_%^`gQMf~Y<EUWlDIp7DOi><M{N;cEl$;?s<NG~etTxAv>D-lGlKbabJ?10<wq5f_
zMGWXunnKUNX=Ob0U}pJXO#f;jaz&q1#PmjOm%acpT#-To1s&4~yeh!zu>ouGe;ti#
z9kR;so5N7THoDw8#whdT`yaLc76;kxUQMh@0BhN`Fki%y7Gko~YdY-(1oe_0sKVOD
zEqFI`<xO6RW>rF1b8Qvd^DtLIoaVndJZ6|<LYL5i<4?eDX7(o;04wk{*@Uj0j~^R1
zg8&RKWu^~Bf`79QZTwbJx#PWk13NwH_*_AULd^4;uB6nyu1f*rgSVHi0^NidBE4Oj
zuGL)}J7;tM>UFeDCZe^^YWdrC0>R#PU!hvU>}g@F`)$is9n8k)0LWa2EB1D`a(Bu>
zTsQz0nmLQ&EWih@hp5LKs52Bc_@!J%G7|{;;76`>#^lZ>09l{vS7M-X`S638Voj@&
zuTI|k0B%dua}T1b9P>03JhTK%pWHWJ5_XuzEw;1#sepRoE#HcRL10V+BRI92WW^4q
z0b;TYlPuh)Y_C5$`$6M<xrB~nN2`z*vcY-LU@39?s|BKx$UaHAnLB;4LLri`9u03g
zipVhw&DcO0Fzi9<YoHD>zW=v8!}DAYcwe9Q9kCQN*JABx795=6&i!;E*U@^bwh|!=
z2LNa3=b1!=YBqEN+*@w2-=2@qk)!FX)!h>)y7|FGFG%#>vA44;N);s6{=JnZhdh8g
z&?8)3%<8)_)%5^W3rk~wBAU9hlxC9z({v^>di-0|><izI<l0YLqFIcL4r$gSg*y8J
zqU}9p7`!ifV=dynZbn+UAqIP>JkJnU+a##?r5T)5tzFB=Kv@KJJk?d;0rD4oZqsFK
zRm8n}Myt(+n}nj%66`B6y^9582!3=%$kE)I2A_Ircy0Lvy0Vte>K#Y{gbTJ|yavD6
z&ZMG5W02{#A7-h66Z=$j(f272CPx+#6vMQd*G#G>H^1yD)rkPn5N|ST3Ze+HcRQ=Z
zD+PUklS?>&;`bS%N49f5?ZtMG*Y;a+8%KT<k8C1FEevLPG`waI?mi2?&)WG&Y)Oom
zT<8L+-s?R_7Y8lDf%R%ZQ*Scbs~D_OeE$weB5g;;;amyqo}8`St4>?kfLILe#7S*I
z3n>#s4BCX4y|;VuB}P(#j5gh%)gQ|nqWAhBG)&b@qhq`sU;KKI`7qL{Db`29ya?uk
zbr^@h2NN+UxMd?3>YfR@&y|=>BxX8Yrr0}B*M`#y5Z<#893HGXY%{)g*kJqq@~y^y
zc=*`;+e4-f9}lteRjze?Qy2a?|B$d=@x!rch-%4VoAHu=-)GA!oRng{y&j;V;YDUJ
z?o!D!q-qsXI_?aK@wJ)HXF`NgYh>K`d-{eq-J$iHpQYCH92L#tIzY^9y3N?^&t#5j
zOZp#aExqcbqR9R2fy2!}rP30ydvCmkW5O_sMz@$bs&Y@#+-HQb78fPA9P{K*Ve49y
zJ4>WuB4Updf{p2>`(jj=9rL1>3;bdNUd~A<u3U0@X;mcawnBQraB|&SMkYfbY{elE
zMJu@r$J?`sTbs#w03ZLI{cZ=<dajJ)P8Zd{t5FR_#SL$z0SIgT+j5F%6VJ%IHFjTl
z`g)voxmZ_vhY$D_uH2!U@28rvD2rDA>C<Dlu}tdDW8+>!DfkJec4gme%HZy5QSL0?
zx8LH*U*+n*|GB5OkMeE0`&HbyxI3P;aT*LRvl~>Mt{}h_%g`2C{5_Kce*Dw^?E`zL
z9n)nXAvsse6bdMl&Jht-(XNrZXqm4`SjCG1qO;46%BJPPBOmG*_b%F1ZuQU8dA++(
zK`FnRqcH~S-%ZT5`kKJuW4xXHz~Aj=doZtCVFyA9Y%gTYS}zrLc+Nn7UrQxqVT%>~
zrSJz$nMC7^9Dc3$!hx0AVK0j)MDOQ65V6h7G)(omYkduA-}V{@xhmIxm?G-bZK9zc
zcVq>(R-G-r5@b5HOyl2>nG^B4obyvF6Oc8nxiqlKZbIe#%TUMgHQ2gu<W6(Gy1}h|
zBdsn5<n5?#iV*qMN)S1R{%yki58E)ChWOgKE2R_Xmmu@d_s4dMt$fU4Jk3AJI0^$8
zENOKfR7+u-r(P<dUXD+`%>5ShM(Cen^UdWUu@5!@*MGcN4m}}S0R#r!Z0yrO%}!S6
zj*z)G`2-Tuqca=ziWoygP^5XzDR-;aC@=z<4X~=^=*={2Onq!|Tx{<v+xo3j9n~^5
zG8UR?69qqBaYi0se8gN@JX-!dv~;1CULw}YJvo?fg(GJinGDo^?jW^DBK?tQIZ}pK
zvbrb0$~Wdue+Dv%gV6*pW%J0#5bCoho7SjR<l2j6c%gtfaEi5ZEq2KAh|BTq2ap?i
zF?M3rlDTU#;CQK&lgLY!mvJ>?%jUcGx7h1HrrNwZ+19oMMvs^X7}G$%_ewS*<ewJj
zpJzFtV^9U+<q)MRIYI&CSr+o0+dXPn_E*atNy`G@acbj$DIGi%+y!qI%8DqtdtSR4
z)Ss|kgsXrIB5Ey%$k)urCp7|c#HgiqL72i|`mOk3{O7~nr+JL;AVCuv3aidU`3I$-
z+2WBb2%zI{um(|8wLGHj>Nt4&5Vtk2!}da56vQ&If^L`w)_pMUFt^;hw;50FSzqA?
zZHb=I_QF#&&XO<lv~67AY>*N(3`)tYOTi)I8k!(oo|V0%7>MG`%z1JK7&@-3Ard_O
zMD6-awF!)Rx(X&fs>^ph=BYlLY{!4O6>E~^;Kg=1(aA?8>@moaq})ZPJ_#0!{a45H
zZ^zdEutHbu?I%Cmo0vpS#b^x}*%DrrbRM&u$aIG68O&xGJjk`viYA*m3ts$Y=I3A)
z`&<s^K&m3`cIy3_=MjKg36&Nfw@z3l%7e(c@+%lcBr^QZ-pFgeAa8!(z!b`Lt7Rc-
z%}zEyH&qKkix-zdS*}}XW&kL`TFkak0k%!WJJm8MCKfUQvYwZfJ@-`F4c_dv4n$WG
zRST)kdNmmzd}bULpLFl}i-eFL@{x%*=#H^)$ONZdl)M#uj)l+s4Nr1N{Jsdkpdi;T
zBF3eATbX-hEoH4YwvMa~ILt+0rkC0W;MT9{ZIzYJn=x&DZ6)MJih7pXso-aEHpB+_
zreUb<vj}Xgnc04I=b7GfDEHNnXB4L*^P|DX#`9oEJb3DQernH95XjTpR(^iXH}gwT
zIGSoM9;OgKissY(4kMFB(Kfi1)0Fq}y_Ilp6$DdnN^3ic=)(lFc0BQh{YZ}d$XJwJ
z7>sRm#W^KwVD+-S_A2O-?O+(YDx=%PIqVdS3%JF%rTih`Og^7PaK37$rc!qndOULx
ztwGw%Ho`^s4k<%#WN7|oUH;=@{C{r&wERa7-@ce*wy`*Bnw}u8a&pp9nvU8YrdZTN
zbmFq=$qxPOcd8Nc&!sY#`Bg_PQW+0Y?U@$%<ad(qh8p)|>*5;nJ706a?44vYqD`{%
zY+s(^q;hLE{E9lrO)s?Kh_BGy49zbHB@SjqV|q?>NlEPP(I$%{^`;Hi%%X{<L~8B6
zxhikcGk5ngZmK}`Tr9nPUFGJa)7>yGc901zpN-<wxXg=jn(Dk(EkI|k8YjMQiB-SS
z!?g0Dykh2xP_58fG5Tuw_|t$;qq=|yIVmU}gA>(n=M%X!bHtbXBG@-w<Vc@-$gWIv
zjY7Awl6ADW!eLyY=dTOVM!KYZ?{A6${LkYp1XJuDHq$08VW&lA>AoaZUlF2tPriOV
zHM8R}^Pu?qY~8ZRxah~x+w(^$pA-IKH*04f9czjh&a~Hcp<wI4*U>rBWSTR-C8xe2
z+mRwKIj)oZf8C;gbKn2CS>j`4=Yb3i$Ae>a1X_YGj*5JmQa+|nLcztRVGGV=8ClsS
zOZ5VCrS{1TMBRvW>u-xYPWkHe4f`asSe$-|eX_ZFw73cU)dX);rOK~%m&dv{N8C4J
zbI@x~myg%aP7Fp30UaM<_JlE`rz2AoX8ic;m%#v-ch;9N!X&EZ%%P@->}e@+?T-@;
zs(I)PRPY$j@%s_lJwuGm$B@p`&$RWSY4XFEdp>An{+wB;+0NEHJ{$&IJ;L#pChXIP
zzrQ3Tmp`&@TDV|2loziOKiP?{WO14MMo3d(J=t4e(n#yT-FzD!lT`YxXNM!0`6<;q
zj))4#hVDkUTTKp%nYH6jRvJxyv+L>4+X-P;skPI-P8=w4RCqa_+0&$gj4aA;u+sH;
z<U#7o$9#vueynM-D~$3tVe@|<?|;JmuknEg^_dMJeQ+gsUyNj7NtEQxZ;BSy;ER&^
z!1L&*!DM@)uRd=xBQiJZ3k<blgM%Lq+++iEVAgr5*HaI57(}P3q&9#wdD!JBIzGae
z@3A!R;SJhm0*_Jc5kOX7&R@>Ys<XWMaw!rQzd8-Ibub;jWiJOdnf#nZ7z3rqLryxb
zqz-zE+K79S`AmgYJuO#??*vlN+)j3;(>cK*&Z5fnersjE7F6oGK51dxXwxSyN0~%t
zL_TFst7;2HqfPzXtZM1d%kwYEC#X5qo7A5bc1{lT71=StxSGyVC#%V}cpoitp*udA
zW391Q%wj!psr4Xmx&&jlo|qA{&F^)tFAPhj-l0(a#xvK8Lwqz7d(I>@vvn-tI_IIQ
z!|I=G=1LkKTE#0;3UuJjKsHeqz+?F7V3BU;zrD>rUGjfD&8JC@Q_R4Z>8NGXSMab|
z;N*uZbRT{~OHr)gL_KkuvbcSSF`B2L$o%=S=F17*QkCB4A=Smy&#06-S`5}Stw!=&
zjUctrzMJ@+)hH|atj@m<G#^#XHI#IPu>s8mYC@x3a5P5Xi{Vq@CKJ+3%^-*e1H#wK
zYeE3L%;br`<J5WnBkFutCG&~qZvsvG+a116HagO=Q63?rwU0GNndQ-t35AmfPb?dg
zk{@}qING?y-%%9@JyZvOTMhQ%I~pPX9+2I$MUK{LIlkhX^+Rjr8NcjIS8KG<=yh_#
zW~gWJ%&x>hB0Ra8i;FiUSPYbxG{vBZ>Lr(*wDTb4E0j7-oANB0<IXai8cevU2e&UR
zqSk7<J1b!y$_?8x`NJa+Vt!=B2m8^6O9rzbw+V5Y?tW?7*N!jqqvc*ieX=TpJ7Vg_
z8%?X|)4Dp>Z$;9{*KzLy^W~kv5QQ7v85&n^fKjKJpT8rBe8fUrJ5IUKQmuWjzu#+b
zhrox*ry~20^Z%bSH2++dkEdRk8wBd9rv|x!(pY}To4jW~N8}CZ)H=(qbhWrkH>da3
zhp8q9D#{MMXg|VR{fazfqEl8#{w=!uxO*nK(UCJepKPBOc3PD1l?9Sp&h!;>W*S*v
zpv+{`Db6v!`US{nYp3U7Ei=NAG9M)UXwL{$D#nUldRITo+-$uJcDp(ojg)CVdVf>(
zMefM&QlyFRC3!!aKTpV{rko=T(QUb0<5Y2<9OR8wuo<fI#;6onya^?@qGS@);Tx(A
z!|ZHsY*^x0DaQyyT--ury5W~~XU^T{UYXA~b+#Aw`f$?xaeyfgVJOSs%;Rq@iCTVq
zTY<cdZ_vVnrEtX}_p&y-^ZozZQU1UFpEucg#$KZ<$ts(2Q3CR&&5<nh0@iP$g<T?#
z2u4;&>`g4ABRwx=NVW>QE*fh=RE{u|;SHRimX7SYz1NwA?lJDEpK?;3eMG5DIKn9;
zuO5HvnepcGq-lxgmh;zKBj1gDyFt-l{>h<gAJY=AJ<Qh%_o71&2*Z$Pr%&Z1b)bMH
z*1)`{>rL|JK;n0VXr@-a>7{A;9C0puaIq><mR?$VGe`If+ayOb?%pRKB@i)lJ@O2@
zVvL9~#Pg-KOoQqtmGaFv@^*9ax8x5d2kWH#=}K^k|Lh+Bb1wetzaAXE&7D){dc6Cj
ztT?IT<%(rxUYoF0kL~pZ0_4XdgHKHx+b84s8vQB9*p(7)f3b+=@rY+Wz-Qgtw!h2#
zWb;KQD9p(%nlJyaLbf<F2CO_UV{>0b#XytK4`cXZ-|d1Jkb7~iY~D1@t`2OThiN$8
zy}p|)8&2E&D41as%1eQ-@yptgpDkeg*rn-v&N@+6|Eom&7l{@PC{{9_$T)Ae#Y>FF
z`!X(m>uQSNO?F)xi<xc8HaBS9T$vsPFIG5}=sI(LYJJ<oNE7pqPY;c~Q-~I-2Nwsg
zqlo`M;PwCe_IYa_hJiz}JJ-urpHtG0Aorj2z|lu;IJ?10F0&GFwZwO3Xtcgx0_yec
zW!=(8ElgRV{8rs}{OR~_=aX4zReF?kW@_<jhYy$kzXI}~Q!M@oCbhr3Zt)}LL0jC-
zMZ_=B|Jbyo-C8h$8!7dW5Gv=R26cuO&N?Cls;>OhC80SeeDhqFOm6_z$s!$aHpUZ4
z6MVMcH2LE0$U2)IT6BCu^S}Hy{a`+`@$st>NAZB#8pG-wy`rRjymT{(TFwn2?8Aju
zdY_Z`%S8*R!A!4xyky}q;~)(U&B!k0=)He`wu4Kgzsw~ZoSTFDSE;Y)=|u8boafdq
zcyl+DLnc<#)42Kx*_cj=!+2OIr^ekv%dU??g8u*Q6!Wm=7=Aw7?&a71x-H}<nQ6;=
zc4Qmidq;xIcc$C2|Kr0D$R3d^zQ&Wo=Y9+MKohgK<6_ZQkmkC(HhTk4;UU@U4Bz5#
z=beQWICpv=QhuoM?EMXghS6~S%LwGznXQY4x5pCcU8e=vi|K2DMX(2i*#s5a$-Qe|
z2@u352l4p+CW12Q)4l?WlifGwD%SK|;YVsLWiR}XxAM;i@qSHV{jFFj2CBKG>UWsV
z<UIrVLyALiU@yg`T7S)31d~()!odu97yQeH9=l_)FQ)pW(K`5A6kS0KZQS`mDbI!|
z&@l-!N|g{Br4#A)<)1$cY9wKn&#4gAtiCcAEF3_^?o6$dv^RIRH<FF(gG}i%xzV^6
z3f|Dqs05NztjrQM_^z{^NY*WL=DGjz>6=u`wm9#mNPbzkmDG9M|Lz<4pLc=h@>T2J
ztCL^zQ_qS#YTN&z96IWRJ!gqSisHN~(|n7u9*Zw;h^H1XT{C<NZ}UR=$+pA){yy{Q
z{zv42^XGaZYye=9Jp(1M))X$km~8cDSrz-I(kK`*Zp*>0Q{*(Q-p*dK48@!F-4M2O
zFyRNM<2o9ug$nIi5aRBIsRL7L<7E%Q{PzXEuq4e~!^*PWzb6v^``3{6=}Q81CkIAg
zM9qSShgya@cRmKJuO<skX)cenwIh$5y?)PaZH{}BYIE~#ACy<gkjB$|s~P?b65CHb
zbVh_({@YC32(?hKzv<K=)o&#31>ATFl9x(YxD!9us?-c73CTRhUydzN@z_Q@(xKY4
zGl}3aNpfGCOXx75;nKXS*&#QJhAISN={=+>bn_tMY)mHF;uE|jssGzS`EuC2Zx87g
z)iX8kD|i4e9JRa%t)Ab^Gvyg;MA3>235A)>$AdREfoBN^qsYb2)Wy1gUQ7^9O~2|7
ztU~N}CxIwJvaDZASYowiCCfd7K%2UG63%J@JIvlv@o^H{*E|HM?{xsX^hbbE;qltt
z-LhMvI?|_^IxGaBaWd37gaQi{i|)(_KFiKKi(}2xF4=d~p$PBUPsOcQt{K1}Y`WX3
zlzP15aRP8|Y(e2abpBgp+oOR3*>|Jp)K1ZPZXiG+fHyf6B<wP`hckFf;|-Yn9Wa4h
z3NDduMXU^wf%AFx3ITtHyx*}#?t<Ai-~C$9SQ-uJT3rpRl<2AQG&=-?0c|#Php3Ho
zAyj^!zxkZZkx_g{ED0Hbxn2MB(D_GIushUbZ+AzogyyD-!oLvZp9E8UXyR0AxQ_ix
zQ1X76$yQ2y-V`gQkL(vM8tXJs7SA_#neJ|OScXcKWtLd9W}W}!-w+L8orn+Xs-~4x
zOMNh_d;e`%q~rD%o}#>08RJtG6SBwo*S|meKRU?2NUAZma~A*10w5z02uUW5VLUc{
zP%q~5W3>zKwp<^6J#-*Xu!dw)sh{kW4l~ju-%wAz9>r@Gz-QjZTr>XTuYbRH?@u&v
zVjtGEtm=@i@{8ALeWaZE?&7WD??mFhF8m*V`Og=0lMGBkYH*7EQkzMsM3@0>he1>)
zw7lV@X*u?7z^bCCli|?Q8Q<klOIJ0%^6Be8F>T-eAL@_Ieik!C<_LF}x!55gQDf@?
zOfF)^oO;xryiD$mVn=O%6Zr?r{ExR#O{R|3;|6@-EjX(qo|7G>ej~QKE`hkAkdG`&
zQmMZtVEe7tc&<Kz*DS21vHH)OUbs@l`)E5*={jnrBQ;P2a-5z#FLdJv7R##qu9ye{
zRrED}y!OAQ^n;h<Cik4QKfu)X_KqCsKt>*Tj;#Wm?Z<oZUU=0kZ8e<r*B>QSNKAnE
z7-OUU=ko?B!si`1qte`+sOA2iCD9wTefhyu@|=)2DouFx+uydAKkxI`ziyHtkV*-R
z@Tus_)Gbrz(8!)qtCMovkBFdLVT!B>Ymay=ph=Q&*GAKf{?z}>B@i_gV;0U{HFzun
zu(qCHNIMuS-VKuJJ!LPsX`p!bsIcIFErP`#c%NidxRa$_X#DvVP0Vu`UmupKUD%M{
zeAkgI?6NKgw6UwDUnbfUM`0({V!F|vWev<XkUtvmu)t*jpnVyvUd2p`mg%J&-~ZKL
z1t#(?1h@)1M4q5`#<<+s^uaJZ$~^ngyGRzbCgS7|Q?oI{QM%TlNW-DZ=3};{?+{&T
zu(P>xiP-|wnq%*wUJ5~$8FrWh_yZxyqXeeQUIaiQX7iDb(#hr^V_CqlZg0i1Dl(!!
z|EYnQW%eiR9EpjRLl@bGLXha#zVYa1973ilXY1TZyBGfi>JO0)-S)xbyk;$Y_zNw<
z^CEchN2px#J3CMxChWVmS*~B<=JX4xI)}}`pW~BF*-dGai`d-u1X3<?kn<eXs>QK1
z_!b}!csx+C1~&jT{R7+ox?{@w!G1bEb7nZ!i5eX@9LB!DkmM1y@OFO|*|0B=`c0sG
z%0n=%niZxj$e%JLu=9b%tmS>SjRz;y+3N-Yu0S6majUW{*<)kT9%>QU{!%AI4IBm!
z>D^Ea6rpZ?M;S#lX|L$bd_7u7$TN{eL=-wvlhoRq=g-}1cSnq4q_eA?gMi+q;A37b
zhk}+#0zM6+8O`*cvns><=BwFHm<_24x9&7fd~L2@(#P;y!$^IwWGX3zD&IY+*wp@4
z<evZt=Us%v<43+!9D4SK{T;yEoAN776OIVBVV1_4c`Q5G+Q&U9OWO%WcKVb3WkusW
zeNx=uVLpi;+|P11<PBJ}w4oCa?XhYBEC&d5$p|5*^s~awO%4jWHz4aEMSM^U%H7ml
z``aCIq(-p1qil=BQhIya)Blm@|L%{yhe#sNNs%hPPk!+I<060z#k&`Z^L67^p!Spv
zWJDPQ9XdwQCqDarWz%G48jHUS!#}<A`|Lai%3a4kKgJxl_QUu`iFsiF01nJrL}Ngy
zKq!7_q__jXEW@g*5#58NE0oJj)BUgX_MkhqEt88Tr?%*hyC>=MXnE8ppU<jW&a^dV
zHq(&d-#5ryO_sNKB)b)1-t9C=HqUOsi^%{Cp+nRM4a&_}m6aQ7(0}OFP4T*nw?vyj
zQ-W#Tj9ih*pDCuP2qQlp%AwX^Q)DjC0v%&Lz_X^CGxQIPNZkwibMDv3#qAALxa+o>
za5`~tHNQ^UC#79rNB3Y5@8yNBBpS^uBB(Ufv#c|Nx8EK?ask6LYTcAQmefCq??3;=
z=nai|a`K)-T2CxAP425^j-yZ(+?Az@ToBTCGm!0rkKTa+j=X?|Q$30D49`Vw6bi3F
zKPv%ceDkh&k6dgmXBO0)F~gQU+0kK)%RyngRLf&;mQ^>A&Q4Ysf;Gx!rXu;k>Z51@
z)U=}Zrk_uN+sf2NVPV|_52}}q8%2E6lWR*qSuJNVso(7u&56=v=SCf4=id43M?8md
zY7FCE64JxQic)H5Z#JI9<5d5Ah>O|c>M_+F`AR7eIvPcUg=5jt)01boItV`I1Olo4
zlk<}f%W!E35cY(y<88}-MHhhAo}Js1Ll-;T!Ml`{Y&NuDnbG=l(*qA}{gO`hjx3e>
znx)ZJ7^A_6lW5sB!%r^?mIEnu0(0w94y?Xg0ySm-P2$&8=d?Nt8LE4Ct=bbLp9jdd
zf(-t&{SAYvzwi#J2WX#(v}EnWQ+!iZP7QdCn%tZI|ACo-v4P_DtPB-yLtk>U51icR
z0LxoRr!bR9-G7JbaM1FpzhcH9G4Wf>b&#rxIF%-xaf}qOnOaM>2H>ZeTI<<inng=2
z{p6Liw^J$F?4jzdhvw^w<SXiH#1=q1bP(9z{FNu#gLk#pZKqEwb>9}KGi*Z5(zKpF
zk<T5qSRV{UmLNGynz{`IO{f|}RkC73`vASAjG*|1dHL-yT!I-u`HK8kPJhae{j@JW
z@vehQ|6CTj$hrAV;u#BqkJm?9!lZ-KCOdk$=<)8QpWF<>&U|N_-miGasCD>+%c(P%
zp>_8l_%_e~W#;Kt<qsFNMr*Vvy92WWaw`Nei1du~{{Wr(rQy`wqZ>fpUPKjtnJ+f{
z{3>Fyysyy8Eaf1M66yv%i_Q;k>oG}q&TlD4yAfb3h-i1zow4k-aV;a}lxA0_4(ub0
z9B$G5Q{?|SIm|;GIRNO$v_2j#QruH2HEWH*tW<LZFrql%Bjo}sQPnX6_MPyeK@2f}
zwk)B;HrZ@pAxq$<0NtlyE}<25$Vnn<Abfk}I7#>nKM@pj;||}|x|DG$>fXe<$<F)G
z;D*6B=Nc`wm+J+QQqQScK*(`|8GHad@&tTuC=6{D(RAqX9Bl{<UdE<Wtb#s6m&a|{
z94jR7rwTct!W`5(L{~m6pqnY`JGEq9%z|1<8ZSW{GvPpOZ8D^{Hb8Dc3M<#u|HIy!
zM^pVpf8!;jq!bN=Zc654N`!7v$s7_9Dw&6j86sCj8kH$yW|52;%Unt3c}`@WD^rI0
z?V}9$@>$>S`aNqs|2+NE(!KZnKIiPS_da`g?R~a$X|(WTPVMfsNEdBx>M^sI2d8_M
z$Z<?nGioYEH92xaOXV+Ow&^@4S&!ydY64W!rp;hCNb+GnPO4s*92mQ5Etv>@S5pIc
zbp;Ftaj&Yq`d;T`cLBD>15#|$$}`5Nc@j>`m5mS`>+p<*Sj<%yzuilQ`?3fo=5_8^
zUv<|#_P9~QDQIIZ<$u_Cw*S^SgnkRvtZaJ9r75bKWtXD#LOxIVC+8mMxo-=u$z1=q
z;ia}|xDKP{r5wOgVl0PS8b?%g--8Ym=OxwOu%jVt2HjLq@3W^*fAnO|i_Pc8US23I
z^AM$f_A~UnCJXgJ^&>Tx9B25C^aFBYJO;8zrOuxFPH&F4TTWQHdBSRk<sE{$W!Igz
zN$<p}rO8%B)qIb+-h9No^RDfqTBm)SwDNJduIw)8K0`%u&&(Sx#VF9OG^-P=H_hnU
zQJK3-w^GR8s!-LmzM5H`XU<G#ZOTl{xt9m}6d(B7r&^UoknWF=OBsm$9@|sufBv)0
zSBiRR@g^3B7!8Yy6w}vt%ss;%V<S7qNZ(K};w;}$^yYbPP$#|G@H_&RFg1wVMBaP;
zEp0`I+L+<_z<#!tP|bRSHIEM(n~|3#&J2Esek)r``@gCoXV+n7VmxMog070o4c&1R
zoOjSqHV2qBwDI-{k_o-(U-!1Ne@eVk_%?3$@UhMTa&m_=At%>8;u~e*J*gSmTeCj9
z3z**n{D3LtWx-c>EutrOq-kld6fd>fe0^wLoxF+fq<2)_&GUd*?=6xDt5)u_NR-_&
zpa%Z)n82-H)?`Y2$z@PT@YZ@@J@NG3;|b`b%^4I#sSjV1mrBtJd7HgJvS(zX)KCI#
z3kkZNA-J%A><YS^z#vR<w*R$mh-COW`>BhRlESu=R1F?)uM`Kan(T1Dgo6YB^Nnx)
z_QA5pB2+Kub51`VpME@129-%UAK?arK1jsY$6vM*ii1<lW<uX!4p~>2-Evo_Kka)g
z%m}?@LX$sff29GGK*g6J2yIECUL)?{=UsN#8_I<Zh>&=~*xhvsLc;;Ijsw$AvA0}b
z_s;d+cP*i&(1V5zQL~$CIYedr_`bc2B3iN!5{VhdH<^q~L$vMyi{|r?zzEQDNjBdu
z>v|2?CF_)SLc1OA8E2B7k+vL72#B=%R!MHD*c5PU{g%UpnR7jG8_(0>);IP;nE+ck
znY@F9opv@1ccd(|yl8qY_Jui|O}-_<W!ClT77wL^gd0`6GKcFlDR`2~f5xq!OXI0R
z9t>GzV9N}+OvB&0hL^O=vcEpMJCW7`B6y2qVehe~og69^@uKo~O#P-H1anW8Jh*RI
z%nN4p3G#|ufSG!^p}TS?J;JWFT>><N51>qP<29nTRo6^GAZENdYP({TEdSc$O$TZ`
zFVG5yO?IfAda!FR91LTxCXC|MGpcy$ofu3tAK1`pYcW+g#8g5{Lw6i+CG`^F7)XF{
z!*A2LuYC9jJq3nhc?PCATEO_tu~Q2v|57F$n0XNl`8Um`(YOiPTx1bSfSN@CwBxsa
zwgwp&*lJjlr*DCqb|5<K309Bo_{Av78MrRzZGX#QK8wf^K5Ym_J_S=foqsaszNOY<
z|3D3QaDoQUAlFM_GrPblaRdNgb`$f3mG#%Y%-UQR902aKeJx3IMJ}C8<4uU!$ngeN
z8Gs2NG$awLm+-(*W>A^0_%Z3{vBAb_>8!5011&;xt}}4p+VaOk5JGQMHaN7}sxdmv
zlI<3>7GagIb|1%Se1dYe68;3%&JFrYJ~iceq9-<_lUaKM?Ixhbt~>N`ZZ<p*{`^%4
z18P{6Jv3>}bkoVL2`z6*)-AUlw!T=u<`)oMBE!5(eR8BZt(^FIuf@<?+aI${x2YP!
z6kGb0(#p1(XqAJja*}t8pj0|e#E(fLnMCmZBOjf86#sh-<;iwV5{`;)s{mP%6y0)$
z_)M+cexE^t$|ns_bm=FY|5^P>nNPgE4?rA_LMRZ6LYJVgbS@-7o=>;`DMC!@O*^)Q
zatXE&-M17<<hlB37jIy;5%`UekM!jPFXSmmOXJ8WbC6B`Nb!}NQ_JkacUjo|5EL^}
z;8hZZZpV&qh(rnvoF0i54mYI~(}k6X_^~o4Uq!(@^G*qX<-FA+t-vzDZKX$t)5W0k
zX^+)C2yC?QwBCl+9Ewx@O4((hG!K<uUFZTJ!DCCUEji@szKWqw!A04nW3uFuE>qXI
zQDRt1kC*=G$5mfEOZ^30uRD(^^Xini@8G;(lae!<Pkwc{B}2pNNkavfqrZUZLDwyN
zHrokN#8y&IJ4}ylbufGSB$Vdv7)c?BeAebSfI#m;?RsOXA3~c$Ddf>9wm}@Ga_44M
zR>er^Cjc>-2|TgRd+fqeq!ITHS!wD1PNIbd-6aspvLDKZikwb3;o(IT6akfMXgO6k
z$fG`wS0Ulxh?^T9s25NibcAqSH|@l+cH)WN<k^NX{loyKS_mv%{6aTZbAcIJgMRRT
zv-+w3!b$c`gzvQg=u;%i9X%dsl~>1qGCT6qb#F)%Tm*NsH}+mbrVf!obLuOj)=d6R
zLGqmkjhyXoD1LJlN`P4Ecfh*Z@)vut`^<vlsgP=<8gEz1_*$=$v2spmP9>Oi!`sap
zlR~&I^4S!nz4!(>`~kJn!;4`jtn&u(4jSPK&MmtRVfm`VOe_1+EiW9@MfM`H>y*G(
zkeZ$x8EK(6p9A}V-<Jnkh^&`uZppClAM}`>eWfbGZ(ok?*J+GrC!Vk|BX0R}{W#Vy
zybUx#)fR4nyQ>BGuv0{!J;r82)37+AP|X<^wE)fOrYjJ-y;K&Wc;#eR@7N{U8(~}<
zZ>;lZiGyt6QFOmuZ+&IECxmBEDg*)5G)f(!3NHN3E4|jQj(m6n@MjuByvf{#bsQOk
zNyW39k$+iQL+W_+#gefl!Ge<zzO_E{QlLT_WTYJx2T~1cB`y6})gXU*0A9}HPpdZu
z7U#3CgTJ(jp>OL)K<^&p%9%~7Po4sei-E+JCuBivH!~;noT_9^;ZdBr=uVT;dBg?m
zkzJ>Um(H=}Tz;1<0ARXFH-4JuSm<N#llI(J1U&!IjNZ`uvwm^v843n9FWS%VDcV@-
z8Mao`qPt{+?FE^h_&kyJ9S%R+j~Y8=-~I^)36tsPx7~a~BHkiB$Yca=+%h?~77AdC
zn9WQ1>!*h5eVC^KRkm~_B$6<G+r%J^O&$@31iPb*<FB8((HGyOMvt^)BtO{2ogX_x
zIyXh%`=UZW+mcp8#D^0+rKjisfxcO$FNLLu?0NQp$wgGI8Rn@5*GqP}1ngcdwSdJW
zIFW{y1PE5g=lz<SiGbX!<G9IowN#@g5Nn?q`d(`CwH||g50C}q(3?XfK?;1rb+5n2
z1@T<H2dZ8XDUPfH-2s#+LsD9kX(|iJX#j9a<^)0uoJYR0S`dIyh?J)B@O=Rp{Uq>^
z6K;jf%J#pg_|XQ<8KLS;vo`+u9w@!Yv1400thY&Fwv~U$8&*kHg9O6a&GbT@wbJWM
zLFug%y!OSZw<ZJRiGPOQ<1y*vZ8p#_-jEST4&sPDK+T~krxePRMU`co(30y$=GB*3
zf6og8#1r9lHWq11;i6@K5pS<pvu$qum9`|5y&2?d_s+jVexG!mU@`QwVH~KB_d$Vp
zReJMQN0C?@|1^I>@+>$~3K`~Ci$J84n_~AIb5#HEh@PERK78&v`!wU_iNQvbH?#eg
z_u}Coy{}QNil9rX30gQf_@pzfhYYMKAm)Yc<kC%+j(<!is32t3-+6r&4ggkqA<NN~
zMi#?h+_>Pm=jQ`Y@|qhcmqNbr6F4x2)_t!p3Q_pT7H*Hxi!4E1_g?zytnWvW>_CeG
z&V<v+4lbY4qZ<#8H-1)~o)}`!vY)h8&WPIrI0lsNjCX(=!&3A}10W0(5SQ7{7Ow-b
zR^^Xd$i~)`J`HCmgBu2Xx6Fs(c~a`&e)w14x=mJ(OS1*jXaOp#v;Ko`WSeP})O(26
z^>{sMJ6t)2j_KF_dwt+SLLa4<$}(fcniOS~5u6hdxd0{T1nrno>nJ56tLcrrXVd4h
zArYDfU4L|(*uuc0eb*8)7bC(bb<hI6w7d6AkwXA>w}6Q!c+fvnilyH}g)9aNFrKxA
z{OBq7f#}F|ts-twqq(af^L~qUJ*}2zgyeQXyRjGNli~i-QsIJr&(sSLCekJPnnjyo
zgz|#geXms=ji?+p*9{)S-E(lcK5W{Op4e-dowpGu!U(qJguwoc$*m0@EnkaUSGU&p
zt}P|Hl2d2JFa=l66hWuIf!K|ij!>Xh1ZrWV{#5b+hgvE}tbP`sSQ)g-*(Y?-JNw!$
z0h56GpD2_Fb<(F%qb(0~8>->+5tpEs;B)hsb%?iJJV3_WP`tipt^LFxceQb!B}rN6
zWQ68n!%Cmc02;F%E~}b`7C!sIZypzN)nTdfVWBB??6lgp%t!1vfl(;`5@Pa!^T-EV
zFsW@j;TlhGg$PN(mlx}OI8+%Sqw%dND(|}=Kd*`zV5+pNC@ySKZkY};ph3>cps9rU
z2y~>*DD-*rr{(1`<%yITMvQ2B%g?T2pUI=AlBeLd!U8b1-`7Mdc0tFl#{<K=P15uD
zdkfPNWkdKOxTN^1$FMOe;yl&v2E|bE4UKPYlLsu-%hp+seC7me+Y317$V-GV>^8g-
z{|U3(&x}O{^F(JP8e-(`yxK<fP))eEI!bO*s26&+dn5m+HgVbJKbzlttI3z=<GleF
z_Yng8>i#Y>0debZf<aIfbB;adhxUOha#?ocop6|p9ra1>fUvqJijt@@f!Xo|U8(f4
z%=jz7<h>9gVhHTtl%GzyFRQLzfdF`@abOpMQ9uIZJ-8=vIakL79wuwQYp@N7`|&3C
z{C=xc5SzfnZrRnvMk*MT>4BGzk`qajd-*QVu=;2^W+3G}?hbH;GVK&rh48y=zl@tx
z-DFQ|{cit-j1b(>X_Ca`hFpeM_K;RbSz<F}58w2xzZsB*6nc_j{lRY{X{{rvHBNre
z2cyu~=k^bU{PA3j7m$cL@M($a*!vGqViRyKb}5>)KLHyalqa@TPpimvk7I9>5$Lfh
zKzr0XWP3rphbr(yaT<zse1|r_Kb8HBMNZuscZeFPioVX@D^Uz*Nf+c{y&<M6^@vVT
zI`CHyKCHM30=w}-X%wZ7(TyenCvRz^TsnA)nL1mH!7iv5br97|GYLE&*z6*tr@U7;
z0}7`)AxdGPd?%}H#5p1QY203@KNSdXp<J|$7R4MK2~{c^FfF>k#F`XxBkh~gd58d^
ziExyA1RawXa+LXoZjUU6%EAMvU=Jm;Anjn3)UIbuuVy*T5!h6mc4VDB<ae~PzrTXe
zk|}SzabVK`<Z`~%q!>3Vf3_bHDYs5%pU(dZn);hzoZ_`UzO_%-5R$jryHDwISt!@8
z8##2=!@A3Y=<`l8Lx0@3?F{l543)vL`w3_ET>rXrgDn|~=$7A{)}KFm1oGy}g7-K?
z)?x#pU1Pcq`6qvYcwZvvb+&w?U9cgC){4^JHoZ?u{FAi9a;T}+^wR}^snQ*}7L434
zRi>R6EuR2zs1UeSENbV(9B1erhAEx3R!8fM`(q=a#Sw%&hw6m&=yS-Ozwe)ug<i>y
zcQcMJMnD&B5fEvf1Bb2;t0R1A{c*4YJs~7&S9>Sf40+^6LhY}u466ZKU0L(f<UO=+
zn&D^q1)4lu+Ly<qwA53Lb}Ed<&Zv@=7+C~wXePdco&OXCF<v?(UVbkQjodI{5?+9w
zk{oJE$&x(}K@#2_(?qwIC8g<zp4=ea!W3Gbg8;q;k*el5lACP!uQjr$rN-L+8gKPF
zdflY2CfY;Ky&NpUk`z96-GopYKJx8s+CjDK=VG*ZrW#`XXglNTt5_`qyWo)FFaKR8
zlxCG?bZpum>H{cDO0)U{s@EO8-m8@R2jpvtK)$zJ`B)@45y-b*RD}d+)wln*DA4NW
zgN(l&)t1ba`M-Gf2h&*d`9E^JGr(b37yAT)WkN4c#Qlcy{sa3SJJJMHhE`wbBY%Wl
z%aIKs7e`<zD3GHH6e{t}yovIL4!@mA$c0;NprEkyd_mB?O|g(nl7>X9Q8q`aD=KD2
zVBt43Pa<TwBlFIdg{nRR{9G47s)q6BCWd<*;5_is7#ba=kgi{2MWIwKQEB7_Zqk++
z%F-i^A!4q<zRcscj^Lh}Ltf3k4At5c!2$ZEF_nG#?J>t$n@WxV+&#4q^v#EBe!Q9Z
z`c&%R6M%NsGjymL=<E%w;;1n70{phb8|xwAp+I(D-)_j~%I4LmfeSVMA>ioh7H2rN
zE!&|Bn)-e>5YQ>vl>xUFWcBAaHbAlpe1Z>F{V~|Wfn#Mka6edF-`o(l2~Nbi4T>jC
zB6l6RF_nODR}itXKqg=-XIiVZKUDS|F3|jBC+nWg1II`umrY8q)ScRw=XH;a7SYjl
z9z%8wN*!u&X3tHi<vftoplDd;T}a?0L~pqgRpk1@DdVd%{UE-4nugQwOcbE*c;I{u
zcI__qDJO12unFzXN+Us`6i?I29_v7J<*08}Mj^Bn$=XHRDvkJY_jI!L0e-Y0v%W8z
z?|u&;a7!#Q9iJuWRfb|+{iXqubk|WnIPZsFk>1GXtZr2`UWz-Pc*1GyU&q6t^IiIQ
zY|a)-Htsp>@nfW9N@KU&fHrc&Ilmn%OGHC9=8&J*z%=Oy>}P)?W3^kRK~$L%Ak$Mx
zJyjMfg6<1|8la=TLTzx;b%K`JIag?1+ClXKkae=qIG7b;;~Q(<DVjoqhxa!jU{HGO
zi@i0eO6AU7LZ!K&8g~SroT{DPG6yHPfaYlbPo0uyTYu$UdR-H|U#9lv=$GSi17LCA
zxnb6!U$Svg4P#Gi5T6DYwRFHDlC-{|=Q~&xpNR{J4fe$Tf-L=as2eK=K-0<RKLeT0
z$Elmc*;@>WLz&!%t@ZJm$85nZ8j4iySIR}*Zoxz>H5#yddok1Un*o)97=+$*9pJJB
zA?0<Jd>)nHRNm5AV}A6p4rf)iVX!A}wY!?yJgoK;m>CW)vr+*}^Zv=x-C*>0CRkK2
zGxe9!CAW7}SWZrR#d1Qz;jm_zLCmj`9Wqyr0eHoi-_RSX81jfWpNmzgyzweVGq%4z
zsXgc>AQ8TU!&8K=5GOK%TAWMSk7~KJt@>YqeBIBctBk(}($5Mo>^M}j3`YUuxv~4|
zrNLoqr2sH^w?@Czw&qL@t@RFv#$YabmZf<Wu~I(Ny`~urMu!@o)MIaO1|)PMOZnFy
z{gtd=eZYer)yg~tv_Q0GPBuV#&u`qp(W^I}bIu{I)`_vSJ8TF{o3Wh0SlbPdYrema
z2_Z;|F7-sNZgOx2`(8%M#o5+q0E&K!p!`EmK=ssFZ1s0T?|o{aPqu#y@@HDw+Z$xs
zg}gSPFlFQ)eohoMg<~-5o7s=C{BuriByk+9=gt6>gF-;y{_?|kVRhIK#9)yh02_9k
zE?m_VVnzGGd6}q5+-8e|Mz`RgpA)YhHmu}0HO({*v22>3_svc%9!a?LZFjqQr{>Ec
zuJ%*xWK|F!%Se_FIbmk?+p_y6(d@yg=O@|X!65$>R-UzF$t-O&hVsN0;Fv0?Q$Shz
zs4j@xhXZT|fH90*6^U0)+YY7F3o|naS-v>$<fXfwePQBjh|Vxmmg<f|by3mn4O9<Q
zqhtfMr$*b)^!P{%yc==WrY$XLBgcvOrhS!IobU1@RN;hHd5}m|@bPshCzR0!I{p(J
z4dH-<U^V;Q=->gcd=KysiviqK8i4|4D3}I_1nukR(;1e?|A093+oo)tQrc*UsL~yC
zY-y>5KwJb!A$O+O!QGOu2_d%?bdLduhqtbUb@9RFyWzWI3Uq<YwQ?}dN(GUlhi;_j
zaS+{Xxw9fZxpo3zxQ8g%lw(<t3k6|-H+!(0i*LeKZPf0jQGd=XVEUN@&<~mCwWc6E
z%MSbJ4e~=94@bKOic5y4NR93Odi{9lV=e6&IW0MD5y9PyZ&^N{Gwh>arupk<0BWp6
zBpYxeRoM<R;aP1eEe*(p5gPLLu9iJ2(_I$G4o#y@ab)K6FEMim7Sp18tu((n%l^78
z^uLzTlKa;2J_2~TD)c0Iv{Lj_h`F5J!J(ox%mT0*1m1}TE1Rxm@5O6WU#{c)CHw1u
zWiqdn2d>I8ka0m(P$;)K{oWLUKAk982$)swf}Zv@H*s*boXfQAyKD{RFw#(swGaBi
zJh~}nuw)k~NMZtXM<L^<gy6N_#qc~k01@m|jP*U|fNMWG{;<hm#EE<pJu$XrInyk>
z0(XS!nSkj${yYiNx|Q(`a8JQLQ16BT<2+`WVfqv6H`B4e%^=~uK6U&CHSPf2L$zj9
zJqLQ1SOBX;azf-*nqj@{0qJLyNpK$}pF1Ei&On~$L~6wGiH)OgF!otgeFo5?rb@5!
z{HlE1A^rtbz2u2Vd2?`!R7vwfuEbl|_WQdtTd2ZzLkg7V_*u=Hb18kJ007<(S&FB?
z=?6YgR4iedaYj^R696Rn>et7lQfcJww_PDwOrb*xSoa%0!?(KKMD`uV;=Nl|<&_PL
z*9IH7X#kgt+5jX${N1+0jX7L1$8kf|Ov81>C|-8O$Op)h{^Z5ZJU27x1KvI71bx{0
zB~dnGr|hLsUpEY;6D$xkvL3Sk3u9FPr?Y-e>4PjxB{5mflN~Z~Ec;F#TVl3_9f4iC
ztB3<TSQ4ZffttQUVf8zGuCMWGP%!8TB3TSpE*mI1ui2kJ3N@Mgwk|FQtG+Q^<W@PL
zg)*;?9K64iYX`*|wu_ms5!SxuB6LzSv)yAUc^kR|ags5?T{bLxCst+5rEJOkU+x$p
z&}96_C#5sIFzZeUY$e+Z50hc6|2GfMzk!l`Es_1@kf6?b_30S~3)hsGfPl{HpN1Ag
z8*W?)mP0q6x0Zcr<otHp<LHtJMevuzVR+M1WAU$F95c6Qn%Hrt{}42**qymFWlHH>
z;Wu5BG%aqrX<D*_t40szg9(@&fKGmQLc}dTy3x+=f%>K4(f5z|@5G9n5pHOq{Y%vS
zZ=LFs)uP691(GKA&ptC>x3t*=e-$B$nm^Pn899|Rr7=;cQT^D0edpGsiFY?7dKZ-v
zrF`x{<Isco_YuFoynb=-7Jsp@l0-_t2<+%TV<*$Ivu$EWQPUMB6~=`xzcGE89z0bS
zG)24L?$(M&=V!(<SNKXH%HS-kD#spLvhZ-gj<X65R%Ym<*KO1uO5Db=z<s!>ZPy&E
z_ZiCFpxBo6w~qt9w!pk6c|?C~?{(ocm40O|wpNoDPdjK^9AltBz$0egp7t5*nT}=+
z=U5SC=fxA9au3n*=dB-`sP^0v=UT?z%XSnSd9Z;|OjMmc!@S$wEUS#!aLY|*1qHY(
z$dhj3W83Rym?|5=R4K{+tbEsgyM+ME*6l|e8D$PQrW`lPh*cLaFD_a>9LA35uv&TQ
z#ovDSU%w5eR<#c;EcN}W0)rf32oUePkieBaB;uY@D4Fo*_Elhx%j}eO$-7nUFCUdq
zUvRM*4<7m&Wz1t!Wy+J;W2?|zL$vkSG3LcbU_{(U@&+T3iIVvVE6TtCO1r<Yw3wm+
zx<2CVU*dMMfyRfiB8l|-<Nh={TSK&Udxeq6^<z0IhlhkYd}=xOK2CkLqw<FZ&vJEn
z_pOs@DuX>q(uB$e@96!&a%>N+sfYOkV2rEyJd(KeW5xU;cqP>TbYWdf%y=-k(T-GR
zwlqLMck^CDxp#L2|MrTR=b0xvM@#Fj#AEB=dITOXbkk5o<)i#8mmlxbvN%-jV~H1a
zE#@#0jU8ecBT~1Qf7uc%P7=^)H|T~x4KfVnb0S>Qc*EzXh{n>}L4XymhKF5yPV*CZ
zCJ2vb+_Q*$=U+RqdE5EFqKmsrek6Ik!NUV@iy#CjN<wsXATf){@UfScFB=GR_$P_e
z?V`NjVYb5fv$ZGT8AtLm2=xAQ9~`3hwc@%fT=R=vmWl*nAoF@vIN^;Zc9o3`IDBIm
z-5vXqlAUd-yYZQEaO=k^MB*e}_w)rXMO;0XQ|)mU8Quhdr#hrCN~sLZYJYYPgLdwL
zIrT6%ujY+r6MXhdI*pMkH_;v{3Oe4w_E*hNp2cYaaqr;s5fFC|QM=9}6$XI{`)9~b
zxBp54DhzGLn*=74Xsa#SGt2dE$$xb>)J-`95qIuGM08uyB-k5dxBVBXkVFnt7~Ds9
zXQucQP~j*(6?!*{cOqu4q~Bk*#LQ&OKm`om1m-2nz0yb|HEsS49;zTC+6=8MEo%F-
zr&3!P@g^aD8WW1B;8un)Q~oOhchj*)fC?v9u$UU|!4efp5EXv(o@)Qh_zkGA18))-
zA)>AG%FdG1EqRG^d7i#1)Od!+jG$~mhKv?Bt%waaE$NXF^)UNW*tFp%cnrzNPU(eC
z32|%pBMa83xH@JZO!xhY>F&mrEKN5SO?TkqsrGEfpD<kp{OKkUot@t_ZP>KG=OqXl
zkH@P*YbV0#HnO|dE>ZU^qAp!6ON$~OqV9ft>Yl>H%u}}nW42?7x;h}JviRHxk_P(5
zZ;=^zkkWq;MtjHp-Una?e|%>6NG3jxnBfV{{<4KQ+t%$LC4q`?7J|}Rc5G@t#)dB`
zhQB9!*tKJovmMRWK6s0B=f@e%lcA2(E$r{Ph3OztEvmb6R1B%B<q|x&_3o0W6&*mk
z_E3*{nAs2Q+PoH8RNF}h+&Sk(tqEhs_xCUsSHf&-v@>#DA!-+K61t0`CXIHjTJKc5
zM$*^&rp?B9ivxH|(mWY_ckh4nw^9xCm^h=_;~_YKmMKz@iBNXRzNK9|gLaL!l%++=
z`o=foH!^sO`xt;t6@Bw0h1ufPyR%E0K?0bGeuW5I5{E7+vLvL)IQG%pnfX}8m2EtU
z&yqA$cZ-lB<2fwx`fq3rO_v9zOSfXW`zT*6O&3b)g0yLhSXxe7JHmA3@TdC>D;)$T
zfJfr-Q2r(A(gJnY<4*t-nJP|iiMpE*bqiQdwKH4>5zeB(r>+eV_q+((+!AMB;$1$(
zyZGA+yqiQ+Syp$&UTXeKqd{ky%h?C{BJVoxo7xko^HJi<^R~q(shjc=arF>!=MS0X
z`ca!$IyGqReeLGqeNO~0-A#5+VJ4=0Lv;$sahoP)a6peEXDapj89Z{39i-fem~Dfo
z#Nb~H2%<7CP#OPp1*Tx!=YA2$WI9i#1DRmGxrsWP)%-5bEVmdcEy>%NS*N%(AIhU^
zgIX`^mU}ITh9Obo6OdSUJRt&wrb>0F(Jn}hquF}SKZWY*s9w^9HsynEGpKZvQ{OCg
z+4`Q3n46%<80}yCPeIX1H?4;5`tY9*aN3j~9~p_%xvhNf$Kge$ph9|GoFVJTz)>#O
z4kK*u<^?mCQXF1~IWo8)66xDNmKllcz!ARAm+J1hMZN2W^ls=Qb-D?EA2G8dES4J<
zj8ig9ZnD~%T`>!*jw49|V&1{vX}Wo^+2tik2u1qq#lD@cbKPC?belI`T&yG|A{!mt
zvBt$<fu+=ND%t~yyYLho4E>P-N71~w=;FGaGZ$l}-DY`Gdwx2OD`DLKo;x}}vc8$K
z_~&KNEiBue1hyy1sW4=LtElGREFg);9$-NZZUv*?J|IK#G7{HaUp*<aI=kdgY@!NW
z)L0MJl$*?+{A)zx#JAnOn#wRBPWYl;II*c5uEe<?L4s7K^4>wtR*-s9Onm)<(ZBXF
zyoh1rLfsa|jRfpM8NTYrrhR6L+x5@pc$)J6xdnb<(#-IxcSYEWhv~>QIq0%iP$;iB
zWI4t~K@4%rF5Tfj+8C#_6Mk6xoAlc6zwZDJ{JWcUbB|nrsVn7WDlNX7vz%NX)V!d0
zZc~F7xYzZe>^>*#CgO>aHm+hHE*}&?KgNDOj;|QfqtM7ic-t)ovoo+8ci}ZOm7z;<
zq%}m}&`qtg4@gAZvf#_aJ-5R9)^fQ%)nUol`gd<7*20b`Xt`&c1b_Q3N*i2-pc{&T
z2%2@=?g@a(Cr?18*r%!<UxS;XhGmEr-qBd}S3zA#0*}j|7?fi|_N@Nvup9Aj>F)6y
zidyVV#K;27Nz5&{oro;_cJ2cj-f(CT(X9cQqy13N=?MtJ$ah&59%TEIkZ=pzjV*W@
zptQW8MDt8nWzS~^_~YC5OGm4T-qL;IqhG)10X80lx2Jko94DrfxQ6>%P*!>o(j2Em
ze@%TQ&e@mw7;AvIi1usA#l^?S_CJTwiUcBC&UVAAi{8?oo+cYL|G56P%wNi2F^@=W
ztnYu#3o(RM`VgTs2<nCzX$q~)A}xC>_8y!$6Ge&5RfA$`(=DmXTrCM|pvAVU;k~-n
zXkhwJtmYINn<GAAE&TuPzdJRN@EfN0hs>$Sph-57zP`SBlYUeYpdL%M3m7$E0FPf3
z`O<<0w)Gy&Q?>G|%EGje@yHC@IyjhNKDz<9`{tFW2_hX`^7)(T>HiWPM~zsAWPPC8
zsWmuha;Q?3=&yTa_$VF#PP32|^%dO-6vR#-4UqL&onk?cfaFP_SF)Zd^Z0x6d?*qb
zHD{3}bv9WH@?JvffpnV;a`k?192;rV1h}|yDqxP7GG%~G-XXEz{@|vuL~YC$b|@i^
zCu@%8fT^=RcDy|j=p(k3w99fqAh6Vk+KdOIPGRnlSbmBDWUv&3^>?BIuLzASJBEj-
z$q|bFq52BIe7{#5d1L-1cP)zcLNfH!#85LcAYr{A^D2oWBDvu|Z9HJ+cky(y2E0Zg
zLF;&9hDDTM8*1+0vfIR6J;VGbeNU_kWWWUr;i}?@vtg$q;{kI01mfj>I*@<o9WT3L
zd9geP@YM(>(PsglG2Q$%`@P+D4}#l;LyR-qH32Fr8<7`t7-O+v*&c<W&#2`EujV5n
zCy4F5N#D~YKl^Zl$da1f0Cbdav4c&1k&B9M0RAkHCyX%BdRmXX8AR+pLiD`E4k|Fh
zp<#qMgyqmB1u|;M1>g2iklDfY1PGzrO^1>yIFY7KXaMU0kUx@5k1C@R5!?xi?oCl8
zr1|WW8C6SIGi^XveXZ6Witih!F|Umpc~E^mYF&5$(*u=S<{j@g*g`W~?a?oJ-E?q|
z%so_pi88Z&8OA>VQ|CVr1nm>df%Pj9`1RyEKMmWi?Wh{KXBX&C%nVo+MWTjr^Ov;!
z3DIEpgLJutALx;^L^9z({Po3e&<|mG!Hbc9EE%Yv6KfKoJAeKTl}yWM>F1&sQ%zb8
zh%5$iv`S%(wtOl&&=e;1!Y3at)1+w|wyN`d&!PYo9-7AoIh;4B#84qxfyoxyj&#Be
zqTDh7u49Gotr&Fb0R3i+Iwt9*Vge6JtuuSg1&x$lQK5X$1S)Mpb>%_EJ)fXj?!Xq|
zQK!z2PYR@yYoWBIG27uyUp2(v`}La;+7e*Cx@zZxZvL{h1>A#7LOIkrMu0Z%p7x{H
zgL<||<6w?s8qAlWI5w0KHeX3aa49srhUhFh1Nz^oV{pj!3eLMX83+3Q^Xa?47PaYb
zsxOU0H+RJDb;4>B^RYjY6^Upxzqf<afar?LCYsAc_Ot>fMv?7R><fiH7H26w*!G6F
zN}t+Y>I~_OT~rNM?89lSEhV`)0f*}YB{6x!JW{2jVZOLAVLMroG|tS{YzO-{tdL&f
z%{vO^o!doZ*Wvh*-DN8ys&!HzIb5yXo_|N9+D%^`Fn>qxld{c0^Fp1>1c;|MB}?&A
z_ED?%Cte@X@W?An_JF1!ACbS3^w5udaT%a4&%$SnpNfjMsl(|&yeHSDUlc4LeZF4u
z1yg`h9gfH^qG10)vx{C&b-1vZZ~bS>Y8i7_ToJVB<v_qjb?DZo000iT(GtM7={EII
zpQn8m)&;5R!@hPUA8!6emn+ysCiDYnvp1C!(xBPBR<C=NG*PI&^{RK-FTFdmZdM;?
zbZt^9zm8>IAl<Y}i)ZA>?pVzn4K}40S}}a1r=!~0G~T916#=5=WYnf=RQ;=(V&Fy1
zrt~rSH&h=TI*!iUv6@8i;~GJa$xYZ<@Glc}MilW3-!CY~gS7pouwBfPxy2zzXvi`b
z0&`@y_rfW)OM(xy<E9N`&`=}DuAkMS1kyk(aEn$z3+m<siCM)c8VA{cBYPn$&fiRH
zV*!^?7NK;NLUOBK9Rq^m>hw<zHePYoEw~o(d7&2opI#*N1kss<ufrit6-wg(pxM1b
z^eTW49}3YCv>sHEBi&OPkZd<@5I+YQCxM>&2SR=n5;I|nW`g&1Mggt+8ntV%l;ms+
zb!bXA3yU@ZaJ8>!NorqH%FDseS!w`k*L?|bSL|UF0xA8X{^wfY`GW&ThR%$=60#oD
z$jEws=tH3loEf4Ya2kteMmst)9&IxQEwPUX<wa0Iy>=HuylklQ6|{&vpYZ6%4OHI`
z@Rw(7#`SM)Ey{->5u$GLX6P^#br_A=d|f;Mn1yGoGGkG7mr*&u7Y&~qkD&$-tl5@o
z^;Q=SsX^Ne4phL7u-}pWP_jc^8uwKVB?5$6eY|0hrJzXthQ9=$h>-RSn$``n{op|Y
zJvPK5vjKV?G#3k&^8fMCh%uT^UVw6B;^Zu)Qw*2bTG<z6<U-#@RT?CshPo;#zEV`t
zcY3JvqtY%gNa~gq^h?|aSmehWp0cVYh4)bVW%NMS@NhHanRxTWg90CU!7kYI3?lt_
zFlW}h{PW*>7UNW|zkpZW=NB7O41M5uJc&BdD2@XtMhZYbLnh_sGZRDRP%L_vwt+{#
zYJUc_?i@M*am>3RU|?{+>#15ei6Qu<eH_G=<BOSI73{2sr5Np--C$sV#lf>6zV2s1
zY4fn1vL~h_J8%eU4)xozv!aRsr=`!+Xv20x8QRR#{)GA%=~uEj5ywRVqEAf`$*LUt
z3>yFb=B~EW*LrAb`vtci<#;16_yh`BeK%&b=qcM?N*mq<y*POC1R)Fm<8}2K`oL_d
zVrT|eGA?}d_nYrfSX2zVYLp6Sz<rQl8a3`{41!d%x>fDDpxgslkwI;*eh1xB0Z}tT
zUWD4SjK#Yc;7r<z)Nn_5z#$wmEx+X5PmB8jYWURcUyHno151S5sDJ@<TY4@aO4)%|
z69p5XAO-;vz?jzQOT_WeI-%x?(HdEq&@xVH%!XpZj2-$BcqqyWo<xn@B4?|N1Zl}S
zIiP=)Ku<#T4Sjca(vb!Pbw@4Th9U<WlTY&%xF-WJu5RpfdBjQH^;xhfWd|Gwgh=WE
zh&K>j|0M$T2$?s&8$h{G?Y<uO=Srvf2qjh-nXWOILS_$S%<}nhXYbV5{N83w`eG9e
zcVyr5IGt34ePSC0{rCw`L2NUVBLYKwROJA&q#u0-$bAV&Q3}uO<kH|kfK;RPYCoZA
zK=_3egUB;bU3zlewHw8Ex9$sv{RuK?9d1Yr)3xt-=T?<!td@K+<OB&X#WP*q>3g!w
zbx>2~d)U}7>cTB=NX33_)OfA;L=~DV{1CCSQ-@OaPS`p=D-!Y@e7?@8JB)q-#EZ3a
zVET=wW!o2hGGNQw@YxgjPP!s=mL@&WrfXRbM~D_>FVO-IQbvO3cIiYVR63s62vBCR
z3$OJqH3d3MG=|ccLEXgG(IcfbVKnweP^4o2_Lap!)Z?!5O9<ScB4zoBy)-v00<yCJ
zKQhu)L{=oUNXgZ1Pdo|`duD)bzV@$*lG|eorH`c(sB@yG!?>63^SDjB4&}~PuG!B?
z`z#r{G4MwC=#0W9J#K+w#;4E`fCjL|`d5A@n|(cntTmD2vS5O|(v>{a>;fqv?Rd1*
zfzsM#?!sM9wA81a!bQkt<auy8pk*B_RU6Ywy0}yqv{XzbG1y4kdcoj&3l|$i4x#e{
zS{7`v3(!erg-(18xxRp6l@*;G6|2AU*|vw?!{Ky!blwq^B?fH&vs8pFgjyo|$f0S#
zj*73d8hZ<Bf!7#y^_xKrXi@QnMIcn;NdkP`aKeY>><QH1M9`JAzXhs`za#vbMY)J>
zQ^T52RP2OsInMXT-hxZ#g;e@$vtK=6UXQjsmOT@xc>gTudu!w_mj$se%aXdI<VR9y
zVd6qia?^;bkJxf0ge3(bEXM3A-HOEoW1&{M#8TNWq6eX;p_Gf*$nd(}A!L9JMIyYq
zc2OnA5x?!HqcW?o3;@t6Mmqg<AawRsH+#Q5=qnig%(T~duYRJgj<osKVGd`mjq`oW
z=Q{P3(Rt8begWfFM?BKtfgra4kFZB=r#5t>0}%CutnPktnt_|<)`!P?xIhfNlCBm@
zK4Cmr;(4*D91vjTL7z}7T}~;VgWKDJ_r{+xf<q(~X;0%Cb+Tpt)m?B0sLsmDdVfJ@
z<<y!FijxqKkPADmxxf+7oI>~w$>b$2BC^}e2?$I9*jWA8yh8v8*f5X_Zcr;-l%i2%
zsOk+q35iL1VH<hWeGIxmyMqy1clZh9#C;C~sP0=$qZfdI{Io&<<gBV;(@p~LZ(++`
z_YHJOR(@7d&_E_AR&>JV;zVzy7}OH%r!5I<bTwj?4|eiRl8Wa>tyG}QtTYZGx#5=K
zlB~hjH_*pP-~B16pnk$x5a&TwWD`69i`H$cu+Or3F3g`caP9>jC_WHQTh{P-6PN4d
zf%4h~8OJ>UlfZi`XA`^tg9G@zjF9!9p7pV@3bD}R{OvnRt-~5E!GqrtFJ^h~6}0VC
zw3`@Y1ytCNrl{Sh^LJ6C1;}_Z7qkhUuUvHsfKb9GRMP>sCdW5kuR6o*tDlD4pk(9U
zynyJppxCWq3D7qwaHzzS$Mq^%b5ZBrWJeJ3`?8f*;7|D(QyfONq@H{^^2Nv0{Q`AC
zKy-3QDfImTO!dt87o?dJQ47J;UyCVG<Or7_H&0dYk<=+*?Wbl+S1jeg&sk0s9%KU^
zs5k;L%y{tHSnpY+b2v1!TLNZK0VYDBL4km?h9MMu@|b=WW>tFOuD64}bofd!R9R~S
zIQ+mHIY8%TO*9_PRL^nDx&iR&(BmS|c8*5e5`4Ex{hZe~Oa1t%N@uSWLs89DF!HM+
zWp*o~zIA@l1$-3}w;aagUi@mdNEBa#TL_K)an$}CD12xqUD$f0m0%UytU5aq+F%t0
z$3vtRtOE$R1+^08$}cn$I)jW~1qx<?rS!Y|31IE;3N4G5d*OWAa7O&rH4+;^>Nx75
zMHX)$2Xk<SJ9q&zLg@hGg_@T9xpC;v&+h(-i**J3@8h1)$hC@T@6=fsXf3gV1n?g<
zt&s^o%KYzj)1!=A!{wUq6*_<WC|Wh8J*?>RG14)Jtjs?L71~;IV-GVH4;r9QC5ma#
z#su)5MP@_IllujZU^Z-Cf9K$$+q0G!%!V~$3-SH^UNGJn{(zQE3rgGxyAdgb6SqoV
zxu8he{;d@G+rM<lrz^Z(L}y&)0wx(ObYl2xjbO)MJhyD_^UKfz7+RpOxR{s4ffyLV
zO~~A=!4!`D_TUPp1+N62E;m`L^<p8E7`_e0i+1twUqUrV>;;l5=p|&-y2I;k?+-nm
z4zab3=kHlBZuKAv$;|DlSoim}ce7!<N`dZ;3%>yJiHiY!^<J2li6fynd-~w!nNo0m
z=%_OnqX5K|5R6*0srK(v?n+4W)m}5_U3v{G%HVILIGjd{Ko9QXd}QaHe8k!%D;M32
zyIHWJ#69cm7TC2C8FJGf)YmRku6oEuqBvLP`vMKo|MC5$98j*Vw@&yhFa&sHpj@3o
z7UKF(|EMQf6Vk=kVD2C;3d!qM7DKxF>0Q$Hjl_0>N-wZ+*W`{bfr4>H+Zt~6rVn_5
zsnLEx4&goA?-FHL+<R?$?s(jrbZJKG=b{k0B}Ou}hB0dS4wAzB-mM9g`tS-UYIc;*
ztv(S8TX~&%jG$i;MwM7{M=5g@b&&^0T7d%`Dx@R`7h%7d?#I!XLYZ^nC+8GU(0zd{
zKvT$B!@T!=#d39^&6(Qe8s19+yvY5ag?Nz{=j9#!<32{Tvxen>t|#lWvB@HeWidPV
z7ew~e>&Bu=+B)jzc$mHyHw5*}K&=;+2l;>{Vg99)ry%Zi%R+F0?G_|xf!eFkR0LLT
z!=h4zSXXs>r_KNP=o%ar9wB}jBI=7oue-9e+a4;SmZue#t$lG>^PeqgA_ib~Vd_}z
zb6RbAS5!qv3YV-AZI4`BL1A*Aq&nh(yV<#qb7uDy9QeI;j@<0;HL=Uqb2l0$&fV41
z!Hq=f)TsK7#YHZxo^jvkP!)aE^_KpR0S}Uw@67z{uq}wK(!|6HLs@|9sm%lzRL0Uz
z#WgW1TIII2_dW@F-4oBDTF&Kv#|u3uAFgGh?{KN_g{CVxhEfi(RJC5k&J`5>n1D^l
z4SJ)d?4Ii5T{<c#xG>RW%YqZk2B|oPQ#jG3Y5e^Q8wr#zS{`*@l3cQP8Hj~Nv6d@=
zmCh*n(&C!6?AMF$c0+-bjz{z&i+O^D+2By5aLRDCO?8ijMFK>?T$AvX;VApT%2UyD
zU=<E`hNr^61X~mnKDGa^-_)`gmRdHdlOhE6qeuq`AHuj_L$hdcue`K^)Ii2VK&uT@
z_jFb~2Jj!)p`P@Lg;eb_g|G$}i)6}5tZdl?{_-+6IxAeJ)E89N0-X>aU}E5%;$A);
zh@F2Ie$FZ^K3g}r?q>cuas4ZKN-T^l_fmFHl1EVQrY7N}AY$xd+=jik@nmkW`4ifc
z@&^^<xAJYnl63Bn<abU<b>3*~e4KbclZ!cd8l{pr_TaIgxpBJd<-Q(+CBtPUqCbM_
z*#(c5ms!}~oGdRNpdlhAqr_mH)~tm;HeJ9b=vL%^VE+66#V;sdNMaMB*ld4nTlneU
zuN<~g4)&3U9bbCK!ZQXem<GjyP?c@y#S0(89~D_nT5^-hLHk#XvFqU4yrA|NpIX8}
zt`HO3e{f1?TJeO$DQp6>uR;?U{up|ZFa+;_Nj=`c@HFlK8QBD1$fpU3KkHqX*0vp#
zgJ;syQ%~W4Oh&1f4sR;Bxh7Q({~MQ!K&S?Gy*a`W))G-|Ae)fh9~L-I(1iiWD0gjy
z#f7kVoFrTvF&Q-_<shq5B%n(1-X^~Xo|vx5mt0Q0KO~rD$23%H%7twu*bOFfvI);k
zCPMfxLJO|~5=w?0-bygWAu<@__zPMOg4trQ!JCPkMXX*p^Diyo&tP{*i0wC4_YqJK
z<|uCg@ABLE@jJnMaa}Mgj<&)yyw%MwqyW~z`!*%noM6IVY}e*799u{5DJ(+?23L3$
z>wv#-oD164@QTzk_`3xA7YzHi?!HMe{@55Mn5t)tJ{SJS=;<L?_7JTo6#-?%m|>=t
zOvzd+=X4O}lzIF6E&|fZ%fT3Sb?;*c$N@jI+X+I`T0urII`$5Xy}iTVZ^dS41;aP%
zy7-R}jB!X0Z6EpW=PRBFW`l2HI!mPpC^B{u<e2LL>E1t#v}|rKLV#2h_b;CzoYf?Z
z-Ku8$bj282>A+MjFEx}22YCQWfo5HY_lhSlG%zb|N&S8JV+8xc5T&tiE)r6m*bf7Y
zew(d%AO6B6R6y?AJDO<mKL)OChc{9DUfUpw|IIFI82D0r@c4=m$fWnf&OUhn@r`A~
zfDjG~T3o2~jukwGk;eiJ*Z3X8f4~Xr08jKQlC!U*ZErl#_NK@~LjB^&0XmW=y$^5X
z`5lv_0-f-%Nt|GeTnvms8e$+#Fk27}A&`NZkd1YO8%qtkja52|fP%o@*iv9z0}V{^
zK7vQQxiAgMH}<rx)Kr^b9p$5SWuG?SYXw{nylQsS%4H~Q2gRC_G;tE24QUx+Tis7I
zAHd%w*gqWFKaNaJ{C&cC!m<@&BD3&^MNhv5<?`BaN*aFvvQ2?7(+aN=XM&l+Bbn47
zecL`=As{VQ1lG{6y7MI?0XfJhdqLYrT^HL&IQsUDWD`nvO$}CV1`fX2<XQiPV2s>g
z7=x74=*G$?Vqk?aPX)LLC{nQ#7&rS6?d=t^6zl?|;^(->gs%(N614(FEKR%4tQ@0#
z8^~pxohm26Acx=>)!lSgZRHc%FsrAm5fAZKj?;nV@g)yi%MehVQqLI}ohruYG5*3a
zmtYmQ?;9H8e+*pf0+Z#NU9Sa6=jCH)H2|SpvYhJ(&RzJ?p(n7j;yg_R8;c8pQ<UZ&
zkLrGc#Q~$Qh0%%hGM5j4CO!^oDq4;+b`{#@0Bz^S`3Uukd=f0L=u0G_#E?<4$N-D*
z8Bh^!#w|3f+cDe(vxPbKgKATBWm8#4FkgAlZBd5;AK?=O&6gjJjc8jO$72FsP{{_0
ze0R7(sD^=7Z($t;YDd`#4Z0wB)i+*(GY&?SI0uTg_EXbN{7LEEg>{Q<H1;Q)6pj>e
zLPBpUf&3`|r=5xRn>azpAef4SAo_9by!ZoP4q3uX#m@7PA6O}U3P2DkmdAvo9S1p%
z*1Z^ekANJYMXtk!f2Mdy_$no2lJ3opUbz{hY;Xw3Yuv(DJ8Ow<f%2qzo$zht6EU#D
zXvxoS2`GZq1c{l>*P~o1OSC|$ngkYaaRQoHg1W8F@?u^&hKm%)WqSYB6oNr=cftC&
zF(xaYcnU0iCergT0p0yrVTh8<W_|*yV}myU*=1fNiQwyc+-(rQZ3p9C5z?HP7H!|?
z9;YFEO5rX6L(xoZyl{+g1gBha7=o%ukB|jjPQc<=K50K8SRCvo3D9t3a38@U$ZAdl
zBZ#IXY+Fg&UqIWj`f5FbcVUZ&avZ+<CKQ_q+#Xf$c~>sdcCE4GRsQ`tm53kY*9m@&
z_{c>ZW+k!ls_!b_;b|zeiKV~;`pQ{DD-9ijEe6e_B>{_a377X?X-^QcDfmbnu)#XD
z&8zJ2ib?#N@-j8{f1S!&;#z>uxCE87`_g`GR?+f0r@R4BNt!CgDYc}+9_=edl*|%%
z)zU7V;0#IGY14DGjRsxk+7!<lL4xi9Z-+Z?sfN!)zgVe9cM$=H>}7e%L?F1x&b5B{
zUG#Tr+uX(Uu_~!M_Mi3%P!URH@HY_3hrc(hSw=fJjW2=K869^S(i4Vv?3FKC)%NeW
zq_a}mJ|K51vgWWP9-BwQr%Gx?(0o=<-MonDczRTm&@)0e7*jzEGd!Tw$bd%~3@d!v
zUrb{-T59U3m>xB)pt6>5&KQ08O?a`P1)gZShr-b$5|dec@Xc*Vq^hJxF?1ztB%Cw$
zDXjO5cWoq|_+YNW8>f{7T<klFL|(=wNjmJf%7pj!dC@F_h3!|A+>K|~<06s0{j%!K
zaHiovCqsLXUF~}SsN&IcEfFjHC|v%eJ>JK-QxGDkiEnu`l&Y2vE%E|d6mzOn2wiim
z)(d0<W!~6<Zv)B9AA7ZJ<8jyTKo1i6^r)b&M5q_TA3IkbxNOtaMh<*u(hoTtZT)b6
z<#7A%wkeluat>13R|%e7g9|~1VvM;e2mT(A9Z!h*oa1~wXQuVlwQ{&9P<`ai1$zA7
zz)G=0FmVG-Msk7>Pay0;AcNK0X1M1tzdq;JIHpChuwP{bb86WjRT-tMjL%=h7l=;Y
z5?N1o4f~Aes^pdJ(EQo$8sKfcQg5~gRlSfQJXa|tl);#3tZKsd%){B_`5JzBJfGcf
zl-j5hofLab6YinWdj`eX8=nvKtV8#Wr+;xK4``{DQ@XP900eijI{0!k8{V?R_X1sG
zADtt%?~t5qGK{|lxE=CIYIc8(6|_BtEZ)r7!}z;efJ~EEgL5SI9R`s298pf!jqp-t
zOSQ?u-&b@-(W2A*V^T98-V`<^gCHdKpQ9OcWB&|ImGc2=xAV6By;?agLgv6y!3O!M
zSlQzlTeuM9DJo~j+!8UX9kDI9&j)M^ltp>TuYAE**df~U=DYD<fQ$fE<-E|mAm~jH
zzx`bX0Il!K2uIcfyX`I$nN)8N;y$>_nf3ul@<5=H;Ji2BL?jW0J%=W^`svW@xByy}
zh~?yqj5v(g7o5+TKGz;}Iilu1qRDb9RUq@|gJ;tzd=;}DKE3QU$kdPwy}nSV8Rc|n
zJ2K3g>cpJ|tqF4uL2J$bn4lw=Xg}=9pD!N;>y#@G(oM~-7@rq(#4x8o1Ujmj-m!-p
zx|EC*?N%@j*?w4$ho!CYBfR-4fH<=%*E#~{lmMpn^Z+y_8_7y*IPqbu3Gjpk=d-_i
zVOn@eKM*=?AZK91@59q9Yl*}TAfjo_ZJc=5I72@y_{C{mKDtpIQL1(YrQBnIXObdP
zxbX;xoPa2mt22m`rk?-)`Mj=31d`O6i6jsEtN`rg0=>~rr|Ukrzrta}dLYJ9cd*AZ
z_HZHdYfN=qZ{NYT<DmMkROmHO5r3_G=JRCsHE1`4BqedBq;vw4J!Oxkz#}ELAGFAO
zHsQNa;d5rwBwRtq&jC2QhWrm>LcPJa@_n+$m{#9Kq&W32*ADsO(cc9^-cc$)Nw*KW
zz08118P%94$JkqbZ23L9r5)~R-!{yuY_L)W|NBUwemRB#0Xhkd`^weYgP?t3P6JWl
zm=HAjDA<`Zb@D&vhAc^ilp!y?xjjXJVdW3Ufo?I=^uxBlXWI5;kKW-T1A1kwq*n;C
z!Kb%1;k|>54|%34?~BPM-g&iJHygk5u!km=xXa}9)dTDHOkvqucq1EEVXk#Z#O1Zs
z@RT_2HGF!nHaQlk-wM5PX+O_)3%EMou@{}n+u~Aw^Vg5}hvNvGN8r%h3UnG7_i+T^
z&CL(IF2&gIS4r(VHeN~Bh4YXcZfn#^BzfFt^>MH3ZFKFt^ve}owg%@3Y^D$wmxa$6
z=!}2=>4J3vbKG|Tkn0DwR0EU2wKBBmPOZypR*}io!$iS-=uM7m=hgYmiR9-g`663H
z#^e+4z{DCZx!wT)p~I>Zqec^pjS<10Sk?T*)Xu@gQrp^twm9^PwOc%&QRgOG$yvlM
zC>WzBc{K!2r(wVR1|>;7!=!|yG8;+)fKi2^$=UnEha&5tmVp6XqQ(7c745pZFaB_z
zx4P$*{js?|xdOG%-@TDgZaq!66}s!yT*j<IuNn<trU|AbNailrg%#*T>Z=^ONbf9S
z_KLgZN4G$ZAP4m7XpZ2T%z)}99&{m7vBL)+yvOFPs{(kIH^3?xKZ>T8fKIcHdZq@d
z^EaUnXo~D`-}&^Of<3>k9RWb<{k&d+kRnV+UJW#SllO=G<x>MSnF_B;ExvbQfO^o)
z@FLWZfP6#ARjgj$MTcWoK0Fw0EO$(OsVlO=fWRE0gPkil%lr~g2-XsX!N+0L8KO0u
z4OBbZMy{2gJx#w|^oMVGr}?)-1g8;9zyY|TUu`!#fys}e-o6~{MzL3*{%o50*v~Vd
z{Gs^e)wM0H!lO?f-`J1txoj2g$IH>po0vF1;D(yFSntz#COGj2kh2Hd`-ZdQ@`BQ*
zQS%)mxSBacvpwjs<5*bxJ1V)s%^wt3ZbAygX`_wxIUeB2Cz&>kqWvcJMM@dGZUVY3
zx~651zj7d-q;nXnY`;w<tFra}2;M0>KOF;PWq$2YAqby+1SMCbplYVp3JP*PcMFx=
zu$$pfp;ZZY;cYpGe#N<6pLVUB+IBc(m7Gi@2{`UPY$|uwCD3Yt0ELXVGM59xppe1d
zXsVO>nCd2(6*kKnoFr@mTbF<!{sw)7v)!YOV}gZt&{x>jyg@%I@Se<tgnG`jmTe|)
zzkP7dD!dLD$*=jSOXSUvJNOiW-3YFOeM~v{_&uJ`ufZV~>#IzrJ}d2DpU&VVo9?fB
ze%nmFV;1G!B$aXse>ortxZmJ>eN$ZhJeS!rA)W%5WF|Ujb^CtJ_{tbH3yM)&no|+@
zgP)PU{<&e|b^^c~UEw}5>UAp(CTy@JNKRFe3_()Q4kaPFnH%;sC=U9Btx)U1Lhw}o
zU@1ZDZzFg&)R)PfMUV~^T&YB2kds;VxdeZ!gWX|6_a4p*>S=s??`4DO=@lD@JqH9m
zSLE-5zXR0BxFGF$#{kw{4&6$HS_{)r(lsS;W921)T$KJ)_CdN9m*o=M9<4l1SC7cE
zAVYmqncM%b)iF;9mv_H|dD5?(=Q%V_CQ}W3#*mkXdF~3JvWIP)Gap;A?@ExDR1D1W
z#TP>CUf7uvMzVvpL8Wf63rAPk1rWxS@ylS8FXcYMJb})du9Z6Wg)B@*$ex!V`ga0$
z=1xOC9E<eOd`vCYqSOwq2|Yq*-Eb-6eAm067HuIy&zP7EsT|L?PJA}N<iG?RlWXjt
zM?c00&bf><rUnmx>Gk+z0i_xNV|sn}$G0fRI>81iGWEW&B`>Ik&EOdR%R$JDfG|co
zU*UfqoC@zs;gr!Qr&=xP_CIKj6b5IJ=yNrW_@@pqX(PO~@!$o5tuY2IWT56(wc|j8
zB5jJ>il@9lg)4nBB}j?8Yl1JEqiP@m_QlH=b1SHs1aD2c_?kfao8VnFseBQf(yJyX
z{--^F*oE@h8N9Fu(w%=!1w8^d#@{1EQtg?RoPN!Lm7I!<J=UBe0&+OQhRo_jk3-!0
zRp^8m{&7Y|`4N25BI`<mNPq$01{9mCJW~&W+FHf)f1w0ys^I8Bz^1dXAwNqM2KF|1
zTsS%8imyi~G49Ag8w$m@!O3bMiV&3=!D@WTStR1D|6RhT<fDOOBW4faxjbu$e!_-)
z_m8L(gG(~*HEqE^b1+zp8tiBDNp}3tU%Y~M8A~ZP<*$m?zy>P-wIqMa$l{afI~>!U
z0#dJBwd4%8r|elFD)R==eP#U5|5{6tukUP_@H}PAi2v<B0V9aY{}Zs~I`f}^Ew|JD
zco_c#?C+`c?*v;eD*sNf<;Kdt6AbU@{%0aCH#?O6Up5il3;ZrE^aA*2F8=e!mh8_(
zpZK3Y2HN7EKZfTs{tI|4J5X`|0v=0h1EO930v^k3&wl}r<!%Y)Uqo@a9{s-?Q4H2g
z=y6)NnG7!MjULM8h_%PTec6~2KmM09n}1Js!d-fF&?C>|&GXH8TQ)D-wJ0C<)r2sb
zrvmSc8oe!aU>^iArCE&$LG7zUxZNK?z)>iL!=Bk$<m!EYLS#9n!XLgjb+jW0q*g(2
zQ<7Ou=frB&JWg0K1RYS&YsSl(;w<6Qs_3hYN4E~Ls<qr8Tp13fs_cGL`{M^#Jppj6
zC}hh+wQ(f|4>oZ1N|JCy1X<4F&u&_BjUC+tc~@exF+j;w`#=ATKZn0rFb`k()rtx+
z6_gNsb)jxLj2FSi(8H&e2C5UX#P*+d2<%zTR4BF|ECAnso*lzOO06Y|fRL@HY*5K<
zXOSY$s#646fd43H>qBTl#p0yvj17vZ=|2e`T1%7#DWtLPC;|LA{K*V{*FHeK8Z?}Y
z4-n5$d+(&mEce7HB>-#99C*Pyiz+WD>3u0dni)!B5UNDrSQZ{FoUjNu5#hni8v-40
zB3^IBdnM3fFMLhG`q~&TO@S36f>taPaKVSHu9&rgUoLc>+4cF?M4Hdx=PT1A#3(7*
zS1M;p@a|<+z_R79LzQS`FOTC-6Q9EtkBHT|V)YjjlR^m-FQxJ=yruldi~oaK|MBZh
z2!f7@B_+tO;PdLjtX3D$WfWNc*e)pJD95T$5M)zU@X`D;|G~J+-?tP4J%nTfmq6?J
zA8T5qG5P~*3T%+|vu{Q9(9-*stQai}{CM3ZewP1O6W%lbp+oRSUz}ks^mMQ%%{P$$
zUvbjvvV}ML;tX*(5Y$b#YlBxO*3#nUckv%XFK57i485E^{xS5TP@zBmY3Svu?4O2S
zI0ffL=AVXMe%!2Xpsm4)ZbUb!t3$)&OVHKy67<m9wA=Xg**&M@d*o3agbG|MJkSQ$
z2~yX={dA+5^;rjv(%W4wK{Z0Zisg;rwqHYS`OvPHL1U(=-sM$lQ_R;H6BB5J8-4Bl
z_Vc#^DYp+{vW=2U54T$W*n}s(^M^kPoaftv{qfrgRP9tr(ms>k<*k;FP{j;4M)M9q
z_w<B3(%rA~X>U&G-rNDWdcMt<pnqC>Vj%;$E1mhr^brM9xbIJ6s$*x)u7jDqp`*=h
zzfx0L%N-|@FMW7GQ)DuiIS@w<*H)XS!>!o)(D6{f<oQ5svg{Fr=YyN*=unN*XzPfh
zfVBT%PT_CwsKUy-xibMOS0LCEKm~1hYo}#C9REBwd_B!sDQw;Ja^3uAY4Yq~`CRZS
zfGZA`83S6+NZ>ZwU&@qSYHicXcbUtjE$51!&z@;;m~1Uzupi{M>yX?u_U?<^><n~Q
zm962Ap4j!p?Re0xUbnoU#PdNFpBl2QvVYgcJ8qwVauUj$%&qEx<}^!(&bj%hngZ@=
ztf*unb?h)TF6FNu)Z3F`3qbmUmJEvmRo%ea>D?pETu=!{@qC5lJC5LYnliEYKo%Ei
zj0;nl=*-|~;$sJn+3!9%B`MSOZ@c*~71Ced^li)=fj?`N)n0AQ3-aaFbwLQ|jI}SQ
zZ;4!QwK|VLlljo?xn(-=?R0JYX6U^r*P!holMnZ2wr%|KAV8(&m)HgPrPz(D^<N(!
z`~10Ony&C(6!u}(!MCIN*TyPOyV@v}i~For*$su-fK`q@O&{yLXuF|5T$C``JM>-Y
zt;vo7_*}@EQck`Rnp>*j?xX6H@%}KV=@3ze{_2R)>C02w_y9~S$C24_kNpA;UBT(n
z2Kzd+=~a0&i9KTA1VB}%X_W5}m<rvQ#nJ*;`Mp73_{`^~3Xg3zwJlaVrFr^r6~pfj
zl=81>?M1$S+;<6j{a*1eIKv!Rv(Z0m@G2MiH_zZUK*WCy+B!&cIuwUDUto~`V3r1A
z(NWx%*Rrbmnv-zlK|Z?CRpZxppEQ9q(ZYJ8_w{V__JfJ2$na~nsbm5(sZ7V2(>JGF
zZ#LMStU3Dm+at3bu!Kj)eoaxzvC(E6XljGXodR^7C&TW!H_*8B&eX*jtBT|6Q)KTO
znC|{bchiu2jOvyM6dIW}=!HEzv1w<D=BjN^i2D(~FvpLeI&G6%F7#5}cKJ}~ToYXS
zSpcO*^`AX|3X-aAh8s;OHGOIv`I#E3vwNm&Z&4g3`^JB6s9rqJXF}Jd>xal>OZCRO
z*UzQLa(>U`__oKVN3Gi;d<X?UuGm|2muxVMO*}PF6J?`N<NtXqGU4~H8CNgm3(;RQ
z+)Q9g(yl?L`|CG28>;i#T~^TdC<tSd*E?r?VcbRBQN#?fmJOibdV9?U8nrzRnSOpt
zsaa4vw7ozCilruczsgWd7(oR>JE)A^R(JO(Y@E39>&Npgt_*g)y}v;D)Y#26K*^hM
zQ{At2=v=-T8nx>MRzPpq^4m5%HEzs7yDsO1{z2j&n;UyH>?}f0XM#5mbXuh#>AQG_
zgFyO@e{nrFdui(bWAClQs!ZFrQ3tURlM)aW6(j{gI@AG1K|s1eLX?zl5Cf4ENhK8l
zr9)!TN`rz(OLuq2f_>f0_r4!%IeQ<UzhnQik3E0AzT?cZp68A$&+9zLi#Er7V#<5U
z#Dv!BeX>Y8OCL$|+aaZc`vN~3wd#EMrpS{E*T#sG10D;&@pf8)2^y~4qLqU{WE$k%
zCADyNucLOV{$i72<cME#d;fU1BzNyqpW!_Yyxbi-4b-;hG(el1_e$v~j+G{QQXB^p
z0S^I|ZOOC<Ce_+@y}i4uzWLiX5IrW(YSs2^j)v7jCKJ&$Q>%R>bcg`H-kHZ_`B7Ha
z>niuQsM~Zt1a?oKj-WXksATo7^!rYSh^NLqDLIVuTkuLz*5D8_8DH$dfln7--e0<_
z&Zcla+$Y9)(opjd3`N<9sb+Jjnig2)Y_5&pkbNSMH<2zLaz$TqpCZHOi03rU+%T}i
zOYP7ONBYY<5l`Nqd!0^5jlYOn`t*CJ+qD>GBg-#?N(7K~D;i*jT{iX6Z)fP&5u+h5
zqf~wP!o!McW+K}^4e!L6KV_9N-r8PHX-WknPObG<M?w>2V<EvzGWmt=<DV1Y!z8U{
z71Slb<}6^cjP!fCfkd?;Gvz}snj5kmWYlbH)v5y`byA=WpCgWydn~Yp?cSo!>M*bI
zDJU4?OgVstj`-lKEkNtl4TM#5bxwVbASUM885kfh<@ysUMUBblNpVLWO3pk4Bk_#Y
zWwoY1h!<GhA68$AdrIrWt=(b6XrKfv{v|0c*zp-~S;-^y_0#<s-B6|}fwT7oP3Ae4
zHMEXT<xLaMHip3DA8S6ch-`A{y>65~PE5;KxaLQvKxn<WsDj3Yf1{oTdq>FyyI8W9
z(&(th+^gV$Qz??4BgAN&I655Xzx8gePagpyUgOu;)yOS7+cHLtW7<yh8iz=k_Rn&}
zDYg5JXQntACtRdmw%gi7HX5{DRBcNryN+IMH_AxPJ#wzK!89THYTsG>1Yh_GDp1N9
zsp6DaTsIFmqLYS3A1_i}E>sy~)LgxV=0cy=d119tq{Nnbtc0C07orCrG`IGA<zNaN
zyPU-puEJUGqZQ;B;7fn=@XG<Zr!*Tq`)F@RyL7KLXS>Nt32^3bE~d~Jj}P4-^519>
z!pg3+xL{k3mhdYp5)~o4a~F)zg8~gB7$C2*PmE@kN*x9(V-n790r$w-i>>}P|Fu!p
ztvOb@HTzH4K2KJYXPNI>rk$7b>X-bEusY9rmViviX~D&hQkhF!Ke{%#42{!bp{Qo!
z!jzcVta3Ll+<NRjCFV7oj_5t5VKhDIUX;h`G%B9dbH`C|vz7&JlV5TTw!KbEN+>ZX
zPf7FF(iWQj+Bk}9)P(<VMW^5{2tFL1Eq(PO`I2nxQ<u$Q7h>>P6@U5;82G;RV@%0U
zCz(LjiVQ|BF+Q8LFi?{9<cJ`5n9JrWb+>Q9i@33k^`b3@RuGS<aORxaL{p0^J^O^n
zQG_r96#CYVMRV1R&w&&=HrjaLUZ`=_gFRC3%I1>!KS#h6HY18Wrp9!m$?|*l^!^ZM
zJ?yfc?ZVf)z?Nu<h8AhCBVs268#c>sF(#cP=9@@WxNqRi*?fPC0fcqv^T*^<#_rcR
z%qN*TG3NJEPm@nxh5N7I&e*)nRkZr;zQe|UCMYaF-L`D-g+)rbs1k=y8wGc(8%nzy
zU0*<7YUe_|6z@SJUy3OmKkq)L{sI9L>E?pe7Dvm<<Po6Uo=s}d4PsLM(bX}_vsb8(
zF;d%oFgDa=Y{leV=<Ta*=Qk$wg$}WWU8yH|a^36*@GMV`&>ALV?lHES7yq&EDH)6f
zCJBr4|3M#dj$_GV`~}?x({%NTi5LC+ORYM+FK-YZ7!J45s7Rf)i~=>3=aX6O$s|f1
zDb%W6UCFR`myLgZYEJmtCv2w4R(w)w<wYEGS%Wnin`<`2@LsgvyjUnX4sK}w?D<GP
z#)z*O5UKlgWNiw9uFu(stOn;Y&fAu6WD|Ax?~KMLBp(F9DIyTsF`WK6i}|W0Rg}bg
zweJ^iEqq26ip>5Zc+^x%X~O($bbBxNjq)(&YQOA<VP8!&Hgim4?MHC($gEj+(Oh5(
zQPD~nYk6iiS80Q5pW&Onj|{UOXM-UhVMr`k`k9mk@m-dX;M(frLoD|Tt{Q&2`}O&9
z(H1tztUZ(!Gw^CA*`^?&;Nv1sVeA;(X3CaL?6ivzWG+@}>wn$o0%4|XSmaE}qo*kP
zy-gi;#^vSNzBYWzgwf#{MVCdzvwMU4NRs*zlFK7<GAu`bG~3K_K;TY97H&86S|^lm
z3TEV|-LeZNjj4^<PFjv9Wm|vpF5u~Nr@c<(%RJ4%*eV;;G`BRyhkHt(IQq}C*@Xv6
z{gwJ&aLjV}n<U%j)`$tt`8D4Z($Vbyl)uTAcm<4gD##9GOPal)!$??`9j$*ZxbXJj
zO7HXM5Dl9yq_H~o9_sym;|TTg!4z`I+@UvyqLGGMsUSJhOitSyyrS;-Opz#0SG`^b
zyVycwbN#lOsX~vwt`}S`@(&&<Ukr{kPUKA&CN(_nOUTZ}^f4x(K;R*8U1Xp3HT0n)
zO9CVFy0hP6_GDyQiiT&(U@my+63vY=P2D9=Z<yjnqn$aBTB*_<!7juJ8GjR6;=Hd1
zU6W65j@FamTo&Y_++cQr<I_wQZj~5&!WLwERO2et>;?nXkDS`gMHg-<(+bgns%v;M
zpWs#x@=&R@yyH(%-lvw$ZNh(BAC~|0vbs?#WPl|iWa^ExoirPtX+CL)kUY7elj5=(
z>~esO|A%u~!xLoMmtuYD%%o<mWF{4jJmY#LEUhRfU|023=xD;hctzhRi<0~F)dI!j
zNs2GL7cju$UlAUEWv;WbdxR8QuEEwguer6TsS_$T?YY-Mbl*zR5y*>m+{cZIs-f-k
zpsA4S*Sw7II*flY?#Llgh~{+MovyF+<xT$H_w9^@2N?<l#}`vvoU+wb>hucUjFY6;
zii~G!K>}a@R7V|1^6x@iLcNTAbwT_K&DJc9uVLN6rNj|V88K=rIW#xPS8k73M*2EV
zsCPop1`G-_%Pw2XZv})&hF52v3-Q;dzMh<2cxww$+{YVe?%qA+jY|C@2nIX@d4%Vn
zqm<iUPwwt<H-WQ7EhE}#AXf8pM55xm!@SUvYI1*Fgujn}5NyHdN(C!kF0^pxOd;>3
zy_X@to3<1fGe}4xMfr;YJ$<2QLZ~5B4-Rp4TFO>Llc_zt)(R0i@6Ylseh7$qW58#%
z4y8Wq{nBh$SnhnRgb8dKc{~Nihr7@H;n>GGpAMgO2jGQx26LzB(O6?shxVL?dp~KE
z?xfNc5*UJ@MK7e(##Ww^!(kSZtuSQY{(N+7{y!;Gq<0D(U%%HC(dppjyC{2sw6!Hi
z4CM1#>@!<(SR+hF6jO({XZc{#u#K%O>WP4!s+CdgFOlt|M<*OM_U?27FbREzu<s@p
z+pY}YA4q<_6<V|!`mjk}@fn18WsDOUkYRf}U}YV#^34hFe(v*3Nmqdgy9CZ{OBdBp
zT_jD{j2oBbC+`k4Pm9JmiG2Cz{&J@y0>XGkqx7naw3@!_lu|6b121pz{_OAC>=)XM
zj!wefv-=!@LbB2p=}w5Pl2{T_3hi@17N7*&{tv-FTOzw*7~c(aW{Ub(nX_8o^*g@{
z9SN{vW|eD7(UKxsEPW`o&N%|zRE0^0dE5zl^nt|*Ilp{qF=2$`T-8yjlvTLx)O}T4
zooZ~U7ErBt3YNkU<7*0=+zW}$or)`P7PX%T=MP-C+m^0J?c}mjLX&l!kJ{UqYc^2g
zh|tEvaS{dNZ_zaU2_FhqznQwBN=|87xdvDjww&}5V!o%g$7S-hM=lk57(y&%*h#iJ
zEuUg9pIj`@WkChB3U?t1dF{Kz*lGcK0?&M%CUmGRh|rOhuNe9@FWS$YR$+=$TSU~j
zT2-$u^oHp8Y<p5$2>%qh)g5m)LL}8RD3}c%)eTfTlKWqqJP{aADRM^S61;2WFTaEN
z?S$m%wj*^blx9c@ArNHc%9lIengKhKExGz4<p?DP4G%w!_Te=cxVqd;E^Xg?NVBz`
zb?@^&wE{IW7dAqFt3^hrUP!)y@U7zH6QU)#yG?04zXbMwKrZFH8}|-IF}bdsigVeh
zaG|pKA=$jaMwAXb2l`G;1`Z9OT_=;lUHVMt9bU@c7-T&H*}RQdcmC%H(9d)8mlEW0
z1k!<Na{U_SY$C%gp`|Il>iD^_-7^Z(sakuEQGV`X^e8ALJyd?*=)QJiYgw00MRFSD
zi#t!W<u-`QyL1#MNE|-99Sm~+@aRnB`gt36o7%7LPp!j;)y1cH*7M^8^f%41&Cc!F
zzK}y@#tRApY3ZR2{kR*?iDx0p5G^cO@dCeb2CQ#Hylh5;q;<z@C7tVg9MabtuI;0i
zTMRjIrS%F=V!TEzF@9qC=#iFh5#B$%d}&X)4>hIJ+QJJId_uh|<Dx0j>#2Zo-yZ|B
zRU{Kg&d9@u$5&uUsWT(fq?0!bSQC(y(0N9pK%cujky=7*Rbhc%Von7_I+$}wKydA%
zjZ32~h|uYLj0uY89R`XV^|Yr-@@526jRvXu=1RAPt%H9ViQ70&7`wLVLhpmUng1df
z&-D9CYk{azeV)zq2#AM$1#>I3R*T_RT%PKr&0FEP3bH;VPxc-0;TZ~vPAg&i_(TBo
zVLTvNJBP}6AGV5)n;%F}JDmV2lvUm98uQ!kT6NCrlVmMw7PlS2hg!8H8%;nqW(I;8
z3f2{?HpP$rh`CVlV-E5znqk9@))$#D5ly8a)ZF|_0$VHv)5gPbo6ibx@eUFvUEykG
zk4VphJQNSA*39OLxjnhss`bPcYJ8LeMVP^yi+ZX0pwpEGX<<1qJjgDGU8&{PsUI8y
zN+2usq;$5oj|ITb*n=btb2AhY$OY^z2j{Gc$kZ+GKt^%eYP~<bUWx1t(Y?WR*%OBU
zYkd<<dvbBlJ;V?yXvVaObr^$F1=UJFs5|@yveDELiH2;`?pfy`IQCpY8up_rkXYSn
zsr2D!dmWp`_VKX*=&`t4H7nDlJBaIjGH$;L>30@Ngc)PvyAM!I@PM85V7d07^PvRL
zFPL>ZEeG<>aD*wk2=QIuY`!}>?Xu-ylvYN%80Ry;1SgI>At_5m0@xM2BV-dkM^xQy
zqK`U<cSePy_CN;lsmFJ^3HP-(_CT4VHk_{l^d!r-u6IXbfb^<VFIZh6w;YlE2skV;
zHDz3ex86r>DIEr{EnoL=z)!3k3Y2p42}KP`sl(5M@_w4vFh{#=Z7#B&XiKAeBEW+V
z_()5NX47m()k{eh{j1;?gM2s@$a&0lC7_ly+iBA+PYud5&vZb*u|C&oJbvxSWC!eY
zT9^HJ8R9-D08u(i`=q@N7oW>VKl1oxBS?V}py9Dlp(4(dr0Xo%2bhvP=qt2oo0!w*
z(eIQ-l&JeKrZ)$|Q*D&sDX}ckiKGsLJ@t_c=PgUUTM{1uQNBUq_-#EIozD@dSdyM`
zTekoCV*yZhl22n&bTd8W!`ZAv%dU0h8%F~Nf8)p|d@xy^;JVRh>k88(az16K6oHHY
zwM|Ew<u@icLv)c&GOB|Dhw<ZaB#sLCT+0_y+-)M)cU}7&0s2~kh_cO0FuS)D3_eSc
zx^<h|AUf<igMo@HupO5k3s5*$X^CO|YHoW|T#0r5_^DGVRUh*F7h<RF-g9KjtN>c3
z7S2?Q$1eGjs5__IA)4)nyruFR4HdWuQ<Ngd*tIPV!{PUY8g+_gV%ir(Pk`dHPX<j4
zAS-zsoT;i#LmxoLybg19q>i<|^JF>23i@*Llh3WwEo7{v<At18Yc$J+HfMd;VlHM&
z@eBr3nusq}Apdlq?tOLGO=EdXisr8i-~z{SMI6sO^`Dg%YV{A-)__h@<zQI3zOH@*
z$|}biL87V@sv=5HMdOJ#^y%^%QP?K`oL#yhTFf_IMTZY^0MmRZ>&ia|S)X1d2yhY%
zvb*dJQT!sG*+CKMBhXe)&qvA0dFs$onDx&GW+0BK5N-~MiC8T)pS6*^fDKFGN?$8f
z`^6XCo}s{|8C<K|pmved`jQ!}<W;+Ta~YPcz?KydJoW4bUYf|j8!_D=E5rjDf*SIX
zP(|n~Snf)Q$9}J+*;=XK27aD|SYu!WGb=gFCD-)kCl7?n{YP9@ZZ!3xE|+b0ymmo2
zEYWl`E&yyhPAG)2^I3ho?oj!SDCbl<DNS8J5p6K8Y&A?78zTvvqy|-zm*A|J__wIs
zd#%?83Wq7Y&TV?ZWkBH~CvRv(7hGTcT^ef!oJ$o@jaDr>NwHLQ;nOJ}=$r^5dk-Hf
zRM!SdW#Tn9dLyH737Ch7XaxCxOn->>@%R|~CjP@952}VWaY{k?mwRm_I>7*AQF=_Q
z6m;WM;-$b=mAz8)7O*I(0uSa*Gz{(}+k_ew9S687iA5GGz#uAG=Cqj>VYR_lm%SYd
z>a9LNanc~?{ZFDRUyqK?=#)drTWJm~`{6!RmX50#;2LjAP;BZTqUwZKlh1VH^Zyaw
zTY$$D3b#ll7eIyLTZd*gaqPrp4T8f@1nm6OOc?T}y~brE?If*Vw?V_qBE#e>?|9<u
z+IUI@)PXb<4Rh+1TEf*l7=qugz4dFs!at0UU7YR~a$Zc(ly4HmE;LjUUp*|=o@Lw*
z)s6UrScrMKM5Mv4)xIDq;-I3)ts>^67^u2@lf^FZa}ZI^0OD)BC2pmc9Wn8<;o*Rc
z7UokE>Qu}~-wo=0N2J2Z#r|i&-FHKxdDfF2G2heu?|@2RKU{yG32lytJ;k{V<rkD1
zT92#Pxuz7ZF~mkXujLwIzl+-I_I%h}d3&LDtgaJc2=TjVAVtVG1xIhK;Z{2=kfp^Q
z{2M2H2u{<3UYA`OxV-M=Qc-DIQW_6pU*BWMt)GBQ^)^d%;5>U6bn7-Q5OeoDB4^Y7
zGswg5bFedZO@zg-v5}nL_MvWS+}%g_puG43h4&vDJP<F;9VUwAhnEt7iSH`xUd|5;
zMpurdpy{EmVkpFejO+vR^<p$-@O`ImxDh?zPBIV*W(qYcr|5@g_=05lNovRRAW~#>
zKGxJhq3RUM)|xmO_}a|wS6v8Z+7mEmJ6s!18(jVdtQTVgCDxc}<i|+qMUS@d(Pv~D
z?i>aN16<XTe_`IsmzY7nq5??4+EY6QplFan)294!v$80g=m7{c$2(){Cy8c}xT<^w
z`#@lK(p~*`@i)7tB@QOIEq}tL(L)DH>k^9#UI?s}Px$8Gk{nC(N7=X)V3<6C`Kqv2
zB7OcF=q2|fmD(?7P_^K}r6i)ofjL!hq={;=3!!I2h-Mb1I`b@<Zd|qLveQpYD?b&Z
zqg-jh_49u8yLQQv0~CpNGoSbAK;W0c8&Kmg351>UnM2m70KEbQC7RXt#>&gpNZwtX
zDeX_Tm^4WayE7#>xH=x*@DI0|6Aad{e-NF0a~kv%^Q$c8--CQLN%t@ma!nRaeIvQ@
zmjyVHtCU|UAOsr78|=QNLmNz2qnGOZAx}CYG(5_FVdU81-O6@Mc^rD#u|uM6?sFUE
z<xpXza9L}JJjExc7}}~}#g%>yFPVW(<b^05DU#>ZxM%OO3BlqMfwsi~<Z#t|3>+(v
z>qLPU1Y=3{;`z{wHUr;Qf#No%hNhAC)1H_Cc^1RPj7TgKVvU0Qh{t0v6qXi(6pz8w
zH@sZI&r~4m(}DVsN68VP{E(n)uTX<_<WWPTow6x#jiOvuWZ%1q$}6C#$u;Kp$2+2N
z-HLFcP<{}O^R_08?_lIxO9U@lHg^ve_I8?A3&_0cK<UXBx^o}u9pq}ib?XHw$Bs-`
zK=57;CSP+)IxW06RA8w!y6Tbi=+o&#@Cu5kTlGv<rveoK5!S%tLNd4!&+^#DCOE>m
zeOUGZ|GRG&>XyIyxkynERVNE$@|6S)Wod87K%`})tCg63Qf)706AC3KYa(LgErwY?
zeZT%pN<~*cvb?-B)nG!<0LP?lJA^F=LRV9sG2<<DEkU4`)ZVk{?Ny+Dt;&=g(5cCA
z_Kg`v$vR~e>}b7J3}!IEyeb*U_Vm?Ql)31F2dk$9DPpTqzxh&tCY-1B<WG_~I2AJL
zP})~{u$tmHL`0wtQCb>siZ*xlL3M!X@n2TYrn%DnrQI~xOJ7|Jfb;fovzMkra41Bl
z51MkUxu=<d>{5ukX!EJssU)dL!9{WTcEm$??3O(=S`)?&+obo`lPId)vD7lKO-zqG
z8#nRrieR+!a`;5yCkWv;Kh%Ul$HQuBRCUyYBbf}jB~NJ?i+Q^D@)pZxc!W}@R3M|a
ziC+}4J)ry7dA`04sBL-6D=5cUdV|)T;1mEIY^N~8=lH%3xPYp<F9f;SN>dEM`PEw}
zSg(92#q0&5?-kHrk+*zVCCST}^hw5Y@+arLkh^Z8G(y2kr!H#=*MFlMPRbpN_pS|n
z<pGM)x@r6n@)XEhl2jA_)pq*<O6$5Z0fH?1qhGRF|1PKUY9}76%+dR233!)2y*x6>
zi4&6la;^T301SY)J1(m*I1Z=3gd0EpH=Ke;KNesfxZ!{c!2a*w<NU$(C!+01njU=c
zfAStGh_?OE>UWJ_<z0S#^QbIn?ih?a*s0(iRJ@?(?{bzaDN=B_F8HYh2jl;L^NAmr
zQCogZBjW#rC_CyIfM@?*>i;g~ze|mG0{=bL|2-SqfB65i0a^LO6;xHrL#8*enxPL=
zu*F8t|G_XyV8F!6PG!&P;(6y{?t_*>ef2BW3q>XTi&B_=FL@p4AjE5?<9404|Ffuj
z$UqLs+TjljL|t<ACU!b6>Pw-@>p1=(p_C&jj-`nNxA4CWRcb`?D9nx*>hOs6_H!_%
zR^9b2|4`o@IMiWTO}&Tq_^w^&q;LLtO|H7$TBXJ;Q0wHc&oclCg}94?{?GA^@S78@
z6P16=eye1Qxx^LL$Q*0ho-3z_<4OKLL(+#3_+}=MAFl^{=6hmsiqEN|B}JF^Ijk*@
zkxs^vO*|m2yY`E3aJ>ZlSo$7Vj??(Ao?OvYlV)DYakgpX$HEq$b4-X{`GZ?lzoa_|
z6@|1=4rx|h!G|1tJ80z8(LcKy+#NKuz-L_+VjTNG{I?%r^X0^B6<PExi)7whL(7nu
z+aHthYzkW2C{nb)y^MDF+>BAmuGEr_?1Gz)x(_6~g%Xci)HY}R_Hvp^IWq`uSpNB-
zc2JCo#v(H^`L`b-87D|ReFPiJF`Ql!<Y_k?Quy1Cn3i(k?OM&DnpmwlTT@mybn<_&
ztDUuu7rQoR#(yky?PI1q^p*3*Z?C67o&D9Wm7atjVM5fY5=<O*UNN}@xMmHVR#z2-
z{Ct)&K!JCwUGME#;u*%+E*A90@C;99TQc!#=E<nzH5p-39|<!<IMMh;aSr6I(f{AR
zhueRVL9iaN?3lk%8w`(_$dF^I*Ngq_N9<FZxt_nWl=>qqV@ZQO=#9(J?>}N|w9Izu
z8|y8L+E$AfO4gs1fBTY9{ai1z9a=O`9d7Ilawj1@_CMHFT9(t~QiW?NKNkFL)cF<E
zESrD(OT^UIWcZhU49D;kSZgpTXlQ5s_9H4b7YA(z%Ia=e_^fxnOZzF45j=JfZ|Be!
z8e9X@M_Tww{}<Co!d+O$IQ#ujWY7>*M*q3A@*k7GZ85@Hmz#37+N5q-Y^IicNn0|f
z{_QOZo6BzLi<COT{Ea8S;(FTDL5qOj{*cM|mcFUD>dz53vQ_Dur`H;YmnNWmRG*?*
zH1adkZ~-b_DAT!@M1{W!`;UG=bE!X96(aCQ^6p>n`RbffLG63w61Bkn@jd-yoxDL~
zLV;h3U3cq1=_b)3T+nH)6skf>*``nadtLtXho~%={WVx$x3kA{1w9U~8C%odIab9L
z7d){p1;R$@)nPGfQZbixD_@2{`9!Dj!<{n5PDtX?pvbc&r(yc4r~<U^NnHS5ijvXD
zfo1C(cu##KtOFj(`{M^3bSS(G=Y@w|-Q8ZHr&m8Ld=<zJ5xT0HB>QnwmI2U^%z%VY
zEQKaNf@p|Ia5l*a#k7R#PP0S31eKe0T2t@N0+v@IeLeRV*^MY3ShyieD*T~ggX=wy
z$Gbf!Pp_peBs!ROUAU^LlMLru_=YR=Z5bqaN8dvQZg(GGPz`MH^#?)!CmrmN1gSL_
zeolDteX$5vJ~ftxTrGFtbSRc7OU-Z!J=Y2K(zuqoUAEro@?@;BXW`qxD&VWUf+>2m
zjys(Vh>MKkGOhyZEgMaq){ODkaGi0hq1DI~p%{%G!0menYdZ4?LUp?ps-mHeUzfXG
zgb*lCNy~MQcD)E$QfTn7AfVgRpb1lc{SdR9Vv5lK4@#U-k0uQ?WfZIBi!e`{0Dw%L
zQm~}*$+TM#nnDUxG`yDgiO4UX*?Z))TE~@O%$s{}9>mug8w^9?+Q)VGzH+F)Q$PXV
z3yL)td2OaT8W(ccTGMp;K{Mu(#GLXRhOr+Y#_m9xF9cnhB&CO+_jt}hkLDQkmpl;C
zvNh5qcT~Ib<=Oh!+oAV6bFG@v?}3<!2|tv4XMqL1oPYl72|)X3pk5sUwkbh9#8W@9
zqXu7CPgv_ZU|=QaZ?@p`n_dhpj>#tSFx*_dHoBQ-)WMUbz8rUAVin3t{m}bme9XvN
zupo2ndR1@)lr}w&O6=(Yj(G+GI*PA$Ap6muQ76bqyj_O@1X&rB!TL~RlOK?BfwFyo
zp>zLswLP)5uJ}8=->DW&7pKly)R(vlsC4ZusJePX+a0riJ|Bu?e!US?ZHq%t$?>{D
z%yS+=G*MITPMwQ_vR+`HJw@XK1OA!f-DUuHDKpOQBu!Gu^q{yGau$`fwC#OQ3-oQw
z*ZH>H3r!kZAUZPWp!e5kprXbLC{f3AJ45BNf<iy+j!`5Oa6}%_MP7Mwn`aq1w{{)S
zfs${DFS3B8EElCTJ_(IY^}Ei1d8i0s)6rfVjq^1vT#a3s1lX^x)(CWwJtG7hnuR*A
z;`?x|$LQD}6(##+mE(#vO&#cxbJwvyXg8+F=F67G>r~@g4DQ0^6j6$SzJwY4j@aIZ
z=DJFBCCq_la5O~SbRz(XwF2*`e2EMSsHZ3lK8C6C<iJ{q6**)3pl-?HBLtAgz#C=L
z07X7s;IPb+r3pQ%ck8$J>W+0UzY=G8K~${U5(c9wMf=M-0B%}l8j%BgQVK10S+(d*
zA=LAQ${{U+S)coE0XX<lgf)Op<py-!*b4zY@ou-2n+N-XECq!|buC}p;4s|!_<(*W
zSTW|#Q$bftMCpUM7SI<nMzBWG9`=F9@p?ivy>S1T!45dkZ1=*<%DLf4^t;pgxI<Hh
zcWrrE^J(auhN^L<{6NM)rH@dXU**ntW!nHtAi>uI%wzU=F0uGLSx%piP4sm5bl}u_
z=;0ynZWC@zFo*Q0X6+YlXRiA)&S7~frls14AAuIe+CC1!`SXC&fEl@p%FLw0;lMUu
zHcQm32G_3al?H`eRV@$ZsiJjWRa%22ABnoHNBGu;e~k=4c+{+7P%$pY+PVXVYsBve
z4F<^eAx`~J|IR1(!mXNfu`p{v4<ld=?;NDv^wH2l<;y$RSx?Q4cl)Pxne&#QX(PCJ
zgsi>In=ZJ#I+-`)E4w)KT@WTqcYA_?_3>XkX`2qZToEvfl@StJk=|c-&Xeg?s1@xK
zVE{#IAfTr_wKIEufcV2fEIU(RxFZq1`zU@21j4Vsze~nvFpw~JE$Up<SioRAe>gcL
za!E~mx-%?O(0mAUQjX<jV7N(axb;u%k4b7-@x7JASKC%ERv`2Q#z=mz;b|aYzu#Tx
zR6jZlT~-Dk{o!z;EHqLlvWx!7jka&G$0w>{vePYYV^J@XRNa;#6UJrI_kVBHWMRx(
zV}jcmcu6EUl!$DwQu?d?c#@)JOBEJo?mormNj94|y7&_Pp+^?!M%&FCyk6}W?wVdx
z2kj9`Xl{BVaMu0_-MB~q8?gjwI<IPXfpt9%rjRj<3{1GTa*{S&57Z7!=acavVji5+
zZ!3#qXMY&*HWF;G!YyY!KLV1~K$ool*gO|kC<){vc?gp+!`PgKW}92tT}#6j6)d3}
ziE=56*I5Jt-@c;4BsQZS*$CsHxe@{nnHd38&YLKG;JB(MbC%Mb8WusYGzMwme?&sV
zi!T|9u(_#{EUfa%CS#zu<GbCc7UY}Gw?LI&W`hQaE4R@QUfNgo8o6&2?7DwhPPC>a
zF98y${2*hZ^aKr5rp*AmLe?J6XTuT^b362^l?5Xu3Lvy$5RBeBBU*Cr%TqoY1@<B=
zy@CBOoZ~fdGC<v35+3_5h}tRrhSc7-4AGldhouP$goqxL?wm$GgT`^@T3QhRAxH!R
zpP;7u#<=E|2V-*1JDIs(bTvbe9P^2QhBDlYVwY>x*@u$_G^Ls*r{!`N6%6Z0@~pO&
zUrEP2;Z&;>z6r+)aD}z*fCI@{xQgHkUjeDZW<F5ar(~V5Iv=@3g(h0kgzkTj${kGS
zwH!GLcw`SE0zLYR6;KdAtz4JP3tAZJ^`msbx$$05`5>)%I4GmS!+zeM+WC|Z5}Q#f
zXBem0tYYD&NUkcDVzH37E6g{I$&)TIY&_D#1hszDM&$~VvFxhqktx)-rmWU0Fdt3Z
zFg8g1^y8KY`6XX2Qnl(e0IsYf`8bOV^2P2ed~}0<(fkq%US@-Y`J!9AA%52GFV{1@
z>~gfglu9mlnAEVA#9?zTj%=dpVGTGs<@{-oEjx-KR4DPQkw+bOt3fq$X!)WzO(e-?
zx+@Ku<IU7{F|p=tiPf((7K7WWr-V^EyR)Oi7JR}u?=*s?u-VS^gn&0l7#fhYCV*(K
zz{8%!6>O>$NqgJP;gM)xmMjaq1PE`6>gO%4TaQ;I3-Fv8mtL5ypdtJGo``ZV1X`&c
zgK6pQlCrc<L{Ot;fM<0B^QW>-=VLcw>Qmy(nQ{XH4AGiJ&S~Ib6s@uNUrKoK4N{|x
zwz`XFc(h~=Z1jKrY}PB(#nL$yoM7YO<jm>vQKRM=1O?3IJn>W%YIb16PC11WHRyhb
zWgla84a|gAB)<tv+gA6EbQig7j42kyw|B#+z&mKnlx<>$BPiQ0eAxUz^R?;%X3Qq@
zqmU`wGk0)$h4g^EFcE2q>QKJ__E{eUQS@WCnV$sivsg{sL#lKpy+qnaO(S8`K@YVk
z6UP4xScf4WO%F*vOlRXrd$#GuQg#tDF)Y3S(!7X`El2ngp19*w)n~p#7)nNw%8qx~
z!E6A0UtfDH7vJe$zDa0ezf!`TvZWg1V5MnQ+iFv@=hmVM3$)*P%wSyMWXj7L19i{{
z^e&VPW>J4bRN@HDJT~a<k<@4pR>4kMqX@*RrD|6Vz>KJAoLBDNaeyYVPVY0aWWET>
zgydX<`akHiX^@c7YC=b{Q2L=dH1?~_lhT}|S^=(w5;a`CPI1O76*Ao5<3X`tD`W=a
zkt*5YZ=;Cz$&0c~W<rFkVC~tx*84VDJ+GNLMV|BH=URZRZG+_Qe47#h#Jj^2il)59
z;N3N}6pyPf2mI1X4GBTUK(-|l&#b!MJom6Bnx>4W1?%u2evPZccF@PHSV1T$nE^J;
zxdVU=WrkNt+yTd~s0|P?#vus+EQu&a`cU&V*Xz2|1%Hb7>Zp4trKh_~A2S%lXbmP{
zGgq&Op#lzn!*-+9gqPxIwH^e02J$>=%MOiq#L5uXYeJuV6ao^4tZ^YtQWT_WV-otb
zp17cx`W$z_AVA<hDX}DO2Kdp6iLNhAj$gk85E!98G;)Kz6f?=DQ~u|3V(LEV^|SgE
zfHm|J_mdaZ!=#cZD7}?o$X!H89V_uygRk^3jx8WmsmfKc_m{};I>3e`?%!#_Poyql
ziYI-Ys9f@wx@&lwidNkiza86Wnjwvqxovs`+!lf=yz9ShT?{FTlr{o8Jmy?{!Uq^|
zYbx09vb=lbgU}&Qwx0&zFOwuJ#=D?DA`mitYz2+gK%QWGr{==;Wbr!(sW{1SkDmaM
zp%$=TpwG&f|0b`fE^p;Ll@DPYax|ro4pFQcVV**#*9XG!g81Fv;eXUEXD$&puMY#t
z#T|{GCjU^nKl~qbRfvD*BZP+wF^K;-R{OxH-%srH3p5?g{i#t#oaFPn4hjHXAj~GR
z8{0S(@FgF3v0(b?zlf|-n{zaM{4jQ&4=;=Zuo=`fhFyv+aODNT73=NqSR8oCjU8(d
zKa$aPkh(ARbd&gBfft*kdtv4<aa=S5O(DSZl?ZLw@B6<ExD}-b=yi8E_X)NcL~$^c
zT=dbthNwd~IE*S--CB79_D#Ox3cbJUYbxj}oA6^g<gL@I&PZrXLu$<Avpm_(1^OM&
zwe#krtV^pT`4A5KPW5m*+z-a*o{~asv-e8fkBLk|^7aCu(xz-p2F&kLj8#ot<ADg=
z7fD-sOs;J$Otf8L0mt*s5>UM5D3acK7~M`6VXKW)gv!z&?%`msY)nv0;hxT)_juU|
zLEdo`4Wjmc%zXAwg9Po-RQ{Krqt(`GevOEcRPKul<M`mu;`1kRhCiM)8<QW@s;dh&
zq!F7)tvv7G^TH__yWA}VYJ?qqSKIE;<t^(9U%0`IhK}!S0j~l0kUN1je-5%MBlWS)
zXuX!#;A7%iO(#|$Q;mk$*{`L_?`;^8Ib8x^Bnsj^r6sL<s!Xo&fYVT6dL}>NxFywP
z10E}3+}dKUimn1QweG+L4ZZdj+3ME13%TYWldCKBD-DrSs>lOy>fT`Wvg(rs)pIRh
zTzj&Y<ZUcWXR{7LVqUHHW}_4M6#4+8z~Ivj;o`G9ZlP$>uod;%Yp<^oQRu2TuMQ?8
zX8=7wIhbojrolyo-vM8!Zq@M%GFt08KpG|fy4`}0NWSP<>{LhO?Vs$JWI3G~WK1K5
z9fH)i4F+oaAYfzdvUu|#_R{sM*!5*!qIfw`mb@lNu(Y{`^&iG;RuXRy<lZqbdYwG6
z(Qugx&_V4dYmD{W2N7@)Wg1hj(9m-I<i+6B>I=y}{BxD$(4dEr_t^aedgLF|W&wLF
z#uCnH)S8N_5gt+Bv^|+gDfp``ko`WFXn`yoxNprC?+=@qdx@#Y#aS@AcdA+b44e{N
z9OlRhuZx`a(}6797XXEfbAoM|ahvSwpbKleFy4}qBPOW-Q*YNLK3cuY^w?i>g;J0b
zdL5FRjNkl`IPSVHa_hWg&PLOZ0tHQjCGI5~&!DkPmC7l8jF+6x#20^YSxCRF<x~7W
zzX6%1m=EfZ4WvPcu0UOfy+s!W_#S>|h*A}(X*dn!>g(zFW(O)`lF)!%uGTkL9!k2f
zhu6)RB75Oo>9ZKOKACjp?k9yz(d_Sni}xmEPzh!1T8N{=EFUs~Sr`u$)c{E4X*1Bu
zJJ}m4Dz#bI3UF~W<>XmHv6mOX?V)c}<b?q1;c--oG=uNXUV9{Tsa+enxe$kDrf6>n
zv`y0Uov;Fo--txO3HJQ~7UmfCUn9LVDEeK@N*%|qG@VfI=_M2N?qAE7$*p4$|4K?X
zk+1YoBjFjikW#oE>XsM)W?IDu6~B2o9nfisfNuzD=W#|HYnTG%B!u!OjR9kD?q`-U
zC2Qe|b{b&|)DtLC>~-wvM;A$pcFaH%Djc!q&s;}o4JlaO%31+%ma-OcSgDG+h*f7y
zM=ON2n)x>dAFrCMZo<EAcc)sf(EMgH5x-t5-Byjs&0igq3@S|=)LX{4DHr35AGt=B
z)dv@-<%(sO>3g5BQO=iBBo}f8pfWeF)!0c2uPT<Sn8+F#3VZ_&Q4uib>G$R2{$jxo
z?1``)oadfnWeuEawenN$3+N1@Zz3s|>H**24qKhZZ3jt=d!f_1PxrI1U-j)SOG2d4
zU#fc?&(&VOaEO0#tf+6sb6?PDGt9GlJ`KxC>Dw<1+2Xn6qRq9w0hqvL&jKol*}Rt-
zzeI9`<u~D_!$KtR4MME0tA6*2B9-|as3d+RxKi=j;lPrFyD1mk&=IhQUgT`=a@1H#
zI4h_uB;3XGepzIgXM~r&T}+REax!UWzNl)Jx{?(t+4o$?`+8D;Iz#Sb@=QOWsPAcN
zK;U_G6*!R5Axs$x^v{jV_b2P(N18BC<qHgdt86Mm_>sBn`(p7YZe`BHMbh?JWxUy*
z^GoW`10}a{ML{-@I=o$3Xpl;fG-ydyX97H!85Fm~vQ0YUecJF#Rr`;ITt+codA&T|
z%DGn2&s67Wiq=^~vI=}YH(yImHc=$;G5PWo0pixst%!&%M+mw}h_J^}dGYoQ-hK-j
zpYuXJJb0QzM?=^NHC$Y5C#;q%VxB8MoTBJ^w>y~mcYDe+%{?jF%R+*xZyI767R+Wp
z*987<4}yH5z9ns(qAp;_a*A7B&3t6;w;Od`lyg=Sjw(E@SEQbg_N6>Ow8Y^#eCF4+
zI0>RoSIV4E9TENCO}g!(msdh@k5u6cYAL&1j)H~><9-iXMYuS4dx0MK|IlBbsn@6N
z<DwIge{-cG(YVi4%<8w7^YVfwZM0Ylow&u%I(}m{r>`Ery%vI$R~fQ+HedV*Q*OU~
zqUCh9%x||&T@G(u&Ej2}{So&1-hFBpp|`5P{Rq~g)6}X|SfvNGUJaQZ<5Fsv-|i!~
z(27>DZb=g_Nwmbuci5(DNoCj)<KIET{Ss+#abzdrX0`APWpU0yFLKU{xioB198N!*
zWh<EDl4Rl;aXPfyn}77f9tiMMR&GW3{r&&{Ypw4q^t9J^mz6&kP<hjk+azb!KAI5p
zyXSjlDlaXq_pEQgkOZfEX}Er}!f(CAr!%fcgs6|D6bbROCtitNVI4E}HBP18dVS^U
zTGy&Z^esx8adMq<XO|?lEgqYR<zdI7@RrRfp+)WWye$pGmfL(4GV^sEBa0{TvPX29
zB^cn#Z(f)b#9uq^1CDcLl`E4~Qf1lEMN&NyZfwIw#_#HmG5(qGvB-2;=O30a#EW3-
z$gW*mTd{pQoBz?7nCNu)M6G9cdFFY?33tM;&z~OkK6^$^`2L@XZwa&ynV%RV2z+!-
z?wTuU(Bq40Cz%KdX#TopLeRDAnF5pa=Xjb*nM9dVlB18$J$>pKbSCDg%JH+S+z<Q;
zmcwc$!UU$fuFUfprlk*eUYQ<TCaM#v%cxb>U#t=8{Y820`wfENjvVentMwD=ZS`jR
zj7S{{o%)j#jql)<Rgh)yCK$gz@nB#Ee+qSEkE?#r)>&d6Y_jd$!T9&>uk$xUR(|Fy
zv)MPgzaLk@$EUzS5_Tat<*+8phHr5!)q85PnO+z=&nW$HqRQ;1)F~3`K0o?~6hD^*
z8+7^Qqw)w>T`znGr%s}@+09!1*~LbYf9s3*%_X)61@&f4H#4*+M*MCzXYBvQ0`78s
zOUSWAbYZ{&@2jH>30gLK>R0BN#r+dqRy0iAkK`LRoHta)CwWGnbovi(0lR<>d?h_g
zk)9&O{FBeaoE0~Q=UGW}H|_hy%A4p4I4^op%8&5k>(`+(_uU6R&H7UZegX76*-G=I
zZ^Z>JL79=<=Np5cR=!WAuv#9*)7#tr!262AI#PMsQ#gZP@HejxCT^6(uY9Kv^6eJN
zOMhuq;Z1Q&F><5X_S@6?XJ1Wa@YVqBydh&$n!fQMKYlGAQoT{SBJZND=8|WLHa*F*
z{m~9mthy%tzZ#^Y<mj7vN9^#SsDGWC=@U=rqt)-CE^@FJ5o@6oFbL0Cd!Jpt_e6!I
zrG`I#0~!`+<S9H$lKu^TyiBjwvu-0maj0-sORU9iT2i)e1gmdpDfy4j2td!c`3C<^
z(98A2$S3qi>u;Sq?KD@!mtI?WY`nB(=$nW4Rq}SfRf+3(UJ~wnR0AAmUS<P9ytRwE
z?AlHyNaC`bscEIVDbSU5$8ioDYNN~I{||<O{yJ>fvG^f;NWsBlN(J^-jvF1%X1|}P
zJ7TyQ?4m5Cw)$Gg`=fuw5yR6*@Gk%?5WNgM%P)5bkNldDmu045J;KL7MU5lTt~(Pu
z5aKWs>c;u}pTK)3nWNI;?%$oqe>^qWO6WXU`Ma2!7W*z%>LMqJZrd*^LetmqJ{~R8
zCAe4brL*v()9-g&^IWhJj%z>Gn+<B)R87ev*DDna7ZacVLu~Q><o8y8nRXRCV=P=j
zC}>sBs?&R8t8K}V>~x-S{)3;TS&{R2-U_)b*GAYM8BbMwn)MO)Vkgf?ZynbB>?BOr
zTI@}>Ege&naSAT{Z$@V>2+oU1l`5VJ?oslr;uo5^LQ{@CLgsZAU3c0`o3Cj2uRC-7
z!yOojMb8*h#iLq3N*+{fdRp}~jgaZxWS%TL%-i9PU`inT`LE0U?;q$r;R)$~Rtw^}
zEm!9~lIvr;XQLgQyQgiupXe8E(B<4K*oe#d&8GxyOcnJ-4{w}PNSRt5=X|0np$bX-
zgzDbH&;Ky=PG~O;2kpq-vLL*Tb6p~=%^hl)eG<t^IrCOlC|xgKXS{`YrGORhUqrTQ
z;T?RGhL;DfdOPf+@d5tXMwbe)+FU+`q-w#Xa&NlQn8=?jXy|Yv7@xd-`%$UzsMEwg
zN%&S{bx-a3^X}4bLfGZ6yXT3!&RKVEZt0dE{)aUlu>wn8aQf^zzSA*3z~pdG#Bs$d
zvVulK1zWKmCwZDo(Rocur*wZ&(y%YR;IB^$5QPPLa9uAN@Awu=6OdsWN@gc%uuHz#
zPIK5RI=<NjQ@$<#@QkDM{_s`oA~hQLyMArYW<Mry_HD3>Vcj7{#WGR<@>0d&ZytAO
z#fa9tSn>Y#?FEWQYwhjjv~z)?v<b!&`=!pxGrbpZ2>W)A=9p>6>`U7opgFTO>dkhr
zj=ZYC?Anx|i9PYKiT8B_{%3Z=6>YffWVrL%M2W(kGdMZHLSxm~`Xa4lAM1-qCoFB+
zSWKXLm-pcKIj}A)e%6w>R^z1tESHLD0?ylIp(jzUB~v+m%}wzRtlKJNG*tZ*0snN2
z!X^&S#)%xdKXFN=bpvO;Q(+f;$y|52GQi({G`9@*#<dcbQt-3O6G=N2wOv&Xz`rY>
zLmebM31+O3Wq^3+U&CU)+JArI<`*Mew*Bv|{nX$Qs^RV8snqrH&QRwGSSbMqMx2#W
zAlS{hMDNxz#`#c*AcpK^%`KYi`~G|7|NcYF@$ap&@ekXW&?f$8{i_+H$ThlJ+dx@8
z<zS^=ipmNuuN>k2ldQ1JwmP`+eQ^zeN3X_i0>bz+XM*L=derVfXQFkfriK5AKP_E$
z$4~7KP!VR!Z|u#mBOu&V6lR%c_|HrH=WQ<S0bAdgMY{6^4Vd5s`+N9!T@N0p)FU|<
z8;$#4xgLilymalQ5l)UFe^i<%<@(8RK$T*?^5MPQ{(tfZAJQIa+j>8l9wB5n51&LI
z;)Ca<Z=HcpBD|I^vh$PVuE8fAykohSfN+3)8uvtiDlzh*w~P&Oj$%2i94B#M^ob8I
zAE_z4)#5$4^K-%Ry2AB3PFO{{^Xf_O`J-%x;m>-C{YPKc<l~J<J`%nK=H^mn($2b*
z>#0L<9=KP0B%`hqm@7YfJWBt+cO5Nw6d|0iTSw2U?u-_MNeEAtBpemdKO+nqM2x%C
z;Eey>pno?A9oK(v(7!im+YbI~;!tw%uMhgy2W{u{{{{*F1_|5Y+`m!W|LZ|Q0X=XP
z7kR_{Kj&#cpCCA4yF*TmO$szg*?@h%725dPK<kfk!vmHI!Yt#7ac%!eE&f;R>-rfY
zSfZ(Z)ty27(UF%B`wLb^Vkh+3fLuNY0D0PrV@<24$LjM$y-$$<IMXEz2oci&6|peF
zV(cmG03o%5#uQ%SNa#ifRZd&y2%LZJ8U_*nlN*bivtdBm$p+NKpG}#1vk_Dsc)snw
zwOy15k^XX`mEBcw&p^5&vm<^d*MkfaxBN8Ru33*BWCWizp8zihKccbvY|5~j4aD|J
z7Uh6%!3@x)VR;W#>RI^|X6(!q!C5Up-2Wp~6(?WHJSj~?!#zQ!VSASfQ1iRHrQe#a
zbv==C>^orwrIIqlW%Smc=axF*Kzi<<KY4cMCQ!IGg&b_uo4a+-&QykhHHy5;#U1E$
zMq3dE26zBRx7<8@fN$H>3*fm>Ke1}@i1?ls&j1M*XL{|<^aUPdFALd|X?a=l&OMao
zfzRH&VUO@gpqR0~mEITaG+hcHE(%~5cL&H;N?)8+e*FGQJ^2^!EpQVUS96priiJ(!
zD44{MNfS*AJv9X=a=*UNy7HE0v+lxSQ`?d^-CY^&CV5Q*r9R5{{=g)Dt<SK*IX7pd
z1tIYo^n6XdbBUt@V85}?jW<?iXTKa`UD(q5Ch7|Wr`!PQ(CI6E!O4f|54tJL!YKP`
z0<U=-pb4M6sxv~kGgYI12B-6?L>>RUgb}E!w6PlS&#ko-38EQ?70;p-&r^)t_N=hi
z1ys}yc*TMyC>F>Tgz(~wO4-W@v++{~AqfpOP->YV6YYzp!QMv@&)GLE`EGk%BaXkU
zv1g#Vir0yN9ZidyJ8Lbxa}^fc(CZPOa!9}EO$(GTo`{${4UmUi#Oo4K32;#qSD7Vo
zE8YfhPvWf4+Gy>3kfj*B)_8Ino#Bt}N+8-MMMPVPaJjKg#$LiQe_)s{W#hrNg(u-U
zH*S*b+L__Ia=`j4>w60*`@4$9N`{}>hRgv1cXy|DnDB9U&VdY(oo5O~a9bo$b&<!Z
zn7E!t&$)-5^H!8tVBtH$xM}}*fAdD?m;=ExIAPCH-{6nvkqZm!)a|<LhI2upGSHKx
z(3ALSLygjJB59K}fEf&9It2)M;rLqxH2gB)eiZJh+pamzop)-#r6^jY4``8U0Xsx>
z9FV2X=vYkh4`~pTz`nVa4d30ltSC{r4ad~@F(w?RK+!CT9u<ooHB}jKNwyEo*8sFN
zyHl)FN-GI&dZzX~uD-<4GpRxFR?Ww2b#|W6Gv6cN3^a@DNHyykiveHj2_PnRp`ipk
z^iRoN9s&$JOHINA->&hVhf{FlWh@S811$8(k_<s~oJuGVdC~%QI8IrsBgZVQ>SAaq
zLimC?-g?t*!eP+(VJK8>ko0AZxhw|9oWaYDJc2R`@tHSx!ZW1OM0KYgOjNxgzoelB
z%&1-floi^XW8F$$DPzcatfufeiWd*0GjbPJPk@R_b?4m@4unEUMtx(eKPTHOQ7u4I
zrUAeB3eXC_zW4cH(z)NNAn{X*z}YxK`(RHp;HtQRrt}Ak%3RYx`m254mvBzfBjGA2
zBnB{PW$kS2xuzeachjehgRRs)pHz(Iyz1YH0FKM+)1hEnPE2Cy=L{>{Ga-E)KI#I|
z8)h8K<pt-o?Zfu+++;<95@mqC4R`K+l}sNT$ng-5(3|iU18m@X*9_yG&knc^$BL3|
z`-4WeDO;rk7UdEF)Hh1T+UsGieBjBw8l2N3994=+QB?e@J3EcSueVqD2U_7D)UH5$
z#s)Z_$SEd`bos&bS9`u%<F5?Zf&l(USvzUPT+{cXm5D(s^PZ0YF;XMsYNv9(ZbG%*
zg+n2exIGrvSE%-Dx62rZzD@5KoEp(vaJ)<ZvLTcN_C*98&C9T@QV6WMMS8g%Z!q)%
za@cdbehTiqy4B~R$RoT1$|DkA_)YCc;RG7}v(le^mBbJnh_|EgAKqet#U1$LlM~M3
zilD{4w7s~U$Dpg#t<F+oI*3;S7uH%qw!`#3lqGi<y2`Wg1|t7!L1Ft>I|tJ@6sC68
z|B!)j11qn;;&u<$7*Zh<cK+WV{{O^MGZ^aB0mM(|YY1E0Zi-7u+p0j^FrFozw-rDc
zTq9$k>WSkE_8<Mb>9E61|Ly*U*g2LYUbA0Vf?*CM6{ZcEVqPFaQL!@HpYfdSdelX5
z@V5yjLv|iC@Hk>%rWXD>V}#RRmxoBJPBaG^lU<9+Hf3xTR3`w*Y6GJl40(36ga+UU
zn>c2C@G!i&$W4gP1ns<J!kx%C9GfmHk3g6}1PSLnEYL<xdG@#J))9xDQ#hBXK(fvZ
zNIX|%RokQ~&d!Maftm0<`AZk=0DRJU;Vi<PmwNx7U|zZ(YZy$7t}_e3+NfFkLsRiS
z_~AF)YPiwo9|U`laUpMAz*(yqA!Ky9-YTAQ;8@hv(Ro@dOkmz7%wCRb-gdd3gWo^-
z)&TbaRwDcM*E_OP4oEGE8x-==L$6$aBwPgBQpB_u{~3?bGYAcFdBX04Q#4%~$s4oQ
zscY|A$^v#uK5dxzZ~I1ZRyax%B416?Cphm6A^$b7Fb5xVGUKcU2lO1c92V5AGSc+n
z?9J9_Zn9tPcATFm_X4y{)FZX4XT0&-x2^=}ogNw7!nIY%qutJJYqda}g?q@<UUoL_
zqGeP}_5!xQOC|-2%}V*UW=xuej*s{bD=Ke0V3?IILwu$n$gU;vhT~&(#pJTl|6myu
zbn4)A_rqkk2MFTmay8XeDi+QKhVa=;y}>TcXTE)N(2J6bW$jeyHf`eNyVs3R4y6Bl
z`V&v9T38CDMK3P5;vAWHNQ)x8&9YB95DRV0<ZR5<WKbGV*a4OIyXmeJph0Es&N)95
zU#y^NOoEvGwPC`*$tv|r<UsnteJ%An>^T@{;Ufp(Cq75D;>0ZR^ynuH;3uXaaSP#u
zNvRvmw1_LWA7Z=*6jnFSFkXh}U}d$=MZ;6*+l64gep7rv61^nkB4%0laC6zckbdp;
z+QPPamq$=rx1L3!P!so5M3ERoX_v!vC;Ql|;$i|cgV-MmdV0lNas!l@#B;8ec$j_Z
z<0t?*5P%~n9yNjP!hY-Mv|$)z@$z*Xpl$p1$qR`rnZd+emRIii%Q?kJF{u^Uy!RG;
z9Owowek2NysMhrr^j$A07aQRqE%Zo`^<lBnse6CZA5^v?+-LhM^tl=?n*-T-Sss`q
zw_`L122PBu&l@A&%)(G{`(QPNAH{8W*k$^U5L&+4Tbs4>c1d4_l`$%n;ab|5?A|m6
z$rjB@llOW9#k-GqA-%u*u<Ba(T|t1c9439Z7rZYN-e6%gn=!%-%J&B#I0Sx(L?4ZE
zEQ61f4y4Q~Csj=_de*duX}30~w@#ysOK79_$l++g<CS!mwM*TVFy?w*R6Ok6yD#L6
zvoPNMZiEN73s8cec=1GkeCM48fB^i2$|Xt%l$N8Z{EX216ssG<LNsw<=C^7-h!d>t
zK~jshAhZ>_<@&1g(r-pB9|ODvK;uN6>S6I8ZMQkR5>+x8Vx=Naqp^sR559yyPuRlr
zht8L$2POsN-sKIZAKz1>uycvwxbFd5A$P{z2PfZSM>a4zAU;hWcm{l8#DmL$b|2oJ
z90Mem+C|tUL-JWaDA|CFAndyByMpJpth3b=(jT?4;8?Gt^cb|Ucrmv$DcacgGMr0c
zFl1l9Oqb=hhNzAE9pH9KXRk&SKO%tY(+f1%SUjCD^Gr?Q89ncAyeIp@ljlmSC3k+0
zv;lhZiVVjR+AdO74Jz2o8k4BI13-u2gHFm=Qv#)^)A{ca#lU7=(h76??G7JwZEAiq
za%B+FsRMokN+%a>!w->&in<o&^52@w3%LHV2MvlLZJ&J`g$$a%S2d|~y9`f8<EDtL
z#d_AwkAzX76j2JJkpJ&JMS_1iTi9jO>9@zbn1cf^-_OFsK-^*i|Bj!b1l&Y{ZqhqS
zhkIW`j)97osrH$5Me7C7jbj#s<6dMws!3pc2odzBfgTHDPj0UZnsjyDbeYqis3yI+
zzB62aSZ5F-99bXkwhKEq<`%NWhws*V3&5m{0wP0gd!{P7IZ?OC<P0pl$lkov>;d)=
zUsPt>+I^u=_IK~B$kin;YQC!=9GeDubmrE*ZhoL|RStP<%jG^=06E1<ygm+EZ$5=K
z$;3kD%GB>B@aV3hY~au@NHR^}Sm;2M$+~4%mPhbd43)xQWT5&dy#XDXbtT79aZLkc
z<BN(P?>U6ndFr-=tEK~LZk-lx-<e^-s$4+zA^eL_&4Ap8nxD4-BAnL&-QG44N>L;e
z8NzF206sVWZlf4j1Eeynlxz6_q-Im9wxs_DKXJ!f0I>oiJ%YA9s-j3PjvGdvR7BN}
zSF&CPmKhkbx-usu?aDWpevq*K7@qTc0SgmAuatFbXV}UPF^GP&qMS}0%*vibPA}t7
zmX{Gk3`pZ46K$!K7H|{H0_R03%wL90yLxn72$nzp(g$xq_!a^fop(*gVRsEsM+fus
zIjDj_B^nC4blOnUYuiI9Y66y~6(k)_i8`Vi3xk8lg4N7#IvkPjF~KAfT(cy^iEz$H
zpF<w-U{KP|6g289@_-{G^{p02{-KOaGfX{EZ(ujgim)2!3@}HtfQLe-9dXB`!R+R+
zL)3Mfs^0>})QP6Kmq(ye2!rRL>WOy;s^}uUV3Mz_I*<|B8#wcib`!fJnu@L6B)PII
z3DTFTpW!A90Z{0YI`CN6bEgye@EAD}k8qO-y9&OpV~C8{`fZROFc%3RlBJ+K!Gv>%
zHrk7@s`-marr)-NpG^hSFr1VS!zls5u|${k@tHOpeU11t=vZpUQ92+Uo(j-_4paYd
zSGOr%9#uM!2sN@ebBFRlzRd>(AgeSVTEoqi%LvA}cfKTKClydCSg?I$goEak-_G*W
z{k;YJ>7;xrPCGDx-b^U`!~mQUujRt!g2S|e$L~u;rF@J_!NeuQWmTu0?!bFAAH>DE
z{U-7|_qq{P+9>BFuP#x6Tp%hGO`Kz<MPikGdEHo(ZQn|WCcJW>Y&zemR-ZSJ?y<Ky
z4ez-7qFArKxD^K<6@L+}%KbP1!y5q;#%EM&Pxs{TV$#UxzE>MY;pE~|bHp|CSPt-i
z_d_tE7AnR#@1>^jTACLR#twDn@(7uyRTpu5lr!>CM0`!#l1Ia9%=~EDF$`=$n>YE{
zmp7tP%qEZ^CGubuolONIED0t?;dprce?(_SJw_)z$~r!6IC66|vaev;$<*5QZx;e%
zrP?4TwH?#2J_Ken&_AH^+j8~jx)dzWk?MorDUgG+=)1Q(Yv+PVbD}r@ldrzDp(U|l
z+if#5D5_h@gZ%i}w%h3CwW7VRdR!0>D$OzLH}6oNR2@h+sAw78`4osbQXrTd&2(=)
zveR=gARBqui0+hwAPBcs!03KJc)p%4*veV|$c^`UjTc-U36Lnm*Kj_r?r$VT;w%Za
zkEn4D9KcpXV98G(a-ZCpDMrPkg{dUYPSal)sZRriLNqL#;m_rT?Br}vc7Dv=Qe=NG
zN`~bO350SyklB3B;25W-(A&d%YUfKr74#a|7y<H=syl11!X&7~dWYIXr|yq^R7X)v
zX5AO$2UPd$ECmZcwr}Zy|D04uu9hAaOyb^a)_?9i(lB9g15Ua<J$cqnoE)Gdk@e;4
zD33sWIuuGylpdx90ai6cJdozNI_Gu=1=-|R?n&@(%#^TMjWs$RY5X7jL|AzQqI80Q
zI3OH6AFvZ`52gQYA-Unv{ouFjY%Fe1y{a)de!l{*%q4(~yBNV;%ANQ3LoTvkUrHWK
z@N8|&_nU4GnWp=L^Vpexd=l>7$^d2(mjzD0RXcfM7KbK$!BF8STUhooN>yA^OAX|*
z3-=nfw_dsX<tY+F)PZh7+f7PQ=LKM!;Okzk-OfPBkYrUq48?-qz6}QQH{Twf@4s9}
z>ooI(#>>=sKD6zb_tDC;GbvkZDYMW(6{rYb%!O;Jm>-TeNMWf8*C?A-3#a1P_$RVm
zBbmk%81M($&56ow2Pj2@m{m!UgiWFvCVe{?Ahd=83Tm9rb*g26`FgXb!)&J=fqN7T
zM%R*OSJ2K7L|7fwar`twcVv^g%(U~vb-s4<Lw`H#o*-3|7l_oW`x|E-TqEUTYMi`|
zaKEK8ym;_#EdZAO@4MT$T`C>Td;s)s!9`RcH*nuN-OW!vL7@x;Xi1Uy-nPFufOa<x
zGzsCP&%jA{yJn-i^Eq(RMPN78OS3t4-mS6o$nIL_c?(b)rCvjP2W-w8D-*hqmTlB+
zQ9-Xys>B_+BQ>YRCNn6d54uF%v#HkOJe1SF)8Rw!i3Q3+qvpGqa0=JC+lA{+xTByf
zO@>-w*6)J+kidrtaFd}s>P8wO=w#0i>NjlA33+RpJW36vxZTsOI~D;of5eeTA~aRr
z`B9DAMfzE5@ERMZftdsJK$+7|aC$9ZSCa~bN}8630rh`hu?PB>UO<r;-&tW`Rivhn
z@Q@D&Jq!jW7c5xv^I{xX0O=#h^oF%Rn6O*U>!$AB+FZ`?|K#O3bS3fy$b2Ut>-cuh
zMX3lX@gSxb2Toiha<cHA24Wd1N3or!10oJ}hy)t;Sc>mlvM3*9AU|X`v~ElOZjPmF
zGVt9jsf6f72hlNro;s)3+FxfPixH;?QU@vgzu5ckc&_*M|LV}7c@j}lr(~0iWUG@x
zAv0SES&6L7=+-dH43UcLy^@tv$QB`Gl#!h++wZ!ZbMCMAr*r@Dy?=lGevke;$EWvt
zjq7!d=k<JsnK(iyk1*oZa|o~R2CvT<sX-_RTSGCtR#L=VuNFA+o{h+vdP9N|9{{8v
zJ0a9F!f;a56(o#~1#xQ8L7D4f;RfUX>H)o?-*aa`9~cBd^e=-Jga-(^Jo|Haka!L&
zj<*Zn(H{uDvFU$S_R3-58UWKun#bis3$tEKPEW4Gyk8AGA~d=!o9HB0<fA~^qyb|N
z$8|tG-~$_i8JHM@MqEQX7=s%Z$gh!T#ofm=!KG`J1oU$9%GuNlGYio#&GL-z@<XQ=
z`rSJp&-{)oYG{2obBuo3<{|(Yvy{zF&xD$wP^wMh`ik$5hH(%0*hDy%yzy8nu7zp6
z#jvm+BBY$5*IN!&26O35>~TVxOFTL@bJTrb?g1`>J!sqR6t>Wt0Q8!Rl!5p^;u2dv
zZv$VZgj(Yb3W@#FQgBqS!+fBB-dn_ZV<uZFXpOxA3Z@U=^s)+FB9|FdQ?#{AXzq}#
zN-__qy(oGms1ZzDBv_n$^6s9>Cv*-#H<K>_R4N(Uvc2+qILz<C>&N^#vCa!!3}ql$
zdIu)fh3bCg(wQ$L*@FG|(YaoCS!T6ca8rEqjE3RBf(GY$lOb9$XGf%4;|B3y37yA<
zrCwso6JiXL`*eQo0|)g<%*f(cLj;WRbbc#II(_42=v1g3T@WVz{WuQ3{4L0aLqY&a
zST8de(?SkHRukk`{+!lpu(?Vg$0suB{q4WVa9>O<T@m{>s16Nu0>>R3JZ{jgzL0k5
z6dK~&)OHhL9ebfFK60~2%A}8Wz~PRjO2-EwXiZpc(fV$~XI+ieAP8gd;g((R6uIgc
z`Gw;kD1{k-|CRISxEF%RcqvLNKQXsDEh)}fd4D&d#08I`fG!lp{mF5JY5lAW)=ZT{
zdzM#TuNHWgdIUfeO(!?0!$mboeHX}OdzE<7e4j6nSoh1O$;F@;+*m+_AX3Lsx?#ut
z%Js!#yNF+~zs>><6nZAU1BAa%GQo+W+@F$W2r_0~WehIDrQUmjogf_qlu15d`4d*$
zKC-`;)<oM-ve!+`paa~ivtf8kDrWF)#5EOH*LH8dN`S4KqMkn<S5z7y=REFj0ZE{I
z%gUR}$m}r|KH_NjYQ#w*Qp~lmG!2tzkUIX<(OhZ?`}g?$*417fKEvuPEYULNbHsL$
zTbscBvsT~ZFz4mGIH;2{yHlejPlp~&5!3??&MoZ?W>`185oLocfUQ4!;;w@RfCH$v
z2d|!DOl&^^^G4CbN1Xh8sG%--yh&Kj=ARydjMiA&i1YWxJ?m$90k-u}%w=g!2i@Vi
zqF0c<Z(hib0NLa3F1OxkLM?>z@Uhi)3=;<Q&=1pa;0&=9>D6Z#CAG&*r4eA(4<Jj>
zkoT6Ax&y|?K=0aeWZ)niUOu98eL##viPrfRG|t=_pCOzf@Hu_hNVfG>s1G4<p2Hg0
zOJ4rzd<bHFf$&N|(r0f-LE6RcJsmTWd#&muFv{Gmt_ua2vCp@*l}`3Ipq2aY9`Xpl
zSAV7~Z=iUz*xWnc+b^rXlZBEYN&};zhKHV6JMULE)l@5zOu?&<5csGG9g`yroOl=x
zP%2cbO1)@zV(!*7^POZ%7@Z_=wibQ>ilaS{|9}+Bmp|Ooz#sgY*P``bynBBAhlcMl
zSx7!Qw1wj!iePJ6y*ssHh&V?hzUMQc8&qsQZgv#_J;ws~5VkBhRa>$4vgRoZ;n>u@
z0$Vxp9h<iZrY7>c=tthR*BwcRs01@x!r-)F$hlkn^Bu8;i>R=AvZK%EzgRwnr4dk(
z<Utjrp<s@KC>6FBh)4f?`&1~emCAw27rwm)tW6EgtArv8Y_HoaP=kMXb06W{frA@{
zJ_c@HDE1)wg4NZ(xd5yng{xm=O4|ZOkxMY{e?I3Lh%S}^E0fsFxk57;aZVG@%4*Un
zJ_0`Z=ACt2gumaH1DAhr66lA^Kx9o1pf@dC5TlIyKYjzp7EbSx%vWLUlg}??UOB&c
za=E>Z=fyKIS65*;EZaJUO9?}cJMA^>G#++QVw<tJHq1r7Ld*PCP-1;C5TB5)v-`l2
zyRSg7>cL=+#FnG(DRW>x=Onm^CI-ygVK8q;)OQlhn>dz8o=Gm%V(QuSW&iWEOC=r)
z+j29e*CFBx(;={66&*Zfe|f7PPR&PDzlG`~hT_Fmj>HCa`knHdb%ZL&DXHJYh%NSK
zIebve+J!*AF2%=lN@le>-NilWzgbrk&r-6DnUJb6+GRofvpcx-IkhkCrv@G#zixoh
z@37d=y^j=7t7Jh(awn5)Vv@vz4Z)Qj@A-}q&ISlV5No6ShKL5E0(;mlI=GU7|CVbw
zzs>$>)4GYT)t-=XD`;AOYM(YIn?f^5it46#0(`2<j%LE7%io3Zn@@xUlsOVB4~zA|
zn_-y~8FTI02DMc+Zav+`8N7?b)R}M|Lxgc2j)G^mYl-)J2={ZyL|Q1sx{TG!aAkdu
z%roH_WpWv3LRrwJkCA<^EA>xNh7rIvaF;9C3>y1|hcE;%>$|mgC*9k${S5Q|V8%Du
zRD_?m@83@6kWT<k?!S+*>k8P9`2yfANh|Mg%3xRQ;jJ{pHxGa}KcD`U5xrE+pdb6j
zZ<g*OY4tk<sYAJyHckpaqfxBq7a@7KI>a#CdriQ{Zlq#_FA7M!0bj(jKRBAC@ey`W
zWLg<X@1hih%l=0+{<X%{4Y0y)uIE<AjQC7mu-QhP*NB<yQ*bKNFq&wuBU6B*ilk8x
zuevfcM(p(im}JT5p(<*q|La2iUstz*u-RZwpM&$66yhL$9t_l}^3dQl5S>`6k#s(y
zC4ea+HZT61qCb7mpHoD(@#ijDIXQpsA}|y`4c*EiEC2H>y7}`g`m?<K(_H{Yg`XbX
zO0wstyYSOp_$vzg=`Q?q7ydjTKpOt1yYPSMF63yr2C}J^<`gXs&s3W`??hFXstA$1
zHkMW$J&1%$i*2p`+IX<+FmF&tj@4oDoI|Ta{sBN>YnR1-Zwg_6^vMh0cJD-{Xskt6
zf87e$ErjDZm^`y-Xn;Uj?oQcPV$w{!2Z64fQCnB%zFga(${|M+4E4<t<Q+#=?N6RQ
z`$PnxP(Y(s=Hm^3<}v^)A&yjIq5}xS86c-*>QHl4G}X8<{zXraMrWrdm-d8thjo{4
zX@umK#}km<|0l{xX-MhB9I7TZfZhU7{<eyScU;(VJu_IeG>97Sod8lB`p|mtw2xO#
z!LXS{t3_D};MNNfrB?Qp^U_mu$F4otSKT*%PUM%xD?_wrN@xD}Jq{EM2y9+}&R3t=
ziQY1(#WuW~%7(%<pG&<lluifw^1~kup=?m@_1;qKXM}FC6)}3Ag3^f1_goJ^q2a(y
zM`BJ63K<9<I(_^ckx29xNSSOq<ZrMLd26djVTW$R+{DY~0!}E`)|MbLA3LolVzT!E
zp6dmu*kRVy0tEO<KzX2H?nCHcp;$}fv0jj!Y99ICwoemkT&Qzr0`>Nq_qK(F1~H`!
z-Xh*~qP>vf9B0lYrI76VM{3FSfM90`h#BQt?$KFdM4^T%L8y{yEp*Zpz4a<OI{K+&
ziQh+XAa(|D9*Y(74Ml+d8btIOR*#JGX}IQ=94~Y{W-(a+R8;Jj>-!*g?(643TzhL=
zWfAcd5Oss_cU^f%2}#_bSRM!DomZjQ<Hk8209mqqM+rKl;4ZRknKbV(|08#KX<iQi
zmgl+(Y*rQnP2s5fb2B5%OFcLVbvM^;YT_0WkJN+Akf5Z|Z-l?^J}b`ccqF{DXnt|T
zg#kc$HS7U7X}#_=+IK;ZO%!npRePO&L9j?IU<($Vo7rwLg<C0L(c|1F_`)||e!Bx1
zykE*W2njyO`<21=vhu3`o$z-^0biQY6)bW24x0oq<UEu;W<&iBqbPW#Wz)jQbGqYI
zqFP2LC>jiW&(o$A9?op&%y_(L+F-*w&1>=E?ScKlJ>r&>t1SUq2yQI*J>w-bVQ<_F
z`ApG=vSS^f0mF(lAhp<V$m|~nRM+n1DV^oO5JNyuc?0r;bG^s)udbY%Cl|9L3OhbF
z%sfkrIsdv>c-=1rT_C+tGx36S29fs=@_?iX|I}>A&zZ0_hx7`<=;ZiA1INWz9aR~2
zW1R>kG33{L#g3RY<S97EWAfim*dQznlw3R3dhKM<o<dv@2}#*FX1gDWxj0va9ZNA{
zTge}G1D~1Y{#+7p>GnR%#`~NPz5ADu{j<Wk9<uQaXJckY1fhfv9SUUt!w+s5jXnRY
zbGG-fFTg-A7Y+ux{zm73QiCs`l-y-oG_SR=w47nv@y<qRs^ex6+rQ`}#MRf_+W*9z
zb$_I|c>KomzpHPiCy%07_h7q{pxaG1iZg-KO4~O|oWD;kSj^15QM&Oy!=XvPHJkEB
z_K7qrnwN`AMHn_!yyS5(n2uN+@}D&nH5mbAMpVfRHAxPs)>P_0)<N%^nb?48S^kI5
zd0@jF*#F(-OI-}hc4fn;$k_@YFR<bczmMlMo{p@4q3)3&x};75gdIM-|3kyogULm!
z7wVP!H+{csaa%UeXxuoBS~zN{hZ>dhx6mFvMEQg&IHn5(-Hu^_^2~vUHQ9g}JVp&N
zif%HZ02LAd*svbd{>Xb~H=xWCHvP|@(*K@7JCqB}LSRT{P9e<jK}4FJ28apjfgs>L
zD0{AGTy{3>@>LWt22`S2TX~+{9^NbRi6HKg)^0xtfC0@NSbCxUhLR&5T-KTg*N&vq
z6~~i+%(gQFSG;1-DPrwgN#j-7sIljCVxcja1K>J3-awUGGy4lx(I`U$gpLEYYBxxf
z22hgq=q!Vge*@sY$2tMk>KPy!yDK12=!~`MoOe%hie&1F?hgvZP&RZA3=f&3xa8>T
zZN4Dx!T1$)m})q^p~}G}>9u<Y;u6cOq4~kQ>R5;47+6ZHfL$XfV3h{{Lqj8!{6IXq
zBZSt$m@JvtGq=C)E6}5xE&WnQM9ay*ci2k>o5$o7M`W_UHx!)+G0lzZq*j<Y#BrJr
z)DSuq$;(h#Z{;9E-M}ICV?Aynac;Cc25*X>CX5Xe88^>{Ha@pYIGxHw9&mfh9%|HT
zZCbdw_Loj53HyNf-VtiG%MZW+Zs=qZo0ZCDR~?&30*$`0x|7>Zg_g&(@&mHMG%gAC
zVuir?o0P#}7*76FL4cWK!WyMPp|aNn2RD1{iPV^z&%SPG*XLtuBJ4J>T=$vwfAGI^
zkqEalX8NpWH*A`xc-`Dh1X}T(`43hBc2prdZl|ok0cU_628gI8AK|z-1rAh3w^FpO
z+>$$Dz$&wq2X$-T+}qi8{qeje^-<~CjN<isuy<h>(1Bc9>e65P$m$ztqi{0IZMd`7
z^2{TU9%k%`(?wtumg9wjPl#_5ozZ2_tUJNj{JbZ(aHfPyl7hB}zc}L2?@mkh!Hs2V
zY^%2lB-dZLfYjZ>uY|KKl5?Ht?PmAfO9!vh4M}CZ@d6(TJ2cX*V^S+Y&f@_Gjq<eH
zQFo)#5YW4Q-hI~AO~fcS-m+4*@#qJCgN^F8ZPr}?ny-qO1+?#FY&J9<J*;&<Stes9
zALftT=`7Dk)T<g4Hj$4AYOm07W3Qg`H{grU6airAiaaj}orGxp_CZl9p1-GEKg&Xy
zjpW#n*9M*Jd`}{qc5FKAedYDT7p>d2OglnvBo@~6B_eQ(YG}tRD54~zCg69t;W?@=
zqfBREF{|=#d>^hWXR+~cvpB~s5rCFBW|Lh=P6Bchj%?VxtruI0078Ec&<+LNniNx5
znk4Tz)@G&X*u$pN&nyJ@%3v%Mq^KB*3US#gG?Y5zr7K4pNk@GnxN*B`Bgu%1-@B<R
zx9Q$G!eJik4lkZ1KkOVH=q3Vo@fyM|$3V47lA;>gJ}8mFN0v<+dK|hy+`|{d1)nzz
zvWB1i@`Tsn+*lL5esqho@vfl|Zav8;T6=BKUK_G#2352(m6q07(To|mA+Xi4IJ$qi
z7{JH7ep$b{k6EtrVKif@rA8<{Glzpp|MJ<FoFcU#m)!+TcpG8WXgV#a&;?KCWfMRS
zxDVn%qFK@G{1o8_akaG!uPf8vC#5Q+#^i0gZ_q;vqClv}l~%!;VD&vay*W!EwHG^v
z03sjrjKSz$Vy2RcZ8o^xoRreLbml?Z&_Kqarvi`@eb}yRd8eg=cTX+QlYBCIVTd9!
z(B5|{HAXyn8unz?u_jVv+o%{w3oNUBSj|SU#+8m@X1Dgq)~iZdEFUJ)e4W@)lQ{7<
z+*XM4UOpHAMn2T=e6Ws%l08+tJU4<mQz6w5%!-H3(s(er)>)nbE5k5oWo@ZMHlY>Y
zJr}ixP@%|bZ?8@LS!tHvdBK2bo-gjO=qlBBjkKz!Gf6yutwv&)EWgs&Fu{N@;ZPR_
zZPJ+czi%R}>-XYdA?YHgBP*oALMApnUS8<+358CJJ+E(X85F4Lm<6>nS|lQJez$ix
z2VN467B7&9v-g=(>wo_4+37N9;`?AR2Tqs=^nT}e3Aix1Cv6q%i$s0p>`9fS<MZA-
zMmbHq@<_V}^FTW*bMWmTp|(E24(2Sp*fVtSYARD(P*9j4taeZ1nc6QPGt8i~IG|b?
zF?jW20vUFqoj-c=9l$7e;x4nzQ=fkM<m4Smip@RQ-7ANnp07v{8|$nPv^<xf|AW#x
zo9TFCyYt=VHrWL?)~d^p?v;P(`s%DDj?gI>p_fNgO{18KBd`l_{T|cpE`X+z8PEU-
zOc$5StPDHErHs8VHR@Q4BcMk)V3Mag)8C*qZrTQVoJs&NO<P`=!BPDg)D-70^DLoC
zPPo>9rZ-L$SpvyXCLWmx8&-|lo$zk*X0X%~I6h{~+5b8u#U~dbd_3&b;c9iod=&l!
zP_pgsaI$*HSr5KYb`wD_hp*|vvq(AY=+^`^IJthKAodmNeQ&e%1q4i7B`eTx6tlNp
zhh`3|8%dP$Y@fAMwB)h~8w_?{IEW8ged{6ujg_QeP*^2uF7k<NnT|48=@bK3rZU^E
zf+gk+yp8I#rmB0=uR}LRi#O;&Uk7Vn4K^a8wylI3-!^R!SrZtDebxyU$EW*!BfsO!
zD2k-&wcqF+9#H5_I0P@lc&3qnxP+H!`*r(JKefLn@%YGr<Kw$Zs_pk$sZ^${7a*}*
z0%x*(z1{NCq8{)*wfJc<GUYK}3DlL#rF96>w(mb9W7TMn`-^(CU-m9DM|I#IBH42r
zVNLuOC)N$r(*${9A6ylT)+wofP;gsxzByBQV(N0jsZ>MgLeb*fVgvLdfmKH{D>?Jt
z09ILbdJOaw=~&(>=4-huPK`CDIA^_hp;3lvM~+Jqk4MJUt_ONUt0KlC9Ns}}Y~R1b
z2zp$6eLGMnI<aD1yT0;zWqQI=h1ezzmQIYY>}q`WP4O%;$IU50-H#f1Vu~Z)e*`aX
zC+d1d33}x0I4JoB^yJhuTl+0I+(a@Asy#psFQLK4fo=#@RWG@cQ#1E}mu5?%LT%vR
zs@3274?|b~U2OL75nGxO=#TQJxVl&-wzPwVSvh$dOQzLqPjDFq0nx$QtS`81@0a!h
zyh`y~T?ZWn4DkQ@fGXHD9i@VF3VuhVlc%7smlYe%dbj3H*+RO*hpM!^1LOV+0PLUe
zl%59Py5xdEh`w}12e|ZStZlNVct7|Ls6$I>y4pRD!A0n@--nz878WsOC+}>nKB2%8
zl>u6^)|x1cdYa-)8`uhHWYw?=pc#phuLY<xEwuYGc*vEYzA>a4`lio{Gwx#U%f4x=
zmJ_FW=I?ypU%*dbp-zpVxEoJ^v$`geJ8mBN7#pM-rKVF+eIx@UL<6@ZQ;EsEbU~Cn
zDh|&mIS%PDcy3^DJaT(8t%7B;Lh)~);0Wm*`{I6d*^u#$Dqt6b62(fO!L9r1W@ja(
zPw$}**8y`@0u9Nk@^XR~jN_s^kuCw$4+9^usUEr7`r-q8QwHq~+;<niY6)=)!C}`4
zr=QO+TLt!`sHD%hrI1nfvoEjjpfbI*tX#;^=%M<Hj5Mr$f$*IvGp!_z!W$>$;XCi>
zauQK-$9Il$JJz(uTdPtOqDnGjV-m0^JqU6gsF_9#8^{YoJE_$7=k^F?x&Ydrd5X%^
zcW0Z6mXxmz!<wPPW(^6ov0{{=6@Vbg)AOLaKE_Ob;|p$zaw2>hHLOLZpgHY>$HD9=
z@_F$8RI@j$NqlkHMy+9*Hq@TES;g?vzc%em7X*JQX~=xcPc2th&VDIQF(ydMhplLd
z>MkAVik^KeDpDk%&}a1dl`Faa)_5+kI!=mwgxv#HClECISU(96vv0~&qulu(53kEB
zM~7S$9lIwNK=!*}f~dPLc3Gtdn}UcTx6ROps~j7$-bwD!&sgYI%><d%H*iQk(oss)
z;e6<y7b#R}7pNhdn82^N2vLA`)HT^$j%qiQ_mm86i2=1_^5CeRS~y!@Vhfet$4sM^
z+)n)$3Kuy)_)l}a?0Y&>daUq)jZhmzB|&PvoYEP)Fuc&>JROwR-f^5}lr!wQkS*!=
zh|qu)%?%MESW(@rU?m>^lF8aW4w<YKia~`gK$~N3t?&H)%g{9L3mTD8QOQcF5U`!5
zNP)OdSiiQHq9IYpXga)e7=wFsBgspQsKah;6=c2&xIee?g<B|&t8Sc-F?{!+&s}T*
zCJ#2nLkp&m8dh_Zq4bmJnrz-Z^|$w21>um^-b>J9=mj#h{Kzaa18KA-i%-zsOAlkJ
zQD$+KN$dj;cm-w91->^SrO^JHRZ9_cYUs5>V;+64-#jKGElgu<z(!_eC=+pOkd<cz
zPty1)p?#aoct5xuTjU+&dmxV33k?~Z4rOqLY-eGvZ$n+y@)IX8OVFZ$YStiX>lD>;
zlhw@Va2N=kCv{teP95eb%L1>`pweV<%-5yL?O0rOW^{|W$Lr&9j$lLY9yodX@{>>(
zu<T>7>m)-<4!g)4hyaYPP!Y33X{MLzn5hss&EoPc%O{bYCW1n;G6B(Q?m>{}@~a%b
zCDPx`nvnQwPKUDtI|#Z$wB^P-uRyT`uD7KI#SW)2Pn|0IIgOJT<9Eh~qQksua+yKd
zkLAF5A}&2^JFH*o__l-z<o4jLtN{z8Gkv@G)&|`Foji|c&$dkTP=G%uX2|niDSiT*
z*@xf_EFjD^kO=G)gwBCoY)P34{2AxQa;3{KRfQ^smp(JvcA(wbCOj&h2;OV%Q%tYx
z&5VAbFPjxI;*Ye<*wCZnKbMOKHJl@UI;}~IylnL-C(QpgfR%acbX2-KwK{W{E3=5k
z(sDSsp9d0{rnqvVQyajA4ilQpDP1?)^`PVZX$GeNmn)=z#r^WTz$uSE5>K4tm%qXn
zvppFIhx4|%3mWl#sn$uYeFUw^9a<{wMJ~&x)DPhu_`(*yg50Y&W6*`C4f9CNATa*+
zpisf#@Ut0>QFZd_tgIZ0fuHIVd3EOMgIU-6ST;xN7u2(hupA07NQmS0uLhG7%+V4U
z6fG^IUL7GKgrue0L?-3E7{o*s9>&%O{c*0JR&kG!!DCB}mtx^5nVZ#eSKN}s4kPMk
z(i&i@zJ0WEQ9oe@si8Cw|1|&|<e^jjq1bk(F_)M?d&hEW_LBq5cKe=qVpk&o-3!47
z_@1OGML3oUaJp&iMDnxhkE)8qeT8j!IuQ(Wc)OngvsCdTq@N7e4wikcW`eyve$lbm
zZe4viAm}w~PPCd+y{S-c%fCEsiLkuNfX>6xoWiN33owwB@yMf)EvzbWk!AyCzX7D0
z+>ij!=WI-S)5{5u{T|aC9i6t)^8|fuP5suny0wlI)il3s*gDk2E@{+~MwJNC#7Tll
zpUMX)?Shp%l)P6jU+h27hY79VuOaw@&^QZz&2I(_2?LOT_5r<D0m#-qba$ygstq!+
zj;6o>pmvSXg+CM%en>@Rbo&Xi<REd~85wiFOAZjRcVRmRw5A+JZb~DAC)Q1BOko$9
zs`q6=D$9B>?eMu!D(p@~ZVT=!b&+SB&Sq?mb}?#-#ufyKYy%r}P%s*DjkNH8*3>}*
zD!*{X$HSSad|a7W#(?t10mvYd-)|s%pFBPK_Fb`%ho=KCyy=2V5)t3&YO{0ukWtQ#
zW}Vghn)eJi;p>Y{g4VdkgJ-22sHji4EesnCWW}`B948i88h5}_5cDwyG-(=`Nbjj;
zk~X|6$6&N=_E;>gChZWv1?PkL_I!g)+hc!z++P;<=gj?FmOuC0|04z^G6fu5pl6sI
z=@3&<(+^xPl`WZpseDa{AI887dILRE``8jwPnZbXy;CfN%<&i`KMj|mFPHl9LN}A%
z+xrL7fFbTQWLhYaWt%Zj<c6N$()fgAI_x_yo=x7`4pHaKu(`_~^qllSx!toCygseA
zddpt$e8mhF)j^M4{aefRZLO;9`ngLptEpqrZ{y8@^ZUB5Z_MhHqWI$-SEhWwAZONy
z(};rw#@LYgy-$8*uIjT=+Cl96T7b@FPpG=R@uKBP55<tR6r}65N>q$>8j(Q{i}7iC
zWu@RWDb;_5qUt_uJchp8BVXajKjRt*xrLId5FVlA;j2ZD!3lT)uEz=uG1e{eT!OW1
z%H7V>2Tr+$Pdc1&jR0~T{u;r-#1D|p&`4KJTH-X3vf4&GYsfxcfFRL(gHH46jxc#S
zSCysR&XDXW!NLb5x~ttb6+!jj5;%!d-6jLs3c3f$Zzx1Ekvn+gd|R*SF=Fy&3{&*p
z!!InlUN$a#F204748B;Af<ynpH93lfs1VQfn|I60M{E^^U}*-L5oyGL4J4zeU%+M1
zvpAw9!?3Kma60P+6q7QNx6%vp&Q~X9e>9%eGM`&iaSrAaN!~z0OF<?L)e7k|R7Agn
zgG_x)Zw8W%J3Ao$>V$~4`V55gNv2{P8JFjMUqME6H&O&p>&9KKhm;`tugE8Z91|JI
ziJA4%+&~Ox+LDElKINBgg98;KgHSl2H!4r2T3~Hg(88!j!eUMYW$Pap48&zlmFUV1
zn1n9&dGXZTManRRx?Xh1k<OsS^mHJ5Xj1;J4*hC_J|av2TJOGL0<WVYkYK0kC|y<N
zcS9}YBQ$m+?}b`QgBQC|ZWQDYa%d==Mw%e#Ij#|88mO#tFDAmC6|0qgve_q6d!d%+
zi%t>{20RjRp?HLi&n@ZabGM2h2NLE@*tvt)<XjXtGEWuRVNONw!6|AMW>G*~bRogc
z=Dh+(>NiWUh^ypJHR{SGKh9~jl)<F^gVy<}L7MU?<scK2EL*jjR)V_&a}db~k+}Nx
zR$)TlmAY(Xpy+Kmh~@d)tQy!qJc!HzcM>`Cp1CY9ar9~rZjdbw$n<&-^a+nRSh65w
z3*>E~D2~9L^uU0j!YnXi_OBWNldW1=2Q=Uc)#ok@17o6rCL}0kPqi#T9+ie=)Za~n
zMLhct3bxB@gBK|0?Ngbi_&6?RgF|e<^F68yY01JCS%!~IG)a%78`w;m5(XU{vhT2-
zUiNmguJ+@$_eF~*IBu4c2;>5uLT(@<Ni2i<ygHVo#~_%@4_-HHF^}r!Mp6k2NeX8U
zFNaEggQ#V><`2Hvj4fEhLeDhaHw#MpWsJd^3fI0N%mk|NVk^~K*wh)iAE5m)tXfj>
z+QpDtyw|HSLwfBwJ=8j18`0(<-9Eij(M+eO9(c`z0#{4JRc-|HJqtHn%?W^Bk3A=N
zu-AEP30vaT-i?`ml#PNFVy^e%DOz3}$=Fu|>4BE|vskt0jVv3!QF%Db^w+L3U#>pK
zj13MoV#^EIxh)iNd{IU5tzmNR*8%3rju0prWt%F6KEbT%SdzIDo2<}qX6)pi+M=9R
zXNv5on=(sHoW&8mpBgY(fR1_7V&?lcFd_74fkjU%4y)4jPA;XA=CS`%1<6$fN;P{&
z1-kE_?1UCmwoeT1J>=Q2Mvy**LYChl)ewovkqM$Jq#^wh9vEd7(cx61QpFKelKs^-
zu&c{hVjd~TpUNY1P>4ahUk0|RBkfT0(Mopz>W5j~elpY#c`GH@@+(72DoA7y)~+kS
zd`!U;x>X@Nto1$Xs+YJIkP8fV8&o?UK$0P0607YBG8~QG?YW1Tj+lN(-7REe`t8X4
zrz#i~y($Q-f~)KN9r_@XCJ1D9enN?}9XsWOxq&ciWEX5Q$N-OZs(JFa+jW_Kjc{I^
z+x3F?wM0fyWoYqR9>S!;Y&SyO5f>31N_?}Ot_DkBJL;q=*NT9pr5Bm0u`S$*?fqqG
z;omk50b{KNa6eLOw!%AM?1d<|(R@5v5q3qcpCx=IV=79`VLZo%^3O~rRI+A@mgkGM
z-UibrczYW|yG1LzQ1t?*3H8BRV*4D~MkP5oQ|yC`2}{#emMnBuZ)T-RH2<-gB0`j0
zNa5A%CLUL0aUFvsQugQP1|YKzl@rXaj^?xIjzgTkE1+@wGsgSppGofGeRmbcXBG-a
z{PpIHInfp?jB>CxD|YdfbQr)HL4N-ZA+HNN0t7QYC@RYa5>}izqpQIp4ALby7qU>M
zfZn$dx@sjU96#7|rv6=`xTeRm!2$E2$U33{LzI$lxd8AgF6M-uVx&t!4r7U(!@GxO
zfxP!ulavA**Vq4(BYn0lbV{q~3JnS2>e_W#knxH0PEcQc=!!?4iBuE2?MusZ7|T*_
zF(Pw)&nOMHvIw+P-^$EyhAQxV^4s;6a#;OOgih}j+w<%;pd^oeD_UNHtf0tKFsSrB
zfi?!U5%exVoPn4HFGPDW9=L0zL%|v?FHIp8C8&6Y5mb@eZ?GwUIv)|iQ+?krLa5p=
z9bBiCFd0bxUCQ{hhBMVu4S}sMt^avgDTL0A@3Pw0=dIw#(&~3x^G<eztCKlsLy}?T
z;EYbfD)VlEYr6bFkLTXUJgr|Y>Fi~-)%5JJb{6VimRdS`rT^H&rN>p^6wC(mSrz%|
zq^A)k1?|^=C?w--6(artY`ErV6&9+fr<s|Dz82#-kB*WK;QGb@dluZv2gddS_>6jx
zo%BTM6pe8fcL)BDSDyKqTqjqIk@bc|>7fK3_p9|b^-@eFV`U7F^A6SXVf_(vl#Iii
zg)^OFK(WTPRp9f@^tDeBDGCy}yShrfJ=HP{tlx*&U?Ll2N)BG?e*2(J{nt3CKE788
zjWehSU?DeC5k%Xm2h0?SRxc~ZEr+AG*FbSm<J!mxN)D<|=0#2tYTx#|7kU^5O7*sb
zO~8)sPB%)<x9j-g*g0f^f^G~NCm5vm0;@!&Gt>9h>vBL=w}J5-7i8?yz^QwKX;AFc
zE}xYq&zjK+HZg%}P|cD~_v1YD3XvVb`TWW6VX*$>sl18linZ(b&~DHsc@Wm$ew|2K
zvD+~r67`!8%xX}G=0n;<|Cb}P&4Hha5VQ%Z4PyXyvVJ7X)ANr2AgY)aF{E{5XSSTE
zJ{C-N)v`Fkstr7rVE20pFQPIyI7R{tm@B+bg|6+}AT}d`ycR8I#lqY209sK3Wv-TJ
z896w?qrd}WD}fWeS*@WZ_wo#!^M?y3qq(eam_<`oc-uysum#49HXtdeg*_19T6}S*
zDMhv+mIc2eC)kNKZYs0-J<|z0hNO5nhGB2CY?kl&$jmv|oT9VW`ADnu^50wlGhbRh
z*zz4z<a22*sXg!b=!7uK&7dMPAWIm74tAEJ*eCP3%VH(lE|x}I+EGU8E%1T25dS20
zE74m}SQ$ofnWe;IaXV2Z3AwQc(#1+>&W7=GXc!On1nbnuMyqMnh|UioOe<>pDfO-W
zkU+0CD3p>h>M<`Vj=(~HO97r$6I^T`<nTm27h2v8UMPn^bBhG$_JBWlS?rLD+1lnt
zaW2DR-)Re=z{Gt0<e?A#-_B*w(F1cIV@76_{+pGA+7q9wCUrIqS$2+XO<^3>RsN;N
zTpXZkkkMi@Ua27D$)0}3`Fxo(#i(gkqY9QrkO997B9vhP)qKSiUAa&TqYtSU$H1M|
z_dNooEv)@yw0J9s)hGTgHd`*2uAl4in43c=$a#dNPUULC*O8Z+izWI-v#FU28cRVz
zU!*JFM27@~UNkASRpWVKN0eKD6O|vbVLu>ys5bF-m5oBN{2*ldlt!Zf@}jB<WioR8
zw&QQNWk#jG0c^?z{=4_x*wF$Ubmz9zut+rwGt0h*M9;Ym`peGXi}_M5NL7kObGYBB
z|5e5)E<Q7}5uEoO<^@wK<pnH#4-6`3MIG&^Wy`fv-z&(JAk&oT^rOikmb*&%(fYad
ziO0gp*~jTEzb}v85%bFYmvS_RNM%;D_9qKL4#l9ktnmsq+ki&nUSF35J!DTMZ;TlW
zgUl-w$|NWR!T~o}QbJxfwQ>6FBLVI`IRIa~Y!(yU<EJ%tR>?@XA(z*=M1w*lTJ3R#
z*=$$Krw;4fInPRk&+OIvm4xT3Q=oJ`^Rc;5#qq{TpRh5lrtKTONvlTIc&1Qh0H6M<
z12D2h6n6BPpbMAo`MF9}{^q^+czRTRhFj)j6L{1e7E+<~Jph6m7T|2Z0d&$CYEqUm
zhPyU6C^#AQCxe=a2&K^wxZ5SlR(oSw&iWkx`fw#{{CMUKlVFA75k~TVEx~bU8_;7@
zpRNMrK2=*4XS(sUZdB3WuoW~B;<ynQq|4e~@5KkT6bB27G7IPJhg2`HeDMEckJ0hi
z1^IG_Z9_P~#ZTM!>Ct+J`l_==kLpcNV68qwyb%~*E{wj-f{H_QQInQ!o85RhQ-%$$
zHUo()>EZ+{EehuXgpk#SoVdtRl?DGv6QDY8t0jMfX+s=CNzo@>2Squ9=L1o5+K-L1
zp-w6kMMPS`x{HITCDh(dcrS+ce~;drk9daDW;q#oGf0&)zJd-1^DZ`>S!t;ohUAF#
zDi%R4p=>Gz!|zc?DnWoLNK0YwP<auTyAD=`n6(Spv@+0+fyeWh{O0&b`EZQ>Q;*kl
zxz{JEc;t|t<$(b+<nHUv80QF(`K^;-d7)$ks2cRFTaI3WXkP8KG9X?}5ui<9HuWK-
zL04`pBf3LjEZqP{M8CV<(lVDLOGkKK{+s5v5hx2)tk0+)K|bsnt)RuEo{^W4PoCb&
zbZmnYS5ZFHAV7L~18c!)3eMy)iD*OS2>sYn-zWB~J{rG1@*G&TkC!8@36xRZ)K{{7
zzZeCDf*Tf?8WnQ2-tVMH3&IzFSEplx3WO56NwOz58W}|)v=>6>?cwAO3gMtcV^KbI
z{N(M;<w2j+Ez23NR|hTTA;uD+RBqQ1+!j?=3fKvCd2Pr0C&Yj<ii^^t;pipMz%h|f
zTarLDVRj^V{2UIqTMN*Qbp&>va{8uW^DH-!4fIy>0F)souhpv5l~cE8t2Xu5ILcAI
zHz25Iz*t76IJ)NpRKk?68O<$+1YO;8L|6RujU<<5`YpSLqgrlMaNPM0d-mDodudNK
z1}#6Bj-<b|Qa|`V>tJO;h`)w0p_5FxxO%KIgK>o{!?p~nwow$OK*xet=!{%8=+p^f
z;>As)-c%S94KZI<*#EEQ_5^X>5;0o;gz8Os2U7f=;mNkjqDO}JD&g7YVRX#FXXvI0
zY!gG)l-OrJtOCI&B*6$74YC51at$57>4nquz%iye?TF+f(X1xw+DW>#o0uMLH#$3>
zqrTojVU&?v*HS~3hovW3#w8f=aS$TYjC-ap^1fE2i0;W&UrxS0OtpN@r#!<B5SU1M
zS5jgvb`^F<+ge^yw8VR5aORU+=?K}bc?#Z7Wup+H9@%H#_r!0l6i;Y(t-6AxM2Oqp
zD7b(}Qk6W>`U&f|?bEd6+=bjIlw{E+O&9dM)0?CIzy<;2yrOwD0!tn+(iX6@X<9$E
zy^^x9JEUCSSak%JPrZV4c{jLcwGoDpjPeHmnqQ$V_Phqnd(R)tZPudLQUIgIw*QQt
zGt~8W1@tfkgq>p%3-EU_8(RW|frn>$7kKSQq4fFABDtw((C4;?4QtkXg3cnV4czl-
z!k`OI`W)pIQcbMb7)bPqWM!?U&voTr*#`-U>C(cp-TveU>r-@!9*Q{FNT}IvMfr#i
z;LKHO3xdaXTK%~Z_=YX@8v0xVPe%uJmhF)oj$GCH*oO%KwhQ=x`{?G7w?{xKgDkqm
zZ9J!Y`pr(cZkk;~sua^v{5c@dfel0Hg9rpOrG6@Xl{~O%G*Z1d0vVzUfw!4j*vK{<
z&?OnrE>#MD)T+}q-=?r}2yQbJ`~0tu<pl5>^RZhEh(6&LSG97e-WE2jr82SAgiwHl
z6@MibKKSG;-)i|e*XbR&@xfufoDcB<`7fz;Cn!r`Q=gY_3crO3g%x7&p|mnqZBbGn
zU#E4CY(Q9ebY5|UWyWKxW7qtGHpx`F_n0I5KL&b5vARW?sZ7=BekXo|F08>9(~v)V
z%PvSkm#LTpyV-+-Lo2Yzx;YnsQHc0nN6CJ_Ly0<cVAzx&utDBbyDQAHmb7|5vMhxe
zuzmzdOAk*ypaNsekPFmQy5nZx5Zw5rGIZ5}xbn?(%#b4ycRaV+EshK)3mw+8Q(r#-
z2{f+VZqY&NtWa`Pmv3sjg#h?$C^sSzQA|!0ts^>ArUn62oC7sM$#RvKdd!okhRR*O
zwJu@W7L&ccCuMWz6Q*)tw!rZh!ZstfvUufGLx>O3puvqgGtl6p6wfy$hXgRw^%*|&
zLi_*{QLz*x@51QAZ`_`S_fuYC`(pFH)yEQ;BY7Pt|IOK(c>h)8TEPvNvDKL}ylzO>
zFp~6J7)FB074S^#>3N9r3|H9Sz<0F1p`-4>>dx&ET+V8%r?YuF%K5pPPpK~h1Iy*c
z*rZ*z{*6q$X$tX#5BND8+3JhQY$Ah*^LFp{sEuLlM`&zk=SzlT@P6Oc(2?e32Qc1N
zz3RO>w2O2ihtZw+Rfhe1Dz6e<2&TT;p&vzii-LMPTXm}?q#=XJ?o=oX|AjOZ+LDuL
zpW2I^2#_)l*dja({(}YJP6D3A$*8fdci4yQ-OxKh;fPs9qynd`h`%8Z!De)C9&~wN
z3k~aWCjdi~VuJs_8_#Cgu^8OSlw$7COS)Fw(t?S6=i#$&*NL`m(un#jOYvRlqq@k4
ziPsEENBBSJ-a=|vr2S=VJA2Wa>f@C+O&+lLd8XNcD*8cCU;LCxHN2r=0pz&#AbG%k
zd=c{C$50su-%!b3jH-5x%1CcL>7>^k11Dwf>x%&bJ7TCTI(Xp3Ha$2&U-Ah8OM1X@
zGjm3Cbzvm|`N%-xHN43JxhPXFzOK-2pm7s{0)rKHvS7dvG-(0dKlcE^lyI3k)Q{3c
zz7Ad(a9ENQ<5cE|q;Z@|(Ba_x)-(ow;2sQ14sx7t4o{*h8k8Q^gor;4PPm{)q98rc
z`B2Y?g(bySBBP*kpv`4?^#Z_sH?<2q)Rdg2fBRVQA9dO5s;9Yy$u}t6jI?tAS)boH
z8X6mhSKto_)PxxbY!!5v9zy9K+}irV&!mr;{uQ9bb+;?ksX~sF3!F3!tP>gK=*mb0
z^6sgE9GtSe_TVj*&C~e++Dv=3j`H_VEmju|qv*g+NT+IGnW^YS<sxJb58~z)08-+v
zD1TYj7llDZgKoFSx7Kf#?U-xI8JAwlm<Ph_-=+bv_}{IghAIa1TgvSg0Nv;VVM87|
z;h0Lpz8ZuLWw$Ctn<!pkCihr^03$%t&mal0`z^GJbAXUXgi}zD^xA+<Gyik^_o$AB
zK>5d0u5suEY}rjKT}StJG6lV;Do-~+y9mEBQ9cIDt2VuL)0}PFJNb9A>Mw^OMa04C
z{Q|PkdyXb3j@?27Gc&t)wHd^Da*}}~-`qQ<{cm3wCoIwkcd%XPA_^55+-rWsZfFHs
zx2M#N2T=WhlJiAmRLz1Lo(Pa0+y@O3<y<w>F=s;*Z0<!DQVmJbI{}5|C4lz$Li13@
zo6p#q1DU3xsC*T2mdjZPom47<wg5Q1gyC>$ECw4(k4%Qe`sgWkLB^b&na%bZ;B-q6
z!F<gaTx1rZjPu`~0&b5Tc|Kb&mUZxUy@Ng>s$1)~3@{Ix8ln;=SOY_za|N(~5UgXP
zsXf?Xefs1nq0@UUeqVr^A1CuUo{P~PGNEy??=)cT(hWZ$v?<}&XJc$Sr7T3S#S5yq
z{g&ZqXWip`CSwNu(*AcZde&_Ru#4T<aKWr0OU>C`SgGWO3@UaFI$PvJ5}*6p=T~m&
zR;jQs%b*qPyu33gV_**@Wkm&j)cq<*7^q(}hG-i%F7|6-5;@9bA}9F-7v$R&Td%#|
zGMn>;*6`5|quWsn!_dEK7G<JX7iahH?2m=TIUjRyI01M{Pcfhk5c7kJqW0;KClF(u
zZL?~gg($5$$f(lX{0F3_qsA(vX1<1@hWZ@4)&!jMfm#&kLP^(VL+-s4PFt00_E2+0
zT3xqvNqSsP=eK@bjY`2IYTzK82MmTugw-@m9tt%&NWSjENRP7gD|QZJvV>e;i@tl^
z_WqaOZYwDC9V9oj{Dh5Mu$0;IHa7xpN#&@6Oz4-M07cpIHtjL<(Ah)PH~osDKfeKT
z)4YYJ>GLcL6M-p?(VtsmM5p5H;Hajo7~clk9eI8)+1&4=-&6`+QxPr>*^h%5Wn?VP
z^2bUUd*ik4|G@)h&_p83h`k*u$8PD#TV+LOyZs^?;nTH#m1j|1akH^!wm6Q8|0T$Y
zP*!+sL;vT@mq(TwfM(B#{#supx;Qt*_M#gOj4Ee22w?3n?IUChCE?RcM>XS8VFcao
zR0y3Z0x1U<(UfknFWCS(PmD}FzoB}D(<H#1M5=R*>qjUNzHh%kNr*~DmvIQ~a&@I_
zU+<p170Ph)q07>27m6Wa;J)nVw+Lr;nr_H*D87h5jy%+FP(dO$^(=)c6_={gx39b!
zHE=*s%9;TT)B@TLXkeW&)7a0xVG=!mv_XEU0F8YYP-50{RTvCP)t=8su$;zvHg`LZ
zocA?wdedM&d@L@0U$DjrHoj5Vf|FNTVt5ap4;^N4`2>fvH2ZjM?{p=HNEWKiffWnM
ziyG>ns*x`8uE_?Ox+DCsl4n|bMQPB%q^@`gMZqpK&ShqS&J*C-*$Rud*Fe4eH}V7z
zLrduc8J#W9#;+y}UgasQ9H5BgULB9&Hcf_YYTeVcd-XQ`sEP?WUjX-Emyer-+NFDp
zE8utL!GZl^YaS$@t5jaWlvf?C3@G5^$PIK8GrAqo8uSM>W|Xeqp?6Vyb%0|_;J5M)
z?4VEdUP!l`?=hH{s+|Jil)P5RMLm=B_Q*V|G!_)JM+4y99-arut*N!F-q}GSl(NPQ
zlA`1pJ9{3VwPy5G)p6o068NM8(O@S6?-SvegXMSrzCp5-kC9qJ=P+d0ye~JmnV|<m
z>x)RsL3BhKjrdsJDDKUaE`AOL|J=*j;hi28?rt6_{X-bx`5*`Ky!S30)yfWQ3ys_^
z45ZT_(C*y|C3f0|<#JwsMIKXJV(V#YiC|N*jqfGF=JO{T6E?5wZ>|OlyCd9~ji~)q
zwQod!HhK(_X1f8IEum((p=9k{D>Dk~N6<O(AZWT`WPYVWA&CV7AJzuHGZ^`nSuE8c
z`2wcmCLjb=>V>sQZ}j{zpeSi=Et@fEFa~nk%(fTCk5#{%aOZM70F>4Re%FmcVnGn^
z7^8%zWkyp}SVOwJQE8xbM&%Hsb65PpOn(Z@!2?E;jkR54`uSC=novHrFsVLpZCD(p
zpd})J!@V(__7*blzP_`KD@6aWGG@`s!1>h4_Evhh?)Uz7OAXr~cT9uoVcR4G%h8&<
z!h@n;y+gDs_bVkB%jgX<GA$USlgGd=>F_A>b)y`#y1%@9yr{fW?e=(W3jtw|ORoz0
zJZc|R<Prr;><`r?GQ5_9xF5PDc2dk)f(S$je65yzQ;S6In%}?8VNnp%To`EZhn=I)
zjna>>`~<z0J7C6d6j!ku+4xcQg$P(&ilRNsY6gbJa|{26{bBFEqHdJ|g*{ligV5Xb
zhZOjzu~sH35e9G6pg?nwoT=Va!svN8@2z*siyh1Q86{RW*F*o5fWuInGA63049bxT
zV?o#ir}3+CU8F=B)XgO+Hk<?Vg)JSYLyxm?6u)oD1}ud>&wP}4J@)I?a1KwlLYZkV
z)J7_Ua_yn8C|JPHkq{J=CVk*Bm7A2z!)QY}4f}Sg%hDNbQH-w@<YVjkFTcHSN*&5O
ze}2G)A|RR@k_dOw3vF%8DIP{I;R1SKFwK$^J>M`Z{Uz4&dsK(M`kYx`-IE?!L;c1j
zEmc$IkvGUK&MqIQN`=*t@+$S&-ob*7#F2D+p)15;-?%;27Xn+SM_H>`xm~kqt~B|6
z!2~cnNW+t7OanOjH|*xS>Z``Nh8ts9=zKirZu0V@!iR&^)c!%&Adk733kb`vU~h_%
z)N`edztHI4CO>xivqWBpm`O79#`6r?n;s)FD<A=ga&uebJoH)4Od(Bw1myBAK))%Q
zw(7Sx(7qW+{-h#&P__js+k1?H!j89$8+Ob>EU_JhzI@}4Lkd14O-kEn!67}m#YbU(
zTn3!HS3MC906)oD*{D%>&h>D)gjC;t-{KBKW80)682pF1(WZ1Lk>zyNqu#5APtvo<
zO*UAIB;(yjibsxhg~8b(uS~=vN3W}1;1=ev4m5aMZU)YsZj{>fdM}Q)meqLM7BBXO
z7`a*lJt>~_sOX&fDqa@x0E4$`0D94@=4<v!%4ru2BY!S5m3LcQj%*)DNt-r>{e|TX
z<p{}!B-v%2)9|nCTSiiarW)sM7XdXPN=e@%ra|#2`m-938J3t2Gu5u1Iox1Mr@>sU
zU}RS%F<d#Z#B3LO;IjKJmA8Kg=U&B#pyjAyuAU2^Vw1$xWrqM`GIZrowLzePq;Ymq
z2tsmtet_h(!+AT|F299v-qvk#HCX-=6a_9-(>s6Ogc?XNt(08#ZD5gkzXBSSbp4ik
zOHCu041EBD>tUPg`{%{l?>`QWq-$GMClT%lc$1ybW6Ap^((cgegKY8?K4}R~@=}P|
zE`bY6zwme3vtJ<l<V|riGVI_)Ms-2fWq_b)ZqFPPct<8w4H&O=zxk3Hcct?%S{@JM
zi3RzJbxEV#3n1Tc7i~xc&<-aagy``u@W7WvO59?#t)c}u3FGKcMr|pe@AiUTmV`<z
zbW4VC6f5_Kze^~Mzji&k7pm00@$z}BtE0M=7jTEO*nOW5;qQ@>>ypN~7vR`*p#F9P
z4o9AQi1<c!a=N(KAaMuEsF3vf<uvfA_-@U-j}hI4+ChLwN;SHu-lNr&d2=6$TGsZy
z2e5v9UuY8BpRTjkNi)1g2iR0Cxpnhi8XZRq?2fqp(fa$2{qho@vDYPYNabUYx^diY
z(1&mY1;gZP*4L?W>hlthzY!*q4@0232eK~heYwQ${+Aqe#tg4hAR)4%LIfeQCpcPc
z=xw36-ODx|nqT+V<-UR=hXD<T%{%P8!0L5kS}9$4$9pnEjbYf9mw(uneeUpCT%)b4
z)%u{plIBW@?-w|RbP&aXf)ITjMK0JtijcK$wjTjT7O*j@C#MFq0C$VnTLBxB5HRp$
zy+Co#o<LR3+-zD(b*LbD`9$1Xoj`(>*%twi{QgLDb;1+SFH~0=eZL_01NApNc@Ez<
z*ggBs*O>}<FKoF^waP{=l4(OzANyWgh$IZcI{<Eo!nya>{<1!Evt&YHrTY30w@9m1
z<ghr%_iiOF4wZFD3*3&7JsAU<7<TL_glhaT6j#RVaI=SW0oh!XxO&r7;2F3F4hf?u
z6oN_t+q0smSvi#z*rCsHXV{fc)&_{(_Mx$ZgdCvj%Z-!nzF)WsCKXVM4c+~jsT3D2
znq%JE``Fe&wfibGA~R#XOJFGV^(9!ICgsUf?zU<&e>hxOy;osjJcmy)6C8cH8d>RT
zPYJLka|_6U$WuZeh<a>1p_~Gv9eW7EA}L9tWP_42^bEvX#(S-CT@NkLd>gehh$bWQ
z9OQvMT79+-TRnMG)X*}}-udyMcqBD;Rmm$et@eOnl;x$s1o(Mb65bse1%Mu*M~Qar
zx-Gv#5BN&=!0Le#55pet{#SzpP2T|yXvQii$vS;goC8~=pTrYYw!shggO$Wc+e3JF
zd1Y*)sE9iXZ4}Ty+_c}Wnz$-l`5{4M)Zp4|w)!pKBiV_)r-|<xS9R<FKA{{=xGh*w
zM%bFtM=J=^4!Urnw^dhqE}#)3FHvGNl;GO}=?~;78=n7?@@e($Vsl_frl>LL3LRLS
z3GVTNgxkVul7{&Xyc|Htnusf7uM2BT1!U3hLu<*c&wM1f;g8tfe{O?z&g$FAo5PR{
zawLnaBQrh#&5e}G*2E!!=0>PIv6eFvTukUEfz`DFX{y@~t81mngfQWm*k~NKHUH}B
zw~TB@9uEyo2XOaKz>rYK50K_=SvPH*R`NT)x&p9vo$@YNUGa}tVRcEs>e}MgOzbw_
zenk-`fS=y3ZM86icmWPc&WmYS>h3Tk97>Hu7XwalXcvz=^z6~<r3`Q_SY0cvIndIQ
zUr9%nLeS#+x#50pI51H^H{8z+xB8U$dBXiX;lS_s|NDefn%k{C(p!hMLE{6X0L^QB
zs{Vt+V1^9<U&Yq6t#%p23&4#SU`(P1-wRt@H(Hw!9Szq)F8K5JvJ2Z*`zLi&;Om^@
z-n`EZIFrCNc)w94YqhroBVLXE=)#7``PF}q4F>n4AnH?2IEU|Eo$QI7=ED5X#jfkL
zT&LKigw-W=zX1~NKC*GR(Kz4GYKP^fcs0DXv1CH$GWel0EW8SE>4UkCtKAIn9h_9*
z&m%rb5vvQ1k#TTosIkCY_iG5|%P*RR$Znd>tdhSWPc$2K3h;>>e46`6nfGE3r8b`L
z3H~H#(V?vO?0~1r%X36ueWjk^`W$`Mk59&QSARbqP;&5m`=>pm>_Ol_s%(c#+|Q4b
zDA$ALE4+%<ueqyI8h<%;ktJ|5b<<bEoCGfqjNU)7o}mZt+Jx`bmNwB%`d%31?Ypf-
z90j~W6c+~5?ek28pbTE7cwZ68WAUFCBH3cIq>>BqpZ{~nfd3|P8>Ush6gTn_pIs*l
zpFPa`agG@3!h2uY27cxAZo>i+H`NKhx^`RIR^l3lm;5LTzYO2jyt8^0!(?<&K?_G%
zSG+38XsmIyp@wnleb;PD@=|wT#OLaC2Pl5{baD9bt~)C!j#YP?+yw6`%Izgc+70m^
zcmd(-wBU*+dLcQYw~hDw57YV|mjQg?e^@(L$l)O>yBTXqab4g??}VQsuH=#~{4474
z_viF<+en+^LH6qoOx5*`z5K+%5|7Nq&1NDjyyeHy8i9RnZyx&t+IJIPqPty~l(B8a
zw*y@DN}xzsTaoAC>E3?Lq?-V)NP?RgGbVs$6Y+|Fst2w73vqh5V&G=tCekaW;TiwH
zBL~lLf)5!OOd?sm;*ipE6XA1H=J@f`{|PmS8zB^~Xne0k^Wzmf8zCR~n)yd;bTeF1
za9D}ReTEiD+#l>RZ_7=)y(C|vTLWMCwW?Z+I3#FXs)1CrAz!bnoG@6hr}x3*C)ZR)
z|9A-p8i!wBcOM{btTnF5@RXOoT1F6JIJ|FUBmAOn_C}bC<TK?}V2`qUo*;Q&csw=k
z#V^GZeI(C+wG6jfGQ3Ed*N2Cg;~RThyr(4Itq!NrY<cGz(rvoE?>hYAboGoikt}y(
zCZhcueDbsj2M~VMTsr)svro%uH3SlVFoOh6gXFEG1d_L1<H`?EY$JcyrgJ1O)eI|W
zZf@fsshXZRq|o3LHLsd_NF2sVDcFjl#}fsKDLHuH6%+U+d!KXo>Qf#rcEZH?h*xw3
zt_|YN7T~8cF4FXW-%cElOS`Szi2oX{-w%CUOxs?QZnwIZa7B`kO$0G+UzziK7^)jI
zr0YlA6V_<vhnqJvi35Vi2q%om%R`o=3%ia2A96%Gf)terbIt@miM+EU-Eei8@b_~<
zygy`!_lxUpg{OSoUSmcwRpQri^EBJp(f#;h|5yZo{k9XgXz_kKQjuo(mU!?|l>UL{
z;~C5@9b?KroVn`gR7tB8kiM4|+M0?a()u<$Ll7SSey!S0(kq7H3S;v`@*h8B4&Lg0
zvxW)j72fbx|3GgD&$x}F;-&qX^a^b6jgO1t{P7tE=+fLiKf0Opk`%ba{#-06<{ln$
z7&%kY*NzF0T=MD>JiRI;ERV!E!A=ZCNAE^issCz64_CPeJidQzkPL}&f=kp-z$J{g
z%TJQJ3R>_KE5>eN65|9DoB+RMe!P3*hr{Hbcpq0b83#L9f<JEcz9V_Q>oVMGq4p0_
zFdsam1K)U0S1eIXQ9Kf-@#7NH8j`&Wr*$t(*IeuUf$hXMz9I*PV0y1@0f|#gws8_k
zc?yMJ&XJtsAJ4#$c(2z`C3#!0XurXC9R4LKj5sBcDd@OO?PwAsop|wi9IK9pgWSZI
z>hp)K=wx}p+Jn@n)HT5`W+DrMBvOoXK43XIXk|o5!-9?5XP|oPt3&E0)G^_Byg4#J
zMtn3LiT~{4Mq0zzs@o0;|LDex-+-YS=aeQH5m#zlqiJg=UJwTaCr1l*m6cRC>B5fO
z1+S%kF+lj>kFWI~A_73xfhS~hMVNqESy}zB_QF%D4m|oqvKL&xLgv&cd2E#T#~1s@
zZe`qWxJBJ0MmPR=i82B+SCZF+L&H@PW|vo5JMYIUvM_TSm}pNrU@y_u1ZwF0c!f1Q
zKJJ(F!=zvGkJzS_UG-0JQ9Lm)5M2<P{zl$jm{!9uQ{U&deba$MPZf5YJH>J<S^8+~
z?hxaOB+skT0rl?vC!}sj#Y*}A%ICrP^m*(dt{#3Z=2M)f>UT%a)-I)t3?%A@J53)i
z%Bpp4YRUO-U*KGnQQHpxGm-x|>wAmINUd{3kghUGE+4+(2rsOdC+|LyEd2fVAw*@0
ze*Ti=Ghs1E0_~(Bm88TTlAl~%VfxqqlWX_tE_|uihnrWnlB}gv0eE;xZtjl`<f}fI
z0IzQ!u0TnLFb8#aV1F@O@E|pVWE;Jp&{%t6v%e(iPn9soYC-ouk<2mNxbz%+N~`2q
z2a=y$Bk*KaBcV3K_c&4tzvv0ug_6cPtpE<Z?(0rU^a~rzA3RPTT7rc<to|9Wrf+4%
z7AH-CkFll(2tQ4u164x%mV8e3dX)=!)lX)fPc@t?)$}EKHMk$-MR@uTCc6C3xRh#s
zhQf9gKVUuLX@Et;c-k*GKWeVcJ(}joVHxwQzL<Uh)XNXv*Zw1-@I75TD(Zj7th-HI
z5xv*~P2oWWpp?CP(D}+jJtg3ZIa*Hc8^1pU+%=pwfTD2``2_z)-Ry3c^+j#H^i*Gs
zJ3t%HBkbVtg8|=u1)Vv1)vSNd%zLt>?=H*7J{g3Wr6X>6K4p0><s%2@uU8I_0_Kr%
z9xCR6BEv><mCy)XWD3=96PvPA0hamvCwUH6KMGLC&@Or^JE#9w)a&st@iLooL1LHm
z8RX>f<}F=82LaovnGFz0Q!PK;W^n^3M#_;vxap(wSP6|Ger5)s$C>h=q?~RjHs8`B
zv4K87q=!W=#P%^@d#~F7)X@;t()9rdqG|~3D-zQW%Y+Tu=W~hP=V3U{4K2%5YX9C6
z2M=CDac)F%W<Xn~!2IV8Vgg5p>$k5fAlsW#pmf_IoSQazSzuTz)$oy|j7Hfx-9}~A
zYQb>XX=vAtGoAwu3%?R}+h>JZ&}r5JtuKDi6{OX|0HEsl+~skuWxsHs3}_r-;c}dO
z#!#&d1f_<)4y|43s)gf$Edaj?EwKjXRoNZ4)i1j5vj6riq}I^@s_Vh}J4RV@0IJIk
zK=V)ODA?1Z?ELR}L1+6F73|it?{o)Q|Jr)GyVo(y)!j))PMebC;7gmLFHXJ_Wjg(8
zOweA>=&YpqS;^c}ZX(Zw1wZm{TF0HBRV6s#ZSMH2uf{}sB6JAA+Vo+`CFv2!DzNL@
zap)S%0B9JW(Y&5b<gxrOTUxlw_WC2VumvDn8f2EJa5FUVc-b*61#~G&T2zbX8dtW-
z(=9*)S^tRt&3G0<o-GyusQC!YUd#207nibMt=+KqemLNgC|}kT-s$*zDpq#wK3wJ8
zvg&^o?z|#jA%=5tGDeJ}FnVu?rZ&K*JoMdO7CFs}El-Fw^3K)TmGY!Rb9Zru{mu?Y
zpJUTq8o6Uz+GbKOK1u3#c4!<c6I)7q*d!M8dM>YMK99Xne=&+lBu@3+5Mr38vr?Qm
z96&l<Ghk4;Z+Wq3Ik@BwO<Eg_cZBkxZw}C~pi;T#UqJZLs^FN4Tu)pt0_yiWJ!-tW
zujtJ~1UFYa6+yLl&ut1wPI`$Q;l+m7dwNl^$i(sP6!0{p?;gZ{VuwF=M##4=0Obwm
z{L=R#oeRj+U1F)%wl!|>URMvc-d~cseWen5wPuWI(x_Hsu7={|oQ$)fdqcI^R_CrB
zM^NO8uc0RC*jnuX<iKHOsm1HtN(>ew{?JV`^Budlv;?=VjQ}KVt#Pws(WFq#Gpqbk
zi`^XqMQ{_(-ax?Ua>FZp3l6s^rELM|*XX7EuY<Ol9kpHg)!ej`4{0x-r5?BVbQyfG
zv(qyJZC5q|B4KL_f*5%ZM<jGjy?<dkMt2={TJ3M2K`Np(#?CAa+L4*emK-|f+lQdN
zmqS~x)KrrDq#UyHYELKI7O**VA7lt|f;Uc>z5%#=1Fjq8GVvXilsiJ_p#MN~!5+F^
z40oufRda1mF@sG#5&z0DS$QoDVR^`|o86Q9?_7|pC>bz$=va>%O$p5&GZC5{ER<ck
zC_cu++56;?tf;l~`A-_qPa(T6xvzCl`+NM=TCt7&-4YLtSK3T^jpQ8EtsUUUZid^c
zIn7+WZRU%jJw3wRbuECG5xNoi{EFPZzJ8QYDRG{-FWn0!j?2?#yl0c(?6Kekn+_mg
z9_cQQ9+;kUzau*NMETP*uAU$lq1>0EnlG2J16%){n%IW^u0#*}1-}i9tb^AsTxH2I
zwE3G0Ah$<wHhq0KK=)r<yT||a;$zu;Z&<%Mflih6xcO<@@%^W)9RkV?83q=>2HP3}
zW^|;x@<7#q#y-CR6oJo}j(EDlSriU1pu7y|fL1Qk%4A&Tx64Xd&}dT@nz(=7+G+8`
zzPbMHj&!cMed}la=#=Lt%zbRf*PJV`GrV9{)a4^Y(P3wr2>2o8J;#8tLTz|(N~Wl5
zueDdjSm=fx5AN!dwx=F<&3JY3MwoQw9x4@HzL}z$<2e{AT0?VPn2+9{s;p|i-cc@}
zJ<la|%bJ~X8U{N~sa82VhtX}PXS{!5zxK|rt`9rS6&9T4z*s!8+cvE7J(|lY&R%P5
z+w$nPa@(Lgy+SwM=6xsw09t#-lCZ88ZE44}qhw`754KvxC*63JZ%C`!ez5NSRD{E%
zEsM4x2$`7=j>t-{&YkRkAfD)L_V1j@73OQewV`bOW(4}-!oj9xzvzX7P^DmOU-7*U
zg|pS>Hsj{aHq*c`!Fp(?d3$fiK8~xkhAAv34q9l<a=HChAiBTbu;^3|&;McXz2mX)
z-uU5?LMq896iUe+$;xaPMZ+dDG9n4ttGg+)BBQLx&fZ%Ch3qX98JXFe=Q`%CzR&ae
z{r-JkpFh6eS0DF%pL3mSo$FlZI`3;)%$XthV?mP>uA9cFi_*P2P}>wEiR-T20sSD9
zP1SJ)#!QHnw!cVLM|+S(@YqN6#3D@qM|g6m1NJuon%OTBXNc^7?1(*u&;(4?)jF6N
zQk#SurUg2>IXr2laOvBZrZNux44k=t6a0N_%RvW+hLj)$1%&H%a0mfu>A!krDcmpP
z*?-{&BhEE=Vp;Zn)O|GGgt7lwViVlVwAr)`Xz<3eq6b7S>IjM<C0-LMaR)B<eTN+>
zaN*4lF@Fx{{ns9OY0rBu&v^Flh_aswR{4;8BTXpSG=h&lta4I^VL4*%y`#l)vY)BP
z8S!rGZv{U!-DCt6(Uxtw1kK$nk8vEtB#kAZ0&q*K9DKs5aX!uR(YkxmvWK}`k9hHO
z#QR;sXuJ)|yppKTM1&W3=Pd1#@GB>;Ab@dBPkd?_>(=;D7m6;IrcZV>6Kec<5W0ll
z)5G=Wo9Z$uu)UIhTB#T6v-{pxIoUkrD`9Azb(Rne&26dF#b+tAps)&_?y<g1p5R4y
zTX#d$Xfx!1$mz-3JHe~LR$&vt1yOpv-_LlPy3S8ubsY0bnS0JoET1=v;+g4`JFc3g
z^Tmr@tsgj<ui7i4XF&H*`Bw$h*M7I@qUun7gz2}UzF9CUyCV=mIvXh2oAZ4noE!oG
zz=W2=DIY%PAM>8bED?(LV*f4QM!x8tot-gLY&eTK|41`C!_VQ@St8qO3iSCOz8g`!
z^UUiX$RFN8Gb745a50rTV4(FRAcV&_n9iezMFm-z0qhrfV#`t7)_HmHR2|vl_VJ+U
zawc<t+enkWY9Dp2qG#+ro?r;RV=jP{F~^trcEq&&sa#&DtX~@M4sh{z40Yo4;T&#L
z&A*^#(UHD~ymrb!Kc7^sJ{|w)w~W1KD#GsO7AKK)k#dpMXZU?((vxsb8gXQqn&(Ss
zZ0+%TXZ`*35ii?dHFVW*h_)(TROpZw`^oy@itABAnrAX>)vK8^g}<=ri094-P%n2{
zG}Ef${w&Kr)?`c-FIxVYvda_--H9@ePGQB{g>`T?kRGZU;9Ct)sys#Q!!qp6h4k!y
zw;$Uh)85Yf>i}2X=Tlc@=+E(t_S{`?Q*j_~+0|^q5+|P?9dIHlKLP6B^Rk7`iHyc|
z!XBIqt)~_u%;6cBLWVKywv#yuzB*Zv$x9kiMvD--Mcr{~`b4eRmN5U3)?qL{c7EE{
z`gh=nz5evKz&SjQYG3@*7ABKtLdyZ;S6KgR%#2o4s&lZ6$+S;skgc7LUPz_Gu(_D>
zNj*0g6&vk{AU~ns5+ZXlN%Q#6>wRtQYTvZiwY~40#xL?7^6W0j&bW~()Uv-U6kb$L
zGq;d`@G<jw`?w{e$&`9(l(Lez`_P5LNIS(?e9fZU*V|MJJjdr2=Y@hg^Ia*t`Wm&V
zc`ui|Yd`4=Q5@ec{8Z*^Pp)-kEkYDvWTpSiXr;<_F|G|BUU@Z8->940V~2vfT`?jb
zr@99opr%+M+23tAFo438PU$-9bD8Qk(eRd%Kwa3uSz0C6ZXC2qcXoan!T<UB%|OyU
zxm;7tj?VKB;L$-^rUNR|e84%`T2_YMdQC>$ts^Tt%9^@2xdV_2y6W?TrgM8Izh9PC
zDELbM=AOn>vM(7;Ln9#qx~X5ZdwkY$<`a!#hJ+U5xf5-wxFMxP$2BHnlq@>b7K9~f
z>=c!APU*!k`kK0kUMRm(lNSgmJPK4OLbif$P^WZwFFJS8`f&EQqpI8(Y41FMN6U7-
zCtFAI_Ej<~v9rsPnB@EKp`K2Fv{R}nFv$teUiaFvjgvnXS^RxfX>#hiw$>qIPfG^~
z2I5-CxYQ?GWZli%3=fqUO&CKs9yd21dN9vG(JJin^2F8U(*xOOX?ho_uO^Mt>m29?
zzje9z_>?Z)H9%d9Hp-ONYQ{kiBzr?$*_iaW-6$Bl#_UyDGL-Z8J=487pvuyW9?6p@
zFlK*>o-}LorAwv4{KEGShA=-6-&Mjn^>@S7`LYx2VMN_yL8C3d;T$$Ro6u-fGHfX}
zucKIMou}UNxzZ>;D<Uc-$9{pD%fS&|@i9J_$h+gY<g1DA9IiQ#Fl76RJz=2n)B_db
z9?$MVZN<+`CgtxaK8$K4Sd}fFUTnT<r@iYViX2E1zUed+470Y7m56_*PFS=xE;DnV
zpQ!D%{)S5T6}u>P4wP6PX0N8E!akw<K-xf`F>T0oHD7i}3d5S-6l!~Z8i1UN-wOvo
zupX#<fU&*M^IGM(j!r@7%@SjK>{`lu-romvI%e+E3FA8RfJK@W;Sr`#w*ByR=V^&D
z96(R3dZ{^8rDuN@nFoyr@Mn~eAsvc6HCk+)V4UPIsGu}$0lSw}+VX61c>aYJbo3Zz
za^W3<GvH865S<5~_7(rkP6*r-73e>TwCPZpEJa40^synFR({E-*ZC|_%Bt%9ZLLS{
z!w_vM4!O`)*{cePqK0slOgwJszI&ZHq-+9;Uh3qflXe}R+&6FS&U4{Q{GiFgdm59`
z8>)*BD@Z-Z9X%oqG;f38>p5$0N-s4qEt6!z;kTEX&=;TO_@CO~rPLIt%C+>-bez?m
zDzG-bzDqjz&!dy~I!C0ZEdeH6$thBspxMkdQH;XOvp`d(GWi^5Qu)o?edxWf%b6~V
zd$YV88eFW*%E^9HjlR=v_w5$)V%VuvhYJ?6AnNi_ADrcS@Ljig)=^w!7;3ApJnHF6
zYX#JBm9~=_Cetm)3S}D(CYbxAO&U=o9O?M;gjc>SCmvoO74e^U6U*J_=711gH5%^p
z=H&%XT@V$Tq9`l!acGCP0(v|3<Dmtt3fDDpWg)%r!IHENVG}sFDcF@-);mqTj=QI^
zpBMYSh88`^z`3BX_BSu){ZhKGOgf8ec8o^cwZAqBr=+-XIE3Y8vOYLG8rOj@%aU5}
z`|26d`-8|`JE(;ForvmS6=q<Z&xNhDd^X*5I$3U%wPTo-<(NF_<mab5QKy3(SJJZQ
zujVQ0(sRb(C0}y?r*jRHCO?b)EKNrDjwQQSduq^_6*+tiVVOodYtHM=H+|hhu5rBp
z))FHf7JulCGLLco&=)$-_bsGn6J}eer#m39i%D+d>Uzh~mocOptWA(n!>UKn*DmWm
zp6J<_tckEh_2beSV&)Dzc<NT!8b1`SXr=TrFmlm}`06)dW_kq54NA6Q9`8I>e&9nj
zzU=1$rQA-kKxk<9He2oyQ|Cw}a8yxUNh6AVL}gcj=JH%)<y?-!Qk#ZBR<qZYGtnec
zx5b-LJ27*3|4g3a94=NiUm`1?mtK^<3~y&*AFw2q>r?o+-<M>uP#Cs?ZHxIFE2o{b
zN`#U?XSUx_`UmrzB-g$tyr<J}DzO-4*Ry)o?DmoCflT-9mon)A+)+78&^#k5X5bOY
zmA-tNh7s2!7JW9C(PiicsT|OH!8#(}y9v6k@)?O-cKV$mWEybTX<;;QrbblgyBr(c
z;RJVW#cwq{j^cNPDp==ChcYTTAJ$0-kG)XoB1`b?%ARSgd@p}_$a_mTbq)^VgugS_
zARqi}F}(ntTTk_MRL?wDrT|hmWPo3$OxX*6wxNr()uPWoa-_DlG6VLExi=jsWz*YQ
zl-uakcb+C@ZUM!cDWzgj(E;6I_78>fU$y2r^veau^ji;`2e&$ykIut(t#V_i6u^fQ
zTb%5btBhPv>@O8AKPVjYR7V5aDon*hWZr31w~Fg`A^MV?K{{#QNfI12MLgu8>qp*F
zYN965JnQ4*GB2TQj6m@E+d<#;6a?JTC3f_C`pzW2Z}c$?c<PYbFi*~f@BKFOP!1oD
zolwc1Ilhr;YZo3K3NHrVPQP2H=#$u~Fn1tnb%(eBJOZtczzWDm!7Q<~=bv*9t{k^!
zTMXdUReXp1BZf|4Q@ipCk_3XX=eKgFo;5z4Y@|kg6ZG3e9G_jg!mp~G8}AWb(65TC
zTe?^>@NL-m`BaHfXY#N<ykk?SNA`Qh)23X<r6BJqyw#}e){Xs0sMZ9n4`SvD#$F|q
zt0oH@U1M>oH;Szcn1xs0*-9@kEOggs$cgJ>OgNtqXS(<w{gfSST$SYR@Tx>SmGwan
zakfjM%%f(yANSw|gOzqfl*Zd6-7L0<%2CQJ;V?OtEWIsqPDEkz36r%3riIG*b&y)|
zjuZ?h>9`~eFTMmg=da{8w^pVDFqFK>ObJy(TbXVB!1X~#Iu?0H17Z=Usg&UQSmS%M
zDwVd13KF>(jgI!&3ag4ao}n=6$*8$+_1&n!u)>jXhg$0-hxeAJDj&~;FbZ#TUMj&O
z8QnWH3~nQeeMdS(G*$;09TS|)XHja$&gYqmxwc4B0sd`uc)w;|8Xgr%P2B>mG<pZa
z9K~36DlRvS!+Ug@0-J_zK$cCRjP3gbDV-se40emwvOpU1a!!RD?PoE=tlfM><JLpN
zN(*qtD#+`%bGzu~;2<^1H_~g(G@bXJld&wEPw%zc#VipKf;z~J_fxWw__6*0^U9o?
zm#uJ?WlsXM52XyNr;e}^cKHmkKLUET&uIHm8YRZX%%)|gwcaSAP?`Mn9apE(jCeap
zm3GCn(&e`1?X6kD@A92z^YwSRRtc@?<XB<>E1}nNBz|;c^m>YY8(9K@4!ysIYWX(O
zl$~M%$3$zHf6@GWksLUlSK*O;(1?QToA;zdQ@9=U021t?2t<aP_qyPRg{$$^yc&bD
z-OTT=rLv#9nIx^WFi7?*y2C^z-gU^_A-&#DtOolc4s!#d@jY@K`KyU)%nS0qr9TUR
z(V~9OPRBpennVbPqWjHn6T8+{+Xmh*jtI_EX|e=w5AmNqeb+J0^OC8Y!tb+SpN-8a
z89OuDV=J`(PXupXbar0u{<r>SieXWajlAx7zjd;LVNdgmKJ8&5INhR69-J(_Fqjsg
zY;Tmi@LUhlM(p|A-(%xvs@}dI08EA;x#MC&N#8|TA{B;D3QMaR<Nyq0Y_e%ELRcRs
z?|>KhG0A|g*Zkh>lYw3Hee}lMP$@x6HU@>+wt&!buT#^()g-TRhb2lozAqUIf&JYx
zGTFF{a^0$R-z*%jf{u~3tesXkP9)C$OQns(m3Hmpw#lF0n<8AxlQ?N@W!H)<x)(bp
z<E_Y~NKtgFAT%j@sT{W{4}hUNm^CECkBqj23DD9wwQAFcti;VhDc}2Ohem>Ix>0Lh
znPoUwXp~>)>yu`I{$h%{DOCp&QB$f|vo@(<9}Sv?A9%5+I0Y^z7+))cQ+Y;9iL1s$
zs7j8r3N;;WeI<{9UH`mGvrJhRi(@_{^`knzkDHM+7ENSPr)n5Xl?Az*oxqOb4nTx|
ze530d2R{xciPWNdw{@-c+aZV~eyds3E10`{&bUCQbwKKBeTOxAcodsFdbvZnhd%x6
z6xDW*Q*H-71k7x6ddupg22qGW%(0}~=bC%{sw2KbQoO^W^G(d?<==qkmcnHIVdp_o
zvHtli=m#3xiVz@u6;_fV{bRHa@0t92&)G95OufCMM(#;uQcC1eYKC5fm%WJ%4xTO7
zkFm`bnP4DPhYU#3)2lL-0L4&0PSwKL&-n00Qt-@_C9Vj0OaEKn%&XH+&AT1g0h1`k
zQzl-3<1{$&)=o~6j+>r5MJ$q|am#(<mwDUddGfqgTzc^!ex9&s4q#sBKhEms65qSx
z(+$ti%di!p;4Wszhn+a;u%D^ZI-Yn=__?F{6z$;zqAqTuf>r=clgra{7A*(t^ri|p
zY^p2B1=kP316Lah2b;ITa@y1HU+D)sSaKR&S!jb;7`l>8j@>D$JlhJG)=mY6b}YI!
zkfX6G^{4Aq-Djk@!)H{=fgoDr?ey3%@xEum#RUMf?9g4G;@EILwR>PtW#^dYt|YhY
z{-bdQWyBJPWrmJva8*oaCo$Do+^CwWixq%Kn<)(6pL_P@IV5@hbJx4ASqB6Y+wpv{
zub8TyK*&b5@t}tO9PI;KPC4<}C&OScJSGJIMRApk=0aV^A$GzNs;^0>&*G|T8!nkO
zNB9{E<8Y3$c3P7x^H<tjG4%yKm}h9E66xBpFq?jUAOAbQVd4Cv>UR)qkHzws>PO;*
zX23obcylNFRxhAW3<>s~C1NV|PI3N>#}6;qbydWAQB2m)AoAV$L@q*HV|so{k|y@Z
zH}jtWa1%3C04eK=IdN;;ga(N6ncPUkN4H<Sjj#qIg^N)~N`=aT#nysv&8p>02iU@`
zreBhEACavyuR4pTF#jxkpHEs5rvlsL{vpW44)lCKA?z0Gv_Az9^uIB3fp*X#uvUdd
z2b9N8W!5|IlH{s$GF`q)aIlM8IQ8{N>ta&fMH~A?NHnv|9a&yg5EqB7o4y8UNS?eJ
z?+^~nOd(!O!AVZKO9id)rc$v@WQncFsM+spQhvSu^=A~f`&C)O<(Gt9yMsF|F0n*Y
zB*-BUx`P=-@1NrxzAM@bP9VM`EmP27UdN~0Y?)KwM%GOuwS!#j`?AHVqTeqAV1R!{
zHvU)N?qh0odG!y!RTtadjbRap(ycRxs|HfcL^A6r*T*H-X4%xp{EoRXrSUxB7+=pa
zINF_BoZ|QRg#Y33<#N62%LmTa>2KN3Wzm^GEpu_LU`6M|p9cqXLu_{lw>P=p{Q%cF
z3KK^k92xByhkjn=q0-V2u-rxIMS;=F-F_4i3$&eZ^to3*GG_)iPen$W^DPz7-?ZwP
zifL(F$D+@)Yw$oqfn$9Se%wO8Qb7zH;FkcYI>#Im{~3N~qZ|=rWodstLLoVsUvxNS
z_yWU8AJ(W!TxfjJN;>UHp|G-r)(cK&RmSx@6RjXob-IrZ6YbgcTPH1r`e1*`u_C`B
zGh)v3*H|*h&IZqjMjrOfTs*;UbFmlUVRCXw<ECo65(R`-n%XVIdS3ocZ8D5<y!AsP
zHkEdr{rsviw%Qk64J7Of_j6^PZtj<gC`#Tp$@-Dg)XitH6=3Ol?FH=ojGew2X~?Aw
z!L_9MXxtP&Q@wuNMb3&Fr^-QT3YkSt_b=T)-9nyn^z$JxKepnq{Jmo<na-QivPZZe
zb2I$&6b6-td;jq&6v7#`O<Z`h+zc%WgQzGqTpIbi7rT1l(ji-A+U<m5mxb<ka<1W`
zj*L0-6h)tDPyZ8ixbkeRlp;Dk*$DMyGN(Kn)CJv%8ZK9RjExwnIo5auoi@^3TspA*
zd9umq>&%=HT?bAseHAu6!v3_P@@KmK)`5^*tV@2+r6<gy`S7zaU2byKA@AvdCI1*9
z-8_rPKW;ikZ{su^g&W6~7AN>C2uhQj8&7Q9pP-#&k1FH=n&`Ba*Yc*4grBLtcjJ+D
zzsX6uP^d#a?$^2ZPbM2csy<{zo%mvSF=b=K6VWCRu=c$_#W{Z2o+LYeGN)tCVSr}$
z1uE^|kClau5Kz)NR#Wxd(M5plk9ExG9D`;90<p$2JbF|v;ilRGfT7U)ed!5r?%T|`
zg%N_Q*5?%HeP~_JpzhUUJ^j6CL-}ob)Q5XO!;raU|8nQ2H-G_{M8ZYzs<Ftr7!R*|
zkC=?E!`CMa8K_mKc5@v5L;%3i1fs-h(PCGVGlgqUE4UE2RhBn)CklOihggrP;01NZ
zO*Y?Wu3FdlNu=#VNVdnDkq6keeY$N&;X^n-v&*S`&wj4rx(t*#*gI|=$Id_bMM5}P
zneuoGDVN<Vsn?$m72(OWwQAG!x(WqB%;=&+Ki7OuI{Wg%6<P?T598tL5nkX3wf{yU
zLiet1WD14SF?0{Y+itkxKk%J*R$<7=Fl&xwAhF1UyZMm-$Qp@><_l!DICPwCrAe6Y
z-|mx<+s#@e&DULXbCfbMpT&53$#>EYH@q~zWZ=_m`q+rAw|ldoo$=)TMY!4=)ewt0
z@tItwS^b7R<J;SI3vSn}3tDM)T8%WJVS34X9t-x?28Uq!_{sUxL*s;+GaqA)oD8^o
zE^qgi-GqY_jy!J@TNziQ3yv!|zG`L{Blkqrbu4%^YaXLLu>GM~4qWnqpZLXakmlHh
zsL1&pt|*JOTkXQOaJwu(FLInxZI4{A0lX}-l-jyhD&H>EYOvzGZ`}@^y=dc9p>h2D
zlzmb+ARf@<`!5gsoju{x)Zc!LL3Wt;99Pokdiz$mT1}|P6rQ7jeXEstdjPsq4n$!@
zQCsJycVp=@Pt3}H$aVBf@l5O{D0(V7`cuMzB4uu|tc<aLwy8u|F^#)8KA2U{-GxK0
zxzX=2gGU*D*=K1n;xRuHD!o=Ze8qJB1s%Oujn)vj<JvcFEd(fRu8!hSkpV|z)cV=x
zXetFbxvK43wW&8G9MQq)Us*g^3q{Oac{&Nbc{@APF3s|CYlQhEJht@>vKvCl%wyZs
z8%&|waXDiu;GHSW_XV9u6!0?#gt>~W<#S4VT9%**e3oG=SJ0@Nq0(gdCxph`Kj+9^
z_lb{~(%;0jQ0|s>F@^pF3wfFMD_z@fRea96xji0EldKQgDs3B5VjO(R1{=y)?%6z_
znuy``(Q{pD)aP|?BquZ`%wkKg<YYGD50g(_;q<x7cekedE@)*nNjbjM-q<XZMSlRu
zvyso$$slz=w-$@G-{pN9HlOmb`=0~sy3hKt1n)gr7P(l7id9e=Z#G#`CErcS=ju>U
zcS_6z3yIQHO2_b}G2IWz_C7wLZL@g%D)|dn(xKtYP1M=Js#>L>m#g;Xtn@JzVIy%L
zb_b4u^qCXOqV#12LxJ68Hj}sQa&vP5QX(0U)VRt-PJbQSRuyojUujQGiLFqZFG}~4
zFMcU_5!n6Hhb34gn(1Z1jo@ysk(;wm6`E^<rczZh#NHwsP=Q-tnWd*93pc8t3gUWo
z(>&3geafbiFR0>!#}PBvlk+?#Xtv+P%)w29nx5ICZDEI>wlvMCv&6*V99su4e(EP>
zP@<5_vb@V79f7ym^o6&ls`%nj^hO!@l^7k?nZ>h$#fw!(+BfI2Xuf35fdgTR;Go>X
zz~o^$JmtZzTn!1Yvek6rh~Md31*dB15^D$80a@89e}cH#i#pA=vQ_P()=SVGquiyk
z^M_Tahrf-T&`lBiBWU6~-h2{<d(r)KFti+HT3CX`!ZYEpAj0QbpkQNd+pyTht(Uvx
z&Z@!Q`=PFrTaKEPajBxnDAKHKN305UBBN8_IG4uL;(<8pxC5;rL(tEmEk#gLuFm4S
zX^Ad^EH>rW)L^k2Q0=&`R^j`me2muClz*wc$f)VDM)1L%`b(~jrL!#FMJBW57DWTS
zb3>;58csrgFhnd%`Mp2?3{Dp;JIE}fGKjog?v4JVbEnen^{QBiPV3I@aa0jjy{SO2
z!Nt;Ws+dE=t5-^zY@T6xrCi6}za`Y&R6T$2mJi;1*_(q4{E~Lr%=^OZ^-4oW=7%yg
zeJV+nNuM=hMqdl1Tek(Vd#VNo!$n_`Lq0&~B?IJJdXv7}S9F8dIfhVZZU<U4FtW{W
zsmzyRdtPwMa6`4B{9616N*U_0x-`q#vdY!tGAjsOo_dvC&X*yS_IU0!0i5^w8SC@!
zQ`oierav7$@Z}*@y&KEy8;?LgGTr>4=jZqR{<Zy&3fF-K-&;b^?X&kgx6hP|ZI>p?
zA_bR1&yCj=D7zhcJJ-X-U#s*Ks?iBJQ_4J6b@8O0ogj2+D2+YyFja)8wBmXYx866F
z3_&f;v|E5J!4~h4UhMY7#xIN|{YRzpRO+A-^>=?RDSew><twQy)4xuNWM$65v3U4B
zH14e*C<n$bD){>mpVm`sfymyaTdvG@SJL;JUYF0<Y|D6W*idQqtieEs+924X^<2T8
znxJ7@9h+q3CEj8IpX2Vj%3(sLg2l3#rfNllg59QM)0#d5+Vh7?Z1^=e$&-p$!Z~xa
zCz`kB3GkI&j1KirZ9JamNX9{D@mSt*fDarI%ivNFa=lZy&FXM(I#HFe)e+G`POfc1
zIrLD}d9gG)nSdDe0gY|sncc$`Xx|X1BIMdJ?te%i>E0~|f-m@-7iQk=;s2E~TAr=B
zAgkGa`XfbxbrG|RuG?5MV{9_J^#{V3GokVLX}`Ge$TCWjC5V&1>KibYPF$}rOd6`x
zfBt?d2O5MZ4m~4G>Hu&rxE{p0nqb#9;L?4m%kgshTt_~tMjCHF5lkG_QVAD^Jr$^Y
z>I%!PH<cc8Z6|S<!_8UFQYM_G>w8(bOa9d}Lf51Ph<fE8H;nsl*RMZV@QbUPblyYM
z4%c9=nY=d0KK+AsRwufh4piph@%ikY5A{C5@s{eE9?XK;{4IcH6XI3bB=<_?i^X%{
zUalBL9aCr%dk=xEejGs;##2LEs)>EBL(mmnG_;|_TJLtEF46tp7UuBwEb^RD5Lz7O
z6`M7^W9PCI8|JjVM0#0oxEfU~3&a~9F~P7qj=I(xy}`fstIT_)f}sK`&uYH<-e9@|
zKsJ+SW@VF8&QKnGe(Ls1c6aW9^f6!F{84uWe+H_QaeKH!WierM6z4l}?!2!0D3@N(
zSKoks)cdt)a`t6(AZj|5GOV<e+%(nLO8z!*m{qSgI)hO&JDuc4U_>kIT5@e7dY=9!
z%;%+&66~LS)8=_SJjU57lgCQD@S8iQMw?~)gk@~@c(|~@?3kXIoQ__Ajt754CaY#d
z+&DT~=d+dDEo!%Tu}>?eMGlT*@TmEW5Lyicl%{Hc4yKz@%JS=sWbk;mbF4#eSt*$>
z5rl;*^HYS_$&Y`zcynjx4v!hy@P+3;Q65HKRjxiIDv-!J()7;!eRZp8vE6)n<OIEU
z;^0X`>c^NGg8h7e=Y0mE7`;i#Nc*@Hl-Evq`yCFu=C(HAITgA)#(Xl$vxMUL0RJ)O
z6yTHH>4lWpkjgLs&iwX1`pm3jl&gib?8sz`Q(kCMsuLutqFWHocYQ1~n-hZPyx@1T
zg*c`Db2-Q2?dpz9lXAleh=nZLbHo|m4UR*nz!q~62-PGM^dNCPc!5JVKTfj@I=baY
z37iv33XpH^NWRR;HtGMmr**`>ak@1K-JF<tipo5)&DP)TqFakv7ly{&u!LLjZ_SJ6
z;vY`T0*Y}$@i80&IAIS4IOJ@3X(Pp&s5ZzU@*-vPO@p5y#{S}yL4^cc1<Q{FCrS6s
z;t4e)XA2vmjV4=k|8JwnXS+SkTS<tcBCdiq25Y%wK+_`F-?BWn?r+y|$4)#?2Doje
zGT*w=PKztlv|ka1;cY<3Qa*h)Urosp+CJ#d)?9Cf2E3VC>NYdsO;*A>(n$sOKCf2;
z5S81MLj~?ry;5h%=wVkjw?s8&G6JDhVC{k3XZ=GLjy^s7@`y#9%y>I5QQS?NM1~y5
zTqQ%(U*WP(+~%cApk!aiD9n>lqbxhPDDdmX`10%aD^2M5<C!GG30Bw+uA?;NU$E!-
zbf<9YB7CIkr+pH&oKYEX9kYU`J};sk>ny=XnOY$mpA4X8xf%)<{F)W4Wxlt{#=?`5
z1%Jt@YicMXi!V{_o;ADOj4BmDovx@*&#_zm-hzko1j;J7PTLA>RNNjS!_0(qLH_uT
zC8{enlwD$l`1Xmt{{xm_AEyT6JjaU)R}->Z^X~+-wx_$#*%9*T+2&+gW<rwU(`1Gf
zPVKS%346!8^IT}<aKoW&<f*h&<K$MvpC`01Dek?Tjkm0aSTQq-cgh7%b&z+^k4z5r
z508Op<E<DSX?oMyAA35Klie*Z=QjQJ2a@1VM0Q;l)vDR4E$#~?5Y;OA@8qW-eJ&&b
zn;U&z*__0d2P5a<Asxlowk9EL1BKGP=YaBf(mm`t{(f|q;vd7VU;){0Jvb((xT@G{
zGbTW#)M(I0woP3`4+1dcaFn_(s<0*338N)W!17U<G%mf$4?u3bVF~f_%>;)bqZG7F
zHbZ*j)k=LTWNmy_`tn$U5nmk6&)+hBp)=UO+ZR!TU4K!_;RluJT8>af0VjHNafpxy
z$&?1+7OxrgY#M^P*JiPU>fV7iJ^sxQG1_0f;fr0M2v0#=fkf7STdpggfdf`}O8QBh
zTG~W7vFf6fXfkSAG7sgRe#X!zakab_TH4oEIXG+xZG|7i@V<}Y1*U4x)*5?JEe;s<
z!@~DB+b~%;sEB&XD|q2JK>2-3h79Rx$jx>4kemJWM%Dm<vx%uGVAsweVM;++c*?zo
zsS8^qi}C`ArxD+~ZnLi@306x6fH2WAwAdZ7|Gvf4`U)!K6YzZ>cFGb-mIox;@-JEx
zD&Rs&7j4q<_Z+8Si8sBOwm`|P$n6j@d_qit{|&JSTavrr5e-FoW!_(N!&k)Nfua6H
z@dekdLfeKB1C3z8`KJEY-_cLdUj6Q=MAu&*VGs(}<?t=Hrfann2Z1hA3dW%=GK8VX
z;Z3;8^Ah*JRtUe+f&KtoPaZ~m{p+JbH5e+3KK|b~N{39~lTm7sxWArb;_9I+#O}Yw
zfhoL@3%W%YBeL})txLf6h`-^z(04dKzye7GGqbUSLs&%w0DI}b7FX5$hrPAv0mYH7
z_r8Bk(Fs$~J33oD-y$V6g#}C@$NTHdRwzAOdO4WlREa{uUp(KAQwtB@xW0uYjQlml
z16Vb^0f)uozo!rlfoDLbwAjTCNLfe_sl=b6{(jX6A8`X?DWAA*o#1}q40`7>(sb*8
zyyZfoXlIN63+)THSQ=JQzDwHtA4cK&FoLN`AKUrY`Y@(IEX7lq8u7pE91O=1RybEJ
z`}f=VLn<)&izAeKu}(j0)}Z_jyxFmz-?y}FfMp$Gg`ZSY62<)wtx^M$@1EygVHc)h
zXS-u*ecZ^W4b;3Yw!7728Iyw_^#&3G;_^%L?Am)St<#DrBFydINu_J_-nOGi=||88
z2g{J&)4y2#p4JS`xWl~t3~TM*FWF#`x5)UqpO@$k#ASU<wbU$FXPgu~6HGZ?Zn+A|
z`TuU-HIV#%$E&dZ5-tRCn~|!YS9L`T7qn3P?zPi@gAmekbR#Wv+}0-y+6@v(JP!XY
z_3`$}l?80N*ws%reGP3M#N)Pd-0P1rvSEU3v4zzFNbzgHNnAWJufIgR9Y+p53+l$Z
zKfLCD|6ZIQdz1O+{5N|QlkC<#G`-{x=rG+OWS7$ZaM-ua>2dcKzUvYmw(3c1j<fnM
z6d9sbdH8_Ry1Kta_k$-Dj;@d{$<Cpf*h@vz#IsmY_b)nwXb64ch>S?-otd^^qsE1y
z(EqQIXq&v+<^P8x`|JIZ!FiCt-ZhLl%xB>1v$G0YBL~FfD%2C2^VNejjlukaFMhqK
zuu9`!Lfv%rqFkWIdRH#3@9iz<z2W&aus+%(5RUe>`t=Vt#c&$E`Iio-T@xp57JSi6
zmTX0e7!+XqzFhw=cdEdcBmKQ(*Gb&6k4b`8@;W4MvGyUBqPT-jP|7i`k0%MT_X?1I
zyz<4r^#_YY^a`}+9qHOYA$wt=JnLw72e#b#ZmQ#bQ62bNvHb3Jef&%K|A`Dc^!QI?
z|5@38R)*!NaPj|hj9AVT0>S_GvaS61Z!cR9t5$aE|H6;WfCVw;zwl%IAhhaL|BD#^
z;>UkM{?_gGzwl!#KmI$HZJura3qQ7QF2Ilf|Aimv5t0MzSOBm<loa?c6SBsLwb1kQ
ze@^m0C%In|prZfFH*VgW`APhkQyngi&mO)y{{75y{mA?vyk(gwK6s~QeC~0){jXij
z`Y4BfPVD<%mp9;II50RFZKhUJVZm+fR+U!-=arg=mKPiOr|o9VdA=xWb_<)wYxcQJ
zHeNs4I&qYlUH<bXh$Z&jNQf~;b)5I=((g9kn@)=zS6XxWRhIL7HQDAfGVTNlC06pP
z*(z?{4s56~Ch0)tywftf-B&wD@5LogE26lzlDr<yX1xKzsXfNq2G<}-4+%gzGl}Zd
zUUX@WF+DibcQD~w#axB^`QOFyS8eWE7LM~$mgWz6Z}k{(6oD&-Wz(PT&8GSq0^Uyq
z9aHdbj?DT>?A7hC+(6G#g27FX@DHsFo<w%2;k}UF07dJ0@lWfQ4jakbueeY6jXaxe
zU+xWaS>2T|jDS?;ANrv|0XgkMM=&?aTXfQcswtdU|G87yBd2vJWy&v3-;eIj+TCU$
zAEa>I#`|uqG<S82E0nE7KTi^pKm8BF>2tN!pP!uNh*`To^E&{vGg26|q5DNlHHmVm
zid}gFmu**q*appO|A04NEZr+B>yd2bT@?YxFyy$qgdeo503zxzC(5socbkZK@85Tc
z_^I@X-sscFv>Wnkz;hg^k&+z8m+Y0Cb{KOa>B`UVm1vgzvt@rGRd!jSt(-XBWc4(l
zbpYMe-5{M@0p==J3>jMaLi*DuPu6ccHio%=aP{Z4a~v&eaIG9aVRjPX!t9kxgfRB1
zo|V-tjyC>-?h6;nEdgmIw}n0(R+(Qbyr36@TNzn#SrIgc*6p1wuhAED>?94=ASyO9
z9$H^<SAQ<W<7T;bB`%!^H)^;<@$u+Zp>PI2l@sYjvHn8ZKT60Ujp5N7;#95Vkf>XK
zad#1E&AaNlgDYC2im6*;{LX(F&ypodwesoLr`2nB=okX1uk)|M)U&SqTm-Ws8U2M~
zP1k=^_7kb(ipoCWe>t@V2_gL&Xn*G(wS>;<g-yHPtt}n@e=OYqhfyEVUiGBfV|8o{
z0o>vF7oM#@u`;<PW>xC;Aoc$>`3hg2@ITm?JQOBB?hzxr!nmg0d(Nyd&a#jDABE?*
zw>Ea)r>O0#YFzjT0i8A7?_Yr>bSo&NU17~ai6+s8uB`v#6X~Iw4b&-<d(3|gu63;P
zlLvdbctm9pl@MT($dm_HZT)ED-Bq3347&@z;O{wwNE6pP#QkX1vG~!#&j~1{)sbmq
zz7i{!qo4Z+j5d<|^K*`!<Xq_IPx!SrmJVRfmB&5fS4vQT?-K&6OON^=OE<%D^+yUx
z$LmJ9u5!-s^GAE$-B}UaFM^4c@##$#CjWwy2FEUqaa>cm(hZdZObh>sH6YbPs+fpX
zm3ylEubZP|uBu#NO!JL8j@6ZE;UHtx+HbL9tn^ADUssm?eYBD5e=Pm=iW9uy@!Vm}
zl50@`(%!K5o=A)w={RMpDkqRii3Jj~%H%^f82S~`em#}hv8HmIKs&|w_Bwksm!Cc6
z@yg<Uc@|Kw-*Bx99rp=iV$}-mJztvE-5e`;irP~(b1P-6I?Nm$D_>|mTK})qoA6mq
zuSf-?zGlgX&}AAYwcuss6EVMu`?%2;BJ>v~|Cgz2tkCX9r@ijsPa}=3H5gz<KG!Wy
zbb592w{`zxa)OCfzJSoz>LA=uH&?%X0eXm$g%bpIE7mySUO@eik9@*-EqK+*5v)78
z2n3d?_t_V!e{NCPEuyjn3@1ru<t3u>bcg0%7e&pPXF|z2Be@pkwt%CLWaEw#ctxKw
zC$19yt53DiPR_wAYX>5oC8^NN$!g?wt8Y+Hn@Kl!imS*td-D-q^}Nrli2^H1P*re$
zI%1C3+_M^Db$<#HlOVVD578@I5r@)`pI7ZHOuH-dCx2$~W<OaGwp+-3-m9xO@t7B+
zf7<IL_qmEA#~|ukXzSU=t*R4doLp96tJBEpxLQ*Y#`q&Iy87_JxPDZ>Q(X4u<l+&&
z73{CVySUm7K~1pz9d@!?S?SWwT$`E#e>+%JAWr+6*A=FXMo0#4M9W)JI(Iab4d+h!
zb~W4^TKP638O-m!Z{d-JD)3A+&*sz(iS%J)Yim&j#{rynL5gQLs_+4YR1`R)!y);4
z{W=#(r-{JY#BlJ5^g}x+Au12C9)DL-{gY{hGi!tFVfgXQYy0-Lz-^dlVwkROPnrjo
z1}8oOGOwSzonwgP_N3v>$+#mrOlun!(r_&oSkkGYhY14!v3vqJk)KJ5_3?dCy6q#E
zxwT~tKEq|R-Y$cD-m8LD5)1eZf>>0DpDX0Am)!GE9Qaa<6C_b${pWPx+8U99yd~s|
zq@3Gtuu{Lql}X*lO064%HH%i~&H%Z~3;PFXx8tx9T2mWSC{H5QSzDuY>V5N=XVmS{
zBLP~K<wt*DgeN_BZo~S*9@-xPhT8<Kfa{`cKWyr5Y;YLOUIepW&keNm$OpwQKV%c)
zx%44s^_F~95N`1>e)SP5Ryt8FL0mykj}pFH*wulcApwS`OT=PpkQDfALn8S@>acpa
z7T(*<7rCYjH3Zhwm-VCRvA7d)6}U6uK6`pn8Pp(Es4DVzVxtn)7i59JORO=%Q2une
zaJ(<oJ#z_5gwJIQ`4ui*ws1##Y`L>_h-b^&Xd$fCyLkcmrj)G4Cq%VKcLMpHQ3fO%
z5%G4Mu46#Hn)oGwgMd!U71o2v_xn`hu*u&2w10TF_sV-92>3npKWy%Ch)iP(CmRW8
zdgiL_$WA*w@unRPHAEJRSQSAbm*LmO{V`A{cx(S3!87Ylc})nZCw>IPu&PXYJV5j_
z0R10-M(xF>KNiuy5-w}WBled{8sK5m&j9ovmbm-|wQnq!v<cxAy1At?4hq?N<;^dv
zod;n%!T?iY;aghz=>dOV5B<#eib5oct3(rD&m>N+x{DPZ!-SLFy^@yNJ|%v?@E6E0
zFqgBQ4of~hS3rO@Dy_AsTBO|1E`)vYgZ|3Cv#y{Xu!65EaNk;dp#M+1?j&F>Q2Gh_
z_>e?iAAkgKb#-Fe=eL<|!TW!W&lhPJ`fd-by5}IUu)tjvTRw>_u)wkdX7ebvFLw*V
zo8zM?Zo%5;3KKy_d+l%sgVA9I1{kJ`TEwDYj~uLkV{$0R7&ab?8$=W7MBI~D?w;c-
zPDM1m&AmzYZJ2P7VK!;$6jtY#!gol)>t56+KoyKX|31MgA=)kUSp$8p8Rl1kX^-yr
z%9L#Be?ZO-@*A}r=AVO2lS}6pMtBY{mK;mNIkI7PajV;@&?j^hX%KbSJLvqnVgGIs
zz(3@nFChi)>ca%}+?U0!)~`hddZt7@fndMn$sabS1H)f${(8TkFi=;^+f!So^C8H;
zdP+1n;*$n;fuSVpj{v5$s5uA(c(zxD)$(-CVL@bzUV@)9a@TVkBJ$}vF~WvVBc?Wp
z1DAIN$S=N?n)+%RMurr~cL@u<j7`2P(r+l)RC5ChaC?n6r@fD68v86)*nt}$$A185
zh9cqK>Xn+?4FkaZCck@yyRA#)ciurzUl86!k-Eu^*BzkqJRtnE=-4~pKhA>xIPI1!
zjK$q5ADBM<X1xIFLBSOU-gEcUFdl6B;}QK9zf!?~Wr%@(Cc3vru;~v&^dIInz5^{r
zx#+qv?Gw*sJT~3?b{umc;2XCoDg~Z{0a`qLlCgfj-w*g}nEteb9(rZz>j?@~og64^
zpu#pVMPy*Mt?;VQ3*>iitXSR?OMp<`+$%4oZ+uk>QZ-yU!<hUDykH@$B?2}DsINZD
z<fCnvpZM0)M?7-!us^0iAL|^R49!3SaK*D7+3|Tx=Pv_rkulF+o&u`<G=YT+Ja@Lp
zCmH0!D)nWk3ss16*;4m)*&i}mIbg5x@e`8KAaSA-Y~<=NK&EHN6t;ay!wNptWsIbw
zUG$MqRjY7a9ahhP!sEzA{i?3##liJC2fvqPXvKDcUQED2Sqa4g=>MSK!}mvY`u_Un
zhzKY|`zLFAg%AX{`@#Pl!NX?1axTcPs5Ps%Z^e7Oz?<u%d4C-{%;*F2rff^9hF-SC
z=V5khH!nqO{=EkNiJPmXGPxuEh2h!0-f@W)YYOQ|nBn(I*9O`^P~dw$hFBHVT64g?
z%N>4K<vh^_(z?Up^*%d?)!=Jo5ZW~Od0l8fY`-nwzOS@84`6rMVT8qn*U2|VLD*aF
z=8QKNb>wvm`1*AVDdZyn@*}Hj6rrG62FPddaN)r!&EsH0`mOse)tnPqfFX6CE5H5s
zryIYAe>jocz0xg-IH?8f!y7kWODL>N5}^v(dCznC-LwX%F9`4C;<ZNX<`FHVcdyh(
z#Ag%i_uGa12S@1~7I#HeVfvPr3L^|CdT@JyH^*?;qXjEMDa=IdZ+uY;1}rWH^hdcp
zvc$Ge4@AFdU1}DJ+}*tKY-`Vp^JAG0K0K|s!rl0p8@SmgQ8;cOa=jEwe4<+Ez~5$S
z$2TnC{ryKLM1$Y&P&L6eu<OXc28>N`@uh=6eqX8Ngs=tp8>xR_nh+}#rB=FS;my_1
z*iT_CvG5n{U#|m_>fq4&gboatZ+L1w8HSj86MgJvTIbr11i(h!mq)a}5u3u0XFwtS
zrBYb6mLsrmyft%6v31!2*q~XCW_YFm0p)!V!j0c}KZK=$dytG0B?^}a_8+o>0X9-l
ztKWfi4eXBqPt*$cfd0f&q9*g#<V4mL2o%7J1q|$^gc1OTEANKiOTE2KgEhQ#4A6&8
z`MVJZy9#%-?b!0{$@L)6XDu)dNi$LKKNg3v#3!mX3i2a2=?w})?u$#Dx-ZNr(QvhF
zcGZf*m|^yNCE}3eIz)u1-vlXNF>eE;Am2=@W6)CxE)VLyTl*N&u%>{+NJVsn`S;1!
z+c5V)fooM_unLhezd^gX9dv{*3z5SHVoWq~>&)Qft4;mK)WOeXU1Ojp^(_+c|HgFj
z1z3xJ2kQMwR!%((75Q@=1M)2m42iIq*L57o4*&prV7?fTAI8Vjg7wa$FcZX2>HDIl
zq6k?)luZbG&mU77Yz1DpT*&{54kZE}Li}}P>+=e~JD{DlLQ7ulcEEfPUhu?u^<^ye
zHGt?w;Z1fwy{Hx?@*kfV2G*;zkpUZndoYwz%nL&q1FqfFeTy_6yzSOHD{sXL0IE5;
z^-fOQ^6n?ro`6`&-3ndWQ;t|R9aYtF=YG<3AZIfc=2+iIpn>F1L}*Z+5Vn_JLWb(J
zXA*Q6!YnGHiCLnT!>W#d9@5-|2Z#oi0C99P@Ji%-Ptu$N;QCaeXPJZWoUW1@;Ep^M
zTDdn1@CzA`Qwy)IA_t$0oEo3UYaZ;mm|!k(-Q%fYsD771foMXPMuWK12Hs%5&I42+
zb<3Ze%%<>im06%K<e^J`9K68PX|Vma>2^8n(Vj%JgB5Hp7abd>XKrD?u=<3Yikg$1
z32JLl+_=M`8Ly=b;8_QMd0o}8{J_bUR4q<L=3oy{duNLCs_t#dQwjxqzrPK=fQ;4w
z&?l9yXkILRVu=G|LTwwuQO<F6N>p>Ea2|j#Ugq^HBwWv|di0&L<RgOdKjJV^-F%5T
z?)1^Tn!TID4}mopHY$PnOXnkJMWZ8x^%y2Z2$nU}9Ah(sY()pco#hYKfOM=&WY_d5
z8Tv>#ES)2<E;|o+w}T|XGZ)a4{Ht0c_cjei6nVp2LiKPOm_Vn2e9lIl&Dn8lKrol|
zqa)Tb5=h7uVIF6g?8;QzPdQ--@=tiU!kFyyjZTY=gMgR6j5{sgxT&si#0oH6Jz}9y
zdD9HRqW%q)mp{yXJ}JyO|AEMn!wJKfsSEEo%dedI{q8ovgZn*b??1L-h?pgCa+4-K
z)WGW~I7P<+-0bEqxec=fUVO4Qe8PHVAN9qXTfiFzfVbz1PE*6U(k`9$3p4XicUt<k
z?(c4zJgOFYSo^ti2#!nWm!nL_%@`iY$nSSPy#4WgYg7DVF0>ML_cJaWec-bFfY@G6
zO3|HScaF+9i#+uE5;N!&Q_EE$-FR4tOQ=?8aj5sAWkgJN_MBY(kmBf_X+Ym6$+_=o
z1~})t{Q1oz<P$X7B=o;UAG^N|2cLH3|DG$~=U33CyM+F06gShexb`T~cHI4<+t&Zj
znY}Q8DW52-ot!qYWKe|d-lFIdG*1Y(9BP@94vq-2ANyKdYX@CE&IRwj+p;z)8d`a$
zpPb5N$+IxhApTvNv@kIh>~TeDWnt!bcpYO>_Y0|0M6#+TcgzEVPGnuI($(PK<8_)#
zW4@=18d#!b!k$!~;B&zG@dZCQ&fP#GW&$XV=q+i-nwIv5$@U)krQv0J5jz~aa_XLQ
zjYJkWb`xwL>S}6aRZ<OJ+#6_M_4|AkYdm|#!hllWwu*~%_eFQ@%=FaT?-hf^0?|S%
zobqV<#_s$xH(<zN7}68Qo?xqvKL2FTi{=!4sc)~j!%}8EHrRUqK0B_T+-&F@{)@Q#
z?_Rq<A0_l1;qWEjX*<(g;8b7_{ZOVXD@<>tIo7?qNRkEM8SkzHp5ec<_o>7N>qTir
zV32MJ5i@ez?KpPQ2Z8z9jwVWsgi0A#g<u2$&Y&jG&Ptu<7S{TIrEN2h$sN|XvTr+X
zGHs#5Y5rdbb>)<4=LMAl$2@Oe-;)_$Qc`53q>{nH4mWGwU95>wVD?Rasdsc!D}L++
z9=7+LyzURXoKOVS!EDEE#7$vV&wS5V|Iza2()Xx?A8BxoVR7jTGyKM;J}_Vp4*oCE
z`|}3&_sBjY-1D!1IY&|uf@jVwA4U7=DG_f)TxMp|Lw!*%;!QRCk_TjepWe=T_X?;(
ztq-BP$QS&<>YU@3;xr5LS=2JG*-rGmh+>$<Ug~4C)IRYmINNc}7+|TwOI2^r>;A3@
zA0@OA%#aBaFXp%Gulw4ie{LLms(kp_;2ZPLg9Q7VbgrAf9l$$+5wi8YDTlY0^0;|#
z#jBPyqmwy=-B^qJ^KP3t(^X3<ckyl84-(v^ac`p_5}pZhJH&a+5+hhD(}iFvUjD9)
zl&X(W81F0RBg0a!05YvB-S64cdKJ;qf?;X<31DeW7vq!?f>@LjMQtbhb@h2mvCQ%Z
zBQP_lpN-OW8x9-k58;>*Z@|$zN6CAK1#pbR^F@OqmH=Vm#R(}NNnbYgv57tWtXMOi
zVg{p4399RX4vco?OFzH{q>;i#zuR|+LzLS3$9Ddok4R*j10Ar|co#H>Ny&=H{rtA=
zL}0P?RyEL$f|-O~x=O#tCWeya)04ek6kK|eX@>P<3%B`&uqLb_(Yo>~h{i5Q6&bTr
zCK&UpLh3xJnCMqb3I`@d(j_q;Kc0&QQKsIi6`z~<ts~eL6%s<5yuV|U0yS1J!}_-K
zUHe4~ci?=Nm<+7Tcd!HSn5EFliHn9Xz3ymdks0jlGr!#UFJnzVNXQYHY*I|!PK=xs
zGuTPU39yrMF)29T-thWnnfI2$Yu{gkn5ZM`eod5C2aNiZ_(d^%+8=V;hr{MP*qNAK
zn&i)l2mel~1JvNNdKCeT^|4CR&Hm=uSZa5i9M;<8e4w2SO(p=7(bBI>HjYEaB+o(3
z?Eo*q&p2^-Y}(WFbWpief@m;pd$vM{x-;AI7sjNb)pOFhKb0_^47N`mod_ai&#fxi
z9dKsdHN(#pK?@W~WXB{BSJ}ZuK70*aC5&z+QBJ=3Y!Ah$<92wzBsXbeB;bg=^vpdF
zc#3(xZMu5DSl^BT*`vG5o??5`6P~zOyjx%Zu6yPU2b<Hwa*Pn~^tX$RoHLxlWm1Y~
zX-zlw5_hN69SJ#1u)(`RSPxU{%|W!n3!p0QKREIe*>t%Ht}?Q+z7lgKznF2S?W5PM
z8@5A){a`j3FoJ7vr~K5vJ&vatX9?+3m;4qly1cZgJ3H3hEW{uIy0=DK;t8OO{NQ6`
zxvtx$WeA_48LGphP9CWY5>hp6cpqIxzjyuH@Fh)rj>s`F-ubEH1cRJ8L4@i~dlT=6
zS{>!i!BL2je*vCH3N$y_PRrxT23gNjK``&48JyuS2H^Rxt0?;>zy5sb)~0`*Ru+@Q
z_Q?sbj&6(TxuswDgPc#d9n4<3uC>aqor>{lj6!x(<x3d;V;fQwl?6R}sU5l>{<2>V
zc7j(=EEr=ihI)T^h*uWCZ~jg(?)o#Am!2Cm-51pWF_y*-|GF893v-;!tO^r<GW%*L
zDU$p_KVFldJv73C^LRbZo07kXB=7R<9Q;L88njuFvLXCE{nvMO37VSZY-$WbwiA*O
ze7c*)w*Lzu^4Hbw0T&<@YA1aN1do4XZ6WkmgIMGuuYCJISje+L@3Rkgef^$gjzR)U
z%P^LWm2bldA_dR)^-fsz#ogd)&zZg*kk~`WV0%zDg`MM8&2a>up6sOMfw;dGj(8(^
z_%S?!zkCM2XBd7%z0^>fH(VpnHd*Tf&hPqI-!^qdEN0!0ggw$GCdc6YWwZ<qSjLz2
zW#A8f&F7c6u&z>rpdiVsVMrkpSfSy^Z14Kx_1@ce?D_(D)wBpsEk+@cW}O(&?s%~E
z?*8Gornk@naLix<2i6uKs&#eh+oqhgB<}OSI>Nf*#1)YMNRR-2zmt!VzgV~2bZqBt
zLc(yz#krb@%T%yWc#hr+d+^}Fmvi=;8h;n|K&D13;=R)6!QM$cVWV(S7Ol>C{@MYS
zmTXc%ys?W3fKCk>$dH5jC+>#e>NT>BqeV8Do^Nl;Ko;H5^T2FL>+tps``B=IY^2C-
zkpiFkZmx;@|AbGEya(~+qpqFREB8B^(Qa#Gx$le2rZ&|epU88REK?bIDs%AvVcu){
zN}G6czUMIUIw6U4VAdYwx@ak}VAin&1G3=al9>~4emy{O>ZYV6QEBz&z_JHm-~&_w
zHW$#qVK6YA&;RAX*TOH6At04-2nd>mIIvv|+HiO%1gs^|;3<ge|I5MIs|uW)RPya@
zrijWnrGtALtTDMr><07_Y|^nj1mV-K<P}#mkApjPvm=0RW2?6W?QO0OMMk4V2Kvs8
z^!?nWshxFlcKc|ASs<J!Tl%o6X7@#FkUNP84Z+(9zCpkpl*Cr$|3*np@ZvBl#d`+{
z+4c1NI83%K7g1Uia$Mw;`4R>+`W-0&2mL`qT~x;DqFk1l?&8H4{TDH#T_t6J<kj;v
zz1CoT3n?T5f)=XE0_bq?`53{hVg~9h7zg~HcUoX<#(Td0`S~f&kusHLeWEW$$nUh}
z<%{q`Y;v^laeM)JZsL8}1P!6z?Rhx>smgB#2sjsEyAO|wl5{%vZi89iB<|C}Y=>`z
z%i`|;^8(bzNZ$@P)?z3-CZjd>>+30t{yKX>?uw1A>I)T`t@c%bGaNCgNgflbbDhDt
zct=B&e|ceC_p>u`O_&4`BLhQK@V4kpZ;DU*oD6=aa-)Y9{#gRCNL=7Rj|HQV(Z#R&
z+ck)TEAz^-*5YN+d~hqJzCphP;aEQ9$E4Pm{qtu<VNk^7H>HEkDem{F9bzwvt@|Lf
zGv9`2r8J$d<oq^_JqgHsNCwC5J^j01Kw;4LOmq1S&U%195V-D*QcXg7p5Q&7#<!AR
z+qLg7gQwsaenLx))j@2ce@VlMs-lmd0|M}8h&K}1#=x6%vJ1yVp%1R@>+yA0$PN^Y
znT=FWT|gwCQ|p_1+ZnNwbkaMhI!t1K>Jk%{3&j5=55r!W<>!N8F=*J6*|%WWGqMZE
zVc1~XH|FcBOV|y=8dg>h-$eV?E3l_eMXZDjxRZVJPd8QtocMkW?2c9FtN;Xhv52`s
zy*|L)GRaZ677=nBFs(SarY*TBIC0Grf8&LME$>bUWGQ%ALfoi$uS8r8jyIvjUO(-R
z-5|91zFm}<Z8+5Ah_iyc@1;~!#aw8>SVnac%q!Mp{DT|5sWHIWVuV5~Iajcv+h%$9
z$5Mh9rFrgrXV|%O=iIKBE*tw2+6wl=R^W0cgc#z*8{ps@D#|9az0<_r%_zVS1yh2(
z?rUGy0H4(z-X0ofiA=Nz*l_-&FCoJ#W+k8L4kyjmJjO95hqn?=i3tC_oALn=j@-ad
zrdm7@e%>V*5pF(WKD<si(o2osj4l2y=wwC-u4U$z*>>HTKR-{Iv<o_XIvuU6tD9_C
z&!hsyn}h9nsvgurF^poe`Gm-OqXPusnfpu(G+s5Vk9jo`Sw6dQ?)x7r?1mL`znf8k
z6^cv?_pTpsHHcw_uYy7Yqw&GP#URbR8c%-B#l_`jZ$%(7Wrt6pvNR48GX5Ml9z;zg
zAKg#Nz`{aGM$2Q-FVD!vwvT?CWZfZ%7H)?%?F*4TfrCGD2?C_F9P>0bU1I`9$AcjB
z5xV>1>+u&m3h~adR~O48&UPctGAI!;oPWCSRgvq#0V*<|xy9)YSEhMrX!$sNj2I({
zk-I9Spy4p7z{tq>to#SnhGL5rgXh(Cd+7o?7>RW7NN_CZpwwWve6*}s9dl3fn1t5C
z%%~ZhV7n^`OzGFXgD5p<dFiJs-QG~CIORvG*7uf^c=RZcF_Js(k!sw^6Uv&+4oS~2
zbhPsoqr@S`ai`>d2^|ega$hr_d(6HuQV$}b;^k`~AsCUc7_>ryc4SoKWvh{Pb-<oG
z+?DoYFBK2z<Z(lub-#do$(ML^h_y9O9YleNEjq+%u#sOwC)YPT#x=YQEEbdf^=1HC
z`Xu}hF=hq_uqZ3Lqa%Y{G@aEp2Eup)1>v8IuK-D+l5l2sJ~nam;DGGghbL$50=!q5
zGcnye2~TR-YuBEKoNitx9O?NFh@6=sA4iM$A}-8<3%TLvy3;#c%C|JFQ6tvmnzG?Q
zL!2)B?*^m&1=^%SKnFi*+>&}x?fHStanjND*BJ^Pi&l#m^#14zHDs{F{xzw$<ot!~
z(uu1`{6Fz-5R6tH4l7sdH@_W;bl@=@iq5mP5;8z&|MDRo5xxD*(YeV1u|-`{Djdt5
zbpx$EmBA)fyrp#Ltg;eh9tT5ywaxi93`&dq8S}OD#_;#7&e9ZocSBH8g`hF+*mUQn
zs=yC#o(CVu<IOJ~k33s9nv?@3rIYaRo*yaD?C?vhgV^pXfLBxbe3Za_Nn%j*?%7~d
z5?$6u?{gczU-CMPQX@}_+;A!qI7_pJZ*uyNr)!>f`FY_<?4{m)1w2-=z;pka2*ys&
z0LFe(*KWHmoXKi(ZOk2X%IhkH=cO-6Np(ZA)m_y?8uuPLgFa7AJ{7H;6w3DI^u}-)
zQaHDS<Cok4CfrBc^cm9RJg~A4%d+_W&rY1jK6VZx^A5@nIpL2C@N_qL-T2#VA@_dQ
zh&9BiMw&<}_ro#%ChWIJwuM23KR-Q%G}YXTyg-I^CkTffC0MN*^(lz-Fptn`FEp$9
zCQ~muF?JQZdqE6bQcAyh90!)Dp>5&Bk+WQf(J}reOzW8d(?Y9eu9^%JO^5U4bzLQ%
zQYa4qCq~n<7tcfmF$o1r*Iis)yU4KemW2_(w>Mg($Zdo|iclPbo3CfCUTS;q1!`!M
zY}zG`wtQW$&4VCl2ODzRp%o#F6lm4Q(e!b*SwrgLRF$TNT2DeEqlWjhv;$=Bx|7S;
zTveY)<8~6DY~?;`!9xl7Hk(?_2yW`qc6W8O*>k{FapO==$?Z5P6;Bpf2&HVG|7ZBE
z+KBAiHJfoRxBy*<Y}6W#o<6w}fCYd>nZAnK=?#)quVYR>VBD5%;iaVMz%9c$M(oVU
zp;fp|CH=Nax&5ilQDjjX<~5Zaa_MJb;<L|a9QBOU`=8P+{kgTeMdZ>PIAG7j%<QEQ
zt4NXH-;rmRaahRq^atvjy<wt{cayy{?J8c0zW}HP9ZhGgUk8w^d(9u0&^|RtPWcM1
z)a;}s;Sk|mT8wk4$c0Rd)!N&~r=RLZ#BfWR)a%o?j&*l;ugM4;m}KOYz)ox#(e8wy
z7(<c>VeI|%ki6UTX~EAO0!JCKwQVFcXzoq<R9~29=_3LXH>Ia1<bqBu?cdx+GzdmK
z*?L*l&P22&z&^9@+Yh(PAks8|G`=z*EkPsiLY_8rHIL<>vOl+xVyN~ny8Ut?>N%Fl
zAOpYIo2v&;<Q?9zLDtMdU<#S1H#mj?zAweu-jh+s*7uj>V>S89OLk#{(5jF8Ma4nf
zk2?VxRKAPg*NGD+etxxw<lvgUeMtl-wdbL1U@Y<%IS(PeU`t85pRD(dGHa&Ob#gC<
zoaYe%dc!e0kV<CvmkYE(h2xbX?7o_gHo3=sZEVT-{|{&H9glV2{*PasQAR~()0R!4
zvYpv`7qTi831x+>v!!Iu$liNK*_u|79TJ7iknH(8UPkKk`8>XV{QkJ_Kkoaw-rnc?
zb<F4Uc^t1qD1g|d1sqV9i2C~WIRDL$1e+<3n|BvwadR;Ie1FtB7@wF*D)ATot|)AR
zjL!h<d_0=-qxqZv9}72;18;mtm-apc#p|BeP~OL=@WiId{{;An>~~J{-7Iu-?;zYZ
zLg+Tgf+@m>I)%Oag*EkTw(U;p!F@o*YZ=>V1`m%)aIK5W-6L>l2c-jDyCz55G6rF(
zX50C5M=%$0XdV7^tEQ^?Z@KNd6OUBR2wr39VQZfuoYIGiZxC8EQmECSnB#`1&_XN1
zIXhm^7z+3H33ER1Blt|N=J)A>U-Q+!x?~w>wE41Ip*UMZDdVR_3WRT23f&UBG<~>}
zq$KmvbBI#)@E^wJg}Yb=94xz4?Rst=lzQkd^myAxML-GR*Lfbi?bO%+MghEhphZ<Y
zM96-0n8YgC>T6jzp|w6eFZqV2J4}A%y8a<==XeBAH+xw7I#j^xh~$_n3>wpDPk8(*
zA`z02g1UMDw{UQg+p05aHza!(J+S$h(OJviLyf9M0)GL_HxM*|lU8rsXc!MepdS}1
z_Tb(0@$j{(%V`}(?&X{ppR;a$m8YkSA9^1bZvUzhykm6#8!;0o<~9|LsHq8H99&zT
z>@by1oQ1+K8tyoI#9u+@?~M&Qp=?s^d%Tt+2~hhjPw-F%&ph*rbpm$C1KR;Lx1Syh
zwc1z5{{17L^Agsou8syU$|eHpV>2_{v|+rVSRg=5dn;Tu!1(^b%W6E3BhYE!Lv42v
z)ppBKC<!Q=Y~?V63#>mOYX$)~vk2EROPaf2>2z^Kl<ac4y5U-xI;T1fSUtBhH=RGg
z@aar(tUu(@t6BBKs~M$Gsa-6*&a78-qluQEbN1C<+TeqXGWVhC*>wj4;I%(o;obf3
zciL+E(eFJ7iHT9bn@0tOXQt*PMQT(xgHu7NV$SO{J%B#cCD}gcVyEFY+nNUM;O%Y)
zj9CJVnM{u2S0-xEI-m@$&ebGl9W`hkjcR^e6w0Mnx0lw%AK9H5e!5l~!=1VDfD-$$
z>%Zl9JfbENe$xkC4G6Le2p%l#jk*Y5NPZyvT692#>+)#6gn6|eZP>&8N9)#Bm(drh
z%B}));ClD&ox|+}kv|8A#lXQ4deqE4N3S!FUMG*7m=z#|AY_bC56<pAt+@`@|H(s@
z-XBU&j&sDE{c`gj6smNF2mc~-c$pyTNcwBt`wvsG(T+H?0GcIxu%F~^t3?^d`Sis4
z14-o&$ZX((`v(=jckL?0z~lMarGHMyI_=aPZy_d=9aVo_ieio{f)w7WaK`RCdPa%H
zZQ=6jt~VO)zn&P?fWsZ%D4VF&CON^1O8G6j_!pyCAMYG|K76F$k<SQ5I*G#0-n-<;
z(H?z)dywJ{^Fqv^(0o<~7fwC)3h=`!;Si>mZ(4ulM-cRaTbAu#0uO2Z`MG6v<8|J`
z>tw>~-1u6id+~);mUwLV65NfhXl|Ys91`-5OZEgN53jYqpw0Ms=T3^v2ajhLZa9X&
z3vXH*9hY^KPMDV%N`a>H6Fu*RwCoHQ<{`bvgKs64m5GRiHxhg%`<#3Eo*i&e?P3%=
z4lAbu_G6<GJe`aVtP&1v@6BXtH6goUx}#^UT4^~sXJLCS@7hIn&c_f7$rj+|IQDDN
zPRUuWtUJ*9_Rm5Ml^UwE`srHW+b}+w8EsSCs2q%|eM58$t}?UBz=c8hTRnnTxaywm
zFXMrAhLdXkZESs;7}|+B5(q%CSDkka-mYAZ+ILJq&HW_b?82}9=2UgY&eGe~Uu;Ym
zNEKr(I<oTt^wsH1+^Mbw=pZ8Q(n<#dHv*!p(BXD63eTkQ_kHw%_AJR%>&i2(PE1Uc
z0(|S3LjP_k(=TM|(7P_p<dq&B`~z{~1p;^+!bI^-gbhV#ONh|Fp4H+9Lu9T=iE(2G
zOuvctgOWk9P14LnkFeEY-h+==V(&U@-6Yv5-LZ41#cGUTr1!wOA)s{5;sm@-;gesU
zrZFzl@*h%|@#mfXN8USaaRRX(5QIK|4LiPb^ns|liX=}e+jt>sc%fa9wSbz`d<Rx+
zn1mt}-C<T!&r^A9)0hSNz4-><Dk^2Sy$U-RmWdP`y3C<8Jm7*jp$_j_42Ou=&duw(
zWOw+iRU}I2+bNL2hlk+vL7g(kynV-oTK|bXxrnnqap;{WDAD~o4N$fCEDfJxu++Hv
zfcck~d286KqZ2))NnnQYhSl#LV%agO;;YZ0@#hIK=-j(qs9@<3)ob3D#{srLN(%Y(
z_?~cxKkc0F$&=cGlG@0V`z@cHwGKUF+4Yt~f(CO7AF@2l^^bf_njiWvNSxR)Aqf<H
z9<u^C0<pOP#pdryT(O94PfM9ftqhC+1wzHCL<WCsEFiJdu!mO>oQ6D@>FCbOw164%
zaw)h%e|Z*FJ2Z{0fIwUg*m7q(-Gk=aiE#7VsHvXRR^`3+S9x#&%7cQ6)@VI=4JCId
z1F;`kF?B$$(UsqsZ>N2$B<e8!6iU^)I7!d%p^W4>9SkuQr#}#W^cT*IY$w70FbAN`
zw;>HnKZvcOMqbqunFjd4M+WP%Uk~lZKdXmPDhS&^dNBKh1&HOyh9To!Z$FUn!)ALI
zKqlMr+|QsRr>}qR^vU0O^8ZnKJbR2!L7ACAbN1OV8~bQZ;-39S60Hf+2$)C(;O<7#
z$*<)aIyz5dnf_|QRiI?Bkq8lj_ftjhS042k-Y<(G>tbObmtN5@6st5YQ|At!x*{cY
z#E3MP@#N*#c*Stt#m{kx9VNz}wlEs@Wmb=$fbZq_@hzqf`nv(h#~rkEoaa|VTAx{0
z9Nvo1PnfRf-QVHe=+2_LyT3vGAePR3qafr(k@=!D^xYdM2-tI$X?^H4BxY2g;!OU|
zgeU%?*_%;kzK5o?#1QbNf+qm)7@zLJ4~C-bu-`Lfe%yI?X8Wa)l(D1d*fE57>DoC~
z-UTT4v&x_OV}>4@@IH}&NrnIrBihhHHKml>G)p}Z{z03mzH{mpgujZ=b<@qj3Oh>I
zNYZ6{I2)8XTZ;S@H4rLqS{<H%AX%E0D5&25Ycar%bic?!&FE6y8x7C>M^6w_bMI69
z#)zpU_CCxfbG0EdeC0s!Uy1!Cq#N%b4SGoIN+?sjX)yrDVvTQ)McL2bf>al*|0i~W
zU$jxE=5Ms_Kf7Njwj?VD^*pfIu^4l(MFn#I9s@4q?4rooBPK9)P)Z~!hH)>|eQeFC
zSW+16$l(F7W(x;Ea3e}X_x9|!>rC7!Sxw|2{)EN~BAG@O`am7Y8z0Ez_m7XwCH5Am
zB7;^W3@S1<4i4g^U%Pe*V`|s(5~76IK#j2VQlzMkccHZ@-*_W;cq3e714aR!fzXL1
zuQwXB&cpPF#atY9h-eSW8D1Lx6qoP#>-)g%Sf=e~{2yR`N&%deaw0Tbs7Xhk2v&hJ
zqg%W;ZFmYc?w@ZE`K=^es)oofAd%p|=upLKj>#|0K#)+WC!Ci`i!QbpuJ17&s0l!C
z(`F<O4fJVLJu|VyW_Oc_uc&E}9{%ghtq`x<9+SZ6;W3Yx;N_E1;7^temCHSoCVf>n
zl(>NZKrB)|fOBd`Xu+pG6<TQPmO5lb8L;kr8-Ijn9>yjTbs%icVLv2O@8l8>4Sjvi
zy$6o@gCUM^oY6FLU$qwm$nZO=!*no^{sJgDF1-C46QO$oww(XyV;uZM9`sJ7r-53r
zP-Xy@BLJc~aEod2P>E&f7i&1rPaI!r*o!w#lp0Gz`Q^=@#*fb50A8na_J>&~!q8`s
zAgWA@S}o93v>7y*%&MZZ!E1deEca!-_!@|y4p!+Mf5M(Yz91QH3<#kZG0LD;>EPcH
zQty3_bMRgL@Clj<1DjgrX%L0UZu)nFK>t3c`JUTBTq=UwNAw?VAA(KkPMj}B0-pj!
z<lKC1_`$coI>~kwS!U$4Dv8<+`v(UHL*=iF3XE9t;V;DnnMRd{2t?g>2J8WNpTYm4
zBpDD-jtY!x@IFna*}9JlJIFx~nNYw`0r64~K@g+%`Zov?8K&Us4n@E-G&)0EchY$?
zsxMB^m6l#aFLVc9sGmOsKNt#{THtPGC?6($wQUpy;ndq)3NZ%%vZ{#F76ad(@TmNg
z&IQ;Z533up1J*=~IzR^-b`<`kz&Ar%H67pk9A)9V*3gPtVnTeazRz$MBNLpnl1}KK
zpY$IloEl!mN%OiL$bv<$8r_6K(3G8+6?{7cuEoA!6c?VpHZ8C0FhAZ&{^D3T!7v^a
ztXz<m+f4)hI%q~XXu-QfxxiVSMJd4`ybiJ_RRFc2<2kU8ta_p(_(%vZloq`lu^*>}
zN5!c_2OWSUducw&pZCG;K}Y^y0J}a13OA?0xldicy$kXCHhopQ*x1>dwat>(zLB{5
z1B>{cOA-~zp*%+1tIP9#A<8>Q;{QIYlpe6({a5hDQnha|JA2hZiEi>K_FM_OoFrXb
z)nk@u);sePA+KM*zHiEgAeqgUw9gyxyleVtZO2jJz6z8ft-k}cnv5R}SHZQ$c5SLq
zX8^kEpfGu(vG53Mg8{%P7@5PlbLVV?)rfWi1qX7d*B{P-L-`?x67HCQFSI8BprBf~
zbt1doOKuE$+_J?V&kmZqoP$;rx<|beN>BZ7+^XA&q1#2S9n5fvUe$*4s!oi;r!al#
zy6j-yo*7=j%XhxqO|kr;V!7{={m>jiTnqw`w7bV=dfSkp!T>16=Frno$efcK-u=0u
zB-q|M0P<ZeG=7OzJtZaMYF*Gq4uAoZxDG+7yp#1h|EVbsBLe=iTDmO^Mu}d%jov6J
z3O_j8ptKI^{aMNfNoz|Tm7##TdLlkx0B^K4%~+)U<;#}>#m8A6{Q1=Wh#1pN_<;o*
z+N4D+t{K#oypV&oY!EuqCUAH7x=NraITY}o8ZVIYkg!AMho`3)JHH42RfPfC6e5|c
zHAb}2sDl8_Pga<iwNZS%Odl>OTk`tQ3K%tTG64|kH(D(r9NZ!zc|tiNv7X_t>S7z!
zquGTg!h@>i2M*AQ-g%O#_d=z_ZX_N|+7)gh)~IxI6?C34WAz~+VB$~%l(n<aA`EIX
zx_|J4K7TMI&r50Tlp7!zG9<rR?%34Y4Ci+j&Yi<mF!0TS5&A>9`4*YA5H9_E#u2~M
zVxX4$Cy5Rq*{rXy#~a8NLmOvzvJ+siXa*!Vv+OP!LDd%|K*Po3^slLfF<S^?cjtq@
z>c9h_60ji%y$%Zrz|Ho9oBet^j;RY~(;PTkx#&|7nA-|uxGeCIhc-!tL#M9?e+cJe
z`G-`fL0xTTkS!Ld5k6EiA5eCH6HBQ<GLTR`D|BLoRy`Msta^@YoY+CNj&Zow(+n>F
zA*ws(*nkS&_OLEn<n<$s@NFK|a>k~n%#8Pxb~xV1s|a0B0{C+o^r9K4aKdt^rD-sC
z7TcykHT^H7u`T9=^lZJ6GmLm4D3(hBP(N6O+pUptfy=;w_2+DdQ5f77&NCB50&UJP
zsA4j8@?QVb=74rqFy!7#tteY$NZiPf-kDITWxxDR<OCOx%uwD>`fAsrJPrDeEyK<(
z{&B=<rb#x4Ly2uC`$R&uTt_IpSNcDk2pK{RKbRmpdd#%!oCufPC`3DtvKde&()+9Z
z$%34+f8xe}IHVn_^W~GRMIN6l3@QD;69pdAT^2dD_|*EbIaNxauuA<U6$6DByOkjH
zqEu=f$<9m;PpG)!|5gU~F@S?{kM4^Asy7-boPHK<zDXH;iRL0pJyd^zj8#KJg8`V_
zIDRoY|4c04HKwaev$a$y$F_r_2bKzj>0RNH!24Ix1cBg9d=EKSEWbs&Qjv8pcizg~
z!gQ9|RF=71`b*le*BYbqZ3Y!eOVz=Zw(F*POx_PX!0C0Z2tASWmLX!O%la|;G>@KJ
zbf`ermk2SjbSV0c$@jO9w_@UHi}sp#)JBp2#6**G;1X2+f>I3#@*V+-%=lyzrVbQt
zCatOJOHpq$pklG>%r{lss0da1e9nIL_XQAK`>Do`5elLno6q?Gz#C2s=w$*nUxw45
zL@T>|?jrg+2#9!j%NTUms+<1)3Ch+^AjI<DvHgV=GNVX#_s2=W4K`^AET__um{nR@
z8nppw=QBdxuJN76zZwbGxZkSxf9#=x#fKC8t8nUbgC|T(GF}9I+Qf#OAjT2QxS65Q
zq9b(o-UE*=7I?i_g)~Syd;iiIJdJ}o6)wdylfB~qG}u+=;3)^bYuGnz6sI0=OjP|`
zu^`@*3gy&|?>)t7U*)lD`YpLC!Ym;&1L?wy<a*gBlXi*~8z?i_Q&x^01K0_~)v1N#
z{eW01KJlgzy?C`TItv}rOtYdXpl}<9bQp*5LioIipt24B9*ku8n@_8Whj_tx#Wpn)
zK|O)M99J8b?bKntO`CNXG{^wcY&6JbI4Z7-HGroxw-+8Zh@RXztzybYvv>5IoSdwV
zzHdUvjo(HIq*Z30Hr!li4c%{Zrf$J-=A+5cS-sKI`WyAwXG+JZS9=HHVy~iq@E%aB
zN3s*w9AmhKg7XZjIu2RvO4?*<POPe2P6Hr6N9pNF`pqhS3P}E1bd!|+Wrdy8JfII|
z*ih=R2WZafjTC>f9a4OvTt~Kn>CdhWdcbIXsRQ7anJo}JO32;>b=dm)kxIAfcK9F@
zU3gUJD?$AW=;VsQqWihqF?GPL=p7bu%ynboJHI-pi6C?U{x+zt=N`^=Ee%n^Ng7EC
zO}Z}IvIQ6cNQ1-^=CJQ0kim@`-{^cMR6_CrLox1#+?6XfVAolhnMw)@3R0*vgOcI)
zms_4n5&X3iR>3>ZsR=}fch2@{Bobcaqr;1_+BJBeT*V&@QttBQIP_NK8T#b~k?0||
zKRyH?+1o#Lhaf6^C0z~xRqI5Le9Wf=-i0jyK!-=0c@Rj8IWWi?k6pVZHc1O$3qxF&
z=N@0AHvKEhr@*heJx82X2h-%D$&3&cu_IJ!R^_u7fNBeM^UQ6f5p}Ln<SC~>35<+5
z6`!n7EuKp0^NwMcAqGX*Ray~nj3`vJuEy(y<%g1m^O}0C{5V^Q(pMCuLFfnHZ=RRO
z(r*4)PqQ04W;lTa_FpUFOh=GO9|4u-Ud<Vdm(9{bzn_oxjgk2Uqjcw9JK!dyHtI29
z7=2a^P18*2`X?WABZbdhl~_VRpUhDvop=KDd?;x@o9Ol9>A>%!Y7FRp9)MVgP&HOd
zl&bwML;HP(c6Gsy1-=B{JLJyt!_XuiNTlX9kVuu9h*_x(>;lT*;@zd!m($2uRDvPw
zuxuDI6)QVCZ*)88ZKK{kVB3MK``96HWRD&=0w_GqMy7Z^4Ch1`z>)n@BKppRAZm}{
zb|@z3WpASPSWc!7*t-DeF!;mc|MvJeFq&iEZOy8Y9Y02qnG3L8h~U8>s~sVJh9pDp
zjdEQ;@v&*fRX}%`el2rqs{PA`r+~^~C$S?_28cr&WpxMD8St$8HPE7*pw_>M<K_r<
zm>E{@e=KGMf}4%d%ov+ppQZOg0kU8qo66zs#Pc8S80rS;#XKGKQSN#QL9zWcwVI09
zq6sS8fcQ~2`mE$SUCWZ>zP^USGxj+a3N(;ZNFF(J<51MQ+kc(~!wR4Z!m>0;fu9v3
zL}UJbj-sxuH1umHfWQpXad_q08sywce&;i3F2^TG8@d1*_w=yt$cX)S<8*%Cot~aP
z_WYqFgTF4q4pjjQt0FSO7Q<zM8x%6S4X_sOoQp<;jub)arQE^#Fx2yIhgNJA(&d)5
zXg{X_k=sTt<@p8~Lq9;!_D4;sdIG{P0lO}_(V5ST@JlAfDd2A9fPo8QQ~3MqAd>ag
z-kXlH=){~*<l%hjfy4u7cnIA!y>$2LC~!-3fb@AIC4wCWfwT~#$MWTN9Q6M<vYQ5V
z5sVsIT1<OMXo3+GkuPb^ezO0IZR7qellGuF?e@2W(u0E@U6ta{BnSAw4sczP*glLx
zzN%(1bV;UbD~n5Ya?*4UtC+*|88WnUd;=7D|I{cR)KH^1#f6!HIP|41%3HVZ!Us7P
zKri{QCXiVXeUROI_W83!J2Gs3kX(?ijerV~SDFp%e#^)E5VOTSj2F@O6avi&E`+Gy
zQ>!7R-f<zjPt?4cel()&`;YMZgCgfGVs4r+N6*~whmk62mft%flPXC#ACXpFevrND
zZ=4U;D{v{_8n~U15z6*9(+AB5jtem*B_-K_sUeI<P9_%S8S?GNE<Cn3IH9@sFA04<
z2kF@D7hp8wXYMD^_j#aKV~1C>OT!Na=)E3V#|$3Q(2|?qBo3dsp|er;aNY!MYb|Ix
zqF&N5??l1>Ir@?nMZuAqvX%PA4ydzjr#|+O90DdPN4ODuU^F8~<sUIsHj?c`WXICO
zHBX<7<B?cB>>^e@M^3}<tjhC37`!=S{WWsac%jgifH53$)Dir7jY()orVV47u7D$j
zAxCn5OEkXt)7qT__(RyH6l$}waLXJio4%F{bA|wvMy00e)CC;?AW>|$E^kZ`ZXVf7
z#L`=o+<hnjBd>ms>mUdk0OD1MJR6kov-lrcVLDM$$YK1@i+$k)QpAxqb!_<{o};li
zw*So^=dQYk4(ziz|6w>VW|XUrmsCZQs|cJsa7@7eW|qQgTvt+<;q|Ow3Oq6b=#6Ay
zRwX2s;xC-s3M$IMBAPhSa3p+h9gtA9)&WbAgSI-<ka_&4Fb;;Om}dD|dtcpry`57$
z_5vE{l-ptkru=e4lPGM569wIS#)pX0^%n)$lK^O)2Caj1(lp!viDAEa1F_#ALR-ME
z-64o?MnTBzuqWJK7mJMJ1hw;M!YIU6RNRBpE0&YZ%|v0Lu=lCHtevgjd3&gV@DG)?
zh@>C1cS39>Kr)QDtLw6R=U5Z`#MeGMz6&j!Az?#U3H;v2&c~@i?1r~BwNY3=yrHh&
zJlFpvs#0}z`S|pd<a<Bb7-#(P5k(w~^VNZ#z{P1Il7HLb{Qcm&Pv>b1f*!3V8B8I=
z!S0eO4D}2Qh8Y!6c`Jkq+=_>oA*!C{B28C;1);;Q=g~mTz2%YP-hPOO2M2t`;(R?b
zt%s;5pfCQUc~l+RjrT_KP>FFA5`C@@!@xJ=rBVPef1>{PP~tZIhjT$iieGu1ATW1l
zP#ss8u=|do%P(m(0#bNMM*;K!h{b^diYBxU6&ElRqYvW`VgEW7qRB#fBocU$SedZ)
zL39_C?N@#5CclanMs?P|_#+W`G1U0@|MBfT_s&JatZV#QZswJDUhQ`;v#E+LCZcM8
zD~YqR`Xk5*)M@0@$$0<oZvYGrRQVxY4da7~XfO_18NW!oX88&cca<aC*`!7rg5F|R
zK_oTk4;3MHo);3y9Rfm_Y}PAFs@rKOiUq`P&KrUz`jcC7FjNO@g5s#1LnBmrtI`%f
zk*vh|R$TW{f%F-;Bb5b;*1*Zi^{;CU<KOgfEh@swdq`-;yMGV8-8ld#szP!^>z&{L
ziV(4rbPp$rNCiOo>H#D%<D%Fgb>!+pxwc@ptrW!aPe+#x2RT@06{mif4E*o@fJEb`
zNAX8?f)4OBpy*^bau8;Lg7h#dbl&aTGncAu!DUqO<suzB>J&m3e-FuSx>}gN&ErhN
zav?W*o9%an*Z@<?b+ZOiyIT;A^|oI^)tF*~()<G=E*L^Un)LkdKLECnL33O>nR^Z3
z;^sKXy%xCK-~?Ge1)xfRE?~Tzwjo%B1}YJXvd6?4Fkia$QLEa(FZ6pv`MEcyjj-G^
z{IX=Ff6NBLhX<MuTP#@?bOO)uKz}HnXdDeTlmYeD{vkE^0Wo!5?q!fiFM45Ja0LxQ
zieEv)u<)Q}Cu~gJhYzA^NA9Df4%HF4FR|#mgKv>=^G`zbz+ycFDc_)`PEHj}qfHeb
zz8KT~!@O@J4u_x_Bh(LTu^SzH#FKmF9TDJusw)x$J1n##NLNBwN*>E3vm(E!eh~eg
z;L1NbTPz*)V_uih-%MnI$U8a@a1CN?lX7!EMH~*DhGS8HAhvP#S8$ptg3?8n-BiDC
zKYe>z?XNGU3}5U^>(S@TEvT)F*B=h2x!AhXro;2yknl3)>koZs0SKY)PRls36`wZ)
z)K+{p`hK%{`y2d;&o<5IsapEO@_vVjc77iER8-(TRAgl6<n*Aiu)Otvh*r!+2U@Iw
z&w0ihz=~WfRQ36H=B?87i*4fL2X+q95m@>VSs|~h$tZ5rzJCziym0v+aW1wB+J$#m
z<ZR6_#SszF3oM`sD8|!%elV<{d<twg{15M63|SAo?a0z&v1rc>fF7*|-0#}8j^KJK
zn7{`6OERi-Lx@`N*k#!*E1L>~<9DBZ(c2IT#6)%Rr`0nvf=1gWh|atQ>7bOZYd~aJ
zP>17RHRR#p2?Ox71_0mj$qzIYl&Ju4jCa3KMS)fCo{7o|nyuHjZ#$oV`OQNUFX0P6
z>Ns+O_tH3G0X84{d@A+OmXS{*sk7S*<Fz;>gSA5nJyMMwPlL~iGl8OxIY9~)taL>%
zX8QR|o=vN!RLSj08jfdnAZ<qd4P3jd_)PZuI-c(_K?Yzo5iDAKHxZEtAb_oh#{H+)
z+$VyqpLO$pwvx_Q_fQo6ZSfxB$gv7~1ufwBM}lK5hn73f20~-n*&870!l9<4!;17(
zK`KJEFZR?%zH}~+a{d_AeTK5;1yo>wHQ&-}G23Q2UV{IG5DVc>2*4&2Kl%Cx6Q2(y
zK;82vc9Krv;1p=>cvArV@kQykdKt7FwTbdU1JDD$c)^(D%5}5LJf*hp)jd~B|KR~Y
zT5)dsvGx$adYRcHf>wa{uW_C4Vd*fqZJ^@Ycy8+(R+S@%=sx?L6HWCgNpjE3>?J?_
z&o6aHzx2hVpe_7TZ}`3+1T;V<o5*q#fGe0xJ?9)^a(NO#`|;9G65%*?1&(-bzxe+e
zk<dIYG5(Lc5j;AXa)ko@!4crklpFoZfq>w{piT*fB~J5&2<oFjSraG^ZTfxAzF#qD
z{dPi4bStC1M9BUVp?k?D8=YG)v1q^X6QaNEqe>Qjpi66KBuo(rAPn!|1$KyS&oG8W
z<Al){q0Kd^4P*ucum2L6`1VKfME}m`5iucCBp&%8G2fIrr1kex>H6WH;8jt00L?jq
zxy8nAs!;C=BXT^IE2~aLO<fPScC#TpKFD@nYK3iYGw=w4e=V{#X+nQP*z_N@bG9c#
z^$wr{pKW^>$^ok0dtg^*XE2uVKj-Y3!@DzqlsGWcfQapT@+;fqehlh3<z<&=_+ivo
zIGVU8Mf<;$9pyWG+QU$bha8k)2i%Podga+$L$;#?FXgJnaH3xMxw&e|O5qH~@4W-O
zC^<^x!jiX+|Nr;f;#>&E3FoeqI{Y*NO(u+MkSf{CHo}CDJY?sj=g(W)ekR`944yB~
z!vX5udFWs&0m(%(fLN6L&10dz3?EyS1K(_=>v!Z?C_JF}HA`kn1Hn-9!J3q+XuD%J
z#n!NLzy#odX0Si~Gxv-)!)Bi#1l;H`-Xk|8QPz4w-19y)j`dF=gmM=t6eR3K_j6F5
zRE9$Yl_`uA`9izWPmu9bp{?bPq0=Dc$fvk_KV3VwW>+f#DVVlsT<$M}Z@nB&+(WxN
zx?uYtHXQaEgd@Rzg)FcPT6pN%h(!a%uCs97{n201$^ir#0Q1n(!g>)vcl5WPpt|&>
zt=Yvcls)e251acjvu}IQ#xrTrgxtoV(fJ>QhaSG1NCvr96Grs-ckz`izkWbGxHOj2
zcr#0D)44!i20eSD+hFDz3q%w_vX`HVng0H&|KVEUuO?YlN250Y4x&x~{^D8m+aDS(
z#O>EMun!Rdv5z>@)b{k^6?mkVyrc{b>gOC7sGVuNud3{f^g<@EIRPK^+RQNs-qTWW
zQqguyDLBAA3C3?}?}bo%zCGJuEL)~&tKQN_WOdSU$pT(95m;OEAUfce6f=Cc*eV#B
zLEteWco!#1Cy$9VlbomiH$~C%E}*&W4O^9hR?!Rd4XMU1A=Rq#NiU$nV3X&Z!0QN`
zB&f<OR(|=pD>=D4Y@QW}l)$K3yVr!Z!NbLdB<6ejhP)3>c~!w=lB&92PfzY;NtI_V
zgCW{=Xc!ybrQTdv0EP}X{@zPXuv+2e2N&o)JZ?iZ>YS2e#)etsgr8CB99Bhhdv<Tu
z2F88!1%&O}^CVe{s$hjJe#D~XxVNw%rq8<ysg?o5L|x{!1D6Wgff%iaaWNhkswn&@
z%C<e+wur_6dha~_4LC{)bRI%nN~=&bB8an=)ih@xPPDV&$L7+7ufW?6u8IxpyMm%i
z*Uh+U8m0jdN=)EEd&L9lh#G_4SB@1!pO(Lwf*z~iHQ7osQ_tW<D`y3UKsXW(>_UXO
z`Y23<YC)$YFa5FD6^^hWG-4wJPOCynUYljRHggYNA{n20iR9j0RrpN@aA+U~215pc
zF`6(_)8e?^no1ASix~aeJ@)E?F55MwIC7%R`A2_wFz4{rL~bx=m|Tq38jS4ptD*sV
zePb4w4yD<_Yn3(^))C%wj8K1ApA7!g8646OLxudMom25niNjBJgy}(zd~sTC+jG`Y
zSX4E#O-7#^BI)Kc$5HdqSU&=L9|Sf&C4|RCBruL~4zhiW)(TO4lklTC0#GAPpV-ly
zJ_%qcq=qd1_5LVx`zn7&^s(*f;Yc7Hgt1Oe#RYqlAzowYRKXU0$jjOsA#l{EJ36M0
zrl0s41SA{8mTw^6Pfd}iF4^lG$@lX|7)VQC8l*Yf6YUOyYG>*mHnW#S=h?L6<j}ex
z8R%Y5Q9G}ZI5d#%${{o{re|(B0n`AXRH3L)<1;uty9M%k)T8)OcF&9k&;diR{;|?W
zdl>qNcd>LNAelF#jJH21=E7OTm#^^O3f#0{6EqIxP%e0JV=A<~_)~%KlmO^jn5b^r
z&@=yP(&SMGfe^@HQoN4|YP1?%+8#6TKwE$o6Ji7vAwj9YU$ZT01$wNOSfG%m4G2Rd
zAX_xf9VorK^_P%G3Wk73-cbM-C@6jTN%XXpIVYxq^yYPXW<5uM%>2=u9$QLeV3wrK
zu`PqF*@lUs>L3OiLAxRFFwxXEuZWHYDsm%r5+WFeTm`l&Mbeu&V8Zir`?(a2kgxd@
zi>|(s+7G^>yNiWNgSY$j+rl*bXKb3uHGnT9LficKcI<X%?1*%l*GhdKLfBA*Lo7*T
zG~0s_qpq28ZUvnIoBP`D7?}UA8Oo0o+lm1KNVji(d9*PJd4U+UGq)#wHK?hhu*uXK
zcGWc(F7D(%|18O5`^cRH$e-o>889St@Zt#N(zLeC1}nV}R?4E42GZ$qA`=@?h1FLb
zfD0vp9K_+-Iq2XvEqi<2NgtYe`q1!QV*XQVQX0X66kQN8n;3olg1G~X#sT)g!gcX^
zjbo42=0Xc>1^lJD7c_$|V5r^Df)sEy{369-S^R5fQ;k+fwQYJ`=SL1OjtYQ^<xKRC
z-+;)Q)%zbZ>Prk#2}J8)7L0|~1h^_Fi#JSY4e=9xZgXS0$W}$DON@ASH5}bF4{uKY
zuKe1_T4-9zJfav-Pdk4$lxnjd@|T3XEckZl6C4auvi^=gDo6-w3jeg(R3tU|T7Czu
z*PseQTof8E?{)d@13O}tF^Kem=^zc~xp8y%Yjc|)gU%1kaZvr54LcR8U&*-b>Oa##
zQnKOnPkRu2EJnugSOw~;K!*&2z70=)e`DJGj5krsNuJ&cs(v>}f1?{6-_0Csa(lwk
z+)s{1k>B2fQUEcrL!%I4UAfvGC@V3F87;g91FeAomn0by-r4*`eYfBj(M)0^Xtlxa
zf;@>uCj{u>3~lH(UV&efhol(dZSmUD*zf5t0&JV-$O}L=JGs}QK{Y}^Qlg#HQGc|t
zWC51^yd|U+b=GPXmvCG7=J|eIY@u&={}7D7nn3jsE#PT)G;J+z3g%}6VIxpIZuj0q
z2<p5p3pWMlzu5L)Sna$P^I$yq*-|g`z$O6bfs%6U<?_<0pygEBJ5jbWRj!!ui+&pO
zOTZ55si>6;jlz^{L1K`91s15s!3Wfaa+M6%x6Ca8Ql&`OEq-A(^o37w&m_JlegHP~
z1fC&RdnM@ilG=8);N<EXxg`l0f#|Cs+eo{djvaxfpmu{A5}onCrs52<EHZaU0N92>
z5|Y9@0KPQ?hy7{ZW%Z={noBXrL<UmCSJS@Se6xQVlE0UNrfT5|>>f-^O=bO&W(2bR
zY1pl2pyi=@Q&aWT3W(qm!E#<PxkZ3I9VP-t+JjTdlug4LQj#7mhgE{IGXB;;3)@lk
z^3lbvbuUf2uRmLVf3+%*eZv+&XAoKUY*i*BXoFMuE}?2wg=AX!EdGxl1dkV<S@%m^
zYm!;*-@h%;9EONHVj#B~Q{}$CY=kgq7uf2UZ#$%truQOy$PA5mkH82>(**c@ZROf5
z54808Lk*0bdot*fPZ__jL92E^P=`kF`e})lXUAAzb~aO88AeG50n`lTYZTE$a|mGc
zZ5PC0OdT<h0UfX&LbvdGNKs@jd9mM;E8fjDf|rtX<YVe2B2ylzEZx%9Kr04<mRN@6
zx9B}g>Fvo-NQzf;Wzsk2e*aKHtraM0!D!d&t2Z?8#WJxcs5Bxi<0`L<!)(}~pZiKe
zrAkD>X2!012$A)uju@wq?GGh172r)PZa7gsZ>S2Cqk3A3K>AOZK@I9%Uz?8>@iRH<
z+9*0-^Y$^h&!wmhd|1M-OS@cbut9h-7=e4BH~yw(?<cmDs&D)8bll`OS{dJCY@@VJ
zZOPH?iP2qeuWsxfp$8@_pT~c~Ihbf#E5{P3`>Rs@>%U{x<sk2S)Ac^uS7D;nhs43O
z>kHL~vx4e2J$uRv-Zp23;-``{_@kx|v1rJ*H)8M`6R{=4Y$`<`{p`s<_g5PZMQ)Tc
zNK2ESFzB1lu44>qpW>sFzYioE5;nnb$s&xYNv~2mfrUniL>k+ZYyv@7TPw|)7>XxK
z#*PVJC|WwvPME5l^!$38LdEKwUO2yb;Fnv~$3x>sX%8qWDkf!SeyGt@<u93fP0Tf@
zB>}CQbX(e)Xr@ai&DQ;vQ$rsBQCO3YEg4fsvjnLf)-u+kl~L?|9|duNfn<nbv)vne
zbu(z_Lorx4cwFjtGeR#109dp*md6K;EuH^_gx7B6Y5NeoxrI%*50DWih>gs+fp`Xp
z>``{ZpUxrhV{~4H<t*i)_~<EKm}@GAg_Thv!N&GsVPQveNHkw@US9@zpHIpGfm>BK
z=9Dw%jBdT7e4kO+YF~bP(vZ<ZR(lR8j=*YKcYUGT6PyTO?s3K9Pp4LK7e+acrcHnI
zq#aFX;G&EDt$FhL%0}7HNhWt&bQIRN1%ZM`Ct|rJW=%_r=9w*KbUq6<uzoDwtp+~=
zRo2|7gq=PR?_`e3wCP)MDW`o=raVR}cbonl?!;n=0_KiO&uZkxoEer3LQ^2tM*Se=
zC0hVCYKNN=Oy|ZseG8TFU#G?%n7!y5C45@};Dckrp+S`Ez#RpFW3>p}U#eDMS-;Gy
z=Sp5b21?6N82an3nP^Q5y74)e{z6k?b6duXha|LC!K-aKZp=`^8fh#q=8?{;i_ZxS
zd3!A5HS79mtgkx${&<#)Wwu?<`tYuIq5H^qPsF-acfGKVxfbL7^b`U6hSN=<O?F1N
z-M_|&fEbe6Hg0HDX9(}$R3ui2m@QgZmxTXm70nit4W<CvjP%IqZlMKG%$b5#>Al!5
z!x)o{=0gu;!Ap>ze{o-5jTgIPuX#@GRmy(lNP#8-FPrcB`JHD==4i#;^wFXQ_4WTU
z&a_3tWi8U${3Fjn^q_Si$Ivp^{bymvz+GZ)IcK9v$^o&_7c@gI2XCfeg~rskrRAaI
z#O_f^%t80P3AQH*0!mV0KnS_l|5~aNbeJ2+eV~LA3;Y8=IxG$9fCVO{2ySe&#A)GA
zo2tx0yzO>8@4sY#$O_n{(DJE8G2XIu>{hFV7i6dtNI(wl;R7$AG+3q~_RF`DtAb*h
zOu=t$3UJ<fBvYn!U&(Vp;z|vTD`oAekPAvu2TQwDMdf8e%7^2hc&B_{fr&KUQBC)}
z9vIN?KXzBR0?Lo~W4XZ0nlYuVwvyYUdKk8;oHE*-NR-m7V%n0a%hykNVRWVS*6SD0
zQg~etB4Sc+DESGX8a3sO_##Lwx?ZUr7t*vv)0WQDM%uUQ%7g=^_SkvaTSx|+*(U{W
zc<zPPAvv{Kwhovv7Ju)SzTO}@ukQqCmG}2L)$;e8Uw_P1JhmzwHJWwK1K~(*T?mI{
z=O~*Z5W5%eoNjOZEbg+9tMAZh8t3|2d<Ak_2raAyj(1CEAxw3JwRgu1-%fW<=x7?R
z0_GQjHg~?CZ2_%68z*NlXyF<Txg#2X@Q{^qP1tg(o~t<^u}m!fqt3xgletUQ@-yy}
zZSE|slV=gjm%S@Ep9`{JiA4u@wITWS(sLV&^Po@kMWqx$^qZ?FZU%)Di*oe)+yis~
zq}kLOH;%nts6IOD`x_+rM}yV@PJFw%_qeeAke+TK0aPJ5F2Y<5n}b|9Z$d{`l_ZQL
zNFr%qD?s=`Ii!$d%<TuYrA+Q4B5ay&LGnS)ctfs4%#qef3wDBYmfzf&U8|4ZKD3LU
zeOL-VBsIV38IiMkPSr(NEvPOF{kKmQ&G~E)AtPzOVx|Y)>=D*k1;$w~CQAQ=yGC06
zzDwB{mDXErw<)FK+|rft*>a33zb<4hi_c1hC~_CS4Gd8c29*J)pXl6EPQx;thX;-y
zNneKfxM0NoKjEn`)5-84p$v|@qa5b!6Grr0{n=9WYfvE_QVl|bN<NopX8m+VT_8zG
z*bPP9t^?Q}ZTCF5?vZM^e|ouh-4M3(G9_eAtQXzrWPqffV+wP~GMCWoa8m^gYaGq-
z)2LQI$IgBzUE%>z3A?wpol~6k>!Y-55qVQHQ;kqd^!jM0vk{+;7GIf?d~+^D-8B*8
zVfQQ_81B-5HW+nu`u2d@?wvP(sj)J?OWGVz%+)XV0Gj0UAH!?`l+E#X$68zBWxNj^
z{QcCbern0J)iU}yoQejyoV}<5zNbozSdWcbNb=fp-H*sB`wjyZ=$maO{zEMs1Yi$Y
zFaf8;1hcExF}TW;Cd34IGtbv|qhqZJ!Crc$f5*tnzA`_EW??LCx=LIuWbax4lXgn=
zt&L0yHDM#(l~wNVb1nPu_6Dnl^CpRqre0o|1Zawf1n>9#h~tiy{=bBE`%Wv#T%@at
z6PY5>8pQSaJnC)3oI5S!M)!%1VHFC!18tNmqy%<sBrB$c>Jy+W;3mOKIiy=3#zXS+
zD8WMlD!+wA*u)UJ_pVdjZdG#j^^ewWnlz5--oJ!<fQY2-(w>#3lViD3iFR48z^E4|
z310I3y8VYvR>^6O;8O6VgSCyC&4)m;a>+3osQy8m$5XW)+@*wRs82o$cut)nC!in<
zA#`&DazN5@wN}-QOo_X<#6F{G0$ORo;G9v7pO#JW(oS^P)H2l=u|XV`EMfnv{uKxS
zBy9&*zd;r}qZE=g+KW~5J9R+o_^Z{nMwpkDojkHM5mG^T`tu5@W1z$$<U~Chf*Twp
zcm<vf^RJ0{_?xv#rU^tort2H*tz&UH)&p|lrt=RgNMOgJR_|SWVu_I^T8+aN=?3Yl
zt%XMk4zaJpU|jv~S4M+4y6jJ<cB^W8i3Mj=dioVkHD+7!!}?TQ8M&OPs6x+^dzFp%
z@8R9`s|PX4p%Na?-tL9W^?8C)?mDrl#HyPPVq3_m9!_Wd_$(IzOSCp7Q2ScY``Vjv
zDY$CXp+3U06SL(_n0_09wup;lWj3ABG8wj?ZL|tgIoO5Cqu*kBILRElR2C_<^)Ig&
zYxbR19%Nr&cDVz<faE&q>6?93Q=E*Cu=pLRwpl<6{c`>BAzbFBT;Hr-+-i3DT-PTz
z_D91Y6+boID*hTT`YL&yCaso-^rwky;>)^H(MzaL1;bnVz)J4Ope*IqS1TUpdj2@X
z;wLrQMy=x6s|(fQ2SSKfq$`rV735w-r5~WwN$85(ckfxcHa;hQ^da*Vu_Gedrg2LQ
z&d)@gY*Mq>AF((SmtEz6b0sR{P1naKd$9iIN?)zmLDLGDD}Qud>lsFM;Z$Af+CGss
z7KZg4_f?CxYwQaOFPze>?S7IZrxi=&8X3ae$EEziYkjlkaW**Z+q%>D2yrQk_v`Qt
z!zq;xtSVj0Ok2BP`4bX*s@Uvh*c&R-7dfih^U{lTJfG*>@^B2d#fguQOhu^Sz}PQf
zr-6x+SU>)#;>9mfKAfnwp_6H}pPb6(N^IN!{*_g?0GIv2Qb6JWaP0c#!Nk{d3iYH@
z8C$3H%5FBSdwq4%eN~~(X^TztxQBVwdp6!($g$>m#syU)!Xj-^OJ?UWAY53*g@%V<
z>WXMEjpY>@Eqos&snzGX<f<r|CM#-sVk#m*t+3$gN1~qz(M9CtX_q1|-HG3=`}&?Y
zCTGd(YoSP)X!-#Ix_PwSRc#fg?V~$Na3Zlq<-F{az_rL*4-Ya?WNPC(q&^4E{^&0L
zSGJ<?H||L@xa~1Crn+w3cmDE0r%}b1kJB??LUVBRG4|UtXEL=ZfzcEjqTXbcs3UdS
z=dewBeY0wDE|jcf29@lzUk|*PSDsmX;I4g6mAytmFwOS7>Tr|gD4Y6&2M=TfX9v=N
zNgwqJ$*9VUEq4wKv{JOsOk+{PAG@$itVuO;6neB@CagWDdLpSrQifx!toSU!VOEgC
zjFPx7+@q4nl9cXkS%?nzh>lGllGxt-fm<iUTVFo3N=}P6eGal^0d)e;RNuVIut@`e
z4`porbrK3wUz3*8)?P_*z21>yst^pjt*e<^^?-(%V#zZZLbmy{w6%bAz<36stSDTe
zNx}%+9m>2*-kJd=!_Oi$(a}OI<ZWVC?On{HY1r$uvt69CGm%r4VDN*WednKs<ZFdj
ziT8UQAnyByu8D@mNu$0%<9GKy?1BQPVE9IVDbWd{A}3fd*u^{YoKNTS)U4^R=mkWm
zJ?!GO`tq_bJ1*v%Y8@J;+LMxzLXY1l+nL^xC~Y0%I;WfJ&Atr1+kL2m_|yKQ3dxT4
zPmHD^FBe@NFNj+ewi_mmGMFf=$cx14+;zUF{;2?zVwVL2Xkj<>pjt6BXyUwl<K3`*
zP;#(UGa&wjUG+ui;3Q^^fd4(%mhnr}>{IL`ejXRXCE4{EiEeXH=MDII^vtr(re_WJ
zS_HobltOY>Leu=|ifbocrof#4BHxq%`4)~s2eLQseOk(t1Oxh2<r$f}$=@Uzc?DpC
z0QNjNP-|T2sn#l4deQ}*sKB>VJ9Au#?C>uJXpPz0RPt1=<d*@=E2HM&To|(C$x`}^
z2v2<|6C>H0;i2VmZ-Tb)?rf2H%L$76xlpHG%KxQVo-9$`pxBmj*BN^Kwnr-fTY1j;
zG!RqmtN5<icFJa;=90}@Rm#kR<1Cbsaox@4De04dvGDoE8MaYa6T3^d0L8&2-KOb(
z*RmX>@?4AT1SwG6#PMn-E9C{N6GDt82vL6}{^}CPtQPYuNx&;(I7*q)^&Gd^47G6J
zv+Fh9E4mt+yJ$%)9BRhb!+(+y8y@@2c~w7CdDYlBH!|C=<+ufVAGc2B(qXYZt+f1(
zzx<B8Ze&@p<uq<a37Z9PJX^L0HlcN0#WS)$>?YoCdlln+*LQTDdwa#s!a2Hy-kRYU
z@!&pk%vZ_<_w*6mtDnnUxlc1rd&t&Hz+z#Ob97U~&VsY8ntV2iRpN4pWlqUbJkB!J
z@qBtnd|emg*V_?Ff2yKNo+XM2hX3NwlmI3N^T_Dw*6z1yttv%Z_Sw=#2tL<m?k`z8
z(&@avkb@VEVT8g@)1B3LTsuK#|6<=gl0?|lpw|d+7Sj~PQAX9(?A&XOG=zA`j}){k
z6Vm$k-c=b0t6WLKi%w?*73Y=I=&FoFrl}*0{dee2%0l-^QAfN^N{|>(MQ-Mu2mw`^
zyJR?i>c-;A??z@)0^{e!unc~oJO|4DgQm^q*K*^rRY7>sMaC;uXuxMr5Tn~3Ov7NX
zrpl@2;2xNv^Z89eMYM7+9xqynVuc+x>4a60yGiu6c>=1Z4U?o6iD|zD9j9XPq8%yJ
zWn7&A&Krc~?7YqHOD!=9eQjp%hpNbZO=|q<-w#1$t=Knm>iy8v`f}y^vN+&M^24_W
z(83_hE7mv$!?9;-Z;V~cU{hilOg_n2^|f6ib9D78O5`QPdvn-H?}b%J6~`yeK}^n2
zV3lTOB1rh{Tq7UnIShZISGSbsy7Xpy7r=!M<D0}~RLWq4)GMRKF@`06&IiKwz>NdV
z7>D+R)=4s)_Ki9~f%+dURWANG+CLv(#lgwgycj5ms|qUi=`H5fW<0$~K=io+{`qxM
zvVqRmgJ}FR-vwjFwPp_NRhQ~L8+D^cPo+Q6*B9~b>6J<cRwO;=_>o+^-(0L|c}ldf
zGBzDNZxLFe!gi)Dh<Fl{=jML1-zh`;Jbw2wEN$wv`VPd0#Bzpnt3~;D%P<26-BcG?
z4_3GPmDC_$hi95wRcGsBLGfmm@e*{vlFMQ<pCr3@ks>WsiLNF|WXjJx)CNL;W2zG5
zUGO&gD2Mb_?M$7^$NB+4ykKN(1-th57Y>3B#hW$7kdMmch(@j-2B&t=v^5N(^Mg?W
zykQ;4=rE~47*Yyg?!DSUXuF(Q4Mb1Lzrep+Bx%}e5fguwXI9y+JE79EGCwmmHg+W=
zr^1W-ipx6Gs}d)0*fW-gLl&UUyI*Gj(J|C!#JBL$gHd}Y+=K2<d94e6_FVY$w|+PF
zJ@s1~`EjNN{IEuRm++W6$dW%{MCLvl$KA1=5%rl&%Mz3>O*<{R?Wm6^9<5E9PTetK
ztDx^w-Zk6dd}9}`@4@j^azv=mX=5CyG~y=>zV*TsOor$K5NZc!#KuhJyQsL|k}jW9
z>dfCd%(FP1O4*km>8F?i9`(9wI2o8b=F0WuB#}qQ6L}CUt)MGI`W~jc_bu=&bLUS|
zsI#iGN`)%0CFM&AGjnOkKOwuAsG-oV;l&wpR+orP>iMI1={c^)3J%|$CMRF=ee9uW
z<Zb-@rTBN$`ufj?NWu9Jf-!u29~OFSF$tfU@MVY<ptaO$hTTU&Q5A~!^f3uhPbWu%
zjc`qrjJ~s=5<DT8!e6p7UC%geCgFMD__z?d_TYN-^L9<*7H2$}XZ+`=aO$_>L#>(j
z$lT2vn+8d{PVJ8Z5bnz#Sut_Km*F7lw6^5G@%d8&`5~ufg#(NGyy-3_E41YE>oA`@
z;?tSqu3b>q(b!XZ+tWq$Y<s4zvAgzfJQ+=*+Ww>p!Zb(k{rhhgC}9{6y{icc<~OM!
z6^>DPKxu)Hw4W|bnBHdF><6@bC+k97IW6xQ#uyr8iV>{OT)&eF0tQD-y3>;hm;`(o
zON7fyPJH|}f4Zn;eog*fYtS>ynd>S3n{15-33O7&-|SURc`ODZLYB|9`om%Z-R2A!
z8gjXhPl9#kPb#m293;M|3<Tk2CWH%2*TULeQHiK!OSajL$(i7E+^yd0Jws-`zL>L%
z3BH%Yc}jc!NDMyFXVc*trdqCVI4K-kVqiT@zUn<+&JI{)bAP<xj2#6nho9S?<g9#$
z&X^;1#y+HpP7pm!CJQb!YpQY?0}YW(lXtf+t3UOHZ+iUfr+^C}8$4^F9k>V2yswz=
zW;!e-m~*pWK4~9zd`0uQnL4?a$xmB1v8CaJA|KxhW3I`H<?JZKyerZioooA)f<hA$
z4hR~68t3|C2PL?W+&kkN0=sYVtDR{7?($I1D$%1V*tVz_=r9Q{#m3^fk7s+CM%E*r
z%!TpkD9JV{{j~qd$-~hRH56$p$}#pf55#-!7&cehq=a5+y_rVaAyIGp`Cd_<!+a->
zh?;M5pbLg{?Y+V-B)hQ^AZZ{Um*JkdFA!snhaqJ4Z>EMe$0#DZGQzhA?}y?LvtkvC
zVCl>_@rP5i_C(%ht}sEr@s!65q+^W@jutV#?N@vV5^&5-$yqk-P7u1~0Z8|YnyF=p
z`RwlbsmX*tm9AJ?*5_823vBqQHB|+E(6SjqlG*eyzhwLEXQwTweu=IM8+?2H;HOB$
zZ7`36i!&WF7X$&n%%8XJDQPoQ1x3^&iQOi4TJ%|(Hucm-E|)u`Wsl^Cimxp&jzFMP
zhCymg4e;@wA{YEi3SwYZ{L<tk%#fd#su_Z;8AwFMX*M0gsmp7k6%E~S^V%9*ML{tw
z+T%EymJ-M>@Rqkegd;3GLc#s8`@+=apk5AF@}`%E2!V$v>@EeVRBExEGk2jH_3>RJ
z;EZR6)*)+2POrnntAj3z#LN6SAv~f(5qo?`TJ66**+U|@d*X`EcZ#)g$L^wr;;BZ*
z43=5wao$qNjVz~v2tww;CpO+Ulg2kn=MSVSh4V)5zvScmdfN@2VDi=)zI~@>CBkt6
zJHsN^_#oPJh;FaUf4D<%z5!%oYe^$eIQv{&f?m2_F?8R)arI@h^;D`tzO6-nbmzd8
zqL{u$q`cy6D7hZ1k~tXXNuy8)`rts{X=5(XCWNi70^Rsk!FMqYC*DAZz{qzWgsr{^
zkb5`h7>NhTy8BqR6-qrOqu1m%n^H~FNE+{>{Fu_hJB(EK%z}P=m#Y3NJ!XM-ENVJk
zC+~{GB@O4%MF6UvO~Vx)S8Z1Y*7#uhyObMU(^8v`(}V$~bIl_uG0`yAnw|Ll-IHe`
z?KK=itC~RZk6#hKJ#i!K7xeYrBkf+kYKeoo4nh)swrJx^xvJk_WZictlz4GQFlv<>
zRZWy$^6lZ#K-j@x={2p;E6{!1qUW%gEfDJW9R(_X0}|170#^R=mfA5fiIPO^vT`=i
znRBi%@6$%QYc!zzJ8CA<;)sVGR5(Hzra2ai?YAn|mMnyO6ge)ez7a2P-7^aN4Zl*$
z)h@H{=WIXt`E?ll|MfKCJS&Fpi-cc^^1y{l0>a&!Zbe5*@o08_zXz!t9uS5vKaqbE
zN#0Mul!#R@Onkef=#@Mset($9_)Qq7Uzno+@|{zh9H_fw)MZoxVTCa=+H$k9C%Xvl
zdGvc;6R0S1*SQm4okTTpTC$bJf1bXtsl{HaBq@Y^xNZMDW{T~h1ktS<`cR3_{c4rJ
zM-Sr|mu$>P>mZB#OpA$Zp&QAI$_BZsQZwDYC@aO?uAoDWPF#O!Jcv{sL2!$}Wf{#T
zI(iF8^gdD70FaP}b4w22qbSR0!)>0H;`=?`z9~Q-faGCL?$Hw4I+p#p;b$x_2Kq82
zo|6GFEn)cgQK%u$t_cZ!0PNVhu7&Q{M~gyT*`-_DLiv7s`gE3K+?LFTtO~4@k6s;Y
zI_UD8$9|Oi_{TIDg_ds+fv*pw8I$>^Cu_$IsCj05(#(j~5?(2VKJa}?VR36sAQn(2
zc|fRD>R?(>V_{eTilA_VXy5EaoX*n9FBk;tl1`b`Zy8e=OK&N=&m+X_JwZ&;PSi-<
z*nL6#zWaOIs5GE2nWC`@UuAosMpr+&28B<T`=0x~^`BLig-_J0q+DpLypy-3{=?es
z8f*M8)i_)%FI}Yv00Z7K#|q{}QF&WbD=fPqlc!_m(a5}W{&^#`Ez*Oq4S<427sIoX
z5MCStSC15@&TW@-HQ>X?e5%i^0}nvFSn8qjEulr+z-ch#gpN#iynD1eO#0Ec8Nn3X
zs_1?Oxsmu?aP!dsf3tbKT|~Ug;&ixO6n7nwVP4%T^pcFpUHNrpu)wYvNcqr2#&{j`
zi^+Rl=`eoN8HDRadK)z1f`x)Ln8wvD{n_}uwdKfux!pz@`CkA?b1=wsw)o_!LY-U>
zy%fWo_*B(yhbAHA;91etGN|Z+?7F!24~L(%Hq?Ly^@H1Ndw^R!*O7fD$}NMOiXB<T
zBPq5qzF(d@{<=~s0B2;-NuDXNJI+okTBjc7&8|Ef_z>QrR8GZJsgm~Nc}ri+R6t&{
zO}1c4urMrN3$&}_F@T+&mP;4?Aj}9UM`%;0JhSb|_!);3VamARGQa%f{q{v_vL-Da
z=0mGb>~Hsqo2&7orL@3}5xk{(G<)0cbMGY*47yr~<+HZ_@K-_eG&YReLAhE~6L2f9
z17RMoF~Yy6*458-#C;^bGWpu(v~`s=9K(TwIR`+bb(Znt{SG~M!Ysxjvp$&xq#RhD
zRF7aeT@PbWK_>WZ0)A-;J!Usmp7)zCUQ|84zk11cPo^2EkBW8T6FxJ&d-vm&8~Y-i
z6?R{=OV8F(ygr)Na&<)On2^Q*qyuBh!+~OV=5CO?TrNxjzJs1|Rbc%<<}P|6)2Fs*
zy`3eoXwiA!*)*Io2nWvB%$PS!>WvABhVgw?sBdO_mnkhPl3&LOjc_wSWmyJu*F@%h
z3~DJdgSqtDpPx@@6|xzS`(SN$@~LUOq0kZ4=Yu6<*#*Bx2(&MJy5RV=HPT|K2P6mK
zHm6goj}+EM6e*7-b!aaT=zXj9p0}(&yTbQ%OnDWm;yk=acPh_bKfCUTc;2$MrE4zF
zR}z!fMT3r9_dk^&=^0~xV-qza1*~B5cwF&kiK=H97W2a0EX9*yRXHW`N2<s}HV9li
zT#WmgF6i9?k{R*cUUA`!C}dsc5xY<oyyxHu(@Pf8g?E)VVB<JuopsnSE_hSB=&={C
zY(c5eecg?4<vZF%LQSajiU~)DCkWUI^ArioVFnt4h?CVs-uzeUNW5<_;9@NZz1;%G
zf3@cr@!<Lhnkugi_x{EtEPr0yt?49xDe=f6gaRZPqF>HbTGC$zDTw?fg$w4L`AK+~
ztC<Gs;+hv5s|aMso;E0A7)fdYlsH<LY4#X*&_U@AhLKRdFKqRoF4S2k09FdEvNO|2
zRg3dq1)3=v4<n(}1thg)SNtpio{&7S<;*9u!tm~jN>q?rNUk>&fy{*vFRcc4P>@c~
zzo<&xO?Jpsrj=oe<S7}!Bd?1}Wpm1{bu#k%R{FE>Fs2`(;ciShwEva0*p~%A`CJgz
zBM5@1_<JT&ee%;ebJCYuj}hWY$zQ@_er#|3i6>=7zoo2e635n(2=H?W8;<Qa(@EyA
zV`46!pP%c#6>m}Na>{p|yx!}!=B+R0TKjQ$6ar1lsYvp5x<jA5X&QSe3ohFAmScH!
z99K1Sjn{5i?KV*Y#a6!2^0=0c_fsdr6hx1N(F7nhgxeKJtuZXl0fG9iy>82vBVdnG
zq1*-~Xhf4dK;`Zo49@(h$yDE~2}(KnO)rI6jce?FcEO$*i;xHufg!RVxJ+J_(4E&H
zOmdb_GlS-7wi3Ey?QSQ3mGZp=oqQF#_C=*js_nCcvNBbEnA&o`=Gc!uN6#u;x?e^b
z0Xtk22kSzl?_qVnz(!I){N!NlrH3bSyUM5A6`e+w%4e<scVLU2BjcWVWP1+}`o=4`
z0mDQl?bE9u0uDTVvJfj*gxK(6o52q`;gSxy-y*P=BK9>I#qLe!yP2g#w51T?z=&u%
zC8d<<HCD~&AcQbtj~Qv{o`Qlz1MEF%yZZ@H3f7XViqJ!Ws(9zJOA95a^RiEB5Wq};
zCSCEodfr+YfzmP?&b+q8$)tit5$m3s;!bj9Yv%*N){e;vUp>pw6^<;QVOQH%v4AvD
z%wtynFHeF8y1IeoI?QjTCPk$@24g}}LWC29UmUtEd10T<JW(Rk=2?j_Rq=&*3e_i^
zFNZCwrcm`J&j}F?0#`iQ6l$VeXy3&#q;(#!YTlNp5mwQq@c<s9ip;b_vV|a_?V`v8
zwv&qJtuwR-?}yDqg|7SAHi;X}G?E>c_j~8MYqj7T99<YpY?l}bFmHhpm*BfF>v%Rt
zl0sE;_n7`L{Sf|9u#df%JoK3hVTw9D&dGkkcnF4dPa!<a6vPn?ChZ0kYAE;11DOBg
zN*cxHb|a7tJ0?%77&+x7qc$O!Q&$5Px0|KstUKz&OA+#%VOAXd0>U|`$@2L`C=sT?
zFE)f;pn}PFS+ofkb1iueDip=hcjqMGH$8ODg$L_0l8(7*f$s1_4D+|b2$+h5-H@k5
z!Y#w_Ky<|joBV9k_`Nnyf-Vmo#Q56moiSwYyYz?MDk;pwF~3+y$D|c(PQ!w_LbOSV
z;Yl)rDmG%+&5x%$YG5PK7Y_fxEzalZe~nr+V<$<@!(}4tA!t7pu#cCoXYFeH|5$tP
zu%7$)ZJ30T(ITOgmL?i#3Z-3?QE6SYGb)w#Kt^Z}ZQ7-@NP~nf?b5hfilS5+QfUeI
z`L0}C_wl=*<9?pw`R6(um#^>l^LdZgJkRq5<vkHE8CJ_fp~vh|z|IT#5kCZ~**xeO
z9+Xt`9M#KnVs1$7oT^u3(1bxAUw-PTUj#;qQLkE)d4y<LAQ|eeI8(%#g?A;Y)%tw5
zPc8hi;u&)8i#)!+mB4<OVf5L%9n0JOWap?$$$eG(+LA}=L~<L}OxNF=)Dc$if?*6L
z7q4UK_66{z*IuTsaK2i}^7wzOAYM+!VCEnm_us%Mp#54^X%}*PpmyEMWp-~{Zr7ee
z&A_b~uuY!mYuHY<ncZw3*!L=@)DM)H=>ZzcOP;aI<DUAZ89XgyPTbpJf;$Ukp^Q=E
zJbij|NT5R2xFoS*JE@5Qno~-C`OCM~Izdj{f^S&-;??Go6km90hsok+98C>O*DpFg
z!Bkyq0z@^qyvOPX^0kEz$tDZk5iwJL#i*Y8Uwc!QHUnFK!?%^e3R)7rjQqz1n$I<`
zP|WI$k|`uVJ`5q>5Xyd_Q+7hT29*J~K&abQh;>^d_Av7@{~V8oxaFZbnu`UH{vi`w
z=GXl$hxMS}SM7x*mt`}-B}lBkq4yT-Wu8`u&c2oMN)@BHAu#{g!)_nL%@Po((&<3p
zO%TzlFHEhp7w~9=veNnc)Z8VfW=J#*ZD$)0sYM?YUan;8I_Wh!bp+hXJz7Q^_>OKb
zHAbmU6H^81H8xBww`_SJ$bUTD`k9gP0*0T>iyYvt+C+;3MjE4>T?hJrSoCsapz^@>
zf4A+cC6aWZ$gQp!OQHfB%zr}Y)Tc}Ss%H{ZWRu@DSmZCh)o2kO#(m#1oPDYUh$x#o
zFP{O(H=mvMhs-A=VPO#PmSg<hZk4G7UjMcRCF}W7EDfnzJincX>oyo94T<*T8nT~X
zP^@`MVZNKfMu7ImHm-s+1CQ@t$IwHOGbZr44h))}zfYrYT|O(}y8dyWQpm~-?@a@L
z$A$wdbFT1`4_lWNSs4k-TfVFe#ZBotS^%J~kEzb1p&%4kFxijgnAMFOuwY!!>+tj-
z#zn*MTXrJCc9&?HPQT=RXAiLzFMnYOQ}qTFqY7wCZw*L<Y<dm`X<WCX@1RRLo)Iwz
z>=*P}cVAqUu;>+~1c;@T^h-l+SPxn;w{M=iOHVazP%<NvM#LwryI)WRVb6$6C<j>^
zS%rh209TOoKC)^dZk3_$3a?a%9)6ibk0Ohmklogp@1o4Od)CE@N)OV=03ZRwVySIa
zkl!OJbMq^+)3kbH0f1G1oS7`LG9yxHU`E(LC`MGM&`X)E%VlZbw-j<SVPZH`fxX&+
z{0PU6{W}Gha<D&fqcKuW4Tc-J7V;X<wPI3To4aMKguJAmuLbj?EnMRQ-KeMMI4<5l
zEaJ8aL?n0{nJmUl>c_+bQNiXNA`usyr){#V@?u2&fw;;i%TjS#j0<1}1U^&nAy8ks
z6h;`NG83HnK*%l9h7lQ6l&cAymwn(&J_nJ2ta`a_-;nxW!HXoII<jRt*S@6}V9O+<
zXgbZzWNPFwd9)c`F-<q$`mZ*?4|?ii_i?^~r4>YrGz5LrYYDmCF(vbQ!p>i9Gi8+*
zz4-sbqcve(1UKr**LT)G1dLB=TlAF)8b-dO<+qwI0DqPIY^U;kJ72cs8s2XvtI@5=
z7#23V#(N|_HzsLY7u{0gApUs@P>czLROxiCu4X>)TIq_93$?rtV}M%ZZY3{@<sJON
zzkEbK*}2cM7&A|ygMp)>RYjhw;DKdj-zkuZa<??~kT22(5z;eKdSj5p=2z+ULMY(>
z$5CR$QMwjA63@+mqjUyGDf%naGyJF$IJ5fjF2SfBUwOElmsbf1z7PV~wMu3$rM)0u
zs0?83@3W-1g!qoT3>FmFxm2=qy?tHq<&3X!%XI6|5@Y%c!`qG)Ah^JxR(vKq(ovwc
z^B97;#@{DNi=3qPFESRpaguz&60Bg>5TP6OFS$YCLeyReBEh_#Abk-skyFj#kJM5`
zz!r03J9+)#@Cffik(NA50*3|k&yB;$v&8q6z$Vk;S)Tp;QKaYC?d{w2;XWzZ6JRe;
z+U{PsCKWhLjfS+pj`eBsFdZ{8<`Q@qbv(@THgyKN(L3Z}2)(aG?}C~fwB<hW%>bv%
zgxFP*B0Fi{g=D*5r@4rn=Et8?#F1%UBTlV<rKOkWyXBLg-|axCf79r7z&-mb<}ADs
zf9rc8+NnU>bX#(?LqySXt;J%f3<LJZjAbPEBf09<LUysyGonn8#NlLWMLNeUVyOxU
z>H*yxr*Tt)i0+T-6BwaMnZjxjLdFP_`i(`WrbmM4i){aZg(JWOI#RE`>~aJM%wEv=
ztVCKBrMlK3@mLoUzaVIeCw?7t{WU235?;fGk243suwMTSS9kPKqy+u-jOrF+)|=*r
z5XUfqhVJ~9yvC3PVv~O#GHULUsvye2!x2z?2`9(3u|k{;uHmq(!WZ6?dFgX-r@V73
z+B7+`&a?@`*Q#3<AArE~{BB-{A2`Oy7)XBzWG15%w)c3d#nledDsJ@B_ClOl4Fy(t
zg<%^ZU^xTEdaELxUPlk47jfM6P<I8?NHf<|?YHTrS0xF_|L`D2h|36Mp7A61NlQw!
z&$x(s{GU(3j;C0?F^~bI${3?Pc#6^)RSsIk?c^zdW)4D6aYtMRAx1-#4M(cWR~Ebm
zV+4sKjE8s^_JET(Y3kNgQ4j!yj}hXf?S-8<uihmP;_88BqvGFt$|E$rfMFoA&YBU?
zt_%N=u+!>}R&-@mjgzubkv~o6c%SeZS+;J<Q_Cl-Bll=vzc<vsB@A&e7w5f(5x`Wo
zr~?&gZA@G+)c&5>NG_w~288_OM+r-;vgY|>9767mEZcLKOy#V#iQ??!XT<YgAFQR>
zjsL%vGH&@KrpxQQ19Eh|o#+t*hQEIP$ip9mVmM9y@1MC#e#X44K>$A^N`B^0jn{vE
zW+@3XO4JjT@<@KQO33b8lo6om@8#HZjq1R3>qW`MH>H~}elgY8Z$7ysgQ+nCqLcU5
zuUmeY-13K~EjHkmN8om;#uHUjQr7w3Ul&8GW7w866}qWs!8h1!qh)Ok+UMckmUlK1
zR(gBFRl+NWuk>-x$=`7pO$43nQ{LZ}YoQlUz6!@z<k%(q?F2}qO0Ld8&lcMGpmXs*
z^orzIeNv5}htj8p{}meP_Z@^XlcUWBMUP(rPZObp1VJ!Oj+~AXs2nEZr2la`h&!h4
zC=i7_s<1pc9Xl`3E?$BKyY}miLrJa5ahvtM3-U1fPpGe~oa|58CE=_(O^!S0496UW
z-w071eW0h~{{8ETiWuo4n#3cE%&Z%Nc$-sGJ2C+0xJBOP%PRw^7a%v8K$(=#Ly?l_
zLdK{5X3gE&Ki1Qj_#9^S>ZLh{$#R08wIRma#=SowfwK+6chYet97Z~LiTN^TlP4{g
zh@Ys}LlpcJjQmY_>X+(%j{xJKBieDipVRI?2;3&`)2g$MRcbpNR1hGqD&S<0!83GP
zJKDnI(=oWJiPEV*SWVJl$p4Acc!b1>4Bca^7`M~*Zsz>^zNbi5@?yYrJxxeEOIGN~
z=!^Nl>vg?9#0p0KjRj;c(Tu8wbg`(8<1oKOa%Z|X+ia~hGC4AXjtI+;vA44i6vD{7
zH4H|wY@Ye0Brt!NuB;kWUH|F%@lycNGYn}6O_WMxcbeA_72oskib*C-#%9O`5PBk-
zLdsAs=$Y>*?A_GSh!pMmljAQy8zD-Wq<vbBM$fsxvfG=@C@$H3fc4v>#i#R4Gy~nB
zIIX2(=iln>IV}WmW*Fv0>7c{M0nxM*O|F}Q6V5gmIwDXJv}C3KpisixDgx1eOzk->
z*ZY4Fs4XKN?RAAHrmP4b#R0oI<{W4JeSdhd$R-47klYoDtEHTj^Y{OKi99`VTtd{I
zD^o!0oMGsw&;+GaQDc1q0^aa0rZuJ6p(x-)dUTOFV<C)RQaRH4%lT^MJ&uWyaTmKz
zU72|l^5auw0freJKUX-w?SwS17Y!7Lk(?F)W%&kMGW{fM>qz`XfcWO(-6n6AU&-jV
zQ=bh~R9JTEhT&5FEQQNOcKyR}`&W-Tpc5Cj)Kr3k@#Z%QA~;%Y)$l$3$l^_XiyW}>
zo0}pHt!!6*Np0t~1nusQ@Y?6Z<9zt&35Z9g)1C>W2o)F(ajN-dNz5J5p@q<F-T=xv
zCcWoWG$!SUr&ZoD<;hN;+9weO-AT=}+r+O=NT*InX>05hBr0$FUKIg(2g)mc{dOm6
zxumkp-~qYsn!!rdo?#!{<R4#!p6ZrB*5WV76Ul&j#=vx~E70m~SAA+S{F?6ax%}=G
z;^?n(di<CZzBr=Cywgawbh}L9d0ZG<B!0Svy&|=g=_N5`tS;3Pe>r}AXb8;y7m@(e
zZa<ph&z#W*Y_G#TQJ;l(DA@|j3nr-r7lk{mqebN=oyTWm?am$BUwi+{yQVhgP6rU(
z1-Crw9bEUwPQu~)m*eQWRgarLY)5dR=KXt^;8IFoT$sPZ{mY3Ta0#Q44$v5HdXxbe
z8kzHOXU9J&M-D>Wl-@pJOLWX2I(8vDil<-5$T}KnAWQ#S{F7gFefAg-8WAP;&?VZd
z$#EhK1yIluvM-uvy(W-idOgygDMxW~>#m%e%_@CH>$cV>!>z>iZrFafL?|p|n(lQ;
z4195TeFVTPOq#n`sM8?42}2#~Z@0UHuD`SB@#;;Y<<T#;FG}>qQjMzi2Kgq1Q%%cu
z^-eWcPV9Ib8G4-GP`2=)z>YPw0ukMUeKHZ<r8aI$SVA77_7KN$Xx~gv5PFi&uFql>
z_8&k1`>zwpgUQ>yoSzyRIvQD81+N)Khtv-SiBc(Cwn-;}x-EDtCnM=%G0lfDy7?0d
z*1NvFSmpwn*<Y7T`HrWXU!t_@3DLgk8%6G^4(7Z_O?O?6^E%iZ7rz1>ql?W+;m!j9
zuLL9VYfx`I;{4WmP|!uvRX(}1fzJ5_6U#@P%i-GZHDs5$OhnbotTF^>lMj4$`*{Q3
zuW~ec_5rd@q2NO(tdd(ht$;A^=Y*J`JX`(ptHsZb%jmjFWtCYDT9}Nh^vQbG&4}H6
zwdHMRWn{C)NW4Xe%9GDP&uaNF?O?6Xa-<m|_7K^c+60UDc$qofdLFp3U%uN*l@=uy
z21o#h5;K_4cc3ok!p;Y4=Ww5uQJWf;Af-bxaM?w&9jKot+axr7g}S|@Yv&N0?Rv@k
z6De}7ajpT%aSV3x99jJ&Y9;h$6GX$NtFnlsOcY1nm7lgM%jihyzJ5uZwhB1i1kFWP
z1TW1jCcTY=I$Y?kmo$mw#aF&q_ZEu7A~m{>5^n~WUWWxjB1F|=fAE3@^IXgU8}%sZ
zwYx&rHCH~Azvm@={KEa<TBbX3_n(NS+4~GGmFZ9rG?6)W*Xa2r_I<Mr53vsQ1Knmd
z5nQm-GW%p2*8ZxsVlm1!VMTL-V=tSLp`H7UgeF$YG)if@I+mysqCQGu>XRm&P;aR&
zXWf(^q-W$u-(M~%;iEB)Q3G-oh?E^|n}&gfiCn_^{b6V@W#1FvT;k$q(P7+HjX>~4
zP@C^EFJz)B#jZ2kEOVT{TBf1|#>vGcKg7ZAMgfUsk^Q#tv%0gZ-Y<5Drw?P>w>n6u
z9b((R++Ypmr0Yfw?7D5D=1k9Q36+eRW$d=LoM>aN<iVhC=7rPrBUKVU>FRa8dhdC&
z`T)A3XK~Ygrv|NxA+F{{>H#cbjxqw9^uWISa1z)Y27yIH{qgj^tLk1~Q9ZqO^r*rI
zCn)m=t{EY1Ue#Ec?^nE}iBxE~WtG#CXw5;~?u*=9MKX?~Mgw?dMOL@)I4{-;L0EBy
zXkRPX3Wq${b)OYCv^kcyNowD)@<I+=7ZqfqujF?ee8Bl+#OdMW(S;#GUg_&`1g*Y{
zV+cV$&(1TSBxc?*Jxfd}S;aLLu_QLhj7jC9=entS#(;0!LB;p$dAp)jq?S*DsrTI7
zoja?nO(lI1wT-)C9D?hu!&PS^0zs3QyTVM)rcvj<c-%s#>xR0^`eHqdy&w&ZRM?<t
zp<L<HuU0bvwL>CbF+c9zXWNfr2F});pQw?CffmNaDOj*(4W(|Dkyo_5!bWlHF3+CI
zkSg_?kK#B=RAZ!c?0k6K_S>Q0tC)T2?^7l`(rcGLTnAsy5BAx`EzPyxj_HEh1Gq`q
zw>9^cFWhHiFY=6@wd~OFIY?>n?VUt;;Z74A;pLSZ>Nr{+Z{NWbt)-WfY(u$qBtDco
zmd_ff6EZmFz|QwvS!OS;P}_e0+VaL8ry+^N*o$|WjRqlj<pI~V(cQ$tLo-%nPH(()
zsa;Q$X-$ND#(m#&jZa-J66Y4~%x{^*zzJzdqQ1uzXaz`tZKK*A?}WeGU*Gjb2v-i&
z=CHt~ySbQy=n@%Hx?!V;=`FkAcCqZu;qn<)09O(gnvR{_+JzY^n7@L`jQdtdd(rL{
z+9^fbI=-K8nTtS?aB@$8uVAH8RUsFr%Vbb`52KsyRyDey^TGp+^(Z5+u{bY9IoTil
zRV_j^0JYHe|LfN>7)6=-hHpEaRYeewUx)KxhN(n4l!QtvH;6q&EHefVeRW7h&B(9$
zJ}Lq1?^=#{PW-+0=03_5t2eTNJ9K&9rpStnA!rkZKM)F9SJl0+kct~g5|YGkK}Au_
zv~kzjr%(2IoGjd;yuL<|B{D4NcweahmbTo)6oHX=-cw_0+$>_wU@)-{*L*68LHn!b
z$sUkCCT0=NCq%aX4n)eg5gS8@V`K{Wep8_dRq=OHHvV|h8!Ni@dN?s=yqCCChwwrG
zb|MT&&O=r70nM<4;gT26DLPwHxIi>q5NIZw@~Iv9T@*6_CLCz^;}>Mva>)reW8({X
z8)mMZ!q@}@iDfS7g(2H*rAqqV%X(_mZmC!~4PmS%`k&(nF~=~*uY%+W2yjfDzY-OW
z2@j|7k77&CEuWkNu@}l?E5`S+P|pI)OH%Q9F*)=4INn#V(KkU-JcJ=4tY4xhR!rs}
zE4%I!@H@=x;wPc$Kb^%^0w}f!p@!I55PRnGSEj>2pfb8;TgN5BI0LGV#tly;-@uw7
z>e!ch>$Fh;8K>{RvXW=nA6}bzJ2IRV?=_)zX`EJ);@4aiTz61`Q|bJ2pY^p|EJA!d
zVY|2U7M!BQ%c*7B&Utu1{zLdo>(2Tq$xW{ricv081BiJWrW@t!fq`Pdnyqm8DuBH_
zXrD2y`FJC*f2JAqJCZ2PbnBx25@#Y}vDJP09`T$GL#=dTexdQt4|G?71iM*Gf4o4y
zZ1i;QZ~dsQeR$ismOgjOU3lBKC@nQw#OCCEs`5dT%#cXSNXWO`2!*<B#pd%**iobM
z4HtXrC5^`n!vENn;3|FbE7xsxx>pOO&iRyA_Bb_ioMU5Kn4nF(VWYI$Wz`=7!u>C9
ze~8lD9`^8AO1fV2-Qb=|W{FGdAcA$}!{b|!;f<V9g<BQ>i_5+#27#Jfk7WH(JwkFP
z!rtgT@k{XeUmyu2$;HBrGY8FIOMFs)NSAt^1oML#kB6S5>eQ!}0<gn87IG`)botUh
zPT6P%tXJ`{v!Q4dshV84_wBu_h8D*&u)4?J6KP?dxnlL2#OKo4p*R!*dijS5NA<`Y
z8hh2&HFT2yD$kJOL=cJ-m7L~lQAA=E+R07ZD#Cl2!X>?uiT93H`yLv~X;A?#CtH&}
z74YWKrQe8GC2U{zHG?RgliAkjX>t(WHflWw4cuo;NXn)t(|5?$OL`MkLgVAXA`YvH
z*nQ8W&}p>At0}|(8l$!|h5(JsSj`Za>@x||H$60+&2WidBJL{!F%zCaMS$g4avs~u
zFS|6M-A!;XWEPo&?V*3_WtQ;&4kn8?AUSF2j2*qrgNDZ6di~TM50Z<#wz^FOObj$C
zxC6#vT-DHhBQyWKPl{8+=GI(Ho8i?jw3BNi%zlh%+_~+$95og%a%O#t;Ny;MfE$k8
z$jSXBo&%HQbq5*1-wA&ytf1~MPm{<`^Ud63uRmqMD($}WU3M4K#LM)^&sQHr3OxTT
zufHzTWVrFYpGAZc)q|Dm%p1fmhnkI<A3K#EY5Q_=`HBuI&aFG%$oTT^bgcehGH9b}
z*~RuVW91u@N@;^bSw2Zmetu%e+i|$$%jp%3YR|pCmh@#Cb<cLTx<B7dId*MQWqR_9
ztL~G~`@Mr64?FzrbML)d<8GKMf3oG_sN1VLn=OJTMV4=1sV?2Zo8|UIr)+9qc5>gk
zN<$alEp>OE|CBC}RWHc1V79p6Zc!1&Gka9=V@G9|^I_Ll)cr-IM$oqMHC>qf1Zy+}
zziT)eoDa`+N^G-!RT+keYsISD8+W~^w3JC+tkWt+oa>iNK)QY?f#*bVNC5;)*u7*T
z9>jkq5Vzx@cVfejmEBoywDMmwB71;l^c>&4qL!)K94NM|vcC82hi=wVQcpqB<kL_4
zclXV;I&%=o^jXnBtY3FgjRV@_a)zaHC$V|)CYZgn^L@+7<S1Jopj0a6X8GD}PA3kG
zR%bq)j39{Bk0<oA#{z!IQsJPK+@OrAJL;_*%}#$ucru|RtXTVD>zq~5k7ct6OZNTn
z8<=$N+FH*sDSnArwOCfX?B}i@j;}B2tefN6c$&7gPj4UV#=AdXY@Po3;rYtv_LAjn
zt|zP3t=M4R#%;)^MzeGO3NMXZ<__Pyb73s4%sC?hE!i4Xib^{#Tb#bDsk_{)bLeY{
z?{zm{qZL6r{15K4R}#9jU&JEsLs^sU<p<8Ux@5Er%4fg0Djk_st!m7fkSsK3PCC<9
zIMVWy=3`@eUaXR&L9&P6C)N#=^e65f{;?%z=)(z<@v*7i{(#rKcP1i?z9g0OR0Jgg
zrpO)wt<HQR>%eYNdmGUsY1;Lm3}DY4`BmF%54w9{04MD>o_zojx=Mu8+bUmQ0j^D<
zGHl0rnP}$Osn1)*y2`?NGQ6b2brj6pbbgEejrw_xcLwMFBc&xSrahGg7y3Nr!-1(!
zyhG=nLLue6`z|qL+ZIr`t9~Jt(sy<9Ie#ehD)dwUsOuC!rYl}|J4ZPgb)i<=^!mrq
z7h<vs9y7rWCa}MzdyT&*-pNzgr}HZ-LQJK2zr>^ND38)jgP&^40-MLCS#gZI6udY<
z=2ffWu1NZvR9^q`+nS|2?mczlo<kOdIm&C=vQJ;@^d#~fsH*Z*1aQpus~&#^_P&oV
z6-zOcx$BMYB8d{_{03PXK7xo;x^PYp3nPqN(k#%togiicHxvgnm&8=3pLUFHgI3aF
zfH!)r#N%_)fj3nm$5K|F{Ij>*`QR&p@*M;=+uzaG{gUb0R+XQ${SazUjoQ1*HLTzD
z0}6pR6fbgQDR>MdKAT=;Xcv5s<UyaJK6X1j8X?LwPD%;m1%8|N2uacWf~XQ-o}gSZ
zqxw%=3sdT>Q22Zn;Y%~;VxD<s&LQxO!ET)&(ogC*$LQN5&B{n^S3&|Hk&BAm9MgNJ
zP7Xq5Mfh`4`q@DdG1%jy!5wyE80Z=L{PN0^@{t;Awns?VTPo^KUybyVW<FLnD)Fa3
zUAMHah}Z0<Q<AM^*2pnZzL75T9{ESBwbS{em-OE^nZgik34XHEtL6BCY8&bOletDB
zIz%f0Qdw)28gG*sC`g1h1p>C1Sq1Y-9~j#43VIDasABp>^rBmRd(#9%j7~B=A>OLx
zgeH$a&i%Jk0P!W(rF|mny`%>MdWz<^|4Bh^RoYBuRnmDLpA}PknG6m{iap7Gpk*|0
z^u)@wb9*w{5p|wBG-&g~iTS=523+rYx9;6m{wy;n2E~2OA?Ov4#1okY#Q(Iu>E=g?
zC5|Xnuvd^7ppd8L<P>z}m_Psj`+9(;(DE_7a_jfl{tVJSe~HG8ds=e3+i@emkIjY(
z_QUO8uNXdg@5A2{?<bQp-dh)IX!mv$saU&3vog$B4-U!W-8IR@#`R%>sX7z#RrtF&
zclo}*JOE%yIDGE_-8cJbv+wdy>Ue5084b!ew<oWzK1Z?Yp9Pc(s=cwP5$_%!j(St!
zdRreh4*JPsJ(b*NH-`jRMTP{tn7JVTsEf<)XlTB=D@tBH0Sc_tsUHZB9$=MKN-F2#
zQSymvOpy#jqa{M&oyCBS{i-ZHo#DhaW0Z_9x=S0o0laDpwG!_HeYE!IK4%;drrh1I
zsWraW_&CeU!Rp=rT6JixoqBrn{r|KwwIwkqF;~K;a{&>iI$0%=_z&cF_DO1xKKRG&
zj~#ZOX+4ah*Qv7+KvB3<`m-NA(4hYC3%bsV0GandD&<Hn{=VTBMS_;cdLcb+h%@pd
zzY}S^Ik5d;b6^Ae-HbWc1@O~+=u4|Y?#o>9a<7t@Gub(TzMfVTVD!;XdoU<&$bKM6
zp^{g!w$m(Qkjd_3qx}uNeK9F6_OUT<*assvoQ@qSH=W#CmoQ^@l19Qks>GSC@r!Lu
zyT>40`@3Xs%BtP>d6e=>xEx%?vd$IXm)~mG@}#VanzibTg3dbkx`*|;>ll-Iq54df
zbjC{BQD?BY)!i$)?B~yC5|mXm;_7#4wG=v;q>|k0CR)B6jC3|`TxA~9C1WLRs^fe{
zP;#1uKh}LmkKRuA2=V?j{^wscir=z}gv~uLtjLmI=KLW4?nAwnhv|$$)^peP>$aHF
z?6?=7T6$6U;nco$5Ur8|8k6@yzAx=l<eUC;70V~h53#T-DuguGUy2h%NV<zd+7C5j
z!s-p+f;9YuXZVq_>sSBG8E8n(P+wDgmJ<g40dj_6G1>-eGb6u|*M_@mq9spMh6-*D
ztZRSG`&e{TL7_i8lt|%|aRO)CV`s+_Y<nx?Uha})ayycXoIZtPZtu8vCJe=}>Ct<m
z6T#=E`s)u8E^4qT&(?oW5QO<w;Y@$B-+1q*jF6Jo&u9<#gA8!Hr_TMLdpGx{;Iy>s
z$?@gOHlz!`Tt-v9&u!%@kpbQNUC!43%zb!zTz1@o`D1?VSMzU|yq0rky7xI|&l)nw
z(rNGNO%S;A+^Kq_{p+zyWlfjGFEd9yU#YvbZ&!RCkM^G2y5RedWDKsAeEgF0Xn$dh
z>dwo%59g-df1qu0C9<`wUh24vLM)3$qnqyQFQ-2xpNq_i#zFXbvHHV|D)aF+uevC4
zij_wD5k$X@d8+ivEl2#IZz#XD(Bp!;!4J<LKkmc0n&n<uq3Mftz)j{Wx}F~l)*{dY
zn43KbD3acybqfn>;1cwb&aMQ<$eZ*`NTQ<qOk2g2a<Wt+FVg;&{pxuZ?1B%XKZ)8=
z>4@R8$G#;oUCbrkBU38fj^AD`a}hn3{IKEPEo2$zp51AlAm|1}&^_ouH>u@z04hKz
zAUctg$;H(L!+CFpc(aCBo`;~`a@{v32&ltR{augk%}rPW6cpYo4cmfaQHe~Ky};JH
zSWfk<c+c-ZK(>S+WdggP9rr|Nr+clb)A$Fq2fI1Ue1m$%t|lhDl<!S2zt>;)zC%Hu
zt1jmSrXFrbZQ;h=A1L|@kn~y3Ih}G3vT4);QE=G6GEi0!1r0~dbnRXBTa(4Z(y4|j
zu1iR~DMdkQ(HHzrSRoJN$oS=x+68E_aS#n@e3vr%vnJ4B&N=^1#_BQ2Pu{wY*Ocrq
z`uUA3b*)wUj>C4d8Sgwa+&6ySB5tb)zDWWYcVvQGQAf{lf_fiwM|-!zgr+T7%X<xr
zMAMTJ>bB><XTC%Knbd1p99<#&`ChrCrX(h+O(U>*jk5B!kG-TbQjv#ZV2)Hw8<FLz
zB7oi_hAec2$m!_@vgsQbT62w_Q=o457}MFBgbdzE4y<dkQ+9NFmKZ-}H`3AeHj_V`
zxiu)jezgT7fSbS4`C>XO`X2q^s;dOE;2>mY81s|vb{>VxSWg89ylv>$!4`!cYK^@R
zD=<N2*A(1)w-y7Jj3;QKcUM%LOYY(F$@RokdxFU5%`RhQVELP;Id1&^U{)1DAf^i>
z{I24$`J2KHG^my_p+r_o$z9zkaM~>Kd}+)UFV7g2T7KSW-Qb}2JUm-ppB9V#mcrFN
ztPyRiqTsgHkW9GU(`l#WXj?t}ka7~BXKe_n{WTs_DefIv9J@)>+#fB}HW`EC%O{Nw
zu?!!|`JHb6PX&vukDl|Ncn}PTQ~TT{B1l8z>&6GUzk3I%KmCdmRSvg7P#L64o2p-w
zM-<KgGh==9q*_%9!*TnwYG-M(5PlMw_XPMoTL@3x`8hCg5?fMJ&1Zv-`mZ9<R%H9V
z@wG{qA)a4Ss`0pdP-IV1YDily7~!!eUr<g23q<f7GeL!8{fQ1!cgQ>~bk*Pe0&OVz
zh4{CZu)3KgYPpeCghRmcRgAQ2w)^b9boe;sBs}NYCCC5$IZ9V9^Z_#q-(0-Sg?OS@
zufJGw%3|lghaS;iKFTg3H}4QQt`6T$8w0j1vqPUKT<ngIgxaf=e-*4!8PL*X!Y+i)
z??N4Lw@N4_acr_7kZ9fx4u?|3>l<1{&cBtyj8&b|J$CJmYAvZ<8S)D8&4r?8+nbbE
z%~TJ~2fK;~+kPzsbY5?hD#L)K;qG8m86Km`FgBN5|NA^57z@NFg!XG<Cbn@wNrJ3V
z-5o~Af3N5dxLM09q^}ziCj5899%Uf{ICDP1cek++Ngh{RI$E}sfID(DoaU0F8JYpc
ztT*nakk51_g2){R$4)8Ea`;JY!-(ahiV+t$bd59!!WMOn+B0;6IOSnvS|sOpd^!xQ
zWeF0li41OZKa=;n#6AaLNyxt0w#@lBZ5*)Gr<CU4wf=jX0Gq@oHv0qviTH(W=S*&f
zu=DIy<NLOnY<H7&qGe?@jIPAkay4LOS*|avEN+ex9ztFrh1$?=b_%6WZytqCU#zWJ
z?G_8z9YqM|Oc5Y`FIraAN<@YScsy>-ngTk(Up@5$W_j1Y;#wFxuyF9H{SN9pcMx2i
z{GM-l0R)NyarDP6OV<2hbC)b-!Y6E(25dvoMu;-WwfNW}w}<SO`l5G}D|G&@rjlN-
z^6_Q_PsWW7g+@Lmk9U%T>A6I6-!G*V<rQue+58m!LTLrF%n*LlHM(L>h#qP;44QIz
zB^}ln%(F&`<{eH-P~Z`7;8I9wov7T59&CLRuCMT|1X%R>iWvSgjRppvq5b2}n^e%j
zt15Wyc5$T9==Uvx;_eZAf@v2oQKvlG^%O^=e7l$QZL8RAd-U}7^M%;%xm#u_Fm*)b
z#kW`@-V4VkWqHtU(|rub+4D)p?9{`O7($0Ir&d<@r3C;>`k|T|i_N%Z<0``j;ZuIi
z{H_rhqqgt?<|jYrfAAempL8&7sh8RLXMKqHsuP{VMX!k2v=y>@7yV82*VKmji>ODy
zdh)717#Jk5-qGXvZssrD&P8jPnglE<oYR6U&L^9cu9~X9b+{%%_%TL=5;A@1gtqQx
zmYTi~e)~dUQT|Q`1|Dtk>{DdOxxW++Xc`pdIXH>-Qk<5#_YN}lM(W4(hwci~lg}E_
zNw_`>u$>xd#|0_)Ndv*{gy~eJPUn7g1T@dZ%DDAF(>J>jrO;Ep8SyKmA(|T4X8(}?
z#Uf%A$JsLz+@SOJyQO$}D+($kFdUTtp3my_$4Ggk{yUh6>|E&$+*X?gx}(b=N-sFN
zABMz{Ei3ssvwY915-REB5Pts=VJ%HSeuB3<@zfEhCa?ptO{-3%9DaW5;RU1-hlbt(
zlR_q(7DK$?x?6n`8Iv4FO|hyjibsPLaiF^5>pIB1s@|=Ik{>5Re43v=XxFIVt<F$5
z5o|xiM8%%V&y4?MYI1=2tU_bvVDZUfS0|mP&X4+c*^{|(#>hNOIoYF~1#$;BYKj9Z
z^(zH;<-n~yU)UbDhS<wt*h{IO7XNw8e8RX;@n=LmQ%7j@dwXY~<cZzlPhFp@X3zl5
z)LT1xc_2lMP?Z%*sP`EGQIm)*i{t2Ox_?3gxrh;k4bjyFwRLsfx$}F>Uzbph%^`Os
zXcP;;;{nfBxBHYa%7x}@T;nB+lG~tGax5PPBUI{SxdW5}5@o$8#y~$(LLW};yI75`
zom~tnYY4r}kkAl}Wex%W!n#$PMUO!)d8C0$WYBzVxyK*x$%D7}T-b494L9Di6YnYU
z8CukT+Ao54uDZ*JMnf?fq~fSw3`|f#PoN4xqo-QxLEJ6QH|2wXoI(qL0T+9aX@~52
z>v)jNnd}=eZb3nG6$u38_MH|4=9fOhV>;yk5eC5+_gAwK(QgYbm^pnLemp!}ogVLY
z`2VZZD#mD{h&!D>Fu5e0H#KmxcUjvQ0Udcp6H*mC(uWR{X8ZR081AM|^w9x^-f5wv
zq=YF6G$=qC$hn&x0H~Er(GDO@%v<GJ1Zh7|@FI+qumJDdD119iq>g}v;W2_mc#ZWH
zM(TO<qd309$R6Y?{T;jgzcR8VEl}xYc>L|>^oMm$U0y4zsdjJbqPMEAcZ2`pK9+Vg
z0QK~CFKOp<=3VLl+>?Sv7{r4V81HF$NMj^{X+%0)hWAj*91)Y@ErP?CEXL*VQUBhn
zXJ?rSd0j=E!>9_n!A)BdW+q2Etv)yMK4Eq0p(E3}%xRBI<F>EfC-X-!Xc1<X`R=3w
zxCE=a4orPZl>PSG@4+`nqgIVMYoAF*ybW*gQ=k=&h?sMt(_83{S(7M#KQzy}t)AMu
zx{!>GsJMV=Um0`}wYwV1S4i~+`C7&txbmi@QD9kUVAu^0u=9SiU&$OM99mU2-;a`@
zxx|0G8R`iBef{UmlaFgDDgH^-Qm|W<$sic6Q^xR*_xIli_6Bb5vG)^Pf;WhTGJoOC
z>SXKxJMtqIDuEE&f|UaZLiRcVv=-7Vaz_?jkE}95Qr@6<dcsGbBYpX=pE|G7I{&BR
zxxNx%>-&35DQI0elF^A_E`Ei7gteZs#|dcdAL)`WBATQ>EyUz}D8@AScN+;$azhBu
z8aK6bn1K1nkb(La!M~?ok`LhW9ose~be0sc*8(xP>$h_xf3hH~nPCT4WJt7!fz^y!
z`FkPoC)sua)huFpcgurR)!a(GS%32Ii~TvC!)uzf<lVnrjlk%JJ!Eu4d~6EWp9Y{b
z8TR~3+m5D67V;&&di^6&hZj-$i;nsS@ocf*)$c=PLK=@TqN7Rw=T}T?5LnCCf>>Ro
zr&*6NGRdyJ@%W$1X^nJ~@3A~Dwq%YQLut-StGoPp+Dg>0+#O8`>5fG7pp`?f|BHjA
z=GH&IVw6OQ2A>28HUG<Y>?*5^x5R>o3)ytIuoEQoQPwz8_mIe29EDnN)xgV(zl|aO
z<VAP_Hr!-w2Z*(k*;o9Z9s5DLK<`Lhj&y+*>T-=Te?CUJBA!%OGb{&du?{!;`|GNQ
zK<xeg6=8zPxD}Y#;W)2jLW5Vame5@J$0uk<)5xm~i+{pzf5D6%Q3K~o&VTN_Cll$x
zpUOPa7fgVC!I0=4;63=C=a6O7#P7`0$S~UkL>0B$8=U5slzzVv$QXp<OteVJg(Wmy
zs+u^=2Ecpkz_A_M@ccrnD#59iS9{0s_HACd$X1+x5id$|X7)>lVvLV$5~dW@h+3|l
z=kNM^Qve$h1Z~eq&lrqfeIqslO%}qwo!q?`D<vGqV>S5%yi=J6r(q6_O0uB+uSpnW
z6+l_S*S~eXGK)QHAxAQ!EM*7TL0QY?uwMM|cR_%oD7c1{#854a5Tn%O#Z&V=FCQVI
z?3X(K_8v(c53T&8TIVSng`8`>8Os5CTG^5d*cs^-lcOX+Q#{^<;$NV&%cSx6v`rt{
za@z<~hBR|U@096M4P|pq&tDnE``8wXkPj{f^E;=a*Ke>_KNae1zqh88eB?{JND~ce
z7G$~-zqGtGR^5w%#{95G2;VUha@ax}J$9N#rh1pdkT#Kb9|~n?J*`F1&8R~kBKa^?
zS&I|_?PNt_JcLcLrqZ9s<2oT)ynlE)1=wlFL$Ez&g+4yyRe2ciKoy+~7a0QvTNz~;
z^*itUnBi!-A%v6IX-DC5X4Qg&Z6ICtP2JNGJcQiqz`#_<Q+t9B;{U&2#)c0ciE>!|
z4=6NbD)v{^6Gx8VIoNc(&@)v6T*?G&70O9Y4pdwPEI^NpzFfi@d1?LpCW-=c8**rG
z@ki?U4Jq{=xiKE1{ND{Bu-s&pgO9KwB4k4vj~&HmUa}#isq^@q`<tG>N@H4&k~!I6
zc(pNOFL<@*kD_8c*^4V5tYYT(LJfiFUc*$*KOUc|G!-AQY2S4edm#ntSd)mpygc1#
zbP4i!V`NXJlo9k?%8CH{CZK>=d_fU%5)fK|o>g`(MmT86rB|a02~dwk^8XLp)5Y%v
zJdO&PbB-&*KHU5e1*#(i4ILQs&t~B7f<W3T$2*gna}e}v4&qj_gVYcjd9G0DQ%mOg
z6GxV**cGO0&jT~S_qK&U1g;+QNNhPdtq{B?Ai3fmuG{yMOs`fzeQ#@^g2HD4Q5|}S
z@|>PJ5?z;*Cp5Lp`j~<O1ABl734!2DL>)%SSB9K0;3ZKRRN@YxnXO>a{xDSvqG6_%
zJ8SkqynvvA$q4F?=xa~f!nqS+cZ+!UWYoTF+f2fg5QO(ui`3b%3ZK#+e+HrHF_ihM
z5W~n_E&)JM2!~(T0fg(&xtVbVR}BWLp&OM%FAvvRg1Fn2tcJH;$XU^co0{t5TmK(7
zwHJn$&sXwV8r&2Qu?caJDV!9=^x8mWik|Dx25n=*qfSzWf<Y7iRGhBZ#FhYQS{;IF
z2LxlonkQ}0A=eK<<WzNeoy>NUD3OeT$o4aI#Cb_}tAH2R<10CW&cM|>U9r$emP2ey
zAD96o2e+dl1F9WXnb(Y0y}2V{pGNJN9jJBE?0Za}#$`uLdF4WfFWV=O1ioV*Yxog>
z)314-9SSHx1psE}$4J<syt~o+k$L?A(R{P#mzFyeK8o9aa?_H8vowaH8x=1{9P(p5
zTf6SNi0U}YUTZc^?KG5)gn?21z+$k%OS*g8g7yL!nPzDdtH!=N&hy13;XYg=H2*Uj
z^7=($AD&)ILEXEw2#J>CXZBCmJbzNwN(>5+VqwtPeBUJ>qSSfYs+seN5}0Z5rZwgB
z;BIg-MHL$BtwdYPn1kavTFqJYQGI{*4bx=9&kpV*?L*hm=PR3&&U6+Y1FqLB@9Q{P
zD65%oC?p587hTL0_q9J$lkO$AX^V*<5lIN<J63rn=}|4fPa#Akhp*vB6WwKK$+Pat
z@sDOd(~B<J?)dL#I`Zp>>H{30G_;<+-2VA;rgGd5x=A$>jB7wimMiE1D+DwXhvPY@
zEDi@hA{=<`CFSjTGSvSXXloS`MR4-+uwuj(_2eohXQ`N&reE{A=3Ow1vPj4YXCLQn
zpk82Z;O&-}dj?d~T3zUoh5`*2mGuSZTg+=0PxT!yX6Ux?n#H>1g=>ByO5+$lePzoq
za5W;&#hUx2B?A4_8xuL&v`N45)S>6uc{kVdn;v?}tCydp)PVSXEejvP56r8yLxqLk
zsyta;DN)NqrwYR$YOe2X&ifi3BVpx&TfKDIigHY9H(hTbQx(j6I;p9lBpw5eRY)5#
z>7RxGdsdG*o7}pY<t%?ICJ#68=0XNt0Py=T)V6P|ABuNEr%@0o-=k+qjlKgmcv9mA
z42=~au>7ZnClFYhB^_+E7i${lb!E@u(YrD)D&#dmthB-Y;;nV_-Wao5Vl=cqUH#Jy
zd?7*VlPTJ-2vQxvIN9`BTJAgTXqcyZn;p=KF<JEhxdcGgBzmXeAyeSuMHS_iX%evE
z*TWJKHRyq#`b@J`-V}txxcGs&=?{JmF({Bm4BWSqNV_Ris0c(jH-W<RPiSU`1_NDo
zqz^?{U7W-x{%Ha_Ida7KHK3-kZU$E=5%|tHk5}fBP35}k7)UP?qC0(E<x_$|v#>=2
zCmBMKu7;w)`CB0&I~Uj$@sBKIu)FRA`#&tYCJhjT*ryr%l#^Giv*!81U6<e}gxECY
zI_6mtn}jCvr4>>){^?5`WpT1Fli--#60dwj>YG|R5BAcIMtiFBCvs$Lk#rDf%Ogm1
zQ_efaA|_qrv@4+{UQI{%^-0%#9QLwgA4)FH*|5fe;K=br_MBVR_=wx~TyJasHiqIv
zFGdFoYn_osRlOluBi8@(mGdBRmSso`DoUUdRH*NbR;}16IaU`lqh4gA{-}t&QNltM
z_AAPF-&DuGXX)Hj|J)m;tD^9!-l7O|A5Q1=Z{G*Io0Ah+NA$c@51Mj~v}V6_Eo*Pq
zpm2%Wv;e35RidQJoIeCtX`~G_j(l3sWmr532Ixgp(ma)8L6trR44hXNJ|EsgIme4}
zm&<et?JRO`8|GWzW6=&e^J;aw_^DE=0SLgRskGEvl79U4&(hx~8VRA`P=X<W<-)3o
zjH#8XD;&W1ijK4k(<#d5pM1I6ffkBy`;`Y*TZS=&Vz-vH_ec5}{Hglr$>W792KuR_
zYI~8ti^YHM@0UjxJ?A197gxAvG?Lv)NP!Pcz$=Ciyd1hxvdT+Zea!+Cx`RCW8u`lW
zCx~;Ae(BAcnY{h$0xfv4+_Qra54;WHd_j4wLm={@;9|;ViXV)qvw4TxR&09Ct+!vn
zrT&Ovj;)RQj2$U>TO@00#u87q#!}pQB89l1;<+)EqB~F$yzZ;4g$|RXV6(F0Xf>|-
zJY=S=Gi@Af$AgC|#g7>IsaGA#&=o5Z%0`L(VG<s`edgcm%5s*32&NC&q_?6_L>u7!
zJIocvIZRtEn^>MdFL<<Guixf^x+dM9(c$R-s3qVjdssI4c7ta)B8qly;AQqS3Ji#1
zu5NEmWA3S5AN~jafT@*lnT2LT2r4GqJO9zfAntG<OWfNefwP$s{i6Cjx|;N@AC5<B
z0U+h(i0P_rYS-7d=As2wZ^Uh-B$P$iGh=WSH`MJNz(wOdmKIDvm3BQbuy5M(R9@2H
zW6VdGzfhibp_t9bFhn6VA+V1r6{l@lRrVT9^Sx`hO$=nrKyTigbK~JlY8EmCNE9Jd
zNdJ)_hDdn^TP5-?iUX#+l9Eh{BK;fKKqc^rdjbCmKzNrhRK*;1pG*R1JTL%`=m}=1
zX%_mWa5VRq@0Q>Sd2|$<wBm0glJx=ZfhQMUFINJbB`1v4;@!05Sju5`Czv_atv$8T
zC&^r%iyb!C)Z~r))Sk~*M`c$K6Y7=E7~M;5Pmw2$bXA=^p}ot{zV5zd24)&qkb9O&
z&knNa+x-~B)ooo#K}cLA)hQ*Jl+Cx$P-Idc5o6bEG4n+laD&U5Oj-o!RyCgJJP<OO
z_LP=c)z|2w>8=Pa0wJf{65g3^b=Rl3QTd#bl9As@``XHq7`$l(0+`j_E2K#eo$qLh
zQ6nbD{7NovK=bO%;~#>24EJB$qqWv2TtHr?!oj``%SmWXM?v8Km<F+T8^YvYxfShb
z6bvM0hw>453Z&<vwGvl8AvxorWOW_g?}VsuxKKSpR0$?bf{;hxGPy~lhhCT*<&^eP
zUZjFyq<fD+THApFOV+>j!zcrt)JcbF<pm3XCLIU}`l#HV-s6V!l&&#Y$M-n*p?`h*
zzj<}8E*{!wo%KDluc_r<=Cf>xtp#h#h&&-Q)J_a>o7oxp`7$Pq-}{B{k$pwLwKjRd
zc>W)>7;fh0o)m(<(t{FAzN?<lI1V@FoJL$zAVe|vdEe(EV>O*kA^jiYSl>x-01wVU
z&&32`o2lFT(i;ugrxhd(fg~ifUw`xj+bn8BMoU;(W|EN_)e5{n9-YfK&mDc7ETy%s
zFWBVxLA1_Mofbgyn(Gj$!P1)yB9ueK3O$Q=Squ52|8bJ}@s&+IVI!Ge7^+Uzg_;;U
zt;py_#+nhh(H<TSbl_8xp~p-h&jrwZ9X|bUV{-&YvG)N_B4Xs9h>>SA>nwJm++`@c
z%)+ucop<&}neL`N=ZKJ40BUQxN|R<}ycVX#`{VtUNkLHZJ6cKsa9S1cEPP;iMfImh
zG&dBxxr9qO<vim375`1<T_BV>0?_qRh-k@>xVCaqPBJw29Kv9_=-pMi6W1;xEXdwH
zTn8AKGS+Q)*m~Vp1I+4bizVb;pyzt3!s==>`3as9J&c#?Z?gWnOqi>0n?uAed{l&1
z%+N!Gz)HX~4V_&q3o8;xa_V}HXJI65o!r1w7^Z%bRD2AWrisoM@ahgC!opb!hCThX
z)%EV|CMP`S^HGvkKE}Tt;7(YAo+H#Lj>N_;5e4|RLBN_!FB$EAd9XM${^i8*n-fq?
zFm@mPiu@>?w+ND<Y^Ylcd+|0?36p{`dian!jL?HTjxb)=pBLrib}sBt7|wiq+3S*!
zk<OwwZ<>A7KPNUi>Qo_e6vU|zizLXI@E=wfLLT^5!paMjh!8SfLn7FGh73D<%2J#|
z2=thk7A=J?z&v{5td$9o*8v0LWaJp=f|ck5&9@6=A-iP>$-l`rrj{c7gl!#w0O1(S
z;2Ew8bOi*VgtB|TIs%{KFW=#Ldtl<T#`~Mh!VH{KpVlmPc2bG`d3360Dv-b|-(q%R
zgN2GdF4Z7K_p`Z*l#&=cc{tOCS^jMlE>GN>4|~j;EyRJS(70Tmh1q(70m`?hSmWH`
zJToaaIcmYL%^>yDI{G&rL;qYs?*_?C-wwnzb&BX($SXW2W#OlFy7g>#K-SQS3`9$(
zJ~dvn{O^J-QjUjLF$tYn`KUj5LgJDgFrM<u3?QJyE)jCfc`&_f_0}(kk{?BY=9VkL
zKG@ypGf*xM-7T7m_|2r0%>1;x-3IKI3J@ML)EJ`T8*k{SKzl=#HYCpmO~_Y&x0|_r
zVzij;>p&5px=FwA2f!*7XrP;!1OPiGW*ofLW#z9*6<Aqmj0Xa0g60kTZJN%Yz8xVz
zY5J||S6S=U`Z)B~s-Xf%jmmk5CD3?DhwhGDD4rM-x<B&=WfYaW7DpcI;)ssW5Qw~l
zT|XPLeU-dX$&vh*tes}4qnNg5kUaczzngZlZHuiBBk!rOc#aQGFlxipnOAaGR@_o4
z36n7O8v9qwX01gpl1FkJL@gRhDL)X35@%v!?W{b@dd17mq8pWx*Ip3XB-->qU(%vA
z>z;YZZJ|wa6F7nukb~t%Ss^}Jf9e@TJ~floSwtdgAx~ISE-@RUuF#sjVU1%0+ZMrb
zl<anwL2W6of88g@J(ee|=x%hyd{I@IpK}*@4W8r+V2&oGvbe(jhWe*sp8eQa#x<0S
zBs;c8q%Uy%)|>%GvhlQ@M3GG5Y$`v+kM*w&AN3!8EJ2}mgPyB8(=6)3X&3oJTj#!p
zsWk`Ec}YVuAyCF`R{I(zxrHv0oxE-VjKX(Xj~5u2Ia0#e=ccBD5cD)9NeAldP3)AD
zN98YT+dYiJ?SK0=QOm+YcRJ2XrncURG<j<2EMJUUmDD0o8}30~rvy=lp#ZCiA8#R6
z#IP};(NjFEsafaQn<q{e2ns#-&UO1}ThdP9{tgPHXyYw;V_VL2!)FF)QR|YlVoPek
zeF?R&yuZ-gHf@Rrzoj)?UA&i(d9U?d<^V;Hri*BCL|nWRv47bmAcEAy(H*b0J4uFh
zeSd1sFCo2YQN8m2aPvl!OF$>6IsABbkhw2z2h+rfgUy-NQQvZG^)FwXkoWmPq3|>L
zLO2wx>!r$xUT561q3lqAD`)xRh<ralMP_X<q|6Uac>JRU7=*}_Q(tW~7dXntcYV~l
zreBPtT&|D*qL74h9ZEx|c?L4CMLqkWRFysSVxhE#CaB0mchMCFl?qzyP&1&t&e+v?
zG=ZAPV94rH+A(fqu{orW%B!7qH#|vynF_c<VWICE=AtuBKCVQfnF#lBBzekq!<HTP
z4$T8{gWsk;hvE3#p}PVvWYIn_m#Z4P%<c;7cbhoh7cUojGqQ(h_Ws;r%X;&o76SLD
z|LwcXv1g&@NV=jV)3OAyNQ{UY74G1t7t2pV&>wej)u3j9tx0{i-B)j)y#1dGYM&{3
zubd>bm+I7v@vxIKh)nrcuHBO5F%aMTX-BZAWBH~R+Kr;`kbT;TwaNF*HA^J*X;-j%
zMlW3AX^4&R#oF&BfoBUco<enHuZ6!^@YY=%@Yb$PZ8B&0U^&%+U+iwv2!h&;=UT0A
zfQ;~*{D3H1g%tB@f|w=k-+pP%I2mck*;^CYviey1%Qb^>FULO>!-DTl-~-DR7@pYh
zqLU&p+v+5f7*Bm8u_dSW+Bw0;oeq)aOZ*?Xf_#s*zrncGJ*38s%DXSbSXl=@>bEX|
zNKoL!mMXrR^F!vaW3-2cx;aTRyMec)DAkT)%$}I-`H&O;!2muWZbp7wIgbH+&xJl7
z(3rjujS0y*USQ>(g!Qp~F*^k}2#~(V3s>gwZs52G<h{O&RsSt|y#57MVrthmoP&KK
zBRKZdyk9+3U++%PaS|b6LIi0SU079sMt9xXwR|Xsq#T)!YrJTopKd~8Cj=8#jjo0E
z>HY^6VL&Cxuo*MZRt3~IpR|9(lvn*HLEB0Ko0sZSM}nwPtlrI3=vV6;;p16Gi8Up&
zVn0z^buQ@EbWo3*7+)nf0r!yg0u_siV)9!b2M?meM$pOG_wlggBFiZ>W`{27WM!z0
zJ3(!xiVQtVhBnoXQ!x|4l#}`P0!7;v;V50Fz05s6@?X0!tG0FgwH=nT>k85j4Mn*X
ze=e;oTN2iGmqly965=!@18Ag|)|y6m%3fO4|4>tcuLh;oz_Z;!8Zx|O_(G18ux%ea
zk?N8LSct#YGMv4#WpG*|<Iaj7+BZeA6cj!Yb8GHYNK6o<=#85*eoCPoYHBHHbuxLn
z;q%_Gh3D+EStL-DWRj(P)Mo(@Qa+UAngo?9|Lol?-A9(P%wS3*5=%KL##HE`?yT#B
zh7tgCYyeHD<$a=I0wWzI%}5@Oi@qCo9e^0exvi&A>$fHD=S}VUH7SIL8Qp_zFU0{q
z&aDw609np3Qlp=lVET#45g~u(<VvWv51-X1S<1%?AGCGXF4{nMgb|gW&6W&9)o236
zIJHc+cmIJR?L!Z>x9_dSTprQ4It}5DeQg%(6b2jSGxL~FIG9HwZBTwIhxuo<H#FrU
z=W|}Z*SbEbEXN^cdsbPcncKN{#riPVr`z*Ea)iLXs)muDzM?gkVA-l2mlx%==Li4}
z(|~tl<aH#rwzDbW55OL7O(b^KebC0^Q5f<M58%`WOOqc_IUbpp>OoC_U{r=bf#TSy
zs+8DhjX=Ljz0M~D-Iq%EpPG9#>-ji=n)TJWjUpE*pOUd3%+k&ex9qi+ML^GmNPvN*
zzzlLi4l^LIB=l)J$S<1Yk}|Tb`QsLZIZgUX>kqB^&$h;rZEf_OQpzPaM3kveqt-Fv
zhG*IG#k`m?`eT~om-hscs0jc2KJFL>rw7A*-6`ysKDNvT@_;Ol&hb)%wi1H_SqX@l
z{4#IcEZ%|kNB4OzKmf)T;|N<#lJ{isJT1%l7aJr)I-+IK8jT1Ve=DE<?ubpJ+0Ga1
z@I)g*4M|NrFkH5;8dX8W-?@}Utq)1n<Eoz+@JmQ{NGf{eyrj{P2_z->Op=;^tc~U;
zdgnK-MY76npKe=!dXWabl^l%^waEhrvV0KDpV-K|IFfVWKDwfN6`~>emZydwTk0V|
zh!HE?KoH(3Az3KR27#=qScLJ&v0uG)cmNR|Q1lM7x=kDjBDh}mcPBAukqBtd*xT@<
z4tLu7sPCHpe07W|PKp&yihg_o+n+{Y^d3%%%#htyv2dIeGn|zE^kf8s{_GU!?-bqI
zBvvU@gbJZB@O&<B3g}r|8zHLi^!^5q>IO4$iB+$0r#{Z5>8x8f)b_+Fmk2d-zC&9_
zK+Li!tMRSdls*AdpoDpjg+Jl&B5G-rbh~7R9V5^)et@2={q4X@f7ty0lX#t{>{Ti<
ze+Wpi4VsfspGt%#54%Mn9c%0O>x1hmz=5?I`)5I6(WDs-q8&udhlr4)8}KW2YY$$t
zc{fqv#U-a+H&lN&9tGa3JvM{d?A|$x5{w$8V{K8^cm|+7HS{d4f*BVipBF2PGXJxb
z9i^~b7jAi+moU`p@~SeS16my=E@IrXrE}(&aE0sekGi+vCHi8+KYxCUZ;(3D^ekA@
zM$CdyUf)*9=ce+q^}Gf8aS0DZ@2ToP(~AgZ)YKEA<6G0fskGgKKH{iP!`*?I*w3@k
zPe)|>#J@>>V4ie(JNPvX5ua6N_tVL@$&Y4-K2Ie@_JFZ1+83s0dm~*t`^85f-_<bk
z_e`CkQ}6nM55AH;$U1hP#l5$M6=LqLefIPF`TC2qz_}3Ea31O#ZpL7*e$?}pi+G02
zR?nhp_W?=@TF)hvZ2s4%vyv)b2GzZ0U*e_5Ug9EpVy7lGCEJpvRI*DJKcYZB63-B;
zaP@W;&)#~eezdF1?$CP5s#24lOapL2{=1b!1$d#N5e&qtt}m-JA7Ay6K`9Fhi#$ET
zhlR!s8U4;tNs8fT4G4%GV$G1WW+xCVQs3BJcLVqCAH9pCUHd;2s=OO$KMs45zNG(_
zNUc8!VVobXBTQzTM?Kqk!Q`!5*cYDBDSU}l+7qd7w^%4>%hymmQMhn<oBgk6U;Ic2
z`G~~0)9DT>mrLW-mt*#vb38=7l&UnQgygkX_F~}4*qL-PpdKw*yBV<U8SDp+AyxVC
zfM4HB_L9#9B9w&60(FH9B5NXtYg;eR)Mh8R2Q_7v5jmk}bGBg7Kd|XO21+7UI5m=E
zVHMdj>5bGvPMF2m99%+CAn!S4el8$^_21PcPsp}|e1yqN`tvR3zcj2LIK3XPRjX6P
zj<hlyuD?}S*A7($#nZD7o?=>yF(hFLksv5}gVVz^Ablmm{c#Tyv@r7jYkD4tI4VKU
z;sdZiwApdf?laqw@r(76Kok|2U6<G{UPgyLkR&d;Gxd|WSKBc+;u7Ny1E~{lY#*y^
zS=@Fevh6o>YwzGWIyiAM_E=~#;l#91rSzIU+W3@&#%R_;OGgpYOzyPoZ2j`$WgHvD
z{K;{;jwP28ZF90t!IB5nUXEG$+5PHjEO~GVoH!9Y2}L2Q0_0k`j0sZ4CY~r_4}hYn
zhwojYYu~$V^Tlh#iJ;UGfKXcrd(w-6$va2#qJ%$Hst8UHqGzeHU(jb6;${i7@zeGb
z527TiN-yMQ*o6cw9^rBhN^Py)fE%Xd#8y*{vfgIa<-59sa%{=c?_KvXtIHt)LH%0s
z$IMQg6n;{|^{W}?7vKO@R?@2H&y{@WWLp`c4zo@$7UeRp{m;#yJwcJ4`raq<726Uo
zMzTW7c|zFtqgKiBniy7W#2FWO8jtdVnsN?0$DW0Uz^$V_;3;;DYOT8FttBsFx~`1f
zO>T{@vHtVYq1}Ts0T90EiltNuY)jw9^X^SuO-(_2eX|$JgJd`pCq$jDF>XIBOK@+W
z3Z+^13u-~%df<yWU*lQMK!qeprTq)XOjd{i=c`IKRS4c-h+2K+I%mz!FuN5C8@;$r
zqs&X2kIF>uETmaV#mRynRNBL3M@2z9%It+wS|tR#$V@KL2X8*qyRH9vG*wd`10$RO
zzp$5zsJ7dDqbp#b_6Y65SJE!S!kk(X!WVRvyx`H%yF4}0N(%pZ!KGXAg1s-5;;XPP
znsi;#5l0*Mn~{$g`;-s?1EP;epbI#x!uD&tfBoa@Yw(=e-O3pi`<LLdusXgD`fxRY
zY2KPGzKW;kKAhdLz9;ooxLi)ybG4X)yFBD0!UttL-T%onGz81;W)_b~$@XOnmhSp4
zeC@6ks$Z$OJhOPN$F7vVH^qJ%SN4xDmf<OhsO)1<oWhN`!hk0p<-NO%{C$pG7up3u
zsNRNKWwi!z*)igPkq0ilf@^TcD6xT)9pAc_PO4wYR-s|de{O$C7!K%+ZF|2cE|(mo
zqaSy@@$2<?@GB*tv6{fWy%JV~d=C)`)<pNoU)X85hZIlIE-7ET`7b~D{lAR)PAoh9
z)tIXvWU(}~!t0#3jO>meX8za0ETPm4#S;Zb>#o4L0%ZDg{a&vmX8f26B?Qi-2kPUm
z4Yfbw?~zd3j0llYP*;>w`Unq?dXzsIq=!Bd4<=EDLp!pL%D>~gl8Q<YYNmID5xR?(
zp{7%XyKRDqcMQrQL{BCGogUjx`<4pG+VL^z+XWP@B5;T~%t8$#buUi+j46_@B|+az
zH|=<aWt|xD6Hb_pZj8Wx>~AJRLVxK2mo*q+E`hNml1Uc5U**1m3%|7rdWbb&K+X$<
zx2ptTmS{dDJtIj0IzJm98>uf3@4jckw8P-cURTWgBf>mHo+kA@QlVV~C^{FUWZg&M
zpaH557A+5=^F1;lHzMThR~@_0kQ6m34YA~bfMO6W-KY`kD~lWJ?fGp_W#`==G4yO*
z7b#%lM9MFZOq~q>@!unJgR>-pFJQuAGX<^UayTcgzcw&0qtv`vf{{3h;Ii1FI(U6q
z5x`i+n07-q{yqfVfPIx{-kD%@6`=yXNnhniz(BQtaSQTY58wIMC5XQM9eQMsE~$#A
z<STd!bnll};aies_^i05A>YSVcoibKybkM^Yx3^+7-Ac^U8Y7ti1Z3-COo)LIWwfS
zJMsCQA#gILl)H3m2gZ>voGHNesoXo@oq$Tzp@|R1O=-te0Ac3X_kBX^suWQ%2nPm*
zjwaQ=f|`Jl(Q=a}YGze{`3!Mh+j2}+)Us*2xcgi_Q7junLGNmhBoE&R(Qb@#Xtz+O
zS6H-2)wrNO;-@w#vMs$Ly?<J&pZW=n#il<m8||nxZe(%nffYEsoOf_iDK(bVV08zm
z(6&}6hG>}GzzF4I&sWi`e)+48k}`m0G`Cd#w_`cRN~C@J;$}JV?xli6iKn8m5M+cf
zTJ}c57J;2ied>U)V+?S+Dx95L%Qy2Jlo#i_WP1S90!SBcS2F3JHQSR(U2?lDXAWJ~
z?L3n0*W62~uvZJM=eB1a&w|eQSqhp%+USt>X@<*mkCr4fUlo3DLr^5RE==wP*{$&h
zxW5zG(9;=;-ddK4<t$Cx8bs=(PN#3ND9lR@rG1ehk!qcW%ymTN#^TXAr3~+&`bE>}
z5;d{bOV&AG`Yac5?LXJpRn1c@Bs1Bs7XL;`7C{BQDuq67y1>%)H3a$+<04P`x_o*{
zM-QJkLIftsq*CqcHnZIuCR4Z~QGTif3F%bFS#gtzlu+BI#E?~o0ez4nIGsc!3Gc7F
zJ*>v*xOJ2Fw&Y)^b-%hDdVw_6H&^YeEx14=EQm<>tIihG4HHqVR}U3`VDPk^>e7}$
zY0C-)!T-hHn}&0_w(Y~Qv_gX^Ny;omRLT@088c;`H5kg6F{DIMNXE=VWtNf*85&4r
zN-`8e6f&j~Dn;*ps-EY5-tB$Y{d~B$?f>Qf;kK={?(+Lx*L7a!xgYzn@B2ZSE1|r6
z7d)U5kcV>(C1U9%x!i}^`bpt4M95s8*ik}zg_YedGaK`~mc}XoZ|@&BV1M<le{oKH
ztHI59D9jX`Ft>}u%L!#y8VU@x-WlyR%rj^nvORMug{ea^p7uKRuMOJYivj25_O{|^
z48ZU?*Wvw~$>n?N=6~J#I3onG(`GiRi>$$hr%iU+5b5{w%P1qMbDC}d#VR$@s!kwv
z)Ma9ARJ<RP$EFu(-1Zcjb`A30;M>6bUSglnw;aKpb=Gmwi=^9Z^2DVYR{snH96rmH
zv@z&P_h`Y+SS$tBGh$CNlcQRXdo?;Fw!OAvWU?lyad4Z2;Y>n5%;BwF<}pB8yyr|V
zKf|mpTnq{82#jOW`;kG(hsi*?m*k0tA~v++!AtZDLwkfrOh&vK_~PnAm<6BahFsL+
zRQkQQ*Ie@ZUi!Q*i&lq*{-YG$s7I_r`@Vd?rT+6bNdA3~6bQ~#nJ-=qOhJhwK6l5Z
zPzuUioQWE_I|lJuX|E7Pd(V=F-|o7E)k7p-wpDo8_vr8*X(6p0^ziRmBQ2O0D%jEy
zO@+SjTs9H@oC(tBmdt(^&w8W<{XHVP%_8B<+rfOV&V4MIAmMT}=mih+7%_xe<&|6m
zc#%*XhU87<m`A-$1$V-mPiWkjdLB77rmH2xhsEahV!A(PZr}H`H|UcvZ!*nr#{7*9
zt>N#+J|vp3>u_QA$I>Oy(-8!=FF}k^5hq@RG~KB*r+u#i|HrGM<Bt<hFJAEp*+-(V
zk0`WdJ7NJaQk2fdsC-Mj3m_|K@kxLwd<KHOMi@C9lR|hWX;=o;lf;_K1`1i0^S~~%
z*28l+0NbVsyVn54NXN#<6Cpz(_Pu2;WCOft7MH+O!dr|1Lb|(qUE4@r37{pbK{PA-
zD<#lSj^?caGJvqrNFVLEa(VNl_VrF=7CWY_pku#+bL*8G_wQ$9L~x&4MMui<rBkKq
zeCnua<lO&wjuf)MbHgLu=^p@{MCbuTRu#!{;E)&hha^^oY+yJSr>xiv?Zxe8H?zu@
zCEZe~X7@1Ge{jm9$9(&da1%k~OvVebsD6xMJrW+`Pc(qzWOC-^d)rKHuaNN?HwQ*Z
zlNrmlr!9ebjx0|n=qa5h!gg!&x4Uf38o?DoYZL{>ZBJYSe##rYTJ`5D*3;uE{>sry
zKs@LiRra|RGPYBydw$ghzb=%5`fJ28djSs34dwt#Q6YFR8HsE50FF`7<TUXyXF9yT
z-z_2u0Z%5Bf@v>}p;<lL&cte5d^~dZiPya+xuOb`>zpA1=n=7~CjG)dql92yuZGZq
zY?R1ok#1{GrL{m3+u@#P4!rlOQ=fjU5A|ypJ9aWq#%R)o+I8tD_xB>-l&-NjatpcN
zL)bSwR<;QPim_0Sc1n_-H7PmoI`s+652NHWw-UR68Zm~2euvWUv#gPz>uKGm%K!`4
z0~T(utkR<cj7_Qn>pBE6Xe3^o10<vv?+NB;bUSpiTw1w{$11qAARC|#HK)L?UJ{)Z
zK*Llh?H(Xwtn_60(p~yfo?l&YAq@Dj0k!vl)3G2tc6n;jrXy1k$x+u<au%`}RXC%D
z5B;2vJD&Up@cEr3Nccl<s~c9lBtgPq-q!$@fslFA3$jsIO+&K1;TjSAB2851{_9z%
zaZ<K#tY#NyefxV&j#AzCJvLh!vy7sWiBv=T*xf7p8)rBGYLqC+7O%|iT@!EZi&V!9
zz%!z=BC~^3Jd5nZBr+H^|KLxREjZF?YPAM&0kmY@QbyO)qj(BnP*FM(ULi;#QiP&O
zXg7-JN$-q$2$V8#HWoVei1AGA>tYIKwEg{+>amq8zLU4JqbFZTnSG|C>^kdmJT_@6
zbf^W$tA?EHhW&Ihyr3nDQ+xmyhdF+Jy(n|y_uS?=Y3)c<ej)S0{D0_z2oG#0h0j6-
z$rmo8`N;0k&bfvZ==M~MLX%|H4>BRSGTQBGP0XDouVWn)B0uUgMRN2yUo6*hVg}#)
zeR5ZHU;%$J{mAcyegePres{#+S+c;~026;Ek(?XK2E~Dvvx`NehDsaOZMWElMZ0?J
zFbaBW0E*3yI2PJact!m1x|*{jydY<!8mjogKI!);GRZGLtpybDAv=NszYTfgKD!?=
z$$iGdy-h<}z9z-TkJRlT=3NXOlj*Dn!XtF2m?JPJ;3o@c3twbNl9Sd=fasEQW;b~n
zRr)?4qYlJukpjuZ&EEW61bfQbJ~!8MzEodG3N@5k?|HiiAzUCf{An`b&%X0bb*jqy
zE?Tq5ryr>~M*ejdFHA;*Q?(4U)vckg`PuAV+ad6!%KtCN#{cb>k+IAl3fx+UgA!~k
z$h(OIhS+XpHQw%yqDUu0g@Hu>t5|Sry6Y-7b1BbP3otBP*ZB2r`t_rVpY925J%oqY
zO_u*%*6XskLszK@b(Oa6ddf$j9u@RwlXcyGplZng2_)u-evvEG8L!;<ry=}P#TQS}
zWfgOXcM(0oM9JV0qmO~PNgdf*97G5&$+6?GU&C3rJ!=<^B<YsH!!Ls9($5UZzE*R1
zIXA?a6UGt*iy6swF;RL9mFh5_rl&CTEz{avtbtEIb`>292chxOKolZIGM)T=xl~%(
zR_2wq2QQE+CHUgc%rsulO(bMb#JVwnW$-smL(hcOb@W`!f#?8IYAc}BZrG$Mj+g2R
z=SMJ=jfju8uOo%Tk|5u^P?5B-i?&{8<EQqCay&Uow*-;C24XL+e7bf4n_^T9_F^l9
zWi7`Z?Oz3q&wuvk_kKr1lzztLNLzScu7p849%L?zW1>kN>mK(U8BcBqXQKXIy+I<I
z$lXZ5>twkSVFHP%1@%P23<{9PMt;JGpuP{8z6f+vW8FpBqje{Uh6Pw5(CmW;qtlC1
zutU1yX7;?blT^ONV7*r4u_cmz2~0AL2}#zQT+2T5SZ33C-?HrMAcPC?H$NVIXJ#Ir
zyAohT?G8Zep6oyD;7&f&Us+zc(vQeF#P7F4=8-2isUo&2%4x7dyYU|5_DFHmBHf<K
zZ#wj+jCbOiRK43xmI|=M01NK%4-wrNFheQ^sN$kMY(@zc_LNF*NBT=ku(8|h{GZSu
zr9w=20I#V1T!z7I3&()H_-D5e6+(9+#ljWbC!1P3wzepATbC3yVVaN7qq#&Is7IAP
zEBRr9l+aHTcBLGh^Fpz5iPn`IdTkwwXHc|R!Rc^UJwpY_#y5LTl~8ARz=ENJQK;*6
z%$Dy*Bt`UCc-W&&B9HgWJ~XXxUvg0t+B9$QDW5Z3u8Q3y1{53Xr>GF`4#x+kH`7^i
zx^nScF7z26OR*krcezoLia``zdtz>dWb5_n=9@_-X;M-8^6xnQ{F-}9AV-}$cIQj9
zw0-MZen_mDUqx}z?=#!`WGUrY&V7D<{iIfT#z|Mvf5Lj6&j0?tfq_To3y;>)T1FXa
z43Bs2cj_N#kbiUa=Z|cXMs-)puG&b0hTqZ)?T4pI7uv-GWY*Ax{odSf0?WsUT3jJ4
z3cV((8wrMI1OZv8Un6Y-?&Obx9c@oB-jEZ7sFRmLJyFSRFYR@`lznMyc7ad$1h5r}
zux-KtqJH;bo&F?=!^-Gcq1zf6as(--@Aa(11)gc|(*X19@ane-rF0~cd%U8_(O;#>
z>O$S2#c3ApOM6?ZQ7eUxD%UEtRucoq{3A0QF6qqC(9!VlDbQ=LP_;?rzP7Gf<W$%X
zN(G^?2Y4-le13==nb{?P<4*fi#F!h|`I)fuf0jy^+RsTx?*%xk(A<sNGIh-s^LE4z
zv6j2`KGXh_1<*dY=@9xreLtRGUM^ZPK{dNzUT}Edh+aNS&7T9KHK-M4^7tNTHgC#n
zaqIh7_OW_pCd6PigXuVZr{0deov)1%&LV@70799W_vO-k0FG-~E#7<hVXuHrM$iMz
z%EN8T<L9M3SM-JEU<j{<9T_u+;^lOv6)M?<SK}z&?>4+T`=#@U<^`w1vNN%Hdw!st
zjwL3$^h7r&3~GZON2Oy`aabmr3QtZ{^nAJZGLA{pZjUz(uxZqhY%cM0`^7D;++lwa
zS3#TNbRQzBi9c^<!e+dg?KxYh$@$8SVuayN#L_>H(d3xkKx1b;g|?D`ZO4rgljS-q
z<w1k(%762M_8_SDXy0UaFpXcg`|B~zjpopQ`Qhml{4EJP*x4422jQBe6riI!HNzHY
zhTnaMszcg|gp;AZVUS8Yv{q4Mf$ffSD5L((&ln1Bbz|=5LjTXzpM3JuGEvb6J2{PV
zb02zmI_;2*<UWma7`wB?k~FWE@)j^Chm0kZ&UrZfgB$Z>)TcE;hRLq;nbE$?!}0rG
ziU@EvZXpq(N)nmpt~<AenP|0bzn*;U-1@#`nlHN+e;+kZzv|VHXm>IJ!!2Rw_nUjW
zX7hwgRqJs&C)`d??$1efZrk42!GDjZ3<X$`>i2snEP7aGGxPVpp!xHTFgp9*uD<Q|
z^KRgX3jrz1D1Y)*`OH*q3;K-?a0Xayj<Bm{x4vA)pYTCqKGqofn21`4t1mBMBkiR}
zeea(c+eJ0K#($5;cs&R?W{)ajdV_*iNr<UQT~un4`VdJmx_y5-#L)>9OIkHEf6iIQ
zy1*`+|CeJnl#Hgwp23hg$B^9F4b0zjv|RXCt=~~~#pT&-SLumEgCk~xARp*OnQua|
z&QRwlVzgXPz^wJ!0*N=#SkO#L%%|jL)|T9N#xL)~Ct^P%iYI#u<Ij%+WfHUG(+O=6
z<S}D*o@Y|~=adse^UvWB2FYB$6pYW6Qk~0De_`mYiGNQ+)ycYY|M=(LH@DBB<24^I
z?YI`)sMarTv`__{$~u2-a%#%0$7(wwC-L89qUr%o=0KT{4K7+yubtfm+kq{uH!v?h
zqe2+wHm~PryuP$qQx7pv!by||dPU%ID-1GSL9!y-)?cJ-k^USyGtt>{H^b>p$00AT
zufztTCd~YpD2qEvVx)46k1}^ly!7da3ODK2w%@g(BgFtG+T~HDmaC`6b0bcQ-a;Ah
z<U!mMwJTe=8rwH@FoswBaIHp@L>RIVPtE;dmQO+>k#W_TrN}sU{2j>n%N`rEuklQ_
ze#(j2=;58Y__A+b-6!Ce4Ghmg#yOasHHRHD&ES-4qb-D-FRkM%<~QJ+zGus`#jY(Q
z_)S0oKh^3bN}MxdIhIA?vP)@XWpM+RZ=j%6{k6)-yGcWnaic(4n`cV3nk`J&IuX^h
zdV1P5x+QXI16>8(2aAeT{?hV?p*ccE;)_+vJph*=3hgcozt!d}s%9_Qfr3pM5i)^b
zRb`KMR<o!&<w-4(uZo-a@6WxmC&9UUvfq8J`ipM8p^0W%NI<V|QxW&%HvnXwai=C4
zi)waej@-QM{1VmdBW8Vv%FVrO=RqkEHaIKDWURZ)B|WE0>Fib4#xeN9cE7w(JPpz|
z&`4Kg_1}v+;?@GZ@?C@UZ^#Y<uQ5Rdc{N*5rO9z#0opmpp8PK_4Ill(V<|Tmo;p?1
zf3b?&*vUt(S)paw`?@G#EKYOw8Kc)W9UxJL0{kj2>8S0(ArfNJjEOBo>Jx0T*&-o<
zVUAdc5-}8}!W|kPwm6f>`D}S6kNv|`%=ZWpl6=GGCXif^2~`^z3`u6oYV>}aAhxur
zl4Q3Mi}X$jF2U+l!n?n^>o{Syp_Mt#=7x1D*y?Uq;BC_TM!uKB;wI1DBR6Nes0?|<
zC7R=RJ~r8}6L*ZvO=tL2mKcdMeZ9RrYM`~uS`5Dnv43~{5s?MifCj4VvzXCT|13qS
zsF$N=sVW86D5_7m#voQg+!jlZ?z~r*vmUM2<&>mD%@C=OX9pjI!4{@lX42ClJd;a<
z7NK-Cp3(%2wubV1t^#I>2EZKBP+H}3HYPUK<|d2^qv?<vU+tPkM{)<{I({yO>QrYz
zb(Cna*p?;DX(iwNo$b7G-+WRhqC2$3h|AAWt0@I}t2zXZEB)z${MqG}mNTix6ISN_
z_j2y0R$!VVWV9<FS;p8jVt06urmTH;{(Kp~X;;}6f4u#C^m{exlF7q9%03g{Z@(Mh
zW5}-x^Rhyla<TI$-^>JRg!3yS$U9i&X=p*+d+rI0Y!Y|M7HRK<pKrZ(h+#K)LDmWJ
zMoXZ;u@JuV1tTC-sPAlBtI-wBmfZ157acBZ{Etg|8v2GtUo^>)wG$H|{ik03w~xs%
zK&FwyMXT_2;EW(6utsD~M63V-y2qkc$ml!SKB4gM;JYTApcillGc%;>p)Bt!OVANF
zLugmQSa6t?QWwTWvJrZIY&=Ih-3l5Y$?-X$&|y4@wH|#Q7`$ZZ{SF2T#**G$C{0XD
z`vYXkYsUzy0pFkjsB;@*iAdmZLO!g{#sx?={5j%{x$&B*m2E-WSB{gk6vrYW{ty!=
zGos{fQ2We0*18=Y+&94RrOcLI%LY49l=sO|Q%$nFFb?f&8M{?8i!EE48WK!(y=&KF
z4GGadJN_aXu<S|t*tU{L^vogRRp>iz3qSou_4plJYK~<q^WsO~uZaeUuSR@1$b;ME
zF@DYxD)_cM!|kTMC#q9>hcX^sc|rLc47L!Rr^2oIoR(n}#`mIw_hUvHY8k6-E`0ZS
zBq#kscD_h7LV@FI#1yxmQxcrYfoI;=4N*v*MNv81V%DnrtKdXmS<GLLVGa(3$CK$_
zCOWr2rcvyT=HjO1rN9GH-9h8dWka4&>-4p{N%r)kxHEK&_28`jP%(Q1wvg<c+6nVX
zrJQwMy~9)R5Plnx{b?JbheRcmj{IchKdAYD3wUegPsL5bLMb_9F05d_bOsO|s*39$
zw>{&E*g@GUrpEX(*Fo&*W)-nVE(zPI%g$cy+w`*EiE^pvvr-4NeiFKcVrf$(uSk@?
zT=9b&BwbiDRXSl+-|jPI?-K1v+G|J`2lE_}16F_e%(HF{Z{h7>#S1f5TzJC9Hs^e6
z&XJczA3I~O1)Q(cL;93&R{H-AEZpx*T1~0G^R`K17>7anY%<|%we3Y<To9Yen0c^3
zhAR=j@})oAc#CM%s_InYL?x)P=QeWk_rZGV(83SRA9~J3BU36)&TT&rn!_#Dam%#R
zyySRg@gkf*0U20`J!S7{Rfni3X?W=!i+KI}8c()6oU>xv#IjkEJcYgc-04utog$^2
zZ^$U;e{~6VP^G=s|EArQexA0!{Y#?VT*>|Vh}2bNM|iyL`9&k51>kKu&z@%*dJ1gc
zeHG5r4D_m9pUJ_y>*|`TXd*1NJrR+lsZ(Kz{JqQfZoy>nbZB<6yS?J$9+=`4%rycp
zIv;5P6U<8_X0<by0+2@aJT))r2K=*KZk`I24$rr5O#VaNxY>n6ge2zFmK?AYaD<M?
zh+ZN0WW#DraAV9RKZ>vyxl;<JBtibPAu_!E5mXh8co$N1!{}8c{EsK`(R?ryld4);
z1;#pruIy;vKZ3)61E>A?dDXN_|GhuTalW#MITR?f=gyn&^A^#~4`W$a+nCU9(oZ|)
zH)p9zz_l>lsck2A!LLg-y}=(9Sp)c-_XS@jyLreIS((ltT02OtS&zmRqs5uV9S=2^
zRL0Fz690)ArQG2%bMa&Z)6m2$k*0hi`{kPh0QSx0pF2q~7lOdT-2_m(kdn3gGj0}`
zQF$umyN5Y3CRe9?varb^DlRj{d1os?J<4w3xeDhaTE${+=c7#qh#7YXt0@lqSdXtn
zB<icg@-8UaRNF*MVgZV|$f7Wwk5BO}po!XKXBPdGJmAlh&&8<`o`Sy6{E^tv-|`(0
zZ7SwR!k-DfWuJ2XFW#MXX(7~{Pi?&=zZa{=|0$CfNSn*c>dKFTen*>LbfkkFk3|T_
zXvxS1ZTmfV7h=J{;ezysh?!LQo`uhQQKGYDn3h=QDO^6pNCpT#@P!xciGE!4M;wmz
zING^T_Pa${{gHQlT9S8lKRk1fc$VIM!3+SRFnzWHJa?}z8Kv~jtZ9#<eKjiWW8ctB
z)<|8<6u#Z@pbN$8@%G)*^*rJAKq?N2w8x*x{ygoI|2}7Zqu|h$Xr9UP(X>knw>AY{
zx?0=^Em0FP&a5h~QQkoITcX6-eVnY+1POL?PpMvz3f()j4zE(VV$>;I{`~{kNi0&l
zp8ymAUtwSDr&j36n=YzlKgLe~MS@pi{xTt$n%6Hh(-PodB0Xwc?zjb76S=A+zYtS!
zm0XzYRv1{lTwL=BnP~c(6p~cc4o5tq2<6)BgGixLFxQv74!4!kTD^Y}k7>rr+-TZ<
zW+ZQ=fJYz6%p_XVn_rcGMn8btFU=!o$8j`j(u6CMuXUb4^IS3pT`c1C3SD_ey#I7N
z^+k7e0UOtICW1P7`w%DpAol|aseqR4RfE{n{f=94hV@(dmAlesJtw=g$rx9pMU^(E
zNVv!o`7$w|nS0$(D@J3h>ohmL%5L(`1y7ke6^EDci*M0U7D1CC=691u7La88SM%mq
zpK6GYvUTo;&c3d6w_J__)SI%8ZLJRp?pIhF#DiPx5}={$!cRmg$tz^0zOb)G$8xKZ
z+;S1M*li3YH`6IMdTu%N@cx~1x|P0Xc>G%3_+g|pB7PqLDd#9|cAwetm}eE?A;)0p
zbgYq#6xQ(%tB7&E_o&TPpx7Ax>73<kt4(a6KiE9AI6-2*&<9K0D>a7gYggJv<{gT(
zdtyl}4K8nO)UB0!JT7_=-7)&LB;20sn*e*oSXNT#L6zkP-f;0J_g1stGOW9`F5pTZ
zN!T5ybwI3SRs)*8YM5j7IB8@fB@G{5K?SL-b0t<Z0$cQ24;%cI#A(c@^v#OsC0LII
z?gbq8<*xGS&?-_q^r!h%-$b_O%#Z@*5f&FMi-%1Mk|$s4EhdYfO2M#PpL5Ax&v$t~
zK1fT}AB(@rR2p}T5NRWmSPOfc3WS){ra0hrF!GW~^B8sxuUJWNQC}`IUqwf==LbkY
zU82Obv>n0j*#`IJV71yxa=IAZB(vOw&Vv0fM-%*3Ajg8n!Rut+^WomJ-RB<FU)y&+
zbYfd+netCvUkLQx30@%G6cT@MAWq2CVCw#7&GtBAJEPF>N_kdJo4yoNw|gkwTr;d7
z8)~X)6nh)uj&L!r(1hTgm(cMDA{3Us<VMY#>~~d6ZOI9Q$g-l^RW4e03tY4r2}!09
zz0VN)_#sBDsV3X>_N9G%fQQm9(DVzdEM`CwwM%!=XYDUzc8m-k+bv!I?^;fMQ~BCq
z_Z`D&c&xj3*T}3}&cJ~cI!sQ2?#*PWg{_e3`9jZvrAA*bN{Nk3H2)?jO(B70IxA%b
zY7jNy+#yEvF}*4PE*iw<Ni4iBrNd?!lp9N)r*pH_L7O*1w0V7H&)X)NVgyH?9bhs1
z6js5e8)q>dM&|#CPTecMW-{T0CZ5Kj#re+43r0Uw4YkD>3ma9<y>oqTVZ#h5oJWW1
z={ZLrRUIfh+O^)Fg>a*&l}*@RZj$D!Z4objH9KE$8UNfDCZU=zlxkgfGHrmXPLK|h
z_`3vcaut4Zr)iR<y>(_Dldjwbm-C;mzXqKpjPW(0CypH#vZnrOds`s8nmq;gW2@Ul
z)(<&o1%9j@8KdBa2t$_cRI2Zq_0oTkf~q>FzL*-p?HCQw60!Sj>M90pPE`U7+k&r~
zVga5Yz_3}bReNbe{K(+nBz#XDt*ONPI%=ObE-tQ>R65qT1$M3T*|;spXCvdrXd+~_
zO5HRct4kI$mG^VEQ-Urse?E~O!Z%sRScRgl^dvr+Q!^20R9I8fx@cVp+7b~!4zM>F
z&-t*OB5o{v-mYSF+tT0s8c>Cve5gk+7H}h|{9`_s3+mE1qgGS;zRGJ{eMx2wcMV})
z8;Izn@Y$+`#?|E0*W%L$Po_aALGB&)sfh1XKZ=c`A`6rkVr5~f4DxaPDR%t{FDKqt
zn$7;g_;U}@f$HdNR~}h<UpX6S9J&hj&75cB=bUs=aelgnLiWA>a8`k`G1pYXvrUh2
z?&I%TWZ3c}M_`UB_wzOWw{z*g8#m+=carNxiIUEv)EoWjzHaqdT;;bD&qjCR|1=}=
zjn320*_OVciV!7ndjY)n?Bklevn9ytyx-Ni8?L5Z>1C+l2eaNpm61L#%hTOaoaFij
z(C2kufnmeHD`NV3^n}FJL<A>Wxp<};4!>=f+l0OMVUoxmR4=@&C|*>>xZGDP-3*<O
zD)?K4K}i-_%6D06(>MH0uB}$SE4`k4+LJd?@c>!OZZ*3j<jALe#HV?&v_atayZi}~
zuhL55GiI0@jbQkqkl<B#TA6h6vO?DoJ&zI&(<lw<K7VpP_+R06CYgq4n`~mLX1{b!
zMQmxQCKhBv2PSg3%FluRYo!`Fln$XObD{I9%M;dV3|LUb`6jYWjj$A0Sjo=^Us;h;
zL}I`6XoGKFo9v%QJ1wwupY?J4gp~2$e%+lb?(|YGS3Elu>d&C<xbjcK8uoaB-q8@B
zEK!>RwJ|Sf4WL3I36CO&DEPP{sFu8u#GUt^=6`%mezyHBb9}!<V8t*R_jovSQHhXe
z&m%wdJ%MzsCO=*K{!J_ka)w_gVur}lSHgkocCm~<=EK1KU1BfwypQFxZbd_Y4Y1wv
zaf}H7crpWF(n$bSA-+>Y$Zkhix^;?=w?@n;Bxhu1HUJB3L%kk{*kyUrWF?6JPzN?4
z&-%!-vq+~UiP}Vdj6G&A5_#1t@+&4<=SlF7?Qe#gAXZfH(Wt!GCe|UJS<7>QH1;ox
zzm6c`0@$O5&~C`$02#x$kp8pWqQZU5y)%%1)B~Z!aZnHR{^&+TCBbV!xkQauG4*r{
zDDwRT0AlWP%n=gjT@Dg|^?8Ee*F=s>cu!}L;kU2P{yf7oq_Y_<*xj9DE;T_67sl|p
z@l(^uW_dBzr7croIbO#Qxl(A=!1#FtxRM}dl985qsM6PFqo(&#Tcw+8Ap5}$D~7o(
zBVb1Bp(ZLkpKQ+=v6b+_fx+^7iRM+pMS67agY#oze^%*MEwV_vnde4%0r|d}N)E56
zPZSu=j;Fwex|6h<pZMjn<A4zH2LxZ5|3N7B+Yg{-3=QHtYO|3M)T4ckIMZhJ9r7Ep
zRhg@EJg$Q|Fh*$DTizZ^*xQM>q4@2(+o#~=axfG30KC#0cpCsBya8lO&BAK~1EK>4
z3z~G4Bq9;whBPxrL?lrp49ouf7ETQDVnWhs@*5pb0DEWoGSB4X98lb^S%xm-WxpI8
zu9`aa+O<Xk%i3IW14F@8utrAejv|*k4tAp^fZ+^MuND3PfkPUPP52<2Huq#e?r74e
z!yj@i4`$+TBvLLErqTAZ^OJMNw>_#+XAL1TZil?E2CR{vLnG`!OuS=%`OUS3r<(oZ
z#=1#*9J$QZsgqB4bbgJ2NUK}s?=Ro`Q#3bHKIEn^i$qABzv`v0D<uA`d~Bn^eDsLN
zdt2E)(|^MMa$C(w;hVEnHo?aZ*>~hH80=Xf8mVFL2jBbi{F}Rrt5~j#334t+r91FY
zx$hIz$>1!q{eBJc``$Me_T7%&eH|*8t&|ii#aZ*UfbpUucCM(|U`wHWpk{IT&Ato_
zagjZm`q@_qM>=@M9`*jeC=-IL|N80j!C0$`GHjEl9Iw@P;&#Yf9<W1f`(Y1$A<oe|
zvn-f_uJN5^D;3p&yAS-l@E>VKYAMqaiv1#<yu5Dtwdv4e)H-^z%)HH<FLw@83TL;a
zo|)?|_T#Mo0i%SIiQMzGI&~OaDxd!WlDh`Fpr|&=YVM>jz(HA(-mgx|by<E+4vM~-
zV(->=EPKH(Fryf7Drr%05Pgal4F|qsAW}Q!x>)%q`GeHdJ`Swh^!3=u-s_KjulD}J
zH=ol>O&`QZaoa~a))6^nZx!X38v~Y6u)90`^=Iefj7VNO)O1nbEqj|fbf0ca{|U|*
z>!(4C<bk9?mA>Kf8gJ}<_i3oM6?;u3-0L?)As6P^sqRA&r)EpRVr)zC#mM#zB`@>r
zs4^!-oUSc;B?l5|ZSZB|oTL+Pa}7Lf3ObG!NlLzYRi>dyRf(<%uxPv9wsgx%RH{!s
z<~!lZj5oON<r2;AtY%XyaaWErjWP|qX5Gt81)J%)7h4MjGFs^*ihVb=U>)}A$M31G
zJwP$Bm&^Jg&}gxrD7&sQ@27~cA_%wca!{!4wLNrZHYRs}?e=H)|LzS7?8xI6pm(j=
z3D!8-EGc@gR7%G~<N&>xJ2WCKokKqXEkdca`%P+}>a8xR%mf+8cSXk}WZvwjk*KtO
zqg)w&X>q?Cq2yk#ctcm__RR5Llw1);_zT+_Rv!s(!#UmM6ZW!5xypEdAV-lQn;~J<
z>wT!I`>#Ipe$DeJ2z#~_X94<!^b7%?HECsUWX397L<m9F!~{IO@)aHk7v!mT`EsaM
z*6v#!4F!wd*tqe-)8)}AcL73+VIB-HY9yJD58h#<bg}hg9)gsGvd>`rouGV3P9S^s
z35%D;e>oL8unOhHp6h*siSZ)egiMN^`7hQy?(DRqQF*+A<MiQ)DeP7CqR)co)OuF4
z%nc2_N}lN+>%`XR&_99LcoA7mR+$?Sbw_S02go?>sVO?Z`NP<?-?YKTe8UkfgJW^i
zzMBVp*Y&x`h6g{~Uq$5oD%A_l_XgK&`Q61{s)U2F^EOBIGR|$Wt&s^X?>{d3ovL?Y
zi>LH=QCB{4y5Co<T9|UL`@C+s5;0yZab_Fx#HYS!XOvZFA8;qwcs~`}n{zDLXF-UR
z2p1*a3B{PY22F$%eUKj~1~3PUjC!=}+B)XLbUWdt<6+)q3(vXAO3!bZv3|Qpk#d2l
zHJgRn4hdXyx%mt<qL9W&Jq-~--x@@pR_i;u4&000=w3KTIu@x$Z_WyV-m@<&&pHiq
zkCxn5?BN`;sV83=AIoH#v-<KXdCpiW%ZvoZea}Dhi5B(moniM$YM=Wmk*Tb|mrnv~
zruq$fXJ4(y;khDvpG20@{0Lto>+Cl2jsEKjFNMDcf`bAT{_9Po7pEi|lVS8%FLhZA
z2yT(|p0=vFufU5lvSb=M7A1!Yc!LW8dV{<zg29~?gH!Am?8#c8ezyq;Bzz>7%T|Gt
zo~W(fY9!C$-s<t-jCx^n<&#|1PpaPpi(&`K88Sj+W<!bk&0T-x-{l<;XcA+h|EL1)
zeX?ANu*0mfzV|g~jW$LBG$(%cJE%-4oZ|5Mx2SVYTAJ7hXsDAMrf~r9a=yMZ@+Q(x
zk%Q)R)sg7gAFyWZ@HlS1fzrU65T_!aRL+k5G#vf96f8H1epLfvF*ZfTCPV|6AE<ht
zc+;$02c@;u_SXIk^jeyTd5ciI8lA2brAN`*sEsI|G)rpVYrmp;!sy<~TR+CM&K90=
zQ>L8i;pH?bRHZl)4io9~R3?8pgC%S@5TXg?O=8TO;T#9KYg<Tr%1Bv`o{vJ%l;N{V
zfSePdBiPpF@)$3vHdfJ{Dtw(bv+@_x$C}h62Y<QHWNW!{Z*=F?Wj{l&ipHFM7m|vD
zzd;q~)b-iQ2j}E0Hq#e4bE6M+#V<29Xg&Pe!k}tG;#pd1K-tmB{-Ebpa`~bdctF6&
zRUS+M-+)t7^0>!5N1JB7N*7bXIW9Dtch=5s%I8S&deBtz9Hv;Jz;U|~=f}34rfgO8
zZ#|60DMJ0!MZZAs1YJF9^JZ(}Xr<cr$?8T3VGk+77x_;EC@2Q*oQi<jzI9%2nJ%Ia
zJ!3B)`?Xx3!{EUxb@Jx>CNjyu`Azj)@1}iSgIW8evvCF(#<btDnVkpevk5Qvb?v>K
z)Lsi~EbDnX>2<X!$xW&r*{3yNuJWtECMcehmSEHxVE*IuUlhvs)gRbEnin<#qts9g
z=>nsk+2i8QM!8j!#2zbkEp=dY7PL%Dw%vGX_e;#m+z?W5jO6SfVxcGh#%uxz@x;$J
zX?;2bCLm*SFmF}hGZ}?dB~mcItU8vi>_yaNHt_>k>O@w$b4*~GrVM=$5jHfxFEh~C
zej{8g&QuYfTx^NYMc?B*cSvMvK=R<7khRX{>skf0Pv$pY-Ybr|%ovhoRe@AM7a|}{
z7S5L9o>aXU5U`Oz7X39B5^kw6%vr5uV;Lr$rE;n$yC~BA94LmuY<NL6o0KxVd=#S1
zH?w?S>@G~CdjMUYWt-N*QT56_^B}~Z#+gF-e?<M&bscJ6DRQ&S+?qvGZQ9YVNV?Um
zf6VuH%~Ikv?p-4k$G<`3s5*+!PzqfXy%EzW@Unia-6co;HI|g@p&u1)m1EwB5)M}W
ziifJP5zI%z=ZtT4Ky=yVUI~@L$&?uVBP|Hh-#YHwL?bYrYj7*kEpckSMWsoV_9H!H
z&RM}fly$$NxUqi0xFX~)n$;V2N1zv(aC6dRq4kpfP)4kKmey6LaoV41x8}DQ704Zk
z6W){>Gi>jc-XhS3n8?2uHpw76aWqjv2c3XmTH9_uf~+UQcHZmcnrsl3%x_waRp1H>
zDz6~X;U~P=r}fhGiZ3|_Rn}12X?{bmS6rTrc}60O&8>%{)VFVTJnwemNs_%G<<+;l
z^>zd`ef=eZOBgk}m|Du)FRP1=VH9$cmzbKVcQQ&jY2UwgoW=sI!yS6Jo{Jfn(ff*=
zL7)u6PKpg|iDSYW{M@Q?=QkQk`DMbGIMdMh`f^CHp572u-$?5MyGPP+*<XU7#u&bN
zl}Eeg>qmx6aY<4pVYxISX4Jk=|8)Ao^t`#t`1ACs6N&<*DBE~Gwpw}}JkkjXya0eZ
zZHvfA!UHk!`J3zFNcc-nIHXvbHw<q(we&II#N1HhcA!ejM_+Q2y}@neu4w(ES6VbE
zBYoS8eMf$}xfThHpOfpQp>nd4_mf+8x<<BYso15U-r^6|dH3573UXc1_I^-V#2ch3
z1P8sR5<BI3Ym>k^5X)5CEEP`+1082$n!V0y&<md?^ExZVEo1Tt=HmzElT1BxO^6&W
zzJ!x#DN^nS@K9h}FY9}{evbpHxzs+I(`7O1oH=Zo;0}6x_ol^&FUyJm8L{m;8IY25
z4PS*YO{Lc16`<LAorg$Q*nAsp2Cgw;m3{C;FzvfGD#6lj5(1SAaR#eci=YT>AH~ol
zWFsiA@dxI8cvPfe7GPyNWo)4QxPem9v6j2mC>?%n<!}s7M?73!qjC4p8`7e<aOKQQ
ztMcPZ&TIVDnaC@cwIW7sxyS8i3&I|C-Fc0C^?HM<Rc9*82UA%hC@44>4hC-ElB*(+
z<G=4+Rh)_~(g@X89x5hGOV2kDhV7L3rETEX8K!#z3Y5p_rv>~!jyzj9AsQePSv%Gk
z)EsqxzJ}MScj{RCCM~n@<E%Ia{3@x-`ctOLDwXz~Amv2ORk^YPVpd5_>ALG^TJNh?
zWmhPaq<<uwc7N}O3BQ#driK~ysGHX5_#$j8i>QP4xBWGBzfCh)f+3eSdJdwL475H3
z-V+VB&@Pv^H4%GY_)A<@aM1EIk77RXTQRo0#BJ}@swlM*ecjiwuh}ROG1pnSD4MZ|
zbN1EI?gwJoXI`%9#h6E~g^KYTc@CA;Jb<*=&@ZsO*+`kNnY@eYbOLd-XpZqk=tT=r
zPW4-RN}I;GFi=JP^4gYhIC)Apyt_1o#Q|J@j)qEo{({5$p}Pfjtf{`t?-pgC6Nz{e
z&qA0*s!KB7pVnA|y87}VR%&X<?U!X>l&bfGrl<70{?IRG5w`|r#ngixQI|i5>D!As
zajBIR@DIFgDq?T$f>yz#$~yLz)`U6QJ0qedSKCiwE4t}*47L!lz6L{kxyTwBu!vf^
zpF(24!(ruU3SFwBN!5ae3bBl+^<V~#&|ugqHnFEy&R~j;x@*<KeMyXT?d+>{_yHE)
zR$!g$d?)g4lH&U8-CiAg@0XO@UWx&xJrB#t4T2T2tOz}AwWX6T(|=h02%7lIA_WRH
zs=f(P%i7$N7Y>kPTrxU`eq7Njsg+iqCisi?qoTqYd`x`jOewbk#3--5`7<^J$f)l4
z=FWA5Y*Ob6_>Yv*tmdbnIJTPESN&Xm>vHnmJ>;ZcTy%eSrqz27y1{as{#4b0G7zDJ
zcCmS|5@Y?nei4uHjtEyMuM55B*E|dTIC%XcJ8gLuj<VEPSJJ9z!k*Z~<=F4qBqY}2
z{WC9SBQtGomGVz$+n>)>0!VRnP7o}Wk{?FPEReDFDohMnaGl*!K$oWVilD`PuU7zy
zR^u);lJK5tzrsRkdvS|~w2x=$SVd}L8yZR^$BIr4AV9$u{if0dos%R72{=3D{Z%F+
zK4Q|D%1Ux_EtHVwY+AZ|!;gkPQU=IKQq>Pjau58taDl^3Z!61wPF3n3;d49$-hPz*
z`B|2M)sLTh9V2;nhKwnZZ**_1puy8HS<cBw$lhOl?hW}JogipxF4x5aN~&~3fiohG
zddV6K^e6l_;;*FTM596!CPX8}H=whfK_lCKS3v_Wb{y3>O-ibxia0QD*6VD?foTZB
zxmj>mvjn!fULdnq+(-Z@eu06~2MKde=y{FypnJW1ac0~UYLah#zW{6+=0BXb^U=VO
z%r{`un5rBy*<X1G<J7q+R<?-U!;Z&Dzq;G)vNVYYhSwx|UG8g8pB9w)_jw2Q!X_+3
zW3;@Du9%PW=}vDV^!C%|rZIX>vz5)cm5(Cw64l-VD^<VjNkC@1`-j(kWOR=yN$X^y
zqdsc|Ax9LjnvboV5idE3L2?D<KmMzlH5V`@>~^#2)s$y}AFkV;0)FD-{3!C7wV;EC
zC=p4LiL-k?(n+s&xA!2@k&4R16?Kx7@kXu)F2Vs=OT?QA(-*LBrr@%!G3m|=2@0$+
z;~iKu_cUz!`Q@H1{g+>-^WeTO6GG~hOIuu=pVQw|r^9-bEcqsB`qe?~Mlv;=|GmcP
zUSC_)4QH0MHa$70B7}hNiUKcRA}A<;`sBuq6nR1<|A(}ut>EEcbl%DL^T&v+#vCln
z+sG*}4DaKv5k&AY^<C8Fs)v<21sNI3_fs?pE8W*6l%@2DLahus{OKg;PL<Iy9hz+$
zx-m4*qalQ78X8kJqwqY;n46ELdT_`&XxGI}wY-K9kg=n)6b>2brsNg92UV*s$K@CS
zHiCZuh4^xv%SxZ!I}ewmTBzVXY^_3R;Ja!N?aQQT96GVsSl>04n5dro?jS`UG`iRe
z)wWD6mC0QcT1xc84>boV;;cnn-YN*{(YBy*7r9hl==`)yxKO|!>z*scLnuUUg@;#{
zDhYPFimN9q=hyh(-_JSl#guZh7vJH%WTLA5!z+MeCa)<XsCZs%&z@=8&R=t2X=-2V
z_OfKZf;z58JGV1_Y`1BE-CJR~-zVA%<7UHypVBm>SV^CYfoqyPBn5s{bbB~l$v=BB
zIGet06%i<;DU;ADHGm8cipMxIA~ubZmER1RtZ#!1l8ZkMLZ`GDH_+azP=>B7fU!j3
zdRmG~buwI(jU=3qB%qC+J>?VVC$)n}Q|?HtwNO$#KnC^gQB!y`>~0R{xuNzQO|OId
zTFYDpGhjF2GXC$b-36n&F0=+W^i)$%Pph`e@@s@kmcPv9mpXGY-y=-m*?4bx(d6p~
zJ*pzg3upAF1ZG@sIt#84>Onn{3+$=lt;rd&R5Ld|^>QtO=%qJ9)%=h0>jOV&9~9M|
zpUUoueJIa)g7IJZZu(`3S%(@G=G*u)4kv5Q2(tBY9yHty_#&FL;nfTViX3f|MRcOJ
ze^lb%FKdgO49fZ+;iX~Ol;@N~tV()i8HaX_UDyw2ds>*hxb|zW!PN<pZds#BDpmYt
zSGFIrAZ6(yf$cirv~xU;XN7pW>mJ4X_^ySFjHvL|E_E`u6f>?gaFti8W?zs$Abz$(
zq0skD1{<@TiSfj9jaWY|_v{-HmIeVb;}=va3u(XZ7EjwQDu^|#(SCtl85x^e(w_pl
zyJ&3#n5e%7EKA~h%~Wd>EY9ZoR;y~kXUb`lTOH8k(i110^X3a1$pO<!9k9~J_v@P#
zgN5IKEv^8Ku`z_3*ylpb&U{h<N>T6mH-S?;Pz0%WFh<ax>ZpwlaqXDZwRsFkcib${
z_Gg}6?$P;t1NZyBf>0a8h9Gs{RzY6ixc1)_rJ4B;s?S%{=zKB3ODOu?W!yx)=jDQ^
zhkfREH6)qZ*4H^{ZX(b5v7}9A%QI07ABsGK6Zt}A1#kP#iPuTAvAa*e9;I_#ez~@%
zCC*~cLkBieLJ?;XBQ8951AZ_jA4-eL=t+&*R2iX~R3e|UWS{BkcPPf%wz}#!|HGiD
zz`}r)1ex;KH2AKRyr)w+q}~4p?F{jxWtxuvUj}@l0y%BQA?s^K5|Tf^&{Odr2oCY-
z;JxE(&b)OC4>OG;`mt6p@^oiSV*7XegqEP;;^(VB*Spn=?cA1jX_3vT>`^?$%Wa}f
zH;;18Hs}AC{ZhRUUUeJ-+$|7OG{>KQx&4v*U+_a{K3CL)ob$5npmVE^O`yy?bNu{S
z!=h&!W&{b}lyI3%boqNK(XFT9q#}--0YOTKST8LWJYk3WU|Cd^pN?rlc_e2?(mzZ-
zHyj6N^OYfpcd}K6lPJ{ujTjE!ZcN(dqfh<<i37NEOnqfA54g7a)D<!=;#@*#m$5SC
zpf^v39cZt7mavPz#zVKGx;)t%F8m<h1}657qb=aXnn`Lp+t1$l+Q^V&z55im64mxj
z!<%CQ>e`>w%RH9!<GaeI>_zi^TW8hSO1jbyZP!Q;sV&>uE?sivGjOPWvH5CLK2Zos
zU^_dw`O$1AT0)mSu7LoFceQ(MH1Xn^&1rqJjM?(@aRtiF-^UH*#_xi(*M71m(jt5D
zrT%34AxFta4C0a_r=H#{7y;A4G@K^Cxg>U|%trT7c%t7+>DAOetNvb`-bx{+cAr12
znel|!P|a1;d|!8ED*{H95DVL_Z&tf;*M1<&C78qB<lHi<#qedgN1l}be4|=vazr<Z
zvLtD*m-Z_7#GImCZZwpa`x<HG^{`)W2Rd72EB}-0cK)@+xzb8jw>pdfj0DdhS<_Pb
z@VtI~`}O@RcKxlKoHje=p?Ms3M^wmFlA?9s5gX$F+Uq>v=c0VqWl)J2&Ey+%n6IKg
zXmmZ|ZuU`b4YzY9t8`e51*Tewt|{ZQc>X)Tq}{ehpVCWpLehTvbF=u9;ya07Q}&)Y
zwTB_n<fryZ^{`p?lA&~`ir&2sc0v7rCzAHm?wlZ%K)tDJmTikS*3%pbPg><t$#A%+
z#^*`bwof+uHGhy;30XH2;B=i!&~<ktYAL}{TodkR5;27X(a{;T-;>M=jis0(&N!L^
z!{OCe9te=zoQm7rwJR&IhTveUmh-7)NfEKYQlnI`>~dqPzHG52f=Js3*9bm7w{zRI
zbL&baGSP#b24+`-cuC#_*q>PT^K1WD_VXjo`A3t9F6f#biPD$OkmvU-17)vhZee@S
z(1PrIQaw*z2*tmXdV`?k9kBt)p8@_f-S=hf6XMPNy1DJddD=MK$04)LkpVK;%yzHy
zbnitTTB&tk<j2~p6%9I{E;B)9O)9-%%s>>4b06<$2#QReNEyV;nc<3w{QX^n%ifHt
zQNC%p<U>Edc#y%|Vzoo;(2E4EXD-U~eI2P4=0=;3wTv;7S8j+CC#>W<e+%rfi#!9b
z+go4f>5qYP`F?_P20HJLt0}O1WrzH`dq$k4ma9iVvTLtg!zQMVl<&wK>!T?oqWPjx
zllD%HO5X);F;nDwYR4IF5PmSDV+?gijVlP}+YAfGvmWMYB};J4b9~!?90DJnQ?84T
z5BB-y9_H7P@cMX-f$4pJWxq|{C>X!Q?6<jqy7AJ?JS8XkU&p#Bckm_Bo=v6~o@{j?
z6M(HC=~`%m9VPpCy|$wCczzR&oiQoDp4n!ZxhzTM|I&#&^h+||b%2s#I5WRlFCybD
z7bVpX6~AW255|^7HA$-_KOq1}VDYIf1IQeL0Cny6vaUu)T9y)IXMyxwm*w4k9_LBU
zNY5`nziSeCF%`P=rm;9D$>Nl;G*Q_|6mWCWqrvxRY*B63$NTlzT{axUAz`{N*f7B;
zJMVZF#v9#Dz^3b=I16$<u@Gi^u&DRqbr>{ZCv5ge%>SlI95#BT)^>N4zYkH>Rkr|~
z452A-GK#3mr?X`j8dC4^kd&yLpeHn_13y+IyRvSAY;3DjtG3Oyxj)2n-pl^N{}=H*
zo_Z76ablOLseUMs8_<39A8&vUCD@EhqrOIjr|rv@+Gl<n|Dov}Pd+hys=iw?c6CzY
zZMz)x&nqXChGe^v)H0OB;YMZX`uvjFe^S?L)qs-QF&eib?4E$oMvbmrNSh%E$hSU2
zfJ#Eq+!o_1e<-5MY58`Z-Oy9o;)9zIR={qOX|C?QkcR_|1Ot%i<OCaMaqMwdA;I4*
zeUYI7GGu5CLJ41)xgwPAlF8c!^)2wqF$_ysll@0f6ZNH-JnzUJnRhK$4;4EO8xV$h
zoeTBXD1utrpZq~B6XIQWZnFGjZ{;Sf^s_kK#&3U1E5)9Dl!kbM_MDuIKf?dD^f;tD
z<gh<a7RmFje_kXTDLg)99KDQ6v1jEBM*~JNiyPSgrjfNvkA#gXeWzb}B)sQncuz{)
zT^5aq^lzAMqZ*W#k2mfRY@%E6S6M{cZ|)0JMSOcwu=2bh&k-%u-=Rh-xAnE#z#DVx
zM;|e<36L>qfGX0mHyF|`A@dYim7+;DXAlEVnu*8(Pc9mI6_KT*ocbS+@-D{bGMsgp
zr;1~t+}<dZBu#Kz_YwjU`-6bsU2chA$Edsvs$v<sQ*WK*O`s|k(IO?eQvLP(w0}|i
z+dMqCJ}$JhVB{%+zdb641Be5LOc-r<uOs^KUba1U%}RkfL?=KH_m$56^ElLk;&R{N
z;o41kgQUq47F{lLq&mQTp<Z1cn8l<M3tv%a`B<l2<T;aBW4J7Xl8RM^RnE&5GLnPE
z<{u9hTo9h>I3r++kJCTD<tm#fKdqZ1Vufksj+*!hSZGg3S-b&JmL_~V`I_}?A{xHR
zgi7T%CWAq8Egn?Jcq)r9&J{cM7?AW$w_sYk;<FZ&JEu;)*j;?j43Jt==D)@N{{KV#
zKSPBtp2`gv0i<~rJlIapi85K|UwH;qNy|QH<IG$7`b#1sEAm*rIN$|szL4BS#7jO;
ziX$XQ^)U(Yk|$G}2&&#aRBweQiS2tB^6JuM;7Dn_VWWTV?KK2iI0{&@b2C)zpk`RV
zn2l$Fgnvrpk}2&GmSqJat;!rRXMYTG+zqWzB1sSLzx@u+nUmI0s8vMoW1UC#@DFxt
zNWOk*SHQZAnFLNMB_|%t#=5lLMO<LS46?1+O3k@~ljW*bEL|b*pDTQJ!pN%$9qeG|
zrNjmi=$0bOBWRbpu!PErx=h}(UDd$fY_3NUoS;9++AjZ=1mYE<x6`Yk(nKC5eGnEH
z%p%e7>U2@whbYGyd=Ywng|5Q~Y%#D>BX!rYFzA?c^ach`CV0HiH*A0%&r#Bo*Q{$7
z`hu(?RzX4}|16|pz*QbgPU9-x(gEH~sT;$bC7WjupU{2(DYZ{=xE+^DUl6m9Nry#V
zBymchIiaYyS97$d^fuXrB!{Uf`sv9Uyz!r0gY<JG_ZP*l%{817YI+jsY3unBEXu8L
ze=uVLylk**r@3cB1`vF($;Z~f3sQYsD}QG_!in&UPOOJn$YBxs(M<GRKvJnDAtr1W
z={gJ%->|JFNvTv?pTSK+>m(bLjNFN#|1)R@0+bL8&_US(8u!@EbvUv2@@i?JU1>L8
z2$NEDwsr5iur_l!N~)+UOU2;h-`gsl*pGyEY^f0@q&4*Nw8#dsp^#^*6Gxt{ThqUL
zwtrF<)*OeDiff&RgAyk+!kGD+n1O3)!mO&PXLmn2=Skb8pf@EVb{3W_uIdfUgQt-m
z!q^}WN1?|MRN*3>gqn2iaV$+h&`#auWi}8V-F*#@HQSp|6AIUh;7OI8%HI5)<W8nS
z07FAJL?Y(UfpkzYK@nkv_*12AdN@G7L3MPwDIX&QHORO^BCQh|3_y9Tx)e&$9tZm3
zWvYD*!<?CWtKySaTJTBipPtRY!8i>GZGwBrwv2TbsOdklk(%LH=s(pAztwfZ1j3EP
z{_{cH!Buejn=2e{!JD2F;nU4GT>*y`ZguP3wlD`EKB|LxX4`v8Pta|*dG_K~Ohu{d
zxr#hHWV5>oK#D2u=nGy&>=qbA@Vt6H`H{y`y_Im2kF|9y#8$j<y-2tBs~lur!<tcO
zgA*;eqR>drY!V{WYlDK?+S(k-A=C;ExpZ=}i;GM!NyY#lXwxg>KS2!E5_@SKAr%=0
z%9@rKpZucakGIb*sFCz_gfneA5}xY*FH(JLv>~{DbBy)2*(o+==U%{}^^VEk?+|r?
z@mm>PRHj8-{EJZk)LxYo0<WJ-qFU{m`O|+lrsOz1NVOnSI;roCKRHMKWa>^K-Y`<R
z`>@Z{^aT_{D*E-RUqzh>;uje1<^<Ky*ohv4^Wez=4TuXuC6sEVAqPAiWTf2W`To_5
zAYIa%BcmGB&y25`fXZPVbB+Yc_VKOl*JS@>0feG{Pl^4T`Ae_Y;N`;Ot=!r1$0bX=
zbm4!nY<`DeMG<d-*9ki<FCeY4Bi4f(^hVj88*bwrTlu&4Y2R9jZn1!QMMaNJoJE+k
z()ZV>;&{HQ8g|7zhpUKwVy|7~aS%FBth+Ee&9MOzTDB%ByA<WVa9ih<J~L*Y(UsY}
z6b}RmB^gRHUaj^RE%^Nb@|z<oIR}`Qkigo1LIS_`U=?#p+`1Mays7_C%`fuPJ8}(%
zIw9NTiEAjWaathMI?v1MZ{Uq6KI&Sx$ul76Jf?1L#=O*>X({NG2=$eq@p%RAXhS06
zY^@D6j7Ji$`&6&7-r>PD^HSG4AyhXbJ6@sQ#tiEhvmhc%dTTYz%%gJ;t1g##v+uTk
zD@eT*`NEFAm)Q5$N3HXHYqZKWK!!{>fT~tM+1=BdSx*L&_S!Jv#8U*AFGQ}`Gt?wq
zNAR$>Xn}E*;jz7vb<K$ak|Ti@X!b55O1VSUO;L;RR%Dl|@1X*~OFjF~z=ty;iK^q(
zGs3M2@IsB3lygnF&(i#4w{@CRkqDC>??)WGNqev!OvcDuiX1<s&qSXJAK8v|7qEY%
z9*ME@J_hAt;i*|W9@@6~am4;&gM0Sz)$!#W!<~?yhAL*j6nUwF#}rXN4zG$E${2KA
zI*_3$0JCqX!D??}j3Y!1!++|TE_ITY`5z|-cGiDtJO8zp$#HHV+uj)7VE~t}1h|}i
zSB!EaV;ulMuISa!KPk6x!V-wnTlkL<=l9bJQbN&lV$XkSQ}}&Mr4gzkY0dw=t^c9D
z_W$s0&0r)3*G6nCRzOk+b~Cnpe$Ah#>tZg{8<EbCRWQLNW&d2+{norI@Id4%zUdE~
zg@~;sc^&Ad4!EFpm%NK^mE+W>z7KY!EnS(e8y$)Ii2pyeyLcTaYE^6>f&g1&ra?W4
z;)UoEZh>XFNQr%)mL&`MG}Lsbh%!CLFBfQs46-3(rAXi!iPbW-;4%3^GP>H5qe=EI
ze%BD>^jEMs^{9q<{pjgm++2~`KA;bTe>)DQC58K`5cDB$1=sto<sY%w|NiC8l{*Zg
zrVZ#)5;&sfO)t8^$d`K><o!vd7ju6ZUcvCaPt5U^RTAtyg=+=$R8U@T5tw(8$1ZP$
z{<)1bv6HGNOSIO^F65oNijr`Bo`U8#Oo-J%5|9wQ@Xk@8&)>n}hEPi)4I_;_?*ac6
zRtQQHaUPH(Q{$o?!(Y8b{2%Y3)1Lx|TCVnI*ncDHarh)1B^`BZ51>y_0{tNvG%m8w
znEi7p^}pU{<OJydi{8FSCO06=&ALP6P9riTiqWi<HJ`1x0B`)wjk3Pd=C4UuGO1or
z1HT-?5XpVeg>F98an0Z?Vk4~I&UX?hM<u6O4IMFQz!Iz<7<8dSqx<UO41adKB)kN=
zdhSC32u|8S(f0HE*`F<{h04sFwj6q^^w8`f18XvY%rJBeNZf4cP^0@dhuW`;Z6!cU
z<KjQy(~naA<gkT<A?D2v#Me4(dtqg?<L~99eNDvp3_MEpu-wtNoeMc?ar%=YX0CV+
zp_q<EC$n*>oESU)nhdmQl%4I+MZdZ-!k!KOOk#wwMR%Isnt|Qv-g_}JMo+__^ibR(
z-*T9h@8-VEAsoK@Ewx+IWkSm&Sx9pX0}X7qQQv|#lYlh-OYj53W<}2Mwg0)x1)-~g
z5^>i?r8r^pP@~ODys~^el&*orOYj?KysECC44b&q6<Ai+Khi{@;n+=z1!5rd$!GA$
z7RAiIq?2$aY8#qPeZ=j!5Oe~QM0lVPaADc$QnIQ8G-1--KgN4c?pNU?&YBQE>l;)@
za8~m2FBJrV|5y<GE~Mpl6Kv3a<R9q3ADRP*<6<e<0Sua(v!4~YdD5^QYd6*G(G_}2
zk`ct*faBKN%ZNQ<)>f0TZoES3-^320EC5wBDZn)ibA}pvONf{2Zndx851G|4rX=lX
zApUZf^cdN*ixEdr0j(~dgT6Zoe$@+CTA4hKVbY~E&$zY=9r9sSoR;t=-9O{a|1~zr
zp!EMu<TH3|xx})an7I*>H1X7VLjHF~iMLpu(_Ncq$y-sxWSX6seR0>!#3YfSQ~FP|
z$;>9Hlk4LSiE*v}hJF&CU2YxCpQLUC2n+wtY;2z~ue5F7j4_`f8sAYA3_&>%y+jWE
z_@qp<oXtFt-~WdnnvOq7H&N&_*o1xLt@FQVc@>zSl--)LLL%;m9@vvY135M|VIarj
zqRV`#APF+n`jO~Pn#CQo5q~q!=9JZKLS4tlar&dbl<MDWHf|vWi3Z!=(`BnzMHJXi
zjdHO{x@KtblWW+%wD2zchZderAn`g5p`a!nuO6<kJ+O=CvgKnoa@UN4Vxiz9@y$yE
z#S=vX8G?N$-VGaIu*nqb&Z7aF!s2S#g<$1cL8_^(c}``;H!FtQKEk4;g0mz8ai32z
z6Dk(ku5okKiG<(H{Faz4T#LvDo5qhM4X}4<ci{Lpy90f26`at`h6ln_M?j9K!pq9T
ztdUR*XIc|T&&y{|iqf2U6Ud^p(Nv&NwRWdsvNU6nB7vh0{{Tnj1Q;a1Dl_C_exkv@
zaHgA^>5IT!?`%N=5C26vj@Y-gmzpfYWi}HBDM^HX63($5laC4qFKwRuM&@-U+UMV~
zu0y0ipIa`c9o}G@+hkm8nvbcbbMHwZh+-F^YiYthBMd~~S}2KH`Df*W?|!9e5*&N?
zo>d6ha`ZYY)waVy3byB{kM3K`e1|21n^ykgi!FyQ8o_L(qIYgf*Fk!=+P^nM1nDei
z*3(lsaIlK<93}1dXGN*3$G)VxeJKAj<L3MQTYQoA$z$)N?j_x8D%xc%Ty^h8*XkPl
z5OYv{fU!fl;+ro%rNJEMJ0tQfy{anD$fUsC{tyfQ{bc`kNAbqiu!PGKAS=>Rz2C9^
z<ZD6DvIE0#?K7N^_Hlg)r;8iv_LMt5IenD!{<h%vbxl#^H&@~}kBReR0B-$uimCK(
zFH$@H#zZ^*gRiQHk}l`X<BQ*yUtxEJQaB&gmHkSy#Hv(~m-CPPuu2;H51mVe=IR;_
zty^oPKgyAL+kC^{F|1%gnn-qJQ&zQ#0?NF=OqW?~b9wx!Uh=aTL>ueRijQ=nc&%US
z#DO1okEo4fc+r6eJ0-0&F?%>jhUL^dO?g)(4Ae`bA!ApdeLHh9`o6nM-1`d~Q_0Mx
z8|Z|;Ds!=nG+E2i6&)`2%_y7ua$|_d%8!M-Vw<+B9qlRLTfj)K%>Lw!404>6nAZEw
z&_{<2Q);iu*;4v|^-9x}dcWILVoatSEzgt2HV2rxc*ZWSVoh5*IcKvmu3M#Hh59u6
z1Y*t%NbF?gA}+%a%KzyRxV*Drvz#0e+V6D>D^Wk+?Aw>WK2~A~)hy-xypTQx{BV36
z+kq)sF{dnD>yt76mBB}N=4nd65T+y}o9QbWnfhS0UqFof<I}5=rwQrMqcIBEYTk|(
zPRjf~4oE6M^KPfEi@!|y@ghu%;G^@nlVx^O_hv=G-f!edIyU~BCy56J;nNMs%)SNY
z>F33A9Apx<l_*d?H5f@(bW{Urvzbw;XQf*{N6$?4nO^3QwNU*a4PC7sJU6#1_iofH
z2Jf(y^_|*<g{bS@3G{NDo>X3j%=d2LN6!VJTr#G1_Q_C$Bu2k;&zX%TNYON~Vgh0i
zAJ+fhqG(9Hz~|rGa8Jg-o5@(#=IMG5WP|qX$3x=@c5P5PcPiui7{+_3KqA@ja;J2(
zng?xM%5VoQ9^>8Z|K>5`Iw(B>0GLXa!7N~acHgH`eT-|tm`qKy!JPbyNwcy7Lbb^X
z5GOe#Pi*-3E|KJUL~{L=gz1YPDp&3l99{@o^^Nb5^c2Z%dy%HN(v$f!9^>%RW0d<h
zkCF4*kNo&dHJe>AYr8DlhO0lI2Xz|;80lHi!#7q6mvOR$sIwNS`^E9yVGF<y$F3+}
zGKTtuW*9!?5T=`b&XdW76+J^s?$lFBOp|cD0Mv5}?d*G#g;H*?DhZ4+zfvb67`%i@
z#(KmOR528#)&afh`#m)GB?tWe`f>O=EC~A)?;$07DEbCe+$T9$?d*Y|$FH4VQ#fH|
z0~&T5=SAtn8rdJtB_P-CJzYgfJ#=B&hZ{#rneYQI%q(T=qO6Il=!5ahhg3u85nu*%
zaF^@#>?z*C`lZ>Nd@G4SbNq^~<{g^Pd<DuEbnGFI_$j*4U*@iN!q&`Xcx2~4;n}S=
zZTo*nyYg_VyRI!Mq>x0)Jj)b9)Ilj6b0kBiG8D-?9wHSgWgZhz2N^PF7L8`gP?RAu
z&z+7b)3>&Ur{}xg_xtDj=e@e#r_TAE{o8x(weEG__Y$p<(EHI`4};hVYS&gS`99-&
zb{xhKQmQz#LL_+9oCo}^38ac`oViOJl7Qr{!WN)ZeZ|0?r-7V$Runqr=5el(D%V|x
zSl2+YvaIL<KM<xfwyFj$L>v}R1aglp!1yDy3m(wA^QDbgCVvYcnzjivrrnQ9!1{uM
z8qT8T@0WOTW#dwP=jw+2gqi|11%5Nw2s7d)G{r;EnPk^4P*~zQ^PFq}#MO%RA05PZ
zcj*$m6z+_n!#2Zt*~^d)cKtZo^=l(Tw@65rVFKjg>Q9lMARbSk^nIFUz!#GBOh7-6
zg`Ec%khL%OA%_s8>=TP5K62?gA0)6$EAWY1@fNrOE@wh!=iiLZ+I_f)Wr^<;w@&tM
zlfE<t+6jG#vCh(;m-lEMIgXrGGoekro@Wf`UrF!V|M2P{*7e0Pi}S%~^VdmK7#BVc
z87|<?kP<pWT^WTe{69iE%H~%AH}N;NYf*-Rhpzr`<uQwj7hogx2IMmhu~l_>NWHVk
zBvSW&vBKJknZ~?^(EJ01=4u%k&T9(3F!ruFcL6U2D~t1o?U#kbNnLk=;+YP#IXV?E
z9y@~$IBGw4&jUd*v>Xm%6dBnSn8wKB^m_ODiZu2e_!~>QPZ{vMvfe8VyhQJbCyzsP
zl-aHp;b7%1@LmOIdsV#NahR~1eh+NXq|GhL^m~i`=ili=3X=A3wS;wu)uc!3J}RJb
zR9_|^D_lP;R7Z6il20jXO%O+FU-?v&|2g&ZD=~|=T{E|v10k!)1h@L-VALvx_#3ew
za<=;bt47o?%4inZfGN=gcpaW5Tu{26G_(YPW}&(!q8}L62Pws6ri<D;Ajuf1PNrm2
zHcjiuHf+y{VU(Zzm|N0ltz^0#tF2dL17*D}v|%P-3+sUvWo}X*6cc*jVS}bkFIg1p
z!xZ69IlgOyyQT!{m0H4~(W2a(j*ljgs+yX{X9&6d0O?}%u^u_#CXXCi&n5P?14>Y#
z`TV+sMRV*Kz6xDSrMKm<oklysJ*oq&Xi`ur0H<BPv%pGiqZ&};%)#fLkuE38hhx60
zJui(kTjtbi;mMD*FH4;P$onjec+K2+`Lr~b@N(Efrff4ysQ<t2mV$E%fh|Urp7{}`
zenY+Qkq47ssj#lCmjnAO)ewueO&@s=BmwI|=Lo)>AGoWwVQJq3T}t;to_4lH9_tn|
zaY2;itp7fl|M`~Wp=tt)vyD#|VZ!q1JrBKS%6zXWcHFZu5cnF`$f!d{zdH%9oC3q6
z1SuyDMUpkCur~t(t`qX`vD}!IIk5JdA|FFVi7!bDAUn+>zO;b{mM%!y^_0G6Q#QHo
zXZd5&Y-57l0t|t**5E$ojgPQU)vwi|rmUnSpH|UQnNf>HW;yj@(Dt(k<&Z$EY#_S2
zm?w6u6f}x_{~jh4Q4SY}@fx&j7A8fyk3Z-3h8Cb6w7y7J8c!U<Ao59DtK#*@JB7%8
z5cvKk;l~CI#Q=J78;;ihamDb1EVeiGirRh4DI6tHK*PIHY#4V0RQ7hCRTR#Gf`XJ6
zj*}kP>=f97WmfV&f@Bg|G<Yzm)|W0=M-03iS`GV>bjhp}ThdE7p*8U~%6AJKU8zwx
z!v~7-W-w%(m=MVYCn=U>@%^f;mhBA`>#jlc64dUgIGEuEkMzBSz|)~F7k5t*1<whP
z9N&EwKTp9e1VQ}eO=k9lZLUKykqqyy(9g>|!q`iHJKA>^e#%wKWs4%cD=JFCKcMOD
z1*QO0QM%-@^yhecd-F6l059{FrL7eN*CrE2tu<oYs4tGCXJFi^?TW9gd7Z@Pp(C=M
zukG7S4thhKY|PHa(q&PSz|<RjH@(7|8sF}Zu`XQe^gkq}0sY8INYlEnl=JPY&}euM
zEyjxGM8G$U*b`arH1--FJix*|#HL1G;R!}JMAufCBx>hxt~4tC8}87(R7jHLZ)i8m
zUS8(a;bR#Z9{mn4n>++r*^?7KzQCs1&jL<_lReZRpR<0UrTvBe2OIEmN}?Xi_X4(T
zcce>G;zEy$Rz)wSdZzSiNX>GVw-MoWtHxvRL}!{eKKr*375$Qhd)&j_4-N_VB$Z1I
zzP+g+yRGMYYogXgraCIfiR!GjEcT7Jj5;UsAN{%h5Jp2u-&AEeZ-iLLEji>?W;N7i
zaa2suPu_3L(b3sY*1*gDD*BMk77<s?t?&V7w@|v@b_;d>2$~M=uX<rMyGzt@2I&_X
zx&2uqo&$wIDv``qr6!0~O3*R;lD`^mNi<;uR2>Diha_;NyjqKSgXvCNy)9S3O+u?8
zH;Ti^JQN@4(<>XW)f+J}auE}%8MD~KRk4j%ce_joS<J}<)g4peR|lWHuOeL@e%~vv
z+}L229wbs|72`b4YICeNGQ3*Y(Cx=iq!y{y7{}*3QG9)}^&#c8oLV#=N7!^51V^!+
zZ6=qzfxc3(JVCLhFo%kh$_*Hd=6fyrbB=hfx7;@b@n|g$ePuiA`|5?Q{hYxOhZtQx
z?Vd(DQt#_qF3or~H%4?MMl#EkU-Mos`~F_HkO3A{fnDfJ-_hP>N7)#dGJ*aN_#g4j
zJ-s-4WQ206ek$I`v#hp!WQW%E{iBZ|0}nqte4@3sJwcWujExj!zq@p9x#dR8w;k1x
zQn#lKCO+r#XLz3m;9MtIjI14_tPGP4Oh<Lr1dDLqD2x~=v=ITW#2m-!Fi56<*u3<|
z;P&xX<SxD0F@-oa>?SBH)$MVAZ*q@80NFl}+CVVY;nsjr;MVC8c6bj~fk?6sClll#
z)l)bWfp=?6%$9w#r^}*2Unh~<1iEwCUy*f{yD-5U(xK0;Iwck%GHELN$9LCXDS^^_
z`MB}G&GO4#3gvK71V!dO*zP}Q8bjNqXs)$i^LW86-I}W0^6ySFn&Pk`$zQhhc&8sV
zn>$4EXJoJ;FFN+WmZ(5idNz0uY`B#>x-YKNvgh%&H5rd>b9%0eC3DvKf<%zw+}cIn
zAB0PKBKFQ7kg!%~r@a8#<mtwF;C}oV2M~In+Y0DaNo#L2ba7#7_5}m3=~bZR8A0+0
zigD>f^s)GG5|+Z;)wi%lrH<z}*C=rQ6I`5j_hH84L1Qv4+&8E1Mu578A5Nu&TW|!@
z=tUT%{D9M|?hf3!I7sUplaC=0B3|QcB9v;nI*Is2W=c|t5DDuhil}arF7OyHfy}wP
zs34ShVMr5*Qs>QA(B*7j{2gE{YJ6l;NL-uf&s{|^h23DnhUAkqVPsbi2x2BTchGgx
zft7Sa>a+`#zjhPsv37Jq#r<A=JFs(NH;t3U3HE5x{G5t2BlY>aRbo+9XCp#%O>`cD
zsA{$Q5`diSrR<+DU=?6!CF9z*2pmbXRfYY!Zvl0}fhtBYLMw=xWmUF*&g6aC9Ie57
zQg4{Se84fhhGx8M3`{|&nRVLg4{6RsW!JFEi~yM%Gk$yoyl@-~=iFy{#urK30i05h
zvEqL+cLwS^PrH}3(umRFn~Wx_%yWnF9pibdiIy=#Q@rE8O#09wApSIrGI@4^UJ@%e
zy_YCsr!&%%c1$@i;GZ^$4M7SvE19sQ@mRr0RE=VaBaiQ1I;UYY`<Fh^-eo#>3kejh
z6i(Jud6a)A;edN%^YWmuj8$-&&FrcE8b`Bzk+v(+WYkG6hWCSHY^YnRW&q!upN`el
z2*y?XhS-tJEyqq98r&u!wPAQy75IcJ<j9HfzKf4WfFs5z*!xwb<4U}N7l;b7kO{i=
z8qnUJz;FCCjLeK+PM3D)({uXcNd{5tL0%QjO=GNv9Zt0@7a5|dSG8sL)rmCTrG9Wa
zh{Zk52#B@jYmmui;ufYd@XX|_p62EWhVQnVnt;k(I#_kF>oFQJ7~6c8wOAEoSWQ9)
zS!nTS$Z+iAX2i|5h0NbK0a!whROncQ9>HzRVjvG4c!f2h_UQ>20kaD)M)j>Lu4pEA
z=8DohI5XsA8a>Nw!_Z~ipq8!636|82kwyG8RC6zcl?U;pZViSy4z{bN>^S!ZbJ*3P
zjY5#^4Q!OxI*z&}hfIK#uSsgqHyyxMiyO=Jxi**&3vypAvF~kqwAKqyM;k`G+p2dZ
z0t^_#Ossh^r258A0sF-+oa|^CTIU#5OwoiwG1yP^?peRJf|KI|LyxXpU?COQ4Mhr1
zHZ%PevK=Z=er=JI(mVAtNculu;GZ$k)6^Gw6EJF`cgJ7Gr^!#^OaF?D4B<H7FDilr
z0ymrH#R<sxsVbp5@y%<14uXx~!HwiBZ{E|meIk0G_Nb6+mng4gmN$64YT*rR^G;x4
zU&(i!6;LJRhC7m4e(9~fqcxz!OcZt~GSq|Vy@eY%UuS^=@u>HP+_~pIUY!8hO9GiJ
zHOP90oz^MyRyZXc0QWpa_Jpb4BFJYQ_#C(lV#sLh3wrkVu;H($`w|4VsJ2m;FLGDl
zpa9$);n4Wrnl3{ZpdHoBc8(zCN6-(l4lcE7WO(-3zM^}0t_h5b9*4+`?5i-Mkx%1}
z19d4&y{rVp?QkFW{;M95P4W(uQ~As`2?8@#l>Qi)4?lg9`!OAqKL^8ak)b1y57m0`
z@Oen3e#X);NmA(^#~T~xFMgosSKfhmPaFMY&iGP#33B*n4KI`)*EvxEY?m^4B6Q^e
z2M0_?*EU~SnD%84CTi<K6;RC=zaNxGc_;G=QFh-6v>&@YNNpgROBMr^a%Cu2uuur9
zzB%ZJ^=DnE!C6BG))v@W6F{$X)Bo%>-l>eM>4vIsy4;d{LFG^@jQGev<VqgQ0Ls8K
zwI0f$(s81HW4&bJWA4M|qrLu^3$U?VcrZNiSW=#KkbmBP^Q8Lr2Z4Bh3@X~%UO99-
ztA-)7=_zMVrR$CG?;P~iS~cvoq0o=v<+BD{pjVyPwdcQ9!@q+BtQrm#kh#)jsrm@p
zlV-%>w2nC|lK8JOdII}c50>M~DRAygW%XNi;(6s<AEs<l?s^&|*oB{8NkAAATX_x<
z%J*96Y9Jqz<|GXL&doF~E2u_KyLJ0cZQyt;h6=vm70-^@DPf(aPQFk#r;nVt-h2yU
zPLRBOHopo5#QL36Xo@rF!^58{Iw?kDIVtlpQRhMQ)%|)oYtFn{m7%wy{+RsFw`-FA
zJ3RW&$oDou$c=GT^}<!8WD6%N%igtn4~8EgrJxV20fBEOI3P~w7lTc>YM`4e40l^2
zIY21{R%=((Z+Vh>Ti(h77JL@8y2wJ1sqzoSC}Feb$MV);D&Kq%%+0adH5G25=5W&b
zzPQyLS*SCc9^s%P2|4i7se&d;@w`z~l>#}sWH_9oo<YQT-&pfo`@2otS?HIBE98N?
zqJ_b&EAPB&plh%s)<N6eF?wgcl=F~E!h;%CweUR>WDBqDc}czhJ5>57siIZMMOg7X
zZwYY+Z0BzyC$wg7E87ELvTsumfG{>YD{tah3X%r$C|Y0@G@W!I;{hwn?^^=^07-`<
zwpcR&Q=9qRz+p>(^0ipIp1w&c^8EJu2Dh6wbc_xsbO3`e0~}q=W1E}c%Kt*cVI2cN
ziI*GX0&%B!ws655K0@r}&dgyNDr;#aYe>K7TsdEbmqii@Uq{T0Ll1xJs>@HE0toP|
z_-cH-DNK&<1YvsNzNIPznE~981=mA4zC(byi%T)p#PP*8|6|Q3ux=44&8QI@kp(Tu
z?qsoww)l7$=hZI}Zn)RpjMV$Cy8S8xfR>VDP)FLI;og=ilrvfg$JH1Ed7MC9v?3?1
z2~=8RBH1X)4(|OTw?CJD{GZ)|YNkg3Nbi+Audz-6i)f=JLR^)}$e@Iv5LyEQt4`FI
zeg>*=g%mtvJjhbFeF$K*Zi-8R`H@E|WSQeQ4&cd?z!R-=w*FLL!VWD+`1KI*N%xu~
z5-(Qu9BSor@bBNwaK5@(?|6PdfH{iGAB#p<)(<TqbTrm#jE_?M)0L9mDff6v@*__T
z@=ri477bIn9;JEg+sO*eKyjBz2kEhRPrKKZK%<y|R%BiUHA3Kl^6cK7h^9{<GlL(1
zZg#*prRXSBJ8+YrC}Usj{v0oR6)sVPITC`p#E&+EN{zPC;zQ#sSp8%o>5HEp#II;F
zU2H3_H60_9RI~bkyX53Ba~~vXO^?dqvRUg(p4wsE#;L7iu4S|%{6;*a7B}w2d_6T1
zf3WsI*dTyfZ?~0(T31r?{}W{3PeXaIMP+(C>8mWaVD3^vt*Uc+V=MV)oglVljfUXA
zl0_Xc=jtHK!+M5%nrxvR(@fue5&Ay9`!PeYCBVAo{A?9K{UN`$3`~CQ?!(G_o+oT<
z0V+skFRuhl#w5DZDzo3iuhhz~?^a0<>fM3&n#f2e{cKeRlhFJRmrFAtU%n*Pw#HUv
z_jignwza9OS!jIgdH^h0|F0}=9jE(ptJn8Wh9QTgHKHuZTeY)^Lk2S&)f4Pn6RKI2
z8r!R~yM{d{@ZNMgn6+qzN5=bZO^Nj|^$JF5ti0Z%RhqE64BWM@X&ogNiZk5zvX`LS
zvUMDklQB=m4dRWq$awU)++H5@D+x5|xh(Qb`1SkUOD60-)c37?Hn-y7o<obxa}GY<
zkvslmxF{*$xAUJXj4~9l+iUw!l$6-@IQLTgtuHIR{u^jFM+mcHYU=W^KQ@l6DK@?{
z-T(tz)h27ev{Z59huXK&97|pY;djER4Sn((OPGba>pR~_0K)eSFzio<5wK<pBT?k1
z^a0FRO-Z1Z8KSK)!CgakXC^V{vovLf=~K5lb-*m>=$)e-;70rmY*-SkA!5u2F|L4X
zOb4)L*TjK3{mDL-;|1^lA3^s#8+ljqDB1?_)$6bODF;%W1IH+N(fP=fmrdMCGl#6L
zt5#zy_snppGzh(N4hY&1uZ`-v9D-Recm|C7c}Oxc&|n9+8rm${lDua!iXHlAE~0^V
zof*1({Ue}8T3EB>mxx!*jKnWPdwMnH&Cmn!NMaIq)pTh)ls4&j56*Z?Pl@qqfn(TK
z4Y;=Qv)tarmM@PaOc%tz-=eTCV;N)RSpryN+$rC7ZiP^c)s3{;q=K@1{ce<3RIqq+
zLO#t|b2aMw&)n?yTj8&O-E382(K^M5N-`zmv2MQtof7~8nD|F(%GPh!M@((i#k@U*
z;>gGQzP#Ua)E+Pbs-s2;+?Rq(FSEek%s20zog<X|h9OzrXHJ)wFg$Y{IKiVRUmtc>
zycfh`{xwX7(trf}EeY;%<(}*rzfL8h4Yw-4cuyo5Wj(Dq-4_o3LlF;WWZw36wy3cO
z@7yv%-OqD?r%NgDApoc4IH*zA&Ode)m2;QR$YnO0;PzJ#e`E0-3b-toyDYBFDsvXi
zxXf?9OV+271XWqeuIH#G>Hre?p3KP4&vh0l_wdFEP3J}_2eL(`<P2VM=r+9W?W(10
z3;L!711M_rpzFA>H4y5gEJTO#OCYn^>{aDO$K$KG>#)~f!)<GOcE!f>h?LBSE#%C}
zmSdunQ{`Y8w`vxtO<K!Mp?1Ay$oTI3;tg!8gERCs1QNLXF;*@Xxdl|O%lY2^*yt`8
zeVv4|Jl`FFa<{stp8b{OZ5BLXtf*=8Rb3>0H@uI*mVl)AJz>VX6_O8;5f5qkehDCR
zDc!>K-nJk7ixk*T9l1wLw!qnm`D4J<Ew)XOx7*ae=PHN`(}`h5sx%tIImO>m>WI2L
zq${ee;vn14DhE${=syLYDYR~0B}&?MXv0{{vu%y)pv#yRBZiqzK%#IKGVQ`iu+d_3
z16*s_hUePIEv^vl&E}Vy!Y!hRE7<dAfs>YD70}h(t1n56u$}yxA5IuQpYw<%?X??S
zUJ<ks<xM!<oF&>6@fk1!qj|xYd&5t#N)VfS0uqe*8)@9vcs~pkmteUtB-onkWYY*R
z%`UEnI<lIr)CFM6OrYw%$%%~zOxtAqd@JeXORBa5?O<v*NzNQlQq~wz(%~;K4~bD!
zb6m~;cf-TmN7np5UU8xrip{!3e9l>wgVaJ;ElIk$?{HENG8sV~W{VU#sszxx>*RZ>
z&V`UM*S|Ap+Y>y+rD#*nt!RaT4$FFei|nM3UtG>5^2k+2QB&mhJ)OP=%5Y{+G0_(?
zZHl2}<YkyoCy$;j&Q`Ndw2TvFYEw6cN-#0%V1w5@fXYwce6@PSUQ2>W0|s#8Dn*N$
z$q9HlhD`x}l5RH*Etu|}uB<7z?xTA{FbG=M{Y9?9>rd&tSw2Lee4*fT{>24UBG_<=
z5U1n~1)xUQ{IB$B*NEA`cp7_|?L+!3hPBf2hzu`T8xQ21<@kEI)(Cn~)jcm3nm_SZ
z<nVW;03@B-iclAYkO?mlqI%->q2RwS%4K%=2)QCoH%Lg-(N@RHd2#*qWC}T`Wj!gZ
zvK+hzw)&%20j51RFQtB>YrpcDWKTE^cNWhJ!ozAu0ID2uvn%B9FQItN0Ocoi0iDfI
zL3zMAWpuq>eb?XV5W5R3k5x<318C37qxQV=Q5~wkzoqUjgur+wzK=wdE;&KY*-EJ_
z@~=br5I$n~Vug+%YU{&e5Dg=tWAq2;q8NGzAF=!~tz!&$Qr?^l#$$dD=$rn5FwozG
zW><V>(FMFREaP=3@NTdQ{ja|Om2~D|BYd+Y`V8T>0ZEtBMTURX(wp(x-;NG`ZGzE_
z0tg=>XieXoecbfdb08ZP__1ng+97Aqk2yVjLhH@=8?k?2OF-(h$>2taib;bY6m}5!
z7#R`6rGI~hHU)gf-V=h1_AHcBw~K%zF<?s8!TkHiK*>5C>P$K*hQQAX5U#mel=u1J
zr+>ZM3-A#<Bei|q!mvm2S4-a&uAEGD{rg+OdeM`5tZ2)^UE>Rn(cP~UMf>-c6zIdg
zow{;BHKr8pF+~yLYb{>%mH&K904j|JiZPPPz!If!WYY-YQa=9oHhoDCYkKH&mA0)0
zJV$ETJF4NRbQFF6d<qEKk}`Bs8eol*!(e+b=v>y-|9g9=6vMl%XBhVE=0?9o5`N2g
z5nkwDpK%#}%dk=KX;b(uQs}pQ;EdYL2!B6|GZub}v8`}}kr8^i7mUYJF>WfMf1Uz8
z&nicaL?;`==C-kdb8N&Qx$yUo*+S+E%YH|K<KZ?ai;F}UadHnwU})N3533^#>ya5C
zBL`_}1?*EFhw~Fbe}72^tjEFVa^|D$sFpOkMulll=Uo2Rs}n)Tvhk^MH=LIPuprio
zg4)~v`VQJUJy<d4XT#vRiasQ&^xan(6P4e+o}cS-5^a!js>`tP{5>Iyek*ti-t#|o
zo<i~ySk+S+29fgUGd@L7Vf-|0ZLk0RtpB<F{m<X{Zw3E93*x_u&daa1`44!Ev`v=I
z@0S9gs~?<&hKI#?-cx8z;~Y~MF){}K{wc4~9;*ZsJNRdhs2thu&Ac@F*H2+bSN7m(
zPK_K?v)agr5REYb0@8nNQzS@kJ=6VeN7R-rWX^^VPen>ttxzENy(IqXXw}CV*f-9g
z4|xygcG%`^{@>@8y&pk6iwtSLaBgkUxn&qx`qwUOMZ3^tQb!tg*eD63EaTR{pMhQ5
z9iFqnz7}o&Yj9YGkNx#N;8_|(;D>fkcS)>l?yud9xP^z=|K&&ql8}~h03|{C(`_A;
zS^j<w1@|5BI})c)WJGWwf<QPGMuw06Kc2}%BdD+lu)%m+Be?#KgbD+cbLx7YS4>qT
zc_AhJWIPt&`su%%Nw^j+K`f<u`X)p4&qKiDaO=W(v$S>2wnKMI8z-c-?zTFtf7TmT
zzK#cN<BVtgaJg7MhQnfKhy1Mn`jj$GSdg3|`eC796o4c}hy@SirvAM<R$<ZhUL=Ko
z6&44d@p`rEz&~#lAklO1953xRd~z|+6k-@T(YmOc_~T!{_8@|-QblT&Nc>OG!l|ct
zs?PGS_pS%8HB~j{tMv_ditw{vPP7KLhl~Fe!mcUgpjFQjRxX1xdk&phxB;kt|JY<l
zGGO!L--fVcfy_P^j5AU)&tyEJ#t^kT3omNTBO?~Xo9G4hH;f%I%mk781emaEgY(9n
z9Yo*+WWhB0lVE4AtpUma$QSu-WzTZL-2s7<r_rhKCzk7&a`Dx;3e)c$jaGqOA8pyr
zn8!V^Ji_pv3R8Lb$$zgfU^+lEKMUZW4zRI>bOwO?nt@nkfT$v>ME5lG*FDX>fVwsj
ze=+IXRva=|ZC``QZKrEly|YcMxBq|BkN}=TfC<ls`mHJ=YX3bZ2_?XR8)mJPaYIh&
zz>CWSprAEpKa3+c>HEnB)a<BXC-oL$lCVLc^j?=`v^$FEL&jq^@{4u<y1)o3@bZt@
zs%Ju2V}gKwf495%pX+>wra34BCV(J%7ApQTK$=aI?}#}LJ>R1p08?FJ=oCnQ@Td<r
zDf|qa$mA(5feU0N6k6^p2OKI>Y&)K$aFeK9_i8$69vc2dIuY#=-Gs_Ebl;wQK!xc|
zXI=i+Hj{%x6|MCYYXYxj54;MA6wN>PAe=a6aJ7)4xcKY!Z@CFdj1|Sp^$oDl>TV$X
zgW^a7d4bRjQgVj~spQH7`y;;>zC6{O_a6H4-3i%kBJkxm;mfB-S(*M76<E=iE8oNF
z!IyKPFMp|-_9tW<r}Tl^E)#Umwm^dMt%?;j%b`B<<=Qp>QLhOA(scDQ&zuGkp{Mz|
z>;PDS45$Tq?RoQRbq+-;!5(b8rE%5<Rj((YT3T=k-TAN66N)1I<G1*OumyKR)MPr(
zr~Bt*B$J`B>!?@oSvD8gmC#gc6p3{kf&T0FplR_H=$3LXX{1B&9`E4Ibw&B}?$Hxk
ziy6ugg#QV9VOl8i>R#4igs^fG!hrgdNbni^`8Tu;NVULGeF7%4KP3VIszYvN$|AN=
zmP^qs5iC*N!DL8i+y8qvshWd6+!m%y7Pr7*Is?_!vqr1Lc)6?P6Yo-P^d<rl`$By`
z3N>6(hN&iyEK|rZ;y$%v#83Wx4IM)lX6XLNOh_=5P$0RPsQza;&Cozv)DO@DyPU<1
zbKt)D2}*lCIKiM{t|EP<A5ahC;uLVE*^LOf&tQbcuvd~TNG@cj@?NbXGE?UHSAiLb
z=>pw<mkH>7?_zr?QgZ422?QvLH0Fcx>th7oR%lvt?^ABr+e!op>JPVwBEac&K}Yqh
zq8D^2fTydvFUMJgoYuI1IKT04N>%kkvN;u|SGqP^y#XrPShj+0xxW$2<U)uA9Gh?l
zcff^_bzVSWUSsOF;-g}^wD9`&o|Is;<AH?tFeJP$Ot+r-SHc^CiWxdu88U#~i{J%c
zxA#5xJ9}fI5Bm}UZNyAK_5_v_)JbvAmq!jfySn0jE)#Lrpk-=04Liy`u}LLVTAVDJ
zGG-arK$%eKJq0*69eJ)LEC8)V56bX`_UAC|TJ^f_%Sf`_yI7AIAR@DU#I;zUc-GMP
zE+f|Syv|#wySr()3n+u*!_7f}<dq&r?`QV|B4Y;lzV3Ji&I&wMzTj-y0Vwj7>)<^i
z@lGPNTN~QAFp_2II97b80Sw#G12$tc+1YrTSQGctFa+Wm(6-K<0Y*`i?M6vapMRoZ
zoi`jX!t>iy<<}uiyZ?}?Dmi!SpI;tw&Bu`2jH!O>2I{0igMQ4;O6uIU7*VqsKg|5w
z0}f9BxbE>YjCyL-T<iRw?idQ6Z!w{6)E{qO%*>-`gcGAM4!bU<CAj1F`BD%k<Mtc|
z3Y0Dy6&>*<5tsr-a8A`w$7&bz5$$_`)>*r>VfiwkDc?-9E+)uBF$!&G9ca4}`zci%
zL)7W6an4?}Oi<^|QyxLhL4>=_gYLKSkvx#nG@BpDv<opTM<@9Lxo{9f@ADs{qj>tT
zfo~#6IO1WB%GAgAQ!x6?Ko10cg-^igJ`-%2XEyk%=5MB--0F>3C2n9#uso}dFG)s`
z(@>>1mHZI`c7<|b`ZkYEPliz6%zD#k351g6-sb_PN8!%6NWzFqG279>wwW>b7f{yS
z$UkNfd$|514=~WuH_Sjr&~aLsB^9t^-8Orh&mv~*8lbwP*%@9#Cb72Czi}3IEDeXG
z8_9+yMaD-AN%OgrmCG;2j02Rf4Fd<R_M}F~^0n=C!AAOs@tEn<Q?Yr~^O)sH0+9@Q
z-6eRF@4gk46fPAtzJT&=ZS<l^7Oa!d<0uk$I|*+8P+@9pW+uN<nmapufc&S5>*@v}
zwLDqWPD-9xxuzmOIo)cukS`sVw@xI^AkLtl6F^D63ZX(Oc@4n4oywKo;CgWaHA@OY
z4|H*T=BY`8v&@^Zk)s{hbVt$z{17Jq89j}RWM7=fZ(}G~2cD2F1eOA<<*i)UUA-a+
zOiU5#eI^aw#|5(D=_sedukiw3yzgRt-o>^(;YY86Z3?<gi>nq@wB*`&+myt42B$|}
z`K+VdsQQXpGqFN{jj*v;vhghp<4-~&8G=1|{LHqT7XhYU_((c`X5-0sNfGl7v1B92
zTL+(K-gaM7q``FfeUXQTrN#OU&!RP*d`(gbh#z`G@5mPDIqEa_ijVw&o^)_jMv*_J
zZCios)+*pV^SrQ4O&rQ90JZ27;MS%TRg7rhyaaefYnu+kjH#Le!-uWazY=zU6@nnD
zxbjE-$&TiNu2t50!W8e-y@Y`d$OQ<sZ|fcq?{d8mvW}qnI%JZh-sTdr^EOX=0Hf)?
zG0IxAjEzxf>gfT+(l)$?wYv`(=JI38BRsdP;HXb%`Td9#+Ri&Jqbx@s#gYGfX^_ed
z6q(zLW34&*YFJ&GkZQsN3^A<VmyE%cVIn2*ZRZ=F6)XZbTl{FO7To--_T_<MuuD(e
zC!Xw7ebkc@48MdIDv$c4g!gV%l$u*bTyw~O-LEQV6y9EaE<?0x0f*c{zi(V$Pd2_g
zix^p*5cjqkCc-X9ncZn<lwCHzdXQl#c;8|pS7Z>Gn+lB(9WAGo*dBk366aCBVKtHQ
z<D-Cx?W`p7Uoj6oceHyE!i%f^^4I%n?4N*FGEIIB%bqYa9wI4vv`!nPQzqv9Ck%gU
zAJ%eL@j*E<E+b23DXEXw^K&L$ho(nfNJMA7I0hHBHV!g_RA~#EKbb)`sW?7{atLXA
z0WHma#HxB7YpEhk`6T>&F`k^M{Z%vglw8S`fE=0N{VIl?)gvPL0R-EXS#@reH;FH+
zHo$4!U>A#QhsI!QHL*=?BD=6f2YQdhAX34IWO_kWI0sv_#O>nZ8-Sox`6?o%(JCS|
zC8vDipi5m%F4?y_M-%G^*3c&9E4Oh87KDUqlw&x2){!9fo|35vuosB?1rGR&@2Jz4
zY{TRpt$$T<K;=8c+$S)N>pjh!Q#;lX4SMa-!#<4lTa6XPfmkf1W@}kOI2jv@OOu%U
zUTCv(W?h7gI}LS9&b>WJ>Wyx%jetl@TvGV=v!>c*?ciXsmirzb%jR+LHOCY6HEFx9
zWYFrReJXLtg&Xz!*>s|G(xhzaHb?3fiRc?IgwTp^hZVI_uz-x{XO#&jYla+0$k7Zu
z5-M5$h3|de3P{RYH|lDq`21SywGB&p-OrhKIb$yd9+tKl_5k>K{m>{u?syt7H8O#x
ze{k8nE)nlTcz(VOkDVxL*L0DN#yy7<7m+*UkXEpL;V!d`&f2ttpG45@l@h>cCm?5>
zYv-F_@krxtGDO8*@i@OHGjyHy90#7nS33@9iEid-^n!D!jR1;kvbLMh#CUw{ZiDB0
zfVwI1_vX!+RibKM{*qf%KCv)Jc;yQT<AkW+a<JhUvIT=X*-7Cps<?JBT~6o{^K$}M
zm1{wXYAU>P=&}^ks4)&N9$9%aLTqBO%YvrxPCxyQ*DfaMhgND*ywY|!=Akq)$_Oxl
zjQTEs2YxOWe_;(wV{(u$m~2lE1?I+?jL>H=q^ml8X6~q*S8f~BZEkQ)#U8PX!IuHH
zZ`aor=??VT=>j4>YHZ1CUxK?G=nKy7=Rk3Epi2VOs(xcak^0$-u{^upNP-C44ch!G
zLAwu(!+aHN-!s*`DE0ce5I(K8p&YoS%09bCRa&Cb26{p~4~6i*zykk>Z9@Z}cHO!K
zOdDn7Z7T~VeOt++#LBFCrq8GY??4yc>*9^uFr3Qpvxegl(e|f#hFVai)XkVN^)>Ip
zW4ZH`Z4a8LR=kkSC-vr@4&zW<6@72nr%s4YSdol1gucnP!K$3dgv(Qpzo{en56TOU
zLD}&2G$#mHex}u7%v+YV$6?^)0uAqRxmL`gaXht9ZI}lRJXk~u{NUH?Ayve{GtiOb
zLc{e9T!&AoWyjcg*o^9js4+Z^AB|n^l>>X>%%fS{kb!u#R36_dbz!zY2HF5;fh4ui
z$nLL>Br@`5pEj@^6-CS8awVvmc)IYYC>7xfgKq*%%l8ER&Q)kl6asrAa+RCt1wr|=
z*bS}hXgg_LxYX8yyb2#Q^|NAdPar%u$XkNcYQR2#atdoLh@UZNb8`VT_Lc3opHGFe
zb5JsfBPkZ>3<TPC5y8%^6R0*e?xWJDNWYnW6li!}cit=Q;E9(zV6P;ffn`y9bLztH
zW5&+m4^c>vDa1GpmC+FKqLw|u0!gop>t3ZxuoR1X*ZB!V%S|q^lJ14zn_=p^Y*k?k
zqDf>3qbfHBbkh)^e@DUz($fiCYdK@C1+v7b4z;v`g)qx&bn{ZR1JWC49jc{LmHZIQ
zod4y-##g*K^{?IAAVTY49y-KMGlnFq&Fbr{%aCkEGjqkagIcQt>B?ygjm<-Sc+P-0
zAt45_L^>r1kI=09;=AKJrzkz12?dlw>D>tz+Lgvv>V$2oE+%~Pi(|?|{}onbciJ+}
zGzx;2MH+6lBiOK3kk_4^@*iFcCOS@02hGgP!8NhGMUHSFX$bQL%|@yTGURQB667fa
zG6x1dC}BEaKjl~&h%g0ag3l5ZE>7;=oD%UnreuIv(}~AMu0B#rZbz-N>}tQ>AK2Eh
z{I=-M&ltJy%RID`#!lW{t-1Sk6Hd?}(COd6l@1@uzEP)gPBN^LehJDNr08H66Xr2C
zVj2YdFD3s@n1ykoL&R^iIbpP#%yEyJC>ufig*tH32^Fi+!v>IDu;h`E)zhlAZF2w)
zd?zp?hH1umdbcr3gAK6>1~g0SrNiPKv6Up<Kr3F&O&JP)0;-I+J{+92(d44zBOXI_
z;i_-vW6Z`m4I2*ou6iP|`!g5;&SPejh?~e*Zw!Har>K<k%`=S<nG3HekXh!z@%k}H
zy9txR{#q9wI0bH8IQ4d!S}WsVZsoCvluyV?Z=)KD+>?;;YDwC(odjd3I=1%SD4s-!
zh^w8*=l<D-VAYb!Ic{<6iX3Rykq&m9)_D+VE31<TM*)W6c_B>DJC}`Zk(6;5ZGAYq
z!8cvHvA&W8X01!2_4ur*Iqv035Gbim@Kl2o|1^PsZCOdMFRz#XXt@tFE^ZW9A`wY^
z&LjF{X!QXS)?DLT-rt<qV<8I271&lZRx^Ofm|C?+fzTp}*9)j21gk%Cm3`<Lhjybr
zkR4BD<V;!`lv~}BY?aUxV)}q!@oykSy4QO9%wOe5kE!FZCsO?V0K<;6Qz@XJ(Bzr@
zGK9IIPI!Qf-8OpfR(u6U%b4F>5jcK%Jlink`&2=J<(@4VGKo?reER@b+UnwNx}Ds!
zRA4GH>7YjR6W`m##dfV12zJFw1WowNmO@|};$yie?IO;%3^@#2;*=Xh-O7;KZKi}n
zl3z)opA2RL^4|HucI3~~*^oiEn{g@O#D;}mNAJ9&NfPTF!LYOBrAsz$UsTMiRy^)d
zUz#H-UJ19DFlW>9GokO&s-~q7s;G4uH3OQ1_c4|w+Ad{sY0q-+bF)xjQ)WxAfI_UZ
zUG&h@`T!Hl&A~M^D{BRSqW1PJN8AlV9tW|1cMVhY)Src;G*68C+~-fux=$rHi2i$H
z2b9G%INdaqz_*JU++h~&+uLT1TXktIO7&$Yx?T%pT<9><@sjO&%PJK;-1&mF59;76
zlipBHc*;KE)>hK>TkZtSpRrZ^ZpB4U9DmM}pz!Yo%soXQv?LPs?N7JYRuF(WLv(!?
zi^*YK?2B|U4pnyeh)n`n?WHEg2AOy0I;<*6d2kUtohYV6tv>pTuZw-iHJcqz3ao_A
z>X9%dKaSDyFYH<&10Qr7xaU8<aO?xu{@@26W%o752!F^Q#<lyaC9x_AJK9Bt&e(r$
zH>3IVweBu;L-73%Qx?9>Zizb^6}H?lJmeMR7u+47vH3be6qeR)ViLKV{z1aHv0q?}
z?5<;GlcCot4Q9V_t9eOt&GOh+ep5O}>a`Gi5w|85XL8Y+IA@r&&My*H3(PeuDaPfi
zV<|zT@B+iDv*p(XBBa9i%9iqmX%!Vz)P7{^-quzzd$h+DC~#DGs_N<}k0T?q=axI!
zc5Ls<DytR$PLLx&=l)Qzsd&}&@%VbdEsM-9qu(<l&ZL=d<1P=!Qk6269HpWDeX(lq
z-?C@th4rLc^CuLpZ&Tm@J<FT@TVaIxmR?<K`1BjhYGVkH;|f;BavhHK-pft*ru{}_
zRCcc#O05q2_OVfFS&P)8iJp3@4A?1EiqtZ%`^^@W=qcL7hUErs?rECw?*|3UCxso9
zcivZ#cTdx~a`-ch>?6>fyw2Anz`SJHruy#AOvc-pEbqhlTSRyPc)Zf17;U=_F4sl1
zF*Q3dfn~jBi*GNKU`+upDgLaZXWH8szDs*IU+=EMX5H(zxoc`WKCh+;!gWrkPWV{$
zsf!TpmIew)-$ZU&aF@1PI7ElPP>eN&>|@BN3$CLz3zfMLnObT4p0Sd6r-*(;e4%+q
zNxm!oaJdJww@_irs%niUq!htRjqcwcsnL{N%-ve}a5>tgFRDG8YGcB_2D4^$Uy^^?
z?Mh{Xt&K#81+TGtA#-gG_hyd%j&$TX)B3i5I)d!Wq%gZkgs}K=8>2C;Wa%g*C-XyV
zv8#y)2kL%VS1Vfp>cphN$sVR<ZB<f(rhboda$64<b5l~gp{=r!A66TVzO>XV|Cpk{
zO0a)iiY0YbHyrv{IZH9iHQDP*Sbv%5+S;B}dCs-HPT4MBj=BPnoz)^Fqr|BH*$1o6
zzxM$&zYbS<J5_w&=54L*^g9WSQ~+uvu8YM17*e~A)v1bd42W7HA+*ZZ?cpZ{d%z=j
zjR-FY%&*BWF3u(xh#s&^cQg;3;;SmD(^sUU=ZKtP=mI9G(U&;;xnR6unRDt@CXSVa
znjPKNyp(63fKnV2LcBG&ePb-YxC&K<r%1){>I6=O>#DA?Z`TKgKJHowKWwjXtG*C#
zw){gC_dv`PfJ3ctlOBcrNV9&hAuC9Xs*(f;RCo>d1%1<+o8IwQH}g=>MJUNt6rx!R
z8iG43LWLbECFrs(S08VqynWl|#wEB)XVP{JXB9g-CO+^OFuxCUx1}njdaqB_etrqn
z0zs^@E$y$8moYNqGkI&%xF=%Oyv{8RzGy<0xjgIvoLk!~&*hKzqH39SeN&m8TT90t
z1s}6HN2zk>16QaWjn+5gkGG8izYE!PKTSAl*5+%si7+ggY|0y!{y~rNN?&?hT6Y~w
zZ|pa3YXp?vWp$B~pm`jZEdz73)?9I2Qwqn`lW+UHOAFdcV<$m{%q6@|dbgf$tIzc@
z9$IJNuW||MXDlCY6c@!%mw81I*VX%iO2^G$V8@g0hT0qUvfZM`##35IZMds^&hTLZ
zMGXsH`=xVN;gDyd7N5W7%+Xp)@1Vw7=NYQf(d45KS0o>Arx*+MS3r+j+Uxa`)CZ-B
ze;Wl9NJFQtd6A%kQsL&@M`=2KT2^~yLKMgKg+6bGl^Sp$zZYlH(Ga{{Vo2{u7e^8a
ziEb_I3Nm(VDA#~M4`vSNa11JK0K%P{xtvqRo7Zd-mJ-@iU4;+sougkDSGj&i{oT!k
z1rP`BWL699oVE|SY<m5058>_s>6|)WV2x4WDXUAd)?U4`^=jPX&7gDkKW1MR$pR%c
zA21`o-!QeTD|Ix#OFYO`#l&L~;Zol*{E|xaE>7h;C!bG_+69kGhp>!SBF<N-vl0ZW
z)Z%SsddmE-6!ccfeSCUuw4T;wpm-lRi<(zo8ojiXbwp+Xh?E*1)@MO(J%xQ2c22G=
z_h#-Z&z#AV!dtfmTf7i1Jj(XZX`iWTPbB~Av`dl_1|pV^)UB<kjSQOv9d#}C>EWq5
zSP|(Z&oR!k@06U9GvD}lE!npE!ZpaVChSGTML+kZ)QE!Mg42f1kYiSSs5jx9J`>@2
zY5i4co)rv?Z9E-G!zOpkvBz(1((F{*m99LFxtw-k<)US1*$6q_we%*ZZaHB}aq?)W
zCCeUCVB^Fy+nf?1`W#8j_|+GVXQUush^rP;7iF{VUQL22=fxl6FQvwcdZ`;N-(#p_
zU#S99R$p1KdRpqkxA{nFhFGYl!by@YMfzaA#;64{NJLiBJDHuRi`B^IWpX%nj7(=K
zI0-VNLt*+$DSgb|U@q?OYBiqe3B`Pp&n~Np1<2>)Uwt+ke^YB0Yeq?QOt*~Ix3i2B
zi_~|*2nS3iP1UgK2F~S@`66e?KKQ6FblitU=MOK}xk-Sl5oc{AqeX<r63(%jGYjT%
zhp#p5p!5{0VwSx0u4a3G{X(nUO4Eqwa4B%Go9y4LtIBAU4NDuDYaL;4*~RZnBl7V)
z(#zILD9*>bq62Dh-$(WP<cI!5(Pl9fL|>vWxaK|UI!e`jv-}S$B7-3;$@;>sT>Ew2
z$S-VOiwsFU3VX1En8mun=fvw=t;oc!@@g^6q*{4Zp={zVuk>T-Yv|)NfoWB*en?my
zmS>NsCpP60b%d;9Jj~H>*Ps~Fr~RH^@6}WYbSul9hzcDaTDz^@c5V*RMgzQuGnTJv
z<>rl)KCe+0EyCgD+8@)Ps1Od0#|D(hNL4R$&`$a^gnl|5d;;~ooVNpS;YX^%+Rsjn
zN{GUdyo;bOO4=FT3DXs{aI)b8ID9Q5rEGYgkLvv)4x_*Valfp((w>2ofT4ZPcG&M5
z>-`&o&UK!oBu$zFHLT-1q`^AiO1yY|?g0{dYNd#zpgRdL*lMCC|0QPiW+U6A8Yfe6
zqi{R4=gt`yi<LA#l<$ueuWO|8h{(OqZSdHn!rkA8+ss4ttk*EJ(<GM(UG4p$ZTAwS
zx}hK_5=TVPI(}(3uY4ZGFdmm3i*a1XF93mj5lvrKkc!Z3QS?CU)~7crDFp2!7NXdT
z=C~Dy%iLUVe0XWoYb7Jyv-W7)Lf!mOTKJ^!2k_(SS1=NDOrP^&5S|m<ttOr8IoexZ
z_Wj$87NVUbzxydp925D4wL_t*@~j3LGc)z?7H0a*t)Ig!mUQffasmppMI*jL)(g@7
z;N!9Ef9U-}XmQI<8=l#R5t644kTwcGj*kuPZE}6RTlwVtA)5nIc4K^!v)s);E>(}j
zjp)lo9376*eZ*?kcSGsG9n%GkfD(Z{=#5RUM!Km}4Be>N!153&P39uk<$IoNpF3%F
zp66HU^_sTJD%K>5dSBi-H`<U0x7+HdHx9kghnSnxcAMaM<B#C|soregzIVQ0+eDy%
zFkXJZ46Jv#Dm2MSo&ZeMFH#{P>v>{4NptET%dKQYhOe-*hI}`d_|Dw=z34+~14678
zXv9))>|ga-K5!YDt_MH#8%3y0w1W|%eAQ|w>sz()f!k9BzGaVyjieRKIn~`TVcI&?
zOd;-6s`rP}7h=bV0PFL!B&WJ+9akm4Kri!vdg?UvQi>-d_4=R#BFNHX5t`+FL8tt{
zrHZK*yU)cM<9%1)fw~^&RIlSIj6X3s(Q8Q{G5h*AH<Iu+QTM3z3r%)}R$-Q%dkgKP
z&OeZFUSkHZLbb)pqwSSn2Key(n3sjP*Q?=f+A;}L-lay+%bmtih*b#p#m`A5Z}mo9
zNYx&bMfFr9Wk}G>@uy~y?XrpwCyZTVsn8^3o-4^Qalh#L<)JK;pwl2QFcQFU$C-qT
zmTSIlvp#bt-r%E5>zPyqZh`;M=QCtamS@+2yltUBwZ0+<`qcuxR9a1oMn7&GLlrdm
z>zqps`f#?K@qnLMuiCvf#riC8tjKyCB71PIB@J)eP*^$h?4w^g`^w07e9(x__mO5@
z+>+BqqAC1I5A1#}JhxB{@Q_<!xX=p}z1L?&lV#P+Y4&`LbD{o~FDN*Zy~He>oHR)<
zlEF%s&Up6AI@Y$+EH50Az4mJ3<uGc1O!mZk;ua2#f1S=(4m54^6aTzn>3LA1p7H2R
z@!UIe-o)iZ85e8zs-59Onr&9O-jxQJg)xOF1=$kynQw!M&~<v|@n)}n*S8L&e_8v=
zwlN1xc)B{i9t*94hRF~wQ{YHk2u_?}^>#H6Z8j_Scdw0$Ee`LqijYAQH%$SrAZ;08
z<_9%CpGA}MB&BNgp~96h_J`to1$=hB{%jGtQ|4j=fAu#&VrK&Cg!5djrdhRG<UQ^z
zmWyPiCg@jwxA*E@_FUv+fs>&<MvCrJQCG%^-yPSV<?d^)FLKIo#;4TwlNFd;d!EiI
zJY?E4<5MinK8(m(C|#MNQ#oZG+M&I8f6tP#FVl)q8hBlMqt0LnC|X*~7}p<qZY8m-
zcGOnPPAR<upsiR@Z*{t*@wdS=$D|*@PZS82+ZxEIjb2`#F?`cr^>t}JEcre(K|JY>
zakYb5VC%e^o_XlZ?)s6()l&J(%7Qpq(5Y+y^xPHGT=eX2dQubC*oxw)%j#V2OlNIm
z%0v(Lx|w=8fOF*ugGn?GLjhe1CrxnCp5Q*NHgo3wWVuj`z3I1>11ob5l`D}5Q{#8_
zpId7eVl1bc|C=)g>Yv?*TD1rIJr{U(k#-~U1wP;<W2s@w!gjpRFGwKFgfwHv6!Q7T
zVDPcKjKZ9%t@gt(?&=(fZb_^m0tc0BNJ(BE-(^KP_1<LUlXFA=nGF!?J&I|}j$*yq
z)Ne}ghCcFrjdh3et=_0RaV`k}8P5Szy|%qw7No^@@6EKxd>ItMAI(!H$bG#lksTRz
znzS4r``A*eoXRd2EG7Ia@7|uF``m6~@>yv9ot{+UT`Dsjhtma*yUMnl7`V(H>5s9`
zJq0t4LY}5KMU9k89gTV0Hl=&ttX*WFzm54u5&RYf4yv2853S3>G@kr1HP|Fz(B@wA
z3f!^rwnBcKw$2*i)lc_4uW5Fcl4iB~vkUg9z)R{V0vuH8(((1P@v?529?<YynartL
z_!PG<m6C*lTr!F$ZxmDs6j3f9W)o}VJwRL<_9qUUPi$lAlJFu7DH!$v+kw+cOW#}z
zlm?_$;FqowUJ$o;pe6~VpUrj$+h+mQW3Btla;K45`}+BwnmsanJp3rY@?ztCt>R=b
z#klaRW}XAvv~Vuse(4GQ&~)!in+Bh7^9rKSnKoB&r|`t>^hgl$*@YK}1-&DA+X6Bc
zyQ`oXV^-PU>XvxRioNg%qk6(kzpp)smS|;O{9KtY=Slv}=NqKm+#N^Uc_>K@MbYlK
zy}3JnFyOEK-W__R89O&!>W(o*)UPHMK>oDUSfGg`-j)p5N&*mmw*fX%oIaK;I!?U7
z0TTNQBYX#tj9g^{3~;u$rAy@1vJlh*-R?+%J`3XMUX_?^Lb=uMT4;H4goRen{saPg
zn^TiQ`SO&h4_1i<v@Qb?Bt}-=gWP5V;3{wVHDYANr-%IjI(vLDytu7}ar}I}IvOx&
zo<J-m0<09@e9D*Shz<182?#8y5C&Wk<a%Yz(#N=}*vdY>2S&*r6}msm(?j?_jR`h?
zy%R-g>**OE=86b1|Ljfi%gLEH(=8XTQt2rCDuC<uZ=qy&+z1I<_6#IFsId+7+s53X
zfTuEe-*P0W0nGM78V$K5M6^~wxUPhjxlbE8Kp%@L$_45;cW+On28$k`d6;n1Dqd%6
za$mj*jiXmVbFn<MuQf(v1SY6YpG6(Hg?&JAy8X!)x$<jpEeCj;V9$D)FzV(4B7aI?
ze)bhM{xIR3F^%SF8d{_A1$wSp<bJ0pux=2QyieMJlIR8el64jB2AT2Osv%Gxx1Q?N
z#nH%)CEu`1f-L8`2A`rvnN_&g`0;~Aj;){#n;=)i?x(yjp}Vx7e+DPEQkP36Q|1-4
zwS)Vfv6ek5hU+GtL^4@xwao%%aEdf>uq<a!{#~OdbZ5)mTt?Np?lo^N;_gkWk*)ut
zO+v^nxVFMP+!9;KCxE!!a3-&8M&<=+4xV3hBY(TypdVo)-MA#`r>l~Cp#Fj7CH_T+
zz43ma*$Hkwfs1llEQ9Lf$`7az-?9xhW-o!Veya_16pj}P%vAu5NTK|keHl;Q8faUn
z>>NH9KZTK<#R~fi9l`vny9iAJGrZ!*v-CoE14MzgC1nzHK7(mgvGp1UE<Xo;l)uF>
z@U->E7cf)fkOY&D6$UD#@hP(LWL3fxhDDf|w2xm4oFJ-F)nd0da8o%`m%1SqKrsHc
zM*vw%X>G(ZGL;yap<{0cJLnoYK$O4BtuK9zTcIy`LH1ZBOfS2p6cHI$?aRm0Y6e=M
zNmbF@p@>@!{~}H~7rp_a%K;Xk|M<$<xOnVXETwk{15SlCA+jG0ZNOy9v%%?sA@ITY
z@V(Bg@r3&PN81YDKqthO>6zo3Y}d)z7RExVgw2*A^X8j4ijDijF#Op`yv4D3ONvDl
z?)QJN*_SXz+kEuQWgUax+x$rl*g>!^1zlZE7z8$f3NAhbn3oA!aFIV+P#Uy8=)d3o
zkSeeUFqErfApY`WW-@9}Z)?A^Qt}QuI}TFYd;zd$Q{nSxm?1(3#yg1FS%qC=ai+JI
zdwmsZgv;%&z$vJ6H{?(ZDzn-_?i`o+(yAN)lcuy%KS1VK>eK@9u;6igTQreDk9mSX
zB$3qxC>@)3^;Yo+!#>BnTSEfJB|?u8Zpp8H0$b8!He4RJq_Dc6?ni?m`w+=suXK;@
z;Wz|ysbDPTlPnfuYaRmAOd{`L)OTQ29>NZ;OTOUO2?bmlNB3n&lTO9*x+&g+ol?mq
z9sEMQokJ29ocvPxz>2UjJ5pSYJ@hD6>!@|b>RVaS-m8|TExSFc!R&$k0bV5(gL?n?
z(Zso-?IbKPjBVX(ujU>IK~JoXv_{1A{;2rQGhOSxnTGUl)*qoXzcJRI)Io)M(ixYi
zqh^F8yaM}A(sek(;JW*7{-Irq3$RzcQ*jI+9tIize2x56V?Vw4%@ATSM^$#VS+$(N
zsYKxTHS_hE+8{T;iqMr3O}Ly^lU#vEcFqAhSf{Q#anqi-t^!ee5~~nEQ~)x9s2BJo
z?_|~@4dUK+9;)=>5V{_q1JshWft6xMWX%I$WHd&ao>4ah&14j(;x4S{5hSr*C&X5F
zxki{?zXEeBzL00D%nGKdD>FNtWbXyF^F)1~Ar+&@-TnK{Q%yAafObsm#tH6O@J9A(
zyBsP|vJ64;>Kd!9&gnjg>AjI843YXp{Xhj#?)8Pdk&)8JDf9BhW-lSPNRm`XR>(35
ztxhr@0g<`5=yLJp7LAd8ZH@21U%+D)IB!mowGTXQoF3ZO=F$p^&MIKvEx&RB6bCu7
zb$nI7e>g)WIb0xgIr<A*Nr~Y2mQjeAi(qb{jbFdIvM|s6MLE!$N(-RPM7A3G=j))g
za?lEFk8o*Fx338i2wJMc3<8yc=)HGyGWxo6l}fTsb_RY@mMJEm5K`RNA+UNtSa@>t
znBCd2nVnox{qPU4xA~`=+nOZKEJ;5Iu_ALmjU-B00iuqD4}$Nh#{aC=71*zmBI@0m
zvgh|j#cb3g=1w(lJ9=k`qpAQVxAEp!*!NY+y!(2WIOhy5wm6`75*I_Id`_3L(iw30
zZy%VSDIjYvjU>Bkj2llCyCJjyzh4VC9Zr%s*8x7x7rsaMyn&V_15cX7{LazcSG>T@
z))Z{LNNan*)O?kMkBM^}E{TO38tPRrL2?k>>+003ym+2)8}rtieL78gocqh`C}q-z
zFl=V@Dn#w~nT&3|yQF)}7o}@&&%*%6*eA7N(KED$18b9=S&T079ychJx&h?7Lp*62
zB2XKV96!iAp27cN`4iAaI5ESeTNHU80?JbzknX5p=*h;U3G(b3R;K5|M(<*>Bc*!4
zLcd#b!U6lytw89pHLI0X>5R3Es?OO>M)S$to7&<ieD>dpG{}=FC~rPYZF1&k3Z@vs
zLz$g1k40s+a`?Z?Y)fqg^7Bu?r9*d|)gg*?BI^v#dw&I0^06n-7jMmUpZ=8kNjm(s
zEW6S;M^Fbe<kDSfj;fLq`7u7xZCe;r$@zJl$D*dPZg@n^T&)wm49wK4qKmVqmJc`q
zdsUVEGk@=6b=3x1xiQETI>2vm=)go@Skz3aGVEy8agu#JpIx<7RB4*ih;w?GV6-&Q
z>tb!Vc<r$(jXvVutA0AZ==qrFcutijK9x9Us`}|GGA_jJ%PO5H0%oVA_k!KVK8Qle
zz}AoNx3>3w>z2&T4J?tUwyKRTE;KdQ_w#top=zIUMviKe)Vq+u2WjZHsb^2hcE|8U
zbHxa>p5dv_XMr_aGP$<Gzo}pN8IzFy;g3+Hdoydi2a2Z+%ha2{a>iuj2fCKL55pYM
zR#o~0D(o8Ix65$S3g@NU_m<1({D=sym(Z(M+GLE}*#nY>2h^W`UPiw;B!fUiKG+xZ
zv}{uzwpRMiW#wcV@#h!N->HFNIC>A|uCDgkLwGgtdqsbTeEuYREU0PgQ$IY)hQFuq
zW7~x_6Xn*6;|^5L|6(xr>;VSz;-|ZGwdmnr-=V^+M+QdydUjCPGMQr2kfwAa`n((1
zFb>kt%)e8l$v~{l$z?X+Z}0L)+<;6#7mM@2MNu32?1hbGa<4!NQo=qALNj#`0X4f)
zg~|UN&4&sQXQl@iRmF|+ofE*BNj&%y@`p0ibw>}l!`UK=HV|Yop(dIDUpRee_g{}t
zxCVwkDkw9w1=gCz#c*WeMn}I7{Ikmk?+8wv2EMpNCp0lD9!yFA1q?@};|q_8Sjg%n
zB>zRbrG{>L&es8^Z5b~&uON8UDzP;X8gpUOWPo2w38$gzECy&UXJM?hbNH2PR4p^0
zLAK>UK)%9pvE@MqFi9@Px?Bwe>6*jw2B^N{89>DVDkZ;PnZxG$1iv}pR+Y{3dt;H7
z>t^XYHCeqsGR<+iE;!AwoX>xNQRG&`7(so_dn`xKy3O`s@{+c}FR|akq~w5v2jJ~s
zi^hKCA^%L2zx7k;L((z*!#Hsu{-FN35NA*f6@zlAIVL8V`PVDm+_quCh#Yqz>-2~x
z(k`hurPO*ig#G5Dv~kQKlUyXiM;15r{Q{t8!Q+7acm@dV?co?N*iaE<Lq;?3gKsN;
zKbHIYaCZ`DH|!iZ8bChREdolX&*yFt{w1i8LW-^Cr;6Qld(m3g1tg7Syk7sk(d@P?
z2d(p{#XOO9n|K0eoM&Z@nFdU++4GQ75SVVzxkQ+jJ6|1)Klt%n7y*7*8Zw(G-&XgN
z4GixKEiQ<~VDarS4W<!Dg2W+0jL_U${0Eud{I+~JOke@c0RIA{Xfuz$@rQ%G_Fo6?
z77}*3GJY5dLINW9p%bm4d+r<keNib`BR<dFqh(;ex{ekYN#g#sz`wnO#}V)8LUTaU
z^T?0~e|?7n`yWaW2rW$EF~G$|`tPRz?{A=zg9ovOHuC2oXT;6DHvh-}3;{O$_uvqv
zzULH1LFf!n;!f}RU-#)ZAoK18zms2gsZ+}6;JDiFn{*TN0sRyFl%E?UEDP9CE}&HW
z&tg0dJ+xmf%iig_-ElJh`&s$P;8}O1L}!2n6J7@%<e;03)W429vQ1d-dl_IN52t;)
zHNdosk@VT$=#2kscn;O3&?<Dz&pdvx_3!1prtsUA!MYl~6RE)yt?QA|bpKEVU`H+>
z;&Rm$#d>(J(!h!zQ962B_3x*w`=9v%UAbv<tVSs4pn*<{qJ`?uTOMg3TNE$@w6!5^
zcW^c^|IY&!8BZdMRFKWkumSmW{aNmn$ADoGHJf|y1i$K6UC7(uPJ*^r-&_PrzFfY*
zuvoa&;wy3|qcH{KP#@pEskxx=j8eJQIxm;ktr9tPFf_&peX!A=-sK$$43EYh&JTH7
z$zM{yhatl%o=TkqyI0!e-C<fK1q_m7&z^cR&ixu9iky@$sU0{yc}w7ALE!$@9SrZ6
z742cv1nzZ3B;}N)K+#Q`A~ixZK?Om>$4^V9r1PDGHnpIGf)*?ll|t-7o*)P60%u@A
zXI_F!G6(^joZmCsYoDgO4=8Q~i{s`k`Ts;2czQiT&53%(xOq?NrGuS;!P*D3a?;}y
zfk^T|mEjHL!Ry#Q@6-YfygaR_oV0qm!Bfyl^Dt$NIbsi<dYWzL0}fMxHg?|)G26<P
zioD^su|^Omx^GKL;`L@=%klT;PfIrFvt*%+g02(<8j=-bsRay?`9Oz!TJO6O7&=HH
z5(jk1%zS3A37}onozg`ta2Fulf}J%9G^BXR(^3K1JcP+uSe?KfQj>Qmx+6l0088N|
zFv+OQox+Tq@{toCl20J^je27=HbxT_Mz(-7gGRGB_JDy^F$mewk_t0S5TS$2qECVQ
Zr~cIPe%b7g7e6xqfv2mV%Q~loCICKJ&$s{p

literal 539484
zcmeEvcRbba`@f7bO3KVUG^}iqeWZ{bT6Sd5?5#slGET`Xt1=o$86~SkDzYOoitN3&
z-~Bp=a`3J9`Skw%_wMm1r*oWpT=%uF=XJlXtEtHCBBmwA!NJ*eQvSF)4i1?;4$h7o
z!tLN4o2mUVI5@-|<}xyBCuL+{YWB7!=2pfyIP%xSBX?+OoTt4!;?Wn3Lv&N6R~e~-
z!=AW9&7CqFiN9BYkVyJ<$j*?CR2Yk|F^=2`3VMcZ`$!C52O5@0!u))nky2Tu4&p`d
z3V1lr46c0gni2C^UWoJ<uAF1SxlNXS{MI&3JS6+4o3i+>0-fK<qh7RcBTCq|?cBB_
z+^UBz1YEs}b3D1H=)=M*oR}!v^UAaHGwmynQ^yJT8FAED-n{qR$3y<j`C|U}lMh62
z5=i&EmmCziMN}Bfs~dRW+Gq!Uipp3dkxJaftnmlRnpbwek;UnFuahoFiSumNJ=5-S
z^*vHuyjQH>dfue@)<EbndPPU_jl2%8n1Fz#Z@?ZQnZbiNYQ2QdZ>@9*Oz|v-Q7*sy
zs!5>QaD?4e|EAKd*Od!Wx!pdBCjOqH{zUxc$IK1z`2~<muOj0ppBP*kU#vE#8e<(|
zJ<ql~D9$9Vg7reooKCwzX>-sVbKPB*@gBCtF1vfl=V22!?wXjNxx#WLus%SdQm0Qz
z$IU(K7!QRV#qz|}Q<-&lIRm4mDxXc3ckH8Mx~j(!s@BCrw!^jR?bO-&IQ9jN0+tX%
z>kwWJwNqj)hkOuG5}g`UDdoXDqf>V>ub*jBw9lu^Jh1Nqv4^zm<zqY_gEgO2N|W|f
zH1n1+CSUS<le}*hSBlJ6T#0PgAcEm30XK~PGX3k?yVngppB->hKlO+=;JvKP-OKOy
zOgu3a7_u5fh6!KtNeOw|a_zkb5uZwB@3U!`+?<5Hse-BF`yV)26x$M(4P;HTNN_Be
zaVi<aUkbV?1jM;sz}cg=jlm2@*MP7B=|d>if&cX~^{q=}gaoHxjE8Z>VffK-;)u&X
z40f^M9yAbWB7RhCqqbw(Po{&=Y#VFwKD%u?+k}f5M#(09w~g)%x;$gB(-%g69ASps
z{$R&GnFu7qTVHXQ6dQiv<?o3H*eLhP*qe!DQE?zXBygl|e-z;DA0lI#pr|J7$kgxW
zpCDXqN<$EKLG%gUoni+AstSL<Z)~p2cZp3)=IogB_oQF3{N`sT;JLjJw;kc}Epn8N
zI6<Cmy8@H(Rl`J2wZv3>O=5*Wt?L}JrxPqr1*ZyWvi4IwqCOHdf2Hi6x!Iv+;vs+1
z<8t@idaWiECbv6Mx=ZEl^CI;kX$@eM_9;>0ZKhES;`H~sUh=>&!XS{BGO8fTC~6`~
zENWYY$6>0E#W9AlnTRG=D_ZA$&UntOIYccXLuJ!Oa+i+Y-TrBBPY6lrrAs0&SiU-c
zmHuk}RrxE<@cy_fjU_GLmd1!&`3iCL5mJ|>t`=SQ31QmLMjk|zenky89x2y!yh+Av
zzYgn8va)L&$2IPW@dZZ7>c~7fAhfrNoM}({E%^g@w@BFU9$-m3bov%)827Dt`S|ij
z7fGJ&7P)zt)%L*1{@k#Uuy%De3i3eegNR!)U*sRA+)WZs^whk^exCEv?OBD#w;zSa
zgxhcw+}oR4rYYpK=SaMkw1h%|3d5O0Y<Jie*c`&oa-8L4Iy-!d@pR(DrbkyEU3_Gv
zCX!%((pZf$#o}In%7?RDnrRvmkDFCxYId4S3Nl)AS??Yax4s;bbf;EiQqY}fiLBs;
z<h|ra-3e6pb2QVmh8}3~;Fa=Tu%p=*q#U$A=v7eFb4Qa*QzDbJ_=ckc-A{J+YxV2(
z<4ze)9hwsBk3SMc>rLB7n?%b+8+xQBhAQS;jG~~6d0Zp?=X;hf&Bw9=vqQ`zs{OOM
zvi4_1nm3y3H!wETG!Rwa?6qnXc`WecWA#+6avfKrM3UG8xw?sJ^1kqa*v}rXHI1G{
z@2gPg7U{lLnB$#Ls8b|cr07#oTn}qbC{B3ak;z<xG>$SnXVY}ysNn-s!78(tr~GR0
zjC}b%o-@m@d{H;+cm3hXC+|hWsKTT*oHRnr3(fud_6r0?r<~6@pZ3Mt;fcsc@m9~2
zxm_epZ)%R2owB~vx;Mx18O8bGtiib8=plY>@mJ2yp_E?&t-~!F3>r);pOR*BKefs9
zKJv~?<;~q%kywYqFGrg%HVtOFnLW0kyQpo@VzN;7=}m~~S!<V;IKvU!q7Q`)%h`TO
ze#I9eOC;oCI>p~Pg)oHZ$KH*7pc1Si^GWv8$xmW;J<d6vp|QjpxJOb}+F~63ktk9%
zT_s&4T~24k?zz39i&%%y(MHE~gW<TBOT*iTb!KR1^mZr9-Fg?wSP(ZKr&Fa>m7?db
z=b2HUSM{ZPP;gp$n$wof_TdcQ$ZpAWNlwpZuTzEF3QG%hy&rkkc%SxGUEH;(u|&5h
zhf9waL6Eh>5Z4b+5zphOC_Q6JoWR~L`Ye`S9bS>6{4Y3W&99xfCiuXbQJlxDEy69d
zo34tw%9e;A@oqI&VrYp{$+HsgUGMw@uAU8$y74sl9AnMClQ(qe-C*Nhw_WO!Qv+&D
zYQ}p*Xe^WrmAonf&b2o^YifHTa*6YM`diDg8{f6b?TGk@Jji9KXStg43}7{UnQUXj
z$Lg*sQtN3wc*$l|^S-pc^lPbH81Ko<lRIt|+R41k{#tWUWjv&7Puy->x{-tJH}pR-
z?R`P=0l!M)tHuZQ?Ha8bxerK*B(74vxo|TyRbf(;i`UwvIbgDS(yOm?U*<mL=-G+~
z6`h8sFXheZMA&?;ePVMx>4ALhouQb5BOb?T54UpqT_iD16LEZSDvv`r%!hU4<5YRt
zr==GMOy1WdaVVB5W^|wLN~`jS8;M`3<DcYjdlvFz{$u0e5$)G{r=IWCiq3y#6#nhe
zo6~rY@y?9s%jkp^c55M&rO&^9m7QtXLhM9*mU`NUdQ?~DruxmSRMXg&yw>*nMbuvH
z*=^oS(^Kt%bET1rRtg*ng;pKbFBbiVc85y^*(!Z$%6@NNmtNUa`MuIbQ9^Mh^5N|t
zkyEzOxk)eYp4Zb^wi)=I7Rf=+;cUC#Cd_sqZRbO)R+}uZiD{nGZyt#?v^6M;Ww|_Q
zAG^Tn(^ILd_lnkKV37OE(wtmP=!^xfMU17g1!w<Lr<YDu({Bg+7OzifEa_Di9P}s}
zCLbASWy)O9DRA&w+BSrDhqhz8TS@inw|>H;bBq?&Sr$*s4F$vLw+H+P&|kbeInlcR
zwVY_kvk;+W!^MvwMp;Ism1+)p6Y@8nCgmO}%wBM<9t#o5%`qy{TFD!^vV6Ti%z{-o
zmL-lu{D`;c%B0z=2VS#E{n5)Ir3|~P%3=$=xEFOA=Q<{+d1m#Gw!9w}_o`X?eqAv=
zGDP1**XHGS$E3MOoef5f^cf5T_Wc%x!=43G{vU&LhF`X(*toZeJA9n^e6{6n%h}<I
zqB<|(`Qd`Nw{dMAiynS60>ee#$OVJJZ~byna;2f+OlQSwyww-(%-Jdzs1&d+crK(%
z27Aswjc;5Tclk1+znwV#RNQItMh`tHGhtW5a-ZI^8w5Pt^%V3%c2O_M{9qE(i!PY|
zu{%R5J>+fh&isf%MxW>3(#l%AM(Ep|g8YLe-QwrG2W7_Of<qUiG9{h7LzWgjt=?0A
zjr@?&t>02e;Nu|}CnCxHP2pu&8V>P3I|X7K+iB*dOX4_LbU0+xN>hztxXNDSN_%Hj
z#T)!o1IclmByf&%;la9SaatZDBdEqyPZW~8Nxjy#t9W`kaE#|>Qd5ia9E%4_+WXG2
zv0M|J4ux&ATB4GYXM`n}vq?`*Gvf@oX?QJNVPnEiC~G?Omc6U6m|#&i*?n26VRmMp
z(!~y(WyuxtUCneh9B>;?;PM6<pF3%ytc=3}UK8TrZ==N_0I#-zzmnVNe!iC7#)gB3
zdLI`DC%_yBfAx1%z)$E868wd}gZ_zk%O7V4_}^ae*EIom?YGJ76Y$nv@5lk4;Ye%9
zoIDABY8crY8{0UT**ZoY%oYM~5ZTF}cfi4+ItcyUc2fPoC>X!VT=SgcIb|hLBU@|k
z3m0t-jk#T|?VxdR#9c+fOKW4t3out}D;oz<R|yu>Z-|1|(AzvLFx0O&T1v2-Q&xk?
z*xDPzj&KWd^Rh@1!(cFR`->)`>c?eQzYhK<!D8m<XeY|U<Kp7N?IOT!Yj4WKCn6%k
z!^_XZ&(8&ZgUi9q#_@tHmyN@I^dM{F95;3_vNyMLG`F>ZLE~O9v~_ZnU}1qK`uPXF
zPGeW|Uo+V_tZoZzkO%sNhmV_==jYhqtK!hRqH5-@##ZN!n_B~#fjJ}v`T2OoQD4~f
z(XT1dUp?($Y%gPL4Zi6p`HSkSUta&=#xJ7gJpXG>J`rKT^>eQKX!Wb&JkZ|P(E?3q
z)VqMvlEmUXKTl1PxF_JV8#qXM^W&<T;3r@+=nq~C_=gSs6THT|d^ko!dl3f*j&t(3
zw5IE}!B*R#*we8)M(fmB>)_dr35)De$3$Ob^1U{rl_ok+o8WJT7?(@(ZBpAGC~HN)
z?8`@-rCH05d)4N_^-~7>4;{FZaOkk?*?M``ObcAWi{>y{PsF#<JocpL6g~VcZ8-un
zZ9S4zBX@gB?uboPczv2%lI$6M6IRkF5@x#?wkUHMXB#d)AsGz*cQ1IK7Sl_U;?dFm
z-7r5_F`9fCwu27w?<`L3=4DtwAZ*vav{|w)gWZ2kF!WA2_L6m}?RjwF?+*Yqo-h62
zzjB)J%kUGpzVsZvf;<16@!&{5iJxctcP~=q;YgViNPp(P8)l8+3GJWm_?M_pNknL$
zN?6PLFNhPgH)+!8e~10y2`peRN@-gEe`h>c`D?~C^8KY8XdZx-C$h+w{N*rfU&j||
z!uxlqj|jxa7tzEwr};OOBm6Q@W*+~NX8cc?t)2Ei7Tf=n*>7}Zbwcr75eM44t@T?-
z&IjtKyLX>tpOhxew!agEt@>CXA+&F5FR|TX;73%z@MWit{o^xSotx|&(Ec5zr)W3X
zCHNj-u>#M@Dr3tY2F~D%COuu+TP$;)lB}yb<L(T$%S0vuk7Fmb^83wWo(EQ<+~OsL
zB9{69@fv8Po!vdwTWBN!FlN-r@^&ID_pXKn%k-tv;M}}SW&(VXD>_ss_F$>$4UPh8
z`gdArZm~=TV1^io;x|wB`*+L`#Y)noPWt?|+g?&_-XbyiHRm#+o?w}R_iuu-vGwr~
zR`~aH`Kz|8ByO>YWc+)vnMC$ak8Gax_Z(530MAnF-lZNDvsq@~`nWZpn(##?vxD?l
zej}J$eFQ&(Tf-fFwl4hC7W3D$k_`nZBW1iSZ4Yfh{Vn$QYY<1G<F78Zole-X#c%nN
z92YYp?;`UoVGG^7o!`BKct!w-%cHIBvU%8p#Hdl#zH`@+ZxZI;4Z02Y3n5r++^)OF
zwm23gBGj-3mV2>e+yrTaS!4y>1KuN>SA!@-GeWQBzJlY}%B=;&U>V7r_qPzvR8?5>
zzB{mh`Oj@in@RNhcY60;bAF%y7{2@0DgK-RpBl)-p>pO~|5J>cxAWU~j7b1-^>T8C
zHxH{off_aOt!^?amJIf64_K@?-O;OC97`0~2w3dv&j+?*`QNYt+z*llVfhOdTdc+l
z%?Q<v{O;F(Gsc!5H9!oer+ij+iy%WS9-bv2BuzS%RuH|JM8AKh`XJfREw{4>#-9iE
z>jfb<T6#x!-51|tJHO+F4^$8*SXsAjR@S%K(4$^|rKI}1A+}ie#r@D?h3tqAZax-X
zuo0kY{L)ThD-GKrjbMv>vz50ok7k6#4CBeYzva;_cJc^fu>6?xpktd?lpu*PGqv)i
z@7MMa*i52dN3fN~NKk||ADl?XmDu7#G99$^cF@VXZLy>+X!>)m+guQFzclCV-$Y!t
zQ|M9mx<tqQZit^_{d!R@3oX{FgF<055eyVi!|u=y+mhvfv;JMgkZpineYl0XokTN&
zjQ+>ogIIDcp(cvKj_B^Yx_L!#FSL=Y&z&~G+DJx1##fiXeG_S6aM%m9^u87j8po2j
zY}ycUpJhAAZ6dDPJ@lvthxsM2WH4kfwAe4j?6)@)q4+j>*qlSyN&_v>)bI19i8aVK
zvu!WnXvpRiM${<&=G3?RC>6zE7J2-{n^%OGM;pmh+o4LVjU+c@d|weyZXzv6McJXH
z*VmTj2bRQD%Yle%Y&j6JiMR;>Yyjy5g|Ei$STY!l4qEJUga+FdL|{h`Yo&~>GzbDs
zeap>{BKgo}wyg#n4cQzouKST#axF{^#b6DN#uQtuXbNp4)!LiyU~MFIA>->p*}H8E
zDs-Tw7r&kN!)67ynI+e~lVAlA_v7c!44a09Uq_F6`^CpeEEx<g1TEGSFGXz&B50t8
zZHn2F<=3rkBXgRmK^kE@yOQ$8Phju_9yBA&YsvCp$+hqVT8IXZ`#!L3v7&w`+&NU9
zSePx0t>28F2f3mn9MyPRP~kNJD%K$^yl{K7g4@iJ>)wHXfruNKmL|4oSj1lRsBD)a
z`><rN{{l0V&iXGf`!6s<tDye^vu(Kl1!h3S{uh`*KG1)G8Sa098OlfeFEAq{`!6s9
zD&fDt4CR&o7nuDQn4z)+{{?0sd-Pw$48)QDRm`B&#D5jD|L+yEA(|f}J}VycOCvt<
z&ck1$_1u1hPc~a-WZxh2a}?p{zeUBPcUMJXs@}0X+BCVQrpC|n<Nl3lqhdRBCYk1w
zC|&rb%<bk}8^$}&p|YswnG?CQiXPL0CwEKE8Z;MrdCWItc!y69x3(At`{fq$y#2PH
zX-<Cr4Rs;%OuCNSD39Bwl1VDa?m>BSV(Lb<E%I=`&BOMdhBAYC1GaTDKGV<RlX$$x
zZkYFcinX@zNYHCvS)NJJwk<dk>oNQiFSyfq)BeAq_kZ-O#T?zHH0U1);(r+Ak>`*t
zmG-j*9y7_i=!70r-VruE?{noIPtoFdmhs1kEVG(>gg;U@qHEcd&>Mau6y^Ne-EB^Z
zmZPd<`IG+oLxp4ks$TsXeTe~#w96N-R@p6%y-NWbWlht{el)6*y)n614=Q00YN1uE
z*b`ea&-y@_$CaYm#aaL2>_)vi!+8VTRFuSnBxOx`PL{NEbdT*cWj3Z?twFgI%Ai)h
zIyHjjjDny@d&{j8h9I8pQ}v1A&L4dPrwrHlaOoXG<Z$7FP2rEj4^=Q}2fvR_FAL@|
zD5+p@>L2HZx1frp3+|5dEAOP1Z<f7#SEtBRQ18~|K!_Wu=WZSo6(#?U-l2j_R`|dM
zKB=Zc=bxLOO-y20q=m~v>1P?1BWm@Z)3oa%ixz)uH?e8IeS?}u1>)Vh{)Wdbjrk!P
z%7F-l)%wrH^Sibv!v1#ah-Fk}I^)K8n^vKx+x$fQijJNULM(6g`;GRUjz_1PjPTAr
z6x<*vu+_5L9+&Gem_+}Bj1ZEUp{gQj2q~reRT48gqC;7Aq!YV*Xd{aG-qBB(M3o$j
z)Zu@^9<&jqMX_2>N8C}uws_WsDtAiSmpq<C>|m6SUm7$G5JucIrFhYM??!E_OM5KP
zjpCgyUK`~ShOV_sV0nx<h}Bv<OGUl#S38-!p`4aC^>v?*m6es%>bkUsOU}Jl%?XFM
zZ{Wk?!|0rS)Q>lFn_+f;psYVLq<whZE37M=20kY(48P*ZeRd<wrQy-(zBrHJlkb>(
zJToOMHbMz$R#ngI&vh*umh*=R3CU>WAzh1zVek6EnUT*w)RcK2Y4gF%sOH`Y^MTK4
zeRc86j{A1?`ED2}RTHi46e1f9u${738Je=~5zj=6Fa7uQ0U(@Xjw1y#$>b_WnQd-z
zY^-@%py4=)(_99}Ht_Io#R8IPB?ylrse9hTXKxpo$55!`d@o9en^eX$U468%u$c;0
z1{tm9h_>Tc?+OJYdG%&mVow+HT}-^s@|<}~j&<K#a@C6a2B$Y7cVy6os0(I_)&|&a
zssd%CDkvjYc1e@YJsEp*m834#r7e0oE!t;;BqOHLK(Xq!f>Z}A=fQ<=JdBqgjP1o2
z5iG=)pdQT(^*Rhh&CC4#t%DLp0|Xleg!!Pg=6Jh?=w>VT+jn3_kk%APe)7aonl${>
zXVGf0wy}`Of)OtsU8jcA*`F-=H%yk`hMp{@ot)#hLvQAhgi^@x<+`%BS?D55v?<On
z+qJp%BrS{-NyU1Ozd!#u!))VZF6hbJ^<`{-0Y?7Wv<dvs);KK>cOvjbF6O&xBo!?U
z*{a?VwpFogOs_23$FOlk0rZ*icHWx9-C&s36%IE@<mzJl0gUn1+^A@!=|zY)8kEA*
zbzbsYw--tjOvH}d?R&hu(TrXM+yuE0Wp_^YVafF1_61<jm+$$xz{)S!mB{ZueC{dC
zx;cwIO*@xc{k+9S02s$A1ny1g5ejTK=1vKLG8*R<7Yktld(R$8+U+x2V!kjvWZqYs
zWGs}*iFstzhUgR1%HNm0amO3!y<ykz35Y#}d1^fgeCcEK{Vtrgt3M@|6ig?aT;;Po
zVV>+W<A2)k(nbT`Phc|NJ{J}i)}ZPhwTD22e*3%l4bu3>J@caK_W`Z8q@5)8<rl)Y
zKOVd5@ad#!yw|jWpX)1o#Lb&Gv+TaT$+G(LEX%MoP}nEkb1xo-Zh_xKdrTBBD+Nz%
z=KbHj1LuY|-t%3kLX<#cr$%EFz6mR@%kWpn&*x2(vMn2vKXSi14BTN`XP$W~jJ;&3
zqZQ{F8yV#-Kr_D+6tWYLHDKdb{>)sSc+$&(yqD@+j@5a{IqQja(VJnC*uX`$glLSa
zHpaHMmIe&^ua!ax1+rlQyo!lrxC=x&F0D@Hon^u1!(U%6NX(F6?1~JHtc?3#k(GN)
zETl;<L>yxV&;jm($=b>LOiS`nGLb<TIAOC~G(2UfEi{kykMfPVqYIE~q3b5(D}V)%
zeM}(44WO8<jV!#KW7FzhUbKLRv7M|{xa{<hGy}GWgz^&Lg2CY|<%xXm<m6G=#rs3W
zZ*jlq@ICL-PK_Z85C{bV2YqO>-2$NZJ6=Em%oMv{zqDXiMapnh9Af507T-FkSnE(F
z5!qc8FI9b14+G$X;75DSY)1N<T$JBVbQuSs1ce7xw=-*>5r}AhFz3aUA#ZwSte7|5
zY#AU_h``tqj0wG^@h@T=TL3zLcWMc6=!{4X48J@mO)3%QhJf$K9r2!d1sto%6xnVe
zc8oK6iw1Ere_mGGh4H~3bQWNLF5pqT&o^G&VSN@^B7I)`0GWEn7t?s<&dyF|OASnY
z`!9gq4M>;>?ajX3vlP)rFfDfMFmEq<zKkKN2xm0(y?S^l)dN#I|BhfN4MmIcrY?k#
z)sff}X6_7x@p2*tGKwTy-3QX1&!=O8g5WP#p-rzE9c6!~;5I7>_;yI;nZku8_X3p{
z+k}tLLH=UC|8ZT8bqfbw6&%AfAXCti>NdRpG?oISb`8SPU42Sg?+*(=%;qa4oI!A=
zzhIFVEym0KWRB_FTw85L$d|asf+gLw4nP+9M(^$~k6-~ZwX=8#JJ3<`hU>cTdc{!<
zRs$j8g01kvu+^d+@eswdBX5ItK?3~oJ!D<TDy@eY_{$QsnZFSA*xCxq^%GWrc<t8Q
z&L277t9<FTGAm*Z9|t{4%p?zYtsiM8I*=^6Ow#YP{{25o2|nRY7H9@o!R_&MSb+K&
zbtY>v(5B>n`t$zv0%w9wb6n^?GJ!GkKbip*;3OW2sVVlQ=hr%%l%ogsFpvw{+Sl<@
zK07)x4gQPB6ddj0tH%f{*ns~f%c8$t<-s}8s?n@8r*30<<4@vu3Q~=s{hFq_0QOxk
zM7?GA)NMdV$Zup9Q)F7Vt$#wGi(u=w+#;?~Qgf9j8mg7-p}7ZIshGdhVl@2hL&GbP
zHQlE;)*sDJPJScaL$-Doz|S`~z4!q&=jg@RjW7_|_qJBh;m3fs@#@?8ZtFEX*($8}
z+=9#zEH{${qVIzf5o1hQy3(XOjK`30dC;lF9C_l@`hik)S3}F(%S7uZ{?k!FWAH+7
z_c=%7p{0{Z8IMN2o;j!s&bCrhX`Qk|TtNFpyg{dhutq$aKnF5+_#^kcC1?ZG)2pOB
z1X_!q5;TymC;a@X`(AZ*66@5)Bj8NE&ku~VXOcn640EO~3xPeD)maQ+q^JYXQunRe
z@(0g)>aAz&%}N~o4ox1UVr6WP`VoOR>#4lL0p5e4+4>B|cM+?|_@+o~hhM9TzER>T
zXRc6ZyHXqkbv_;#$B2Tnj;%VvR$|InYF08-);m?j>GQ#I<|L5Z&oX%JE9kSlxNn7t
zYe0w2Vb4)DgQ@<8z7VnDqobLa?TQJ;XcL%ag=Z1@e^Cg75<d?XaSl@G?`qAHV{!;Y
z?mE_<H2?7-ye~-LlYC0Nx4Y<6y+&PA(>Z}zlTDt_B*<4mDuC%mFXejG_D^)cPL9!0
zpZqAj7J=yWqbvHg1Fu$kkR{40DvaZ{%e?dmA$$>fugRLeK+bgeNPTZN_x6Q0VS4%d
zo2Y65-2y?LWxh%9|8T-&=R5RQmWt~9LW^$eEwrDlOHW}`conyt>$hDLTt(nRsYq&n
zbofPQZLn}#RC-?jBgd|ra-TyjWjF4$m<;MI-PVBo!-Shey<9ngysP@*f*OL<mtOSP
zlve~z*Jw^|0au!Sk)&5AT{~HaTKgIPjX-6!FrX}B5|puS)-1=}v0cQ`LV~wi8Yv@}
zTv%1>OCQ0bALqT;$&@ma-y<=_9J2{$tacyFMw_XOWh{w<NW~2xw(qTPA#$9}V9iO{
zFW2M5xbQ?e;Q=h*h~?)8)-5?na}zy{FjL}<1ceI#$%?ro%=wc?|KW>cL=_ZW)$G%$
zb8-(FOnFW|{v;}$AhW}`JU$6u<g5KhuD<R&wv5jrzHQRLm7v+$%zKySrZd`FW`3>*
zQAS`w2d30akdZT-jX#JlV)gX}A2>hn*>fhFzKah$4Kjm0x3QijnZSJ)pe1K@;(osT
zs;_?ZHZQ*UPqvPDySTz-F(M*pW%%S^8Y}7BSTo2W{CQ{^4lvEBCDTvRq}2MJm)vK+
z1w5AHWBIw~)fcd*fH8`1+}~JNSZ9+qeg*we{Xt$cq)dtmf|TKUzof%1dm1JtRaJ>8
zO|7Q}(oQt8n;J?k!eHghekas7E#Xg7`E#kAUYqH)l*3N@C@~kC`Kb(_*iT10gI;e7
z+H?@*O2AVyAxYNR$diB4sXJv7`e6Yo!MF^fdzXB=X}atxk2F5Ge*OCLo#{>H7-j@K
zgnc}5wVU|x4=O}0y!a_l6;nixz4DXG(b~uoDEIp*Rt(b6BbAJ!WpNT7wM@&S0h2B5
z%R&X90Mr+LR_|m+(enJZ_Hj-SFU!ix*48E|)&){>)&@vUCJKs*YRMYqt>;xZ320x>
zva88n?_X?vmYA9zfk)ua&{Z9+3Q5Cs{7V86&jIarh4#g6Vbk^w>B713k31<Lft11^
z=KP>&c`9SaoWC6Y2dZ=6;Pbj}Db~(U4b;7An@WFnrvambg_r{9_>OxPo3zuW16;ok
zaApO>^^`!xDnbkDuDa_>A0j?po(1x*P!%_tPVBAlp+kpI8eY$9szJuH#0Yn256!-P
zC+UQ&TqoME12a4=n69y&8!mDKZk!;{l>8?YS^qna$Wq116E&X0147;=N;weQuLYW<
ztrZuRpLxs-XMwWB<Y>umrf%)!;|a02BVHpObw;5Q5ump9RRu?Peb5OL06Y}U6=jbt
z5Ap6qBm*8QAOt+*=+jNnH9A?Rno_4KRq*QizPQ0J&slXHYfkjl$T7u!yR)p6P=K2f
z>#XGt3ej~S71Mkp0i%v^kp)mA?5g@E9{;0kvR69-*%XGfKwp)PFTLa_V)p~qcu)Jh
zQAY2)<@tfSq1^Vi61qYKKnW<rP)DvRezUvsPUA*OkVBL>e48I@N+==Nq)B5L2}B(a
zsDC(Ee&JmRL!UtWT!rJe=Zoh0vn7QqN4@Or?IUbwX+bjMt7DyV9}p|&S9=^)*KH5c
z0nC*z4-(s!NSImxJPYWIDqMzfLi=-BW~bcyk|Jw+d(BMyOqN~(l!b87T;-FZksJ8U
z+`It&5$Dz&o#oh9o1@Vizg{0*fNq=6rM<ebRR0HM0tcPFso+itgwPIA9e^}32is=|
zM8@kc^`ET+boNLe0@Z6{(Vn{w?%q`-U=gF?8J-7Q-FjlJO{=BWv-TTeLVLnjg2Wq3
zcYpH9phi_cdeObsI~}hvc0C=DQgQ{zmfG1_SJP;dCATlekx2oJ?q;9cXHMYaFSFYX
zX1+@X1$CajYx_gsm}T;9fC%_H6bG>;f(-&NpD4FAF|-7`;g>zh`itMGW0Rt%dfJz#
zbAdn1n$-HrrD~o*{{Y-TtHY&eL02o&uyhhn3S%xf4PZl5w)aoFUVnDrjLcMExv+o`
zO*&kLdpkrHyI7@8QtP>1Fg<CqG#5MK$=V9|?)h>d&tBNmqT;u2kLPvASWf%&3a?j6
zWQ5E>4&U;3>A(Fjuq?20$nj5_aRYO}b8rvt5IqiD-m$mjbtF9bNJ!G^+P=JUu@(Js
zTcb8z_m#VQRU{Ev6x9m=6scu!`O3FcjM)}XCJU@*R8sjXGHFt`jQD=+1oAeEnqh)W
z7cAh8ZJ|gj@YW=3<K>vV1iikK$M?=j@IZm^Q|F;($CafSRjE--xA+Q3M3I@Wk8>YK
z%CrrnxML7!gHhP9b#NR>M7Jq=$&QqN1AHaVlRtV)>)FMVRV_UrVQ=ORm=4ANpqTsw
z)U{eYMhfb7FJkskUH~D$mGgo5B%t^^GygDd!x8^}0q}>o2k}LZ@3IHkbdp+-SFfG;
z?6y4DQwQXr8PLCK#2tZyFh^wgEDiMKwAkG7?!df3f^a`L{0j9h-D|Lb{6^KO%^UpP
zI||h0BGt?m$P%c!m1Y0qLtO4B|6O#o$;x-o`yS+Z>a{RE`LfE(W+lKIL&gzgK=&Ox
zEVFz9WF&lcIbp|j3A=Dbwu>I$X%AI=+^4^o)+w!=dV}z=m>F*E>k3m<fiPDHpK-Bm
zxDv#RW1&7uA0v52#<=t`>LdtJd=pWDXmHii{h-|LsEM*GQ=eu(*L%}|%wY-4y=_z2
zis`hh*JF_6CfFeE;xe#o7qJ3`Os^qJ{Q<Wx-m@i4$slT(T#Tb8Bx};|xUHdbvPx=Y
zae4O=X2$vN7(tuBp3w0cRTr<QXpElcSgD|p(C3L9k%<a{x-eDAFB2YmFt~vaqQ$mI
z!n$eHXRy^H|Fm(e)aVF3Tsr@%7BLG;8`(%Ffo8sD+nAmSN3?Pq&5;021+Hwz<8sM0
z*Y_&3jVQ&qzOSd1r$6a1?k#sfrK93@o!~<^Y4IYJPjd*mX%Ua1=T&W0S_?x=4xicv
z+&<4sj*L9(b(8Sy9^CEIhbeRiAij$}(*?=<+%UHCF8B)m0YUnZKqQDkOep=;a9%}5
zg+P@!xwduo>29|}WU$c-h=w!TTw2=NM!=a~93{Z4r^5CDh5J5>rTr@4J(m^D^)`9q
zp{_I6YabQwz-8Yq`T!8y^wV9MBNm$7XS)H<g|KTRUyL&ag?A>ZqYH!2J8KnrmPR8b
zF?GE<VEbDZa^5$9ytlVKyo;5*;~;(zs1rNNvzK4EW1VxIHO|%0?J0MmTW7GaVMj4y
zY<?iag|8L>8%4>_nJ0xCE-%jc%*{M}ekmPu5k|n|34Si(Ri*Ayi#nr4xVE!@xR{TG
z56F6v>d2DQXLTuAV;yVxkA*LZx{Y@o1t}j+IxA2zGA2+t`t1tjr+(CP!9-Tof>;(Q
z%Ve*A6&CO{w;C5qqjwGgqQ@k*#~Fq0hqbcU&H$80lKz?57NHg^awrzOdVkk`xWNhB
zAs?dNxt{pYO7OvnNH0wMBN50U%BSvg<353uiF=jejGfZQS0GR)wChdn19&-4;9pAh
z&Ne-b2gTqU@55M4y_fxVOVYd=DfC$5#r+{iS!@cXK5rHSb|41tbGkBsH#s*D9f+OT
zBRi)DOTq~V=~pMRcb)G`G5W}rJIMC<dpv}3>x|I0dFYytz$biiuH#z(QwYHMxRp|S
zVIu2JLUee<Fvlf}l=1dRvB#1QEb2S$dV3X#rAc26<!Fn~cK9(IJiRypAfY#=8ug_B
zHYPT(herN#@i?x>Q1;mhaO2$^Ng-yf?*QP%179K*DM9kmM?!-SOI|eJanx~=6|fSW
zZ9ZQJKxu{C;#hWje=zc`NY7o3F;_eyvU8cmUNf!KKqRXIdFCT9nEV#Vcp!BY$Il~S
z0iM&!4ZkyYn;A9cv_XS+!FLUhoq;lP8PrP)(<(8EHPWO5PoAqB)_K8uM$0@&NoAyH
z#oH=>tc(w&WR)g%9)fq2-~t@7llWX$7}IhsrbU+50-jLq(0kQOzVzdDoZ*|YmYY|x
z?j17|gg5~p$V0FQ@<U43u9LGm-_{tb0^BG@D%zV6_>K;oMj(Mr;avC8{?z+tJZo|>
zwv2oLr0YP-t)hGQB3&7#9Gg-^zcqo?O1N$iN2u=iqu0{X`u-zQKO|8m0JQ|X-Qv8!
zM~*`;VH}sv=Ea4ISTSMunc;%P6YUt|!k<A}(ptHYn7j*_;r$Tv%+~F9OZN!=oUluz
zIPtz*@J{<CpOvM@#70n_z$&Xoc5HcJ#N}{_DlYybvM6yk`ye{u^uXw3%*ppd+EKj-
zUPaZVmSCc_#hoj^;ZX!GAq6v-d>5Gf#l$DE&ICWwezv6V@%nQg&VC0bAq^#zhk2d?
zxPY9crRC&6qwU;{4MOn%5{lNiMGDq(W~l;o%u+wbYB9JOO&J6W2?aY?f@(0aeUpLT
zF31ej)YjIXQaB;i4kUp_+|_PzI=3jzEMFR>9`4=X4g^36m(A({Bsb;rB<9#vfO>FG
zQpZj`oS<McyeuOZPfOXQHl)u|G*(8}*<=(t#MX|Ye{Xw{PgYJ&+2w6`kD{zGsVoRd
z!F@3`WY3HvvP6}2NQ;d9<CcfL#s1N!gG?^YmRffD%I|P0rqoIO^i4c^(TG;y=Sq6b
zzqR$AxHH@IPV)09*-|N&7VBKs-oyY7MqAi6+>azdOj44Elw-0M2{v#gH}KWF?kV?C
zvzIe3F+N28pL1{J%)(7^-AL@s{DH#uKEhJ|t$crIP+wh&%Wbu>Dge_G$^mKSHQ5-F
z=9uWNyz$y^2aivUFs6W#0N>@Mna2DHVte@_IS%X~M%B@9J&+-XxI;e}TwZy^^2{KR
zNrr=(ixa>zz5X~o@6q?M?I2*xFCKfx#H5j;qPy&Ciy?6^VIZ9Lb&s-0D7%o(KAM=8
z6=Shm&>?{nQSV0|!v|JiIz?(vBkAcXKJ(RqAv9yQ2dMBpe%p5i0$6V-WfdP12eF$Z
zYvn{ds3wWZ2GNXtBiZ0gr9%huNVkuQvz)oLEp;Qj`S%P1KZyHqyNI9zVA1w|hAz8O
z%Cm3#QsUdjW&?cy%%(l~QfJv|@isUvegyT=E~JcMS<t?Z`lFcWnl1&vsOin&HkrQk
zwL`%lFetT!?}*7Vnk+~yq#sT|lI~+uWC_RPCwnsr=PjOOmeDGbq>koV=e7#P0bD7u
z9GxuC+W5;u?aN%wtxk=Uy5ksGPQotecJ7dvaQlb8^y<SQ&Dbr&kqfB+E#rsB=_*(N
z)HHJmxXp+4k5jj2t^Am%xi9C!_W|Wqnyt8)O%HyF-E}D#V<+}dGLJP_p#a<%`p`w*
z>rY&ONf}XUGe>Ct@ui1=3>n^2kDI&~hCWhjbbkPC0p!B1jvCRA3lG5)_99S*OnkIh
z!JihhNO2ufRr%6x5dKR~G)V<JRmDkVrU|Wpa>nS3?Eq!h(>fKgQV>r+8@$gszl7n>
zl+hiB{F(i-LDwL@%$vU<`LSvelPeyuRX3~@0_FbL@{7+@&yLSC*PV3rE-``bmlp-u
zn3$d4Gbs+Dn=I7@TH1&9$L<6oN_T{~n&ZQ#;(=u^J6km0@T4w?6>hK(g>edzm~JvI
z9K)=Bb)5&S5%(%mZOE5?uZ6!lcC4Yvtd;@Wv5oZ09gR>}K&WTe{!Ebn?rqG_PsVLo
z@e#2Be$Chi&WwvOro7v56$!TQW~7Hy7f`Z?muhV>_yB1MsV*CR(NYq7`4nmn3_tKU
zCKde=xKK-VhX|ovYBDEN#X~t{N&8Z3d#TiNr<4izIOuV>4mlVT;^c&6=Q!46xp0!C
z2s82-28Ok(@T&P?KonXpHG)4*yIzGMEC}2=%B3&MWHX!>%}DX1Z(cEc4KiM2cWm<d
zY6CzJ62)iq9dspJLIoiSng+#}x79Q^(=IK{@Dy|jU`*BpU8x*$CsMr+h-|;V7Ix=y
zkvi20;)+2DHsSkVC6Z&;Ro_;5Pd~R0<yr7r(3fmCyB}-_(DugfUVv`g9$ac2JL$Wi
zwIFu4JqB<B-<}nSt_zpqa|3Bo^XU_(F^+8g?lD%s=7Vrhrk;k@p-{@n)L17?n%nBE
zP47MXEu4DNXMYuN!C!#t41bjXNJH($KXcwV9P-4T6iheEW9&D<0N51b&(*SAa=`9)
zNnnRWfLr6|(RIO!6NS9a)#$qOmtO$Eu+Q9$F3^dmS9AZPcdqN$ca+!D=(BY83m-Qb
ztomg!(DjUGbuOLg@zcRb^dNLVPP*K<rIu3*DdR42YR{G%$5{S4b%r>M+prz5Y56^!
z)~2|f_D5G1CpvvexXX2#DuvtWL0Zxg9Tp<s9<Cq0zH~4u*(H1&Gb2p^6|IVsg$z&n
z($gjmM`L#+7Oco45R(oKzRJ^3B@mhHZ{Vale2xrQ*8WFVex#2SE)0)#`0eJNDc}J&
zSSWP>pDY+aBnzLtmfavMka<lvHm5pV&RkiT8i=9t|1ScpQ45}g8x9d7RWr-E6Nt^X
zrzNB}n3j6wqDj0!9ANnN+Mej1X!zw~bp)v0@aMF-4<zZ&Vwx3VfWt{t8?yRp2t4vn
zZDH8O$Mwe&9>W4kZCCiTe2z%OEzix|^J#C(c75QX$yTi)O*+%=<1<(=(?)qF)3K=?
z7}qkF!iiYaA`nWrI~ZLJz5;JKx|;^GnmGqxBOoX_sGSB2xTt-f7duNsNpzdUzL(I>
z%xf_qv#V9=gtoT!i5Y=ZXI#+!^Ys0pJ8y6B6w>G}CHdB#&Y$bP!)Mn~Jln~%a?bs|
zDK0(-MU;exa}c9M-kTbK%)??N$ZK|ZIK$YYzchz%p0b}kL<AZE>TY8${!gp~EJ-aa
zpcXi(i6NdY^7SL8s`w&GF@o%6p-k~^do+@ju7iA{2ujb+!VS90Z;sv2U&P(9cIqe=
zq0j-U?KPcF#t$O<)X1!!HD&Q9mO=AZ?><BbEWmynDT6Agz%rjIiq9ss?`_Goh0;CS
zchVr;_5g=nR~(cZ&<wN#SHPs=nhdE3{=d8opXh8}s-i}!cfnu8Bk~tgp6%Y#r52I=
zj=Ns8{M(DpkQ<=qdLeosR=`wcFmE7zZ8e|@;RgA%H4XaR-!RcHbJV^4J_`*Vp+gzV
zi^VyM&xP8F0c-15rA1*QTnSOQ4;lD*lI?R&H-M$ArgucN_xLbMI1e@fhzH%8&IP%6
zs-YFhg#e8419_0mjG$lbZlMG4<(rQhuP+d<(`FdQ`s*0k_Q($*-A=kG&t*nPlY*qO
z*JA(Kq7Q?!?JG;|nV<su4!jrZmQI9Eczh?VARE+<4kaY*ehmTK9ymGSgPeEZy6hC5
zSg;f!l`ntdPRx}l;H!&W)Kk~f@P1;9EIHDsLk`v71jO{5zZmmfcypy<LJ0=%z(br2
zJ5)M5)_DnY^01$i$NlBxr;#PJmQVKFpyCQ#URp4<x3aXnp)B;}0H}jSn#aAyC-j6p
z1(i2h;152@X|~v4`A*`iw}5Njk#*F-b5)^B!4mikf*8OF>>2D*pMj3es#CcEbz?sN
zrs_Q&cC_O%-Wnf#l9>4lb-Zpo0;07ei2r*kfE)sKNx<$ZE7#H2JKjOPu5NJ<3rr5R
z2tyVK(*7Gxf&5>a{{*rGc!B$#n^b-noVnA!c*viWfmSi%@O$w_d|Z5%vS2#uFjiGr
zX$&4xKC~tWm;X`@sH92jQdC1h%)tM=9fZ_$J}Wam@vHj>78okZ&<I)4kYbn%Gk!$<
zG<YaM*CYW2V1Y9BI~Zd6Cxr>W4z~!XQ2b!|A+&Xu{W;K*bU!=$$oynq*n8Wt@1>cq
zx6{4#S?=*ML3!s)g!E4Jr!GxY3U^`z9f(5|^46gMjj5u4Oqx{0`tCFA?CfpA2hQdB
z5Au^2n_6K3Am_zXI9F+Fl5<DW%N>O8jw@s^wb){SMxlcAwo_Oo22Km1L$b<(8DW3N
zf`cGSm7!MvrMviCM>{H<Jq9!7D&i!>P}b^N6X-wfZ1Wg7v4I6R)>z=$U$TH0Kyfp9
z+I<DL5=6Jg0bjex@CMhK{F!{ZyAR~RsW+Uz`r7%Q_}lX*vVSqt=3xX9e3gUF&E_ek
zxu9DVr2=7w0SwO9faghp8SV&X!Yw>NyOsSsKe2tedFSmd?$gDP7EJY=a2%E`(|z#A
zkst3Y{OCd5m&&2%Q*gSwR^U7UOO_=_`+bLGK#^xo=`x%EH@D5*lK#@$J(NL$S1_17
zkfXZ6TM4*9x{=)FiRoXnJ)%G(_4z!^iXn#b#`r#3ND#8b5EtI0dMN&?85RIF9DWiV
zcvA5ZszG%=2vhkWUIpm^p)VK1T)$nVzL%6lfaxy$i{$X7KLbo21k8Do{aGuzGN;{p
z!4=@31W<v{wQ1OG(6PR4!)d%+JB?$1dm6s<Wg!f7jk+$nCK&-t-bZDFX;(Gz6HqK2
z8_-`7ZT8dL@DRdEt~EIb{YyEBfq_TBey8TM?<%j*i(mNK^BvSBtI*_!x*4J3=qb$e
zM64+wg}+lkQAf%z03{O)?mG~F-T=90G=gSo`AgF+(LFKngf9eV^u4CctgVBOd_H&o
z3;PocG!*{w;WZLO{2dASERiL**)_gHE&Ntab3Y>Aoi~kLJp!dDUgkq8cdu=@7-8`Q
zEL%bd?B0uyK%^_{(g*A+FKR5vX6A`o<1yn#5wL)Uv-%PcED=hvA9|dVlhXu{K%Jl(
z^GUD@TpGXvac7j2kM|QGlt(d*OtJn?e?;T&{1Fo(d#E1V8NV_YFCYeW|7QX*Fh*6#
z_dum<rD%QvOc@}XUBP`g1Rnh}yOQ?4kq1=38h9G8tAMewLWHn)5~`ay?@J%&HT?yp
zg(v9UdlP|Tc7ZDN2oq{EbMxyx@jjA503&6R^3)r{DxZ@@pT$Q(R@EXPj2gN1o31Is
zu%H{=<b-_4E`L1REPj?_mpzEDCBFAat+<&)T_vF|^QW_+S@TOOxkn$WdT~G~5g{?Z
zDGV<hGVYLiR-nBd2}E+dmfZ<EQuOLT3Oe>|3;kKnUG`DH8q>RV-Hbj}IGvqQdV*T_
zHrhK`9FHst0-&rt8_xoXo?o&)ZqXBjyzzRN&3c&DU<*S5DjvW$9eMKE<mH)@$Py4C
z(WRxO)ugD#D;);6cIa6>eZ75*X=RKlx_cim5pSXEJZ7N`4F%@V1|n!9dARYOrzdkc
zGn1G4!HpE(V)|c30+9|y3T>wgD(wnxt@BX2XCr;)B4;a1CBx99_zJ3pnZSzP-*CDg
z(O&9>msCGGNkFN-H;e;27s>G;#Fgp&*8T&!=!^_#N@nugvHMd+W}}`*DFmQ`Y<2Y@
zZi~$;Yw_jT&dN^)OBi%eTjfe)+VJ^CkTV=X&SQ%1uMKZa9ZlfuGEaPX#66vw?n_T9
zF<JcrC`N((;5H=^KF`@vw6pLX00*vDd3Fo1VUAu*On?ehp1u)ET1kQhOnN-P@=}#>
zl%IGj=7wZ4$jd-?)BxaCe8o=v`P#_uActH5nTSg6r!}&4rQp6LiURj33uod1Ovn~L
zT0yYYrPdcwbs^mY>anap0gszB`s^8}OgSg2bqSZLG|gUBpLq=^Tm>k2soQ$v(KrZh
ze2@_udCQ)1x3-iGKp5BYMM5-4xI8uoIet!%8PBtC9KPU7Z(-V}f^lenOrX-TLx&=(
z^<aGPlP^6eJ5qGM3$4=WpDCOv00<g#$bW=CC-DSV00ir=%s19#=of8LSiQg%gN#S4
zAm<Oo|D^(?$&F$SPM+a%duh@kpPQcG8lSP7p`HgOzOT$@tcWrKUwV~6q1Fy?w}i+h
z2Ccq|Rv39^=N?58h&*WC=Y)}>`V)VWm7^Hcs)5?U@g*#vGgL}4StT~cttM1_Tt0|i
zgdg(b1z#0EdYn!R6v^qUDT1$JF?43amo?)htoC=tD}uz{b+Sgz2zburR_epv(Y6<H
z5r@cUfv%*7n##h`nlbV7;NcOdNKUh^6u%pkq7sM{)-lU()e~A;)$kz3Y&1CevZyrB
zm)aSG5wkyde4^c~i&mCKjH(l)$EI6sP2((q`J(Ctv1UcQ)C$kavfy*SY)rdhj24m9
zC${f{_d@6v3QWl9pBOZO10^Ea*L;<=P9aOO%<JwgEj(F)vVfn&Cn}(dQ}rY+bfuJ1
ztY~&01I~J}xD5Xnrw7K9hSvZNL1{pPN2Ga@wY7ET7n3`sAa5>EDG&t&BWtDS`|~PE
zI*@m;-J~p5RhHU?PIWqvXUYyq#Y}@?YPik(33q}lScVN{Y0PQbu2FCtBHRF2T5)3$
zTti0-+*nm8mv82bkl8IEXj8N_ZDP8?l_-ayUCN!6isq@XfZKVAR2Z(tA9<vK1PW{B
z!zoO=p8?5SiUy5HJL!ZDPvGtr8+;1tO|q*?5OL^!n|kVgOzi7<3_S(&LrW#UAWu@_
z|59qw1i3=yiT<<dccVXU7a7cL_Z|z2Uj&MQic5<^#((GagNhLs(QMfS^5vc2mNhdI
zTuV%~&4Ct7wG-qTQ7q-mSGhM`gZO(&z-I^`6TU?GV}%r~;wwOwU8<JeVXRjVFn&z!
zvt`1oigI!Gom2=O>CuEQ`!;y@{KV)ygAty_KJIenaFtZQ%_shwci=N9FQTA<!pIqv
zA0amms&qy~L@1E}WcQ3=Ig)v-|8WMCYylo-lf~n-?+;VrES$G*Fk$~DEf@gKi&Z6v
zi2Tq+z3Q~wHd^R%5Gc$9qVd(PWY?t%zITF~JWI+|4RP-}yAr^m9;2@Rqn*MKm4pS9
zXSW<U?h`N#3(&TCd60@*hsH9az`2_MLfv1B7BDqIdD>Jo{1@f(7so%=ymA>aF?os!
zwM<n<8?Mtr&uZ#GB0LW-4m&Vzf)GkVD<6$}MS_(~`xF9Yx$?s_VrAm{4W2TPh=?-|
z%)}>LJl_F4j37|f58mMB#=W=aS-T4$Rbh2G*_mTuU^#_5u)HBmjRd8{4VB%B<UEii
z;P`2Ql*)7`-Mv({?-D?mpu-QUb08ID^xBso^8?3YL4l15To^4Z)xjo0_TV-S*?te9
ztvxyn#runO5zR!9(s1VdI1-y)INuL-W(e?f0QLw(;O!`by8J2}lws8sKdZu(N(Mt1
zW}#XcZO$B@CkDrWw9KjTI85Xy#&2)IJqbCegBj`{cgvlI`+G0V8*_X2*FPK+_L&fd
zDkZ>wnx7eABB45b4dh53NAG(Q`)dbayLrGdf=U!NLRqKk2t))G+8i+S+8<47{TV3y
z$=20%Xtf4`Ul5dxvQ0ms&4sWb_Uu#C5A>N&@#*i@Rl?L%Ce*9Bh<5qE$VG@VaZ;ch
z2+#ue$S&&wK4BKg<Z!DPs0mQhSv}oGhG_+=xmL;jI5zms;qt^w({b3<?GhqT<o?kd
z)2F9D`-a-EIm^5<>!RhsYHLKM(YrFIM%~7tg%=wh8n(7m@Sy4cIqeH}N;V6{_^3?F
z)MstVAfYQb2xr9f%AiO8!xTe+O2)^mZ{z2!=9+ABv-pI;of<SQsu(5sKXSX8B}&fR
zy(;aBY`q4w`MeirP0l?!vNtul7{XwHrhrFZmcCxY;HVgFv~-JjG5t-7#dGjjLRuP^
zP`l@ZD`>mRqv_)Zs6k=~{hSGcH?D<n%o`Ri!y8uBZC-eAEL(YE=5ic%7t1A_K^<;G
zUd#_ksf909hQQj`MEuZVMZ^#eKKsZ$RxPu;p-azZ1C^W60UNSzM*T0G;i`As_#sEX
zUz@E7^{z(+LFdkNfSV+)0hGJzmyE>?Tgd~FE=p{(pI-_CVvBv%9+(6mwmRB1BU8Yd
zmLZvCzJnq?=e&AaD7+Fs2c&bLR=>Q(mD^g+cjRHBGXz&d5c~ONQ2qX&Q`&P)*HQHf
z0GT&s-LJy}PtfcDHF(!S0G&wEU)Ug;|Js)Ccx(|br3Rq=xBi+bMptkL6eHY85mR|=
zp?xuw<1R9&T8l?0ehWh!P!=7YyJ~CFoP}(j!2-&@cmQtxlX71-JVFbVcaN}i;6Z&2
z$%+7%GrW6p{<-0|H&+4V*Y<$+JaC1mfP)?i6(uu4DfEzb%9~#fMZE@EE~f^)cf}tp
zPxOzn!K|A9X*mf9NIq;~SLINF5W=4j^lI<F1hNb7Kvy!|_@V_<GYFe$#GI`>L0cBo
zn^4r;67?%_kU)l((DFV@(1nl^tbwJy-3w`>$lE>Wi=f!DigOF6(vE^I00@EgQnv?+
zSR-|v&))@L&YFgJPWxB#SAWH9y#S{oMUtDH?Q|)iSr35DwX5=4K{>89RRU9i{UX`c
zh94e^Itw}Kfo>GZV9W>r=;crm`EmOSN539B)Ptb|{^4`%tsZVIUsq`sS*-w)>ff8b
z^;gOfjYqyp0bQ67M8n4d8qjr8r=~{QrJ&S5jj)Z5<CJO8F|MzMKX((O2ll^`lbzkr
zAdi&MeQ7zvD@~gH+!3=e;`i|Z;Ot-uh|TwNiKA<%^@<%)Mj_V$H1DBa=;H+kaF=!H
z3xK&dYI=IX$tP#WuUx9x6|{uWj#1TjM1v&G<zA{kvD`Wd0OGP2C1JP+RzrRXj?vHB
zwxyu_L~dI+XA&xzQX{Ny9De%yEk*9@zy;;Xo1c)qZ#m`pOVC;=d;6>ML3rEhbv$_v
zT>f<_N#Gn(MzlbU2s?2cMkU$AEcwt7x-RLCn{tfTZ|;}tPOOiqKk7c&8zt&w0kxwX
zc^QA_7az%=LJJ{#;X%s!1LcXh2V{O8&_8<71&8#u2uuTg*+y?fanL@2?*Q~Q@L6J9
z$SwTznd<=AbW}xZX%Zyg-)qnH|1$a2NvoM7aD68pzQ}_w)Tg)5gumf{a!|x&o_I1v
zhmIb`57_~)<E<$$mighjVeX(X-iX?EUSINtZLgMH9_H`cHvp2{dn)XH9m+ppg6ojT
ziJv-Tj9M@vhDgu>DI;cB<h}GAYO4_VbaX%Hgt6%Em;E&wl=FjZq-^1?R@WTh;&w!1
zS6={v4HORR@%!k9%U18ltz_(~Mu=L}r$z}`Ygy%dF=5c3`*l4WEO$-6JwEU+=(oF2
z2OifDfMj`syN0ffb-t4Wb@vFp(O*wZK-^sgc9`<0$-M+^@&^{N>r_D6LR;+7)<E|p
zCqUTYk90%%6QB6NJ;9)DV0N6^>lXwZwqtdNuN9x)12-~0kuP=mMeu*Zz~iXG(IU|!
zG!p9~XbgL5m;vrU5rG~za1?aX-anh6$H~~G{fj}s;iCXdHPlIupW%x*9_$juPFg)7
zX%%O(c!VZS0|2r2ezk$374y2}(rbJ6i!$~8y21`|Wi`E7{qbfJg*55;te{^q^d~O>
zq6RV@s!P>{tM5ZCGc_0-WIDrHPz{`&Z!BU2Z79wlNO?6tB>PJ-m_1$9j*G`M7hI)D
z8$Lw;GMt!daBDKdKx{)Aosx8)7DYHU4{(y<ccuX*2nYs8yc+W6&^Jn^g{F6g18KP-
zP_LxqQlBbG7Mr{+*J_4_4XhiDR7n3Tw@x0lUSfcFPPF418Y8d<g2Lz2U*BXLsq4OY
zp7AFQ)nd^TJ{mA8F(tA$b0wkp`Pn~uk#Gp*vcJFg6%BSca2TMhzjYtbX;$D&qkuXZ
zAS?vYuZJtZCsuhWW#rypz+mD+QCyz@UEWFDSewU{3T~8w;)7D>7VERQrtV~G4EnF-
zX$iwn=ME2n(88(bi6Zg<_+s&31@r6L+iQtuGYfQl+coL*H#nDxn<x*KLIvPUziDaB
z5L8ABumW_$>6spP*w2l*q360ia=P{`)JDfxik&e@JqKAbN@^SQkmV17l-w2KPJC80
z9-i?QJfY&2+6jf9%?tld42?={z`p?!Qluw2=~n|tCnJQVZ;c-SExfDJ2%5hksGwTm
z*^~MpZRhAgft@5K7($YGR_6tJNLUKg$kqJf%7?bqysdmHOVUrtUqxk&fNR(OXKdz1
zcH}Xe?5*Jg4+zSWY6P9Rs}}tO8UdE*nQ^NSB!~N^*)~33;Wwrqd;~8=3EK-h^pJ2B
zsL3Pi5>Ys~a52$@?JA0Cz^wp2Kdm5W;QMPYAF_gPva^Os_I9bc*zaC5Z;q?Rom)Gi
z8|Qy%sd$F+C@h!C(7{YTbY-XZyV}QRF}e}fQW>oOhPMuEo!X0t&tB5t)>f6_HHJfQ
zqxr4c_+u?SP+!=$0GxsN%`~mu6{Y!+!a(!84V-hI0<DV~tQ>1F!2d6d1|o^p_Y!L@
zn*Y9o3Fu+VpqQa2P1-kg{KTq=t|__Le^to=5WU#TpYclGYN!^;rRV@SP;lWM$1jKE
zr>gkPi}O$=B`({$6RmUF@YP9C&`~3BL%|Ku>LxZnJ#<I7Ir*UrXwkR?I_U)~-MQAT
zX-4i?-Q%~k^U}cp=XUw~hUX{z$vlEk6NG7`MB(Tx3O~^UaPwkq?OuQfum#Ul;^PA$
z3e>gCpH{m5`6iNNYl;@M82&}Fa!t{ygKHaCMpml4h8->eI3Sc}A~v;>{zbR`)f}aJ
z(Y7P*03uE+orl@vx2C<;wDYg8qTfMmN0shg(=(!j5_g>h(3M}}zv~8wN!cNYCZqo2
z9lu5~qgrF1N(c`AzhZ%Z+Q@{P(xeMRx%yu{Cn}+&3)QEFb$59Su2x=xJ?ZkstfWa>
ziv=+I!v3#q{WP=3Af>B%O_m+C@dP=1(C2Meylux@5cl0vjhB>o#x4BIJq5)E6tb(l
zuZ8X3P`>c}zVxyo^bp=Vu+VFP>Qmji%{^GHx>B?Sa##S}Mi7Hu<Ues&QNObn#iYK+
z70^2B2qoyjEP=9Es*-aTcQS}(B@fsMtP{HkwDWes+}3^y1a>;vilFB7>u7&(z994T
z0hPJ2N(8NGK!6awg!N+O-(IyZ*+40y=&3dUOsa5Dl2=SfN;;V4tGO2Vu&p}h_VdsF
z6~|Z&^#5Kc@h^p%8bR~iaP1iA0uP<84gmE$sDI(GUjB=ZBHJHu@tRd9(1_Syq<8sx
ziO-}W3S=+5%rj)2R(FlaBFLM6eg*>dQUhoEuR*<WXlQajGqgZZW)vzMy~^X5_XH$p
zS)m&_qxg*#>KAw-owTBE9HFF<aCd~RWw{3Kp7=ajX!G-;rBTFM9AWY|7~8qEI0E#>
z25iII{`TI98~1G63ypVC^XSnVl?7b~PXsC&q@fPn{}i;fl|sqT=aoX7pYA<bgFYj#
zp;xlkjqh0mz_dM%8pDoZYEc{-qwomLq1oiA;NcCR%f+bn!O2*vW$@U=BLHRE)JH>o
z;~+0+wu6dM_dzUBsuk6+#+bCLc-QSkWDSyfNSSQ=ZtOf|c}2)mHt$RIN4Z$_dI^jG
zqp-5H4Aou_-LX=4-Zs%w6$%<3#UU5-nUM}#GTftB&Z9cu*Yl=8zalCd(ma)u?gD(N
zvo^aiZVFpdIICXVX8TfY1Zkp?%&U3m%VXq^L(7N<_u+M>NG+MCa7uzq?00a3=!j#`
zjZ%sxs57=L->p>V?$0a}?E&-}wigh#dOd*!3=_womG(asV%RQJkehf9eJ63X1MU=|
z{R{Bux9{LqPM%lS0BS-E>W$7QnAQRfs1({A(A_9N)F?e}o>}waQ}3ghqQ7ZTOHu(m
z$K*)sjsJ7ko9O@?;Z77p;V6+R8a+I<r?(G0H3HOpeF;eJxs0?KQ#*oGSSbL)o}JDk
zDW!e^nn4Z0GlJNckEN~oN@}cVI|ORA+1^300IeH&On*Fs^{YXYp>mLX#Bu1XRzGVs
z=nRIyYn^XMcm^G%@5=@U0$lowIzDbH?T!Gc#S5PO8u1*{ENN?gzna|Y8ZzgyJ#T|G
zSmaS-xdv1T{L<dlI%@y^>M&_ifG7+kAfb8-dJ{fycZ2)0u7MLuA9SLQkum75h=!n;
zN8TBxt<_1?R-K*t#6F2JLhull#Qr$!q_iJ}otY<bMJGj5GvKP*MNm*=V3}R=62Qfz
zeCaQ{V^rR~IA?_bP?A}2H8VBd_ML3g$Nqm4W2V(bA6$tfX-qv1@cHkG5qRJlu{Jib
zva8=optjgc;I>3+pTsk0Wd}Ni|37f&Ls$Tq^Qa!UzqEI-s0Dhy**o);*fa50-4&kU
z0wk{OAh_}2{pXamE?YRbb9KSNb-}&4xw${@ur#UDkM!0Oe3AN3wXy$HR?%Jn<N(QZ
z2WO3<%OGmI5pl%!vjIM{0Z^__O!v(hZEMOZ@d>3ABb~!J#i;AW#K(?jQk(xYa@)}@
zgo)dDq`4T`%M+;rhcSDL{=_T^W02;u&A&#BT06XocoN(~*<R>n1ZroQpS3K*xt0N(
z>Kf?XdyeAL4ghEgxgYZXX#47bF4L^-83t59QISweP!s_H6=^{f5eyLNMnR-ST3Rd=
zB&4NNkq+rpq)X|BA*D;YzH|QoN8Fj&;d^KQ*j;yq8J_2U&VAxK*SQXN6c>q(=Vq+s
zUQa$koGat|!r&;xYjd`F6rz};aG$udj?f*p3*2Rhx@Bm|RRDAF3d=a;b2*sndh6@w
zkkhc*>uY_qOd_fF|M)BlKru|UF7l>2{@SR%EjbXTonek34TV;gQy@5p`bm?d+D|u+
z;(zc(VAgk2Y%jDnlqGk*n{zySnwzpmH4z8an*|_21JH1aZ^BRE-pB~~A8#~OgGocO
zTwsV<>B97Pu;U+nns6G_&p0b7xo%GuqQMOp2xOE@q|lZHb$lDRnrdNA?2F|p-7GvM
zPe0--g7UHfjMbCa1dwcASWS=ha6&e7+;|!=_)J{QZNk%+P&f(P<#=Y@JY8_>KwBZD
zUfNMc#hMK%?<LJ3jW>f58hekNv5S*9EuAjNzk;NbL!K<-CM0rJcpa_SW3fMTNfv4i
z;<H-^Z~ohv@RMO`4FN%7%!dvhz}7z0d+J_WI@0iZTR4`twrT*J2vhE&Le)+ACdm;&
z`Aw0W;1BD?hvo$S&Cwp%+Qq0rwY_li9&-|2&f4oIF^U5ZfaI@>fh{kWT7p{M(CVf@
zQg4{>9t;=ew0=i6Qk~dxO*Wve4=k5bTtIy>*5CZUIf<=BOAR3^pXIru)&a;AQqbQh
zxR*6bf;Miy#Sw++BIc}-r1gM;#tKloM3tpxX!DG5H|C!0yhD9qKc{5a-7TwHHh=6d
zpUUnV<|qPP9oCOR4@Vp7J_sE}ApEVAnhy7qqd~=X1w}F$)0g;#B${hR9T<5|-9*73
z&jNd+n~{I*rU>pMim02c+hpfjV65VX@H+egjI$I_!`28eLCk~1nyZYSOk=>(d=C64
zFN}2(aP_((#_W$C<A3rHKBsP9yzbALSo;yFHTi%(z<Lk?nlKT3oNGS@5x8TR`R@|P
zS1G%pr&9=eK3hNe-+vO<H<+G|q&P9gq9X(N&9;c|c`*%v(up&W|1u$jfZIeZU=We`
z@euwe&oJ9*yOur>Xvxj=bweH17;`_7sD?q8-A<ApGGE~h4}qIQA$Jv^ZNYNcAFAuW
zKLE^W4Pro3C3?^GHf+l-V~&#Ffd%GpTCK=~sUH|0E+`EKg|V9$n`{D==pLYFI>fE|
z%jn`nZ&n<2GQdT!OaU%J3iZF5T#Uh{mqTZ0krD#5zv&f*dQ07V>LhM3S++yIrD(e{
zGr883wU?7ID$91c^C1?NDeb|8hsTaExCR$tksUa_Lx+k{bL2~Llj_k>zRR{Gbv_5x
zehp3JyKwr&j=D=~jyOCd2@i#d_lJ|dN}4c>vmGlK>&S8{KQlMBl&hR;+V9V*kvqWK
z<6EIKrs2XfmR_S0NxVsc5T|ubMSB;;jg9KwKwmNC%b9H|E?VH!%(|IdL`bfe+IUA3
zxGxYiDjJEP$XIp}1R1;$THvW>>=l;}F7r^3)zp=k<SOonEMdmYvkw=e)~kkL`0$D2
z@2-IKjdtjr-P`lh-|)jnk4=Ab+>qvr3FKFLOwj;B1Sn2eEwy%5DdZ!06<&*PC9Uyc
z-*C{R4@M|-5kF;fnu)&*yJi+GCn~$=>g#p1B;<{F(KgHCH2xOlGdI$_)}3zt5Plsf
zgntclaqex3;wz@I)J;)~;#O8Mz}E#r!fAu<PoSwKXuxu5W;2gDLL}UZ_Us8hIB@J3
zzTt@){O#&O5W`EjnLnKx!F*&pw*!$dlR8~}{xF(aP<DYl=15D54k;Q$<hd-Qg|7(g
z<VlcCc=P^;1lI#Ed=#y<_UJfK?+MH-ZknwlY5&zxIS5eevAia=_g$Zt<8Q|qBJ54*
zapoxkghZw9N<7RXBQY_IVekW;UakAni!V`9(KSlp>sJ{JK>Xx}d0Hh)O2Ke(l<~Li
zeUB4VcJFiUD4#RjSixy>Ao7}P!Qr*}V#Lv!3cSBC)t3l5!JnSs9+-0ea<c^sHW!F4
z|GwAu#ONPM1a%>nmSqu)krZ-VlZpUP{;JS6>*DVH!mZc(3y;2C7t$2tuxr`D1^Yyf
zyt)3G5b2?I9N+_*k^!h^42(9yV;~(_q;U<ZA<;F;k@cIe{l7R7J=X%LyhuPLDr)Ku
zDwJVFH?-_D_cfU=Bz1!jEn`wm_3swCZ^J_G*Y+~pbQNGG*89tp;ZHBI!wrmS1O>0F
z9`lc4Ym@gPr0Of!WrsT&lZNiRpKPv`dB<t)s|sCEk~)HjamE${g$s`{(9Tl}Wq_LB
zxD1N#+jtJUZbHl_iFhl&Uiw(twP150nkSm%kT!QisD;7GOrs3fWsN~RB&CHW;cAqE
zzgwrnrFE}GRO%#=2)zkn!I3|xeg7T!-L>z8D8@Sd9irAj8#dBN5!V8ETk`X33)R!2
zxpzw4G#lfTKtP1ly!h)0Jv&(PK$B@UuH54I{(H%>f7kj+97)$Q^aO<6+kt&M2k|Rb
zA)RZgcZK5(u|mxG1)E~A!RYcGcM;Vx5Be_vVP@0Km#eH2S}%Sp1>gA|$7bU3;ZGno
zBmU!KyxM$>3<Cx)D2YR#TW|9*=rNBWakGB?G5XPCp!Ejbpl6}yc>vP<j!D4s)<4L(
z#jjxoAl>RUh84T1{wa28{ho$AzZWQV4YVAip6PA#85l9oAS2$l{tS1q*l-PnSC`HK
z(A@+)ifK{Av}%INXL@!Rz6Ygg6Ad5Gzaz(N%RN&ceZQtPN5$ILm=zLL1#>{5G(<TA
z-|Wl<f~U!6w<&Frf;p4MH&2b$Tf&R`B4M9nbs%p~DP2%Zl0PKZtgvs+nW6aWNhOR#
zrYNXmZ86g88%El^=Y5jm1p>xn+oWTOwZWm5W~Z}Bt>dCmr0@s@pmswD6OytWrAXeV
zbD_!%iLW3dRkDLUxJBs(J)xk!^ka}Gdd=Sw2owG%lE^jHp(vHUI{ybx?Vsem!ReS>
z!$2o$eOHDgi<pwDK$hvK>c=OzU*I=pySD?rX#gDYevA`EWE-S>7?C6VJ<a>yJOxG%
zA5DjP9xxee;{vUj9P~*<?3~zJP8`1xNWr9#da@Enemzr%g%QH=Dk#%<)gcVqGrxV@
zuKW)U#&d+6%UZY7UhF8m^;Rx40lt9pSN5<COelmxBgcHn#3MC7!dsaPN_P{jAL>ED
z6YJ5{VdD7Jt3W;uC*7_N7X#e+W7Bwj&FkI1V|1`PtpHf3hhRAHona+7j2CG2j+FBk
zIbW<wUp7b3F1<Y55PQwYPI{B*eL~glJul+6gmZuj(2<<`+tt6gfTD@aiE3ty|3ATH
z7X|7MmSY)6^=CS60kXd5OsP$Xj_hyt(z%8hfqXr%5{0oHM}D0R2jAoQ_wRB2-}xR>
zI9$dfO{bA_U}$X3gMtX;Frz5KlF#ChaooeSiU=sVJ^Ha?H^mjjdN;h6us$pW5VZJx
zyQ(~oQ0^Ho7O1oy&<wGFO?)q-&kL!pOvjm_R0lxsB4`y&L#mUBRZL-YeK*y)^ew@i
zbR)Q12*6GK7n08s0KuS1hT3$xpiXDrxCGKDmw+@ng>j(!k6yk65$@U#72&^c5a9OE
zJ~9~ak1R`}vYPlX35;+li0fK--$t5Qk<fC@4=tgqn{j>62{K<{mhGxyNkW2jt)I9o
zd3|o~swg!y=s*^aQYq88G^+vK0eb>URrnt3{t{_`BOQ<}p$zN0F);i)#z6d-0ukYn
z?6&9d2xQ&};$6R)kc#VMgLYp;250mpHOIqs!&}$*T?03zyj`KdNNN;G89g-C+2|hr
z{Dq+WFrIfF28hzLhZSL2l5S!l$!YKr*=oifarr3UnvVy`fghvSzy2w1)l=I`$I2Ii
zfHRqhAA4n16-ccMjd$drc&=mz_k$TWNrY(v@;t~~Q?`je96Pe%D!&y$pYmUDlnL4x
zv_H&tZB;5RQOJdA5*nw{7?VUg6rwz%@3)xbW)vUX14`2Wf=S*|2esYHPTe5s`2qkU
z?r}1}Y*w0}4nX%($qqGcLHP49oDoyI;2(bS$wr0zsslEl{Tp#hdIv_P7Uk9qjdPG@
zo<Y;}>%kMGY5AA|cHeRm8&Cn!pvb>=5(UtV+mBRVJ7$<<6R2BK^~j+hEl1Tc9+Ij#
z4gCgTo5juM`7R>+jl1Ez^hVtC5e365&3}^a|M4?o9+;f!#D3y+c7ryJD>j!RU?qrj
zAtbd%NyT~p78)E*@oD}nz8^`TejD%1lk`%f#Ke2xiR_>et6xjK@r!hf?9F)C1w7=L
zb)OERDCafS&02BOcsKr$zbKDn{m`>m>B)0)V16QUuN`g4ZOHr$KH!i_nrAXAWqyKA
zoI>BhfC#$JAPIpWPWYcU-$*nBH=l-(WHW3e|2ts=1QTUZD>4!0$Oe>JCw?zbg9dsT
zlpn@8=yITu(?RCXe_T+8%?o-Nr4s-5OeO)<UXV<A4s(Mf50W;BG<;cAa{ze;D_~S4
zwwtzkc67&V&S{Bte|OlN|ANE*+=UG#$Ntfl6vskGQjYW#j&C@Jfj2%0qyi4(KXe6%
zA~sC1to6jjf5A_m<+KAhf(D3ytz8XzANK2KPhK`2ZrJmEbe@0z1~POc`-K|}j1dh!
z-R{gB9>(}Dnw#PhDc-8`=R!Puq)B#GtA-Y-av|9+JR7P5fL{eOV=+p{JV)ygK*m$X
zyMR*S6BeSjff@;`U^Z67X2DICx*4vP0Fu~#xN;;MzVKZO;Ctxezkq`l3q62!#zz-I
zzeNS1_@E3Tf||8&Dx)e>p|vMAjs<}$uRH&@y)kps+`qSwKA0+{3)83$)XCdN{M}{L
zU0E-~>ITomtvtr&idoSK#@b$O`V^B^Je76l+SPkN7GPr}Bfyn)7(6Fx$yV_>!EKCW
zu>(0|TF_*aR(Wd_=&{Rqw1o{eXuNXM_W0hR79-tAxxJ%(9;pHB+I8f_)=gdi&2{kq
ziUApa^}2(HWsW%45nnDm<j2GIVp1SMWsrxbXGAdBvAVbU7j{w-0q7c8Ey4Yr$H8y6
zk!{<t+_x5?D+$RB57=G<DBqVD<ybnVDR=e10+*d^k@6?aEdE=S8-N6xQ9$E-%p-~h
z&cp-XzsEoNLnNFNpoc7urfxeJfz0+n*#iHk7LAVN`ge`_Ql8U)oU}KgU$^q7A>sdW
zmce|Os@6KLJ%cO|XuZ$ODu0%QuG@$=kuK4>JE~|(Ztn^9-DIl#S0EPi>+|I?#-n4G
zPfzX&zhsATBGXT#&$_Vn8x&dWs8S*UUOfCw#(#PAEk%3bu~R?};9U7u8wQ&n4WRqg
zcM{1ye91H!)@(~RsIqYy#Gs&quWnwU$2$wd1?uHqg$cI{R4aT=C*Jz-NEYCdv-%z9
zxJ`zLkeupSfLe%H64|!KGrN?HT&$r?FJX}mP@ctS|G75(Z@w>4Sbnzov?|D@Iw5YB
zo430P0BlXbEi7Z|Lye&V0-lo~xRh7n{XE+8MMm2P(;=nc%wcz{g;sn1ce~J&Bau<W
z-tqSy1GLZ?G|SFv1YQ2=J_pZkgoN3~Q^Z1`3{|KuhW!^&_>bJ+j}IE$frB)MgmZt%
zo~ngb=7WI%GX$!<ego{rqMJgL#EzsaRU>Pl>gfQwH@EukpO=7=-n&9VI6Rz?P!m)v
z-UeD|w}Ao#S9}LrdI0pIlIFMTqcne<P-M==mtSGKUbr8cK?aA2zWa(l9%swf{xPSB
zG=u2)fZF-|>b&8=N~ONd0km8J^}Ilj_&~WAQ~7KsU}p%IxjUy9jzNFlLhm+$mb&D3
zzi7Y;J;t<kzK_+Qqp&->ae2?S5t~F_a225ZO!1S(!Qa~EiT!pfbDie{Z59Vdw0+NN
z5u(lEJ-#||FOrgybeGV&4Ah^eUEv$VIFdn6aXI>3yA0?18So5GJk4q*+w!lc7Fqw2
zG$Lj~Oxhp?kAG5NJG72t$0q=<`&7j<P61lsWm!obIhMr;=S46-VB9q{=(E*^ggS^5
z@9UkY^n|@xQD*5TTP#1n1rTf08qygSm93lxyhnuJelz}UdECGFQVxn;s3{Jd>3+pR
z5Vj*k8<${)s&AyxhZ%XgKxWE%4_|N(`f-ue_TpK|v0J@z!$y#mZ8f_@S<O-s;(VWA
zvxWDEIsKU_uRoE;^ZX?B@=yVL0+dl3GG4t%0h&eU1x;9(y?IKi<cTB4<YU1L#5N4{
zV&lPv#NWw905#2yuGj>NHvZkAEdwDlE9>|f<loz^L&T`<K$WHt?uCq2Du`Kdf?&A(
zJ0$@Nh-Be&Es-%35K?ka!$+96EPwf<sZ)PW_zo0-GUv8`@h^7x%<0r<$5dZ{7N|xU
z^u7`Ttn$_O;}5@y$%wRS!qL>S6$lxi-GQ3znI@=OP6Uw^q1IHrX)mQ9;8j}BNP&{-
zF(mXKWV>qKKTs2KEetYq)JFSS_fA3==EyL(=uDhwF;XOgieILReu%Z?wjNI}_UBEy
z(8j6aZFJl4tn7w$po=?mZ1migm;ZZTU}ua=t5Tu|{mhdpx#tv`c984tLuJ%hnpW*`
zzU7GmM88eaW3Rzr?6r&$Nh$QFjQT6)VLv*_UfyqF=U`z<pUqCGm}rHxdj{Pv=q^KD
zt_J)P=Ej`kk14!P$`QhHU&C_%XHB*lRTn~eNOk4DfK(7|U5Hw-b0XT@gr?`v`tR7s
zl!#iG4hO~<F#MEM`S~H%yS?XD?)vGGw8LiZ=Cq{K3+?xtZrdYDY+BO0%jX-t1e_dL
zH}m<8ca5Q0Pn8m$F@N(7ly)ufC1KyP#{9orJn9@QUdvffAX@5tmF;f@v+0)r`w4}4
z;yyglRl;Q?+GVI`T7$v)Yu|-Or5qEJmL}oov@%k(S?EJ6m*hQPJt<utGs|;y7v-k6
zZ1aWfZP3)-eee|Z7z{-tb=O%%lV(jI526y%0MgIK`H90ogP&~CiLZZPrMH6}qUzmB
zU*aoWmrR0EM#Yq@w<^Uw%=g`BJL^no$A8FxDe4P~65H`D;>@RD?*b>FojtIdN@}*e
zs8l8QFu2&%souiAUY441;F~tHlMll0xz;}@DPG^Q50UjLY?mjIR7gz2;9gi-fLJ4;
zfH~YOp|(e|c7>!W6u@K(I*RLmq8H?2`Os1n)G|-0y_Zya9L*DI7uTEQBl<hc5La0~
zkf)d;`B~nr?(s%|vv<rrE(-aRNTjd(Ht5x2*;u72A)R0-YQPn47SsiVVsb2N{4$K7
zAcLmEcyGG!&=eZv=P$u3YBk7waJa1DTO%bn2UJ7@5!<k80U$U*KtBy<rF5@<+%pSy
zJik0$&1g!jn+G++B=ELa$zZW?pFf;}6X@$na@{rIJ`d;mwJA}pioT}THJKkJ`agKt
z;J^(~hF59+Vi4?e4H2x|B$og1V}B79cI^`m%K!KjUkx@^89-8`o;weBb#F0}A<ce@
zIvMCHdJX4&tGU(#$qsYlT{<OH!JSj$j*=tCLW^M_l<^$D?j!cO))uH1&ftRj0+Jz8
zZ8iYR2N)!a(aK;<i=-Aad=0}A48<{`8iu?`Xtme>hzJ1}e%((lX=t--s(s4xO$$b3
z<?F1rzWiUl6?$`UNk~u^fE0~SfcW_`oy0YG0CtkJl);WjN@#a(veIx|7DliH4QA>3
zIT_=kriSrg<Xyd8h$TG<)b+agIwtw|&*6J~0w1embv@251+=PH=kkIU+P!oFrCQxu
zTGI3nQ*s+WNAMy*x+LI|DxXN6R7lmW58+DlgCxZLxQGMolOvr2urmHRVmvQo%0yek
z0$oQ~v?r9?c|glhHufp^PYYpU-tgzHjb|IS5JM^47ULZRV9(Qx2|7qa9!RimGI_RG
z#9tgr&5_U0h(N3hNvjM_^F!v4pS@-|IuLCF`T6V5&jcMw)a_9V9BBe0pd+>NvIVBl
zv$!Lf#Ey-|7+^LiZno~q!+dl#(Yi)8cM*1!Rm}s{s~h^(HxRCjRJpVVA~}P>2s&5B
z%+w&oCw^Np0u=(6DEDBvFct}*Y<@=93A^tDqqP_u^g*=0gX^w(&xV~Wl(k<y@4n9w
zDuGkEpdx5!4cY5+!~s@pEenWK$|^$Z3O%#9Y>($9gHCKo(}Nx-qggs9Ef|^&-@0sp
zT5Rt|Do(|`=*#((+MX>$TlA-N{Vxs;bFld>q{6wotws`yy%_I+%zUL4{Q`{EJBbEx
zdk&YU7(tM_n|muqyOSxcjSxQmY?R3B(Y|u;VzdhZao~Dcs+a2WFbx{pUp1~f=7TrD
zEN6h#;gWJSIHZdEkU>m4%VipCp^W`?rvMDlRP)5C3h6zffb=;HUXLS5f;MC)x>dmE
zIiG1f=%yEcuLpIwBt*H6LDBDJra8qk5Bj-PtwHgKWLC-9I|RBbwMTI+WZQ3?=8PF5
zqx@;UMSG*3(l)FY=LGM8$Kv4jPpoP4XQFq}+K%!r;Lm`^Qi@0!szk<|cCY=&!WomP
zE6yt`<F<!2{0%Nk&K;SGVx=1p$!+5SvWKn_R0gcJhjJkuY)-wSHqrF~2}L4`^_c;N
zMy<EX*il)jVDgdP-2PAUVMu{BPe00y6WZO~V4~4q#=~!hT`olU0_b4IXqV}#z(K{O
zcRQpHMMbLym^8d7*n9q*+lnuC`<)x9Wko<_II_^r2_`5dU#?%}bNVO(o2H!VN9tkq
z+-Xk9g7BYqun=uFf_jr0)@b3{b+#SaPTJ2#jAY`vBg34_u({5`S4e~vDKeH{r~1mn
zj0UP+Sxs^+(9#e9*TYs7W-{Vun-qU@+h<Q(L4#`|yU+|)(1S7;h}`PFZ*Bxpb*`<>
z_0--tUukv70#P%dvi2-kiJ`-w6gSFEeTndl$2)l4I#m|<&M%bSPzA<*;#lTzKe_lj
z-UVJGAZveuGFSRtys7}4m0PDCg7<V``WLl~9VF^4)*xeWLe=cJ!P@e``J-~SdVDkB
zXm%5%oM!!r8CWN^0cp)#jqB?HNUIOO>)Upm^Es@eLe4K?!hsby1g?i(J>i6OhUW`N
z*`5=!s$E7BaL^WbMQH@aU-()h@O689c@c6u932nTG6dknI#GUyd<)XXo(Gy;(n^?+
zi}Pq&CtsF+m<`5tt;Dp*)f`9J37m#NjBMf&4K3)#zpUn4I*HiI{wq>!RUjte=K9?8
z0^34S1Ms{$DNOWK((m6!;Qy;9L%oY_u$6SO6INVhWKN|bbLntzOjHx$tV<-a-%_Og
zVkZyb0}+$tm2$uB2A%Rioe;w9c$z4zjNJ4k$O(%1AkiI4xY~;J!xW7w*<R1~;@5(g
zjH*}3i9h>$;~o{81$ATasIc?HFFH%H6&{il&8=jx^UNGYImxg_eU%%#9hy2uQa0J@
z0rx-R8dUSn^^V5JseuW-cJ%{5p^q|?{V|1GD00cT(Irw{FbhuZt!Q{9E-qUyW7RUj
zaAE^Jkj?Om3VLcPT;EN!If%NRVH|^`9uY|$zDaKzf9Eg7_lJJH0VKyB3IizJ`wA{3
zBh`Ti${t5pRKZxA4Kcb1@_l6k;ec5cVA57(crz(@&CT!OXgg&z1_{OS@q?5)qKFC*
zpW*4tRf*|tpZoS%!2H_sT;3Wb3RTmyjp04u6kaNsHXoZ>;gfH@Kmfm3gEo)V$QmGT
za@f{cUGJF|EBlxNZ`74kc<Y*&eJ6r9K<2K(*y?vMZ}T?QXzSPidvr&ec>=bO3wbju
zC8<b`0cMNjoTDsXJpfa#e#Y@$yPg5na!zDyQ424k4+6``3^Eu0{p#xKo|pB2Fggh)
zUIZzb(z_c1E4$8WXN&5#jah!N$f8QT^QG+|$8wb+oid<k8P+`54Nvk3ct-ZYeW>D^
zLF)}s(Qv_rE;es<5ut7V19I^CQcvalVN-_ZMGHe_RnH>TJ%%UV7ckJWZ*_C|WCtS0
z7Idu715YENh<V00Ko%;Y8Mf_E?rxc2@7O~|1n;E0Jfd;l@lKZcbbr058T3+$VFW*^
z2T>J-b2+nzG+^B_$j~o<zOxHR&YuP;crrh|R5@CjeaI*Q(Kf07y)PA_Y%?dcwACPs
zX`ZFczKtRX<OPtzE$)|npMJ7qUeh%DoVN72nX-A=w_4>DFFGye+kAKzVhsiB=C1Yk
ziQapAVHc7ZX9w8<L^rIOLBtWXCu?9NT2IXvE=9f{u*drq#RUKtdM_X<_H*u?kK}fC
zw}`zx(EC{|KHFp)MN{DsS1X!K2x4r^X)vFQAXD{XIu@$Kz)*N?3;Fauv=jPxen?Er
zNuIm)PGB2%FCHQ|zdRx3SFT5=(m(~he6+G0c>Vj4*l>mokB;=Msz(u%Lrcls5jR=Y
zK7Hj%e6K5*ae!HwlU14G+sur@&DHX6k0{DcH_QEorJpk+4c#54@d_B`7NS)dWBCO7
z)*KFyD_cN)8N4qXCrAJmI$jaknwQT|0~$`{O8$`^q>`UR>;7~of0s$|zr#h#Efe8C
z2+3e%(nFC9&`s0<ipce4M*?*iB?#T*N^Ohh20F60hAQG!=NmulVR7*#4lAm5B<@Ql
zFAK1)QP^APv$mY5sE`C)ZNeANu<$WKc5QWOAYcYazCsj+B{b$sI<+@dE)ZPt5hawL
zhDjB58y06^*7jEgkb?wq31AN1D$<K)I*{+a!tUr@ME3#tqTZphR#T9%Ekf(;lJjz5
z&3BjU0seKR$M1F3i3GsQupYD54Z=~>G3vMf^9abFh}Q<&c?`Y?k{0d-P9;hfLctt7
zz1*y>%tK`uTc_;z3@FpgXw0dFLRHrehi^gmcJ{fc;fZJQYlk>y%E_u%Th;;zr{6lo
zorO8>V}yHfNHBnk5QV{|4mpT#Gfj|3faiOoCbENzHUy`1ypcxW*`=g6)#ul;*(0R5
zQhngf`o)gL{FD%jQG;j;&LiEgwIG=B)$HMxeG7m0iJ`JCq`eQNWFan{zbg|LuC8z{
zk>Vua=d&e8U0$N42E<uOOQ^RAur)Qe>4?ul`6SXjx(?bAtwFkA|2to=8akOa`0zJR
z5l?=2Vm?_@BKLaYnN~7s<D#NU)#}-2oz~Z@AnWM8^F@|_o%zVO(3{OF8hDP&t&{iZ
ze9Q|Bgm7tXb>--S5)bqkFM7ynIjaH=A1!d~`kt-d>DC@Ve`SA*++zTHsP;@oDQsQY
z{A0QOAFnPJgS9rJvL?msPqT1j8aOV86=2>e;@B~jmOO}C-DpKM82H`AEaO2!M8Y3|
z&-o=*A<eM(B+w1G&BiEtik!+**Fb0%Ei<Ks++_mmnisRnezpEj<zaV=7{{v}GVr})
z8?F@PObrOHmO@VEdQDJ)W_^qS0)@9(8EhE340s#*)M1}55J0Y&*8^yS69_moym=|_
zB4X((;7h6PH2TZdbVO_?d}kAWBVlzU&nEKzZ{PJN2(X4>0y-tdXbP>|Yo*hBWiR*F
zVAQ+6aMcDrBCE<BWv=E;y1n$+q8zmyj-yxM@Zl<F@5@y$Irw}jRhZ<IgI^zo@2P9W
zWeEt$<~K!kBjO<R224T~?uZyE`nG!9c!zrLgM-GDj1U-8=tD%+YAJeMP3^hucz#0f
z>cl}VjOoMfY9<ivNpqCm`r<#m%Vs^ZZW#RXI5sQ8D#b8Aq!O16B%xwhNG}l=%gl4w
zX8!gFaBiCg0F9hN#YWN+`w;UfwXyC8>UA*>7KUv*4)Ji5dBsOA-7$Ce=NJupuwr|o
zvX&gw9$1{5v2ZaAAE?T*lXF1(CJ~xRkdKK7!Hxzh*i|LJvZ@Or1;<|=!3`H))vAw@
zl3adN3^g4U)HxLy{_l5i5k3CXNJPA5Yg0HlymY8FEtl{1??JTEik>TvKw0nv=R4f3
z5&iGcFP!}-8e&#0K<m30KvS;(BS76=vs`om*V)<tm51=fgJ`3ku7VOf^j2&Op$Fu{
z`wif)mw~*N(C4WWy*2fXg%x>i|3imG<i)Zc51+EAm`?i?zF%O9YS42$`z$vet?QFA
zq>etWixO9~Vglo@pHtf#^b`pKGf)kVLMo6OR~Ky*V_nmQBtgnqgu5D~KMn#(rpiET
zx8VsAd;egvtuF{YW`K(hz7G+F{&I-2`S3Qjs@0rC><^S%^uQuH%jJ!G8LpTCMT740
zSE2I^jivyK6+s1}6c?MyLl<By4*r_=bpd;eW*9<b(Yn2qfHxnUZ#*I_x~{yB#mS$z
zK&bNNm0HqzzrwU!;>nNs1*(&e4w+YjA{oXiFh~__w&)8O4Y8+kfLU(_jJ@ygoq$El
zM^LAvk5>;cB&2cfpyteV91S67_b`|eDYyzGa!MMf{#4mFJb~E67Si&|00=Gp!jgGx
z8_CH#QsIjfh%0$Lpw#<0h~M;HWeYMaqo0IzW*Lf0P=zYdcZIY)WH6<&{`O_Qx!Q(=
zF>j<3Fxm$&;TOeuQRYZOjIMe1TxmAkh^}Yec?9=;xb#cq?FN%ZpG~&1n#E1EF~;k8
zJ~k$l-DB6NV#{j^AtRf$4R8*m%hV57qIhTrmlZ70L)k(xG;Iw7*%xXqPg$lje>hM`
z)GKzKfQ!ZBr?#uuEfltP=tb6odSj$aN9I<$_&dY5dyj|GDQ@Ghu1w>gjNpP~YqRnU
z)n-SUF)0eWUQ5}VCb$l&u>Lv`1YXozwUZE7#yOGyTe6B~2khPG0Xkh-g^YU-d5e#r
zV|jPw7o%K)dp?KaZB8+g*hr-|;8va2Zk-?8+af7vPi@a&ZcM^|3%K)WWovJM7G$-U
z^>padOLLUmhN2aXIrQ`PfgtIjwdwO~?rC{Z&Z>}(9=DIGnAuu?*j)t8xDy_CehY#c
z^|!oVPi|w{S~wsdCKJ^l^Qma?e8Ne4n;?S^EqMuZ;mcHPgspKz9BuanGE%L0%Dq{t
zD*~+K%$t+cn{T`DEoyqI+^-_g<tH~HJ?SKqOQ3g<PrIR!h9FsOFp2n`+zPi-9`Rn4
z4276Zt9Aw*U{D@&slvjg&C7G^RLkIVBIj_1WaONDineETR*@bYf!urjPCjV(xJ~=O
z(A-IjnIX?)ef1uMu0e-j3_*iJPGrOZ+$K?V%I?n=cN?O;Nax@yu+cLP_^29|Wc>0|
zQX2e!$kernWwRd`oYiSKsnCHMr$`;%G_Sje#-4%@WWCDS=~O&yU{<KdvU5o}Rj=L&
zO@)JoYd35;a{>gN0|a@}wap}nDQ@1o^;^?|RfEiHxDrBDFK_k?6i$7+!{JfeY4ld4
zyUd;YGLX1gWWob>9GyVRb`Uu^dw69!XpI7SnU2*VZj`wKrRF(^WU!2X_64EKyy1J0
zHTxUA*KI!IHVwRq6B@QAPEQF8IY-~uPOX~V0*uuKxByX^Zd0?Q?gAZxU%w}?>wmyO
z{hw5Q>wqfXnqoH`NtO>#-+FWjrqKTLm>K*P2f06>O}Ia$Fm1X$3E5t+lEjqvxO*O1
z5cb)aWdOUb_y`aViA^RWCOTb^m3SgovMmq;E>AT!8&P^sKP1$pIB#-0`<u!A0?2#{
z=_E)<3h`Lh7R<tg2BZOB=+^eMd4&uSt(G3Y2(#~92f2+kJ65B(-Gcpo+k5U5D&eA@
z35q<T5apezONR{1uqwHbMB!8T6?eng`fyZ4J3|WGOJM=+RSlc!{*j8>`Dm3DhI_dS
z{r)+?(td%6AsQ+oWR9gF$A?d^YaD&gi(=xyt=1D8TV3XNeN4U`;*=|&-8eLVk!;rW
z+h7-F+4J3wQY7XG-~wMRb?^F=`j8`YW+ap}M$pGA#snB%T<7$`H-2ZV`?$&StET!G
zyC36D=YdaojeCFIs-Zhne#SzCytqJSj7+3a4|a$|#wiYx@E|q4+xC=G%=Zx%3YghZ
zw_hH$SRslQAjqZUt_eemFx`ZiurkYAYLBv4PjhF{rTv`~S{)^+^~Q*tj8p)7DV4o}
z^1_mL4y}~&H%bL7LvSQCQk1J!htMHqrhNU081-}?sITo3G<rh4ljEn|BR`qm)<m4z
z8Vk+nKI8a4J>jSbxd_5FpY1&yNa=0$&I~zU7}~tk>gI&}1_H+sxbgm8J2xVsAi69P
z!tN{boT_qDFYg(38TeFwRHG23(OmdmB%_9N<Ee6rRwJ{o7i(cw_Dd*(RSFJf?;3Qz
zrmJc*j0PUSbq)h;%)B_J<+M&i4Chc}ST}%>aw%NnreIs>!llF4o36ruG7zzS)6UBj
zL#;)+vN%&-@2M9JlctcgQsgbx$0r#{wxM-MXbmT^pU!)g>jZ4_u0LoFqU7@y8{!s>
zDlq1*#`ODA<dT3UYz&U^+f6Zz?*{lce~>(S|I>b<Po1l?X%&&Qy0>8&Og+W^*9rEh
zVd8e|T9TSA>=P`6@Ey>ul7Vm_cn22M=-5=m1c}vU9t^@KWpZrf^QVM#rDAzmyL8G;
zcHAVd>2#*GSx<#Lnk~Nai^8P<>?r)5!pBFOlQotB;dg<RV3Ov}P`#>S5hJQ*U%C_z
zc=}VYaDd@483tJ_XnxS6zq%rCvg*)^piA0c_aLy3(NwA>9?nYQ1~2hWiu?Q#8Pe>_
zFrEGKP6g=bCnvotIu90B<alONrfRb_|7;zNlXY^;VBZToF!q^%p8)*7IjDV_&#|tY
zC;(<N`IVB3P=6T_V%DJo?~AwA+v+q2u14e-PWPP&eVb{&^r<q&@=9}n>54%*N;WPw
zWTe7B4MzHHrL)f&-pyt;y?2t4F*Z*r?Pj@2&j&_Rry{3Z&y4W;_M@w&DgqhPjjP<G
zurm#9_(YrzDH3F2Xg@Vl7WcF%wuTkaXxOwE6cR|Uc{9fv208?w(Zi3V5EY59#4hk*
z1RbH$`vUwOD}in#A_gbMhNcy=)ucz!Tr*HUdU#+KFRE6uh3{+dW3-nHlcg-dy^R)r
zL`LXhI@(eQ=&s+WLdMPO1dKYfKlQ<5u59ofj>pfE`75{<kbPof`+HP#OWhRwC?CZ=
zV8gx~tbL0#QVwYZn&0im&j%1%E8`VXvOxpq&8c8G#h1zTJld<%USabY57rz*9FY-M
z5%y-<<W49sbQS3JtriPIudhb}`vU{sQ{p`dIUFu3R2?)~MToG(RB`z+5n2?uJZHOR
z=G2f2tn5NiL2<)n^x-nP3-P9Oyd0X7WRxlfVsLBS)ZASC!11Pa|K2z2$F4_)#=Sw9
z_^2B<2Vj86i)2GsMJ$5{P-K=n#ov3&e1V|UzjS<-7CSQPhPY056R_`)mPj_7Y2y9g
zKUUC`j@$R<9YbT>5%dnFA>xx;*<%m?k9=3nuUJ>oKPp>}dM+FpjUp*kxiJnD7C`Gn
zR=KUG-p%8D1$3bZWU7*fZkwmM=Mi*5lNOOFO)CKIqD1mArg?Sjmv00fE`YG{K&!XH
z=d9k%cT6nf7WcR`yuYT`FE+1!KTas@2%w6eZ_HbjnwWt#7X1bVdpJJ9WI~Y9nC}j*
ziU1yt1Foo{;y)epq9}6**Ya#yo)d&E&6KOYoCd1LFGY6K7WRrInl5ei?ZKZny0D*R
z(gXXUPjKV2=+-vE)}Q|gX*v7A4z)dPfJ-AY=!$)cR%J?rPl(b)(sTs6S|rd%S1`@X
zbW&NJ@-~3m;!Qpaqh_C=E`<v#YgSe!{mHVYzo=YgC&zAETWPD)8dnmTBdSc&n!~l)
zY_$N5c?*4i!@_1QS1iO^!ftvAK>V%YPPEPsJpT(w>V%*oHL6mmrMB<wLP|&nY?5`G
zll+b9BPLN#AE1^ZC<A2%C*V|Qy<pNw4G3BhHU4Z`hyr^ZVN?5FDh8<<lt^X$G^H&j
zyg|w?iB@ni0+9hB-M5>=SbyioHliun`+OL)>xTY9HTKDDfEA?VQND^U{)i7e$WbJg
zIbvf%UP@{MU>H%5&7zj6=2YXb%v^ki2Au*vxe61#BZp9R=@X9Z*$*Q=d6D;3GL`Ym
z<whlDR}Hc_2lHdgD&3)}R8Fk%_5&W;35ry8BFE;GMIO36e9}92J6bxn8|L)sL4nz@
zqjydX7)d(f7d_S7Y<E`kjAzdyA+Md3#$wBm&BmO~jf0DIsniOM5ucP3Pn+=Fa0!y{
z-gxdQ>`q8hKEJhg%lF-o9?<D#Xe{uDhU~A+^jqA|PZsm@xe;R{6_yI`SN6LP5dwKO
z?z5kz8H@%xKs@9{sY4SKa7KIOBTP2RuDkNEHzs}_HIN|N>0mhOpff$U*TsoWIY3%b
zF#dz~cy_vu?#yU^Ulr=Np`*EgPn$OYJ4;Cs&Z$=DK%9~){Bh?!XtJUbWwz-eScW^_
zT$|DB1V7Z<vUG0IMc(f~T{4t9cL<hX!KnNMwf&uLl;7JxfNbPGs0Q}z0!=ZQtZ-F>
zNK={}Ti?tMgy1kuM{ebk6l))9`xI*&|810hnk87QcPpMlFyz^>``|u^hSB?h48qin
z@I$o71npLoF|UBy#|_s3;V!hr#M5te7aY?2SHDblTe3QuAHsQO?U%}=n!ljLQ0WPW
z!AKk#@&O-t$<rViI}a>nGGH{sH=xbB&sCPYNew>Y(tSc&q4$<yt1Ie=BS93y(JP-T
z{q%DWp<tOY>pMbAa{#$8MWFLxXeE2EtOg7Nf%Z*qcoAe88ZH^|je1}i8Y8J9Y91pF
z*%`ZY&2}bDvM^<+2w&U)s1rp%+=&hd0czoE*B1ckr?e!NK7_;kqPqL1YnHOXg-gkP
zsD<{0GRa)r9J~4lk&s`2L^${C5|BQM&}wy;^Q#YF8lwLB(FMbMgkN2;d4S@ceYu!u
zI#?Z2(=tmw2buFeN)_d@qb)ahG>18d?DZHi_YBU@E`N<B0Yc0J0Gql20dxko$S1@t
zLfXfW0iQ(jAWas-v0*xmacF(HjF#7O90lv$>lJ}3mE%wLus2a3==u!I2L_kuI3t)i
zef{uxcy4bHf}?GKUx-0ggkXd|6mj;^TGiC)O79p(xz6rFM>R-DTwruMRGeBcS52oT
z5SyEJ7^Vd0Sl4J<fxO{JTM=q9+>nSwT_8FU1NkYk)Y!9#$Wb8TBim0FTEB61qYX|s
z_5;${@FsU%)8HQ*DZ*?R@G%|-nqLP`t6u7>I>%J%Z3>4we?3y*t!Tj1O9rR@W944O
zYi%>Pv)S&7YMqQEh@S5(&lDVX@{{I1gJ#4T#D4qF-)86We^L?wl0Vw?1j^871EFzL
zp>pNMD%qlo?Q&20*HFh$t=ck`>}s=T2Shn(fH9bR-3*#uJ-|{0Z!_sK2{D$<i&@Ec
zJmLBhVm1H_X6>owkYhAilk|1&?cVO38~wyf;@V1U%#`fU{cJLfs(MfcUUh`+Pl#W_
ziR&4$@MNX9losf!)qL3w4-hseg6Lv*1#9(7z`F;|1xvgYJ+0f6fS9FKs}4?ZukN*h
zde-ZHsB$=9o}Qu<{}Tdh5EXR?0mVV^Z}AUg@2(-K%HDbLwUhFSBzaWdY^+Y*wt2^r
z(zaN|qw4oHMaa)kkI@WXW1%eb&%`LiPHv<LvSa>uCU}OCzo@Pc5pOlA|NNBee?{z?
z9m0-*WZ4bSRn%0;xb?W@B<z5_1lVa5g}3o)KS$mRkS{sIa}PSriBMYKopZls547PE
z2WHX!F7?)Gt^t3T$X<1AXPI$&s>0;xIR4;#kyU1Jowu)?O3&p^2O6lh$<L~A3|OEF
z@_+L^<ZnT1I}xJ7CBl9&_iB{bAW%A*()Y<}f54!c4UnELaM0A}WZKbS_ihT#Q;<26
zU>ImhdsHnuys!0ywh9%dzf<vAwwU|+{)(cBVR82Gbs*uI<JbSmlm~Uh`r*?=a9lM}
zhcrY~cw2nuG=4cbxe+Ka5cybrCt6a88ry;{RJ`4gjXHt8_$g>)dySq)1vQk-Wn%-a
zNQ}7q$pO(4Q^Zx93_HAX0(xNC`TjW9dLFe36@_a#mYI1pbwm{F9bPhRKRF(SE<wjd
zlRvQt;ddLCV9-fiNe(W-ftsIS@xRR&N|^3%pfR4K%cVDL2+<09lu5PD0yZE-*pV<*
z3+Xk?0Fp|N^{;S2eai_fljN1PrMxwA_&Zf&t&t~B8Fjd&ccs2MtzIPK$phN$a_*NW
zCp@zRlaD)1#~lY}G{7YJ6T1W@a(4eW-7UK^ENF=#G8ZaP-Xw!}K5f`rB5V`4C<h=4
zYD&CXzRq{0sQ?KKN!j<d*?_bVP(-Mja}i0X>ipXBNUl>GR;>n$GeAs6KC?5`;!De4
zE&U4PmvmM7lF+$(77qOhB;k5-J%4rm=ofzc7#PkAx6^-{&Tsv8c2ii0?Enl*s^%gg
zDsx&OuOy7DmO05K-z|It4`;}+$<P6VIS3T#5nziok3rdS;vNV~z+*3CNod%3;u`s!
zHoLNVL9ofF@*PjnJxY$aq$wzFl+_ctx2BVrR>1#JdM-FSAMsVuz|)&VNSTm8E=u(m
zE~t#RZ*ZsYwsrzD>nr3Rrsq40Jm@7tm9k8m$AW-i_c|kK4|^2Ryzycv=i2P6)0V3>
z48_FP2A(lBys(dJ8XdMAQIPWThioePq%_@sdOSO#jl_?rgO-gRq=7^{UTztr`p<u4
z%U|&zZGwmF0WhKJtSLZjXts6gx7-ee_nJ4YU`oZQ)SXUa5wX-qPJ8Z<(Ub*J90HkJ
zTaq=Z(V%Pc5_=S>&%}ba!H34%1k)tU(Kh$Xgea0^{5Y31S|t_K0nLt2z4hCl+_~@#
zO@r`}9q+WbK=hA6onp(wZ8eae|5n!;Hh^0PYTt;IpZzviOGgELr?uhbwY0d*p(tR1
zGM=$4EzgV3p1O({&7MVlI%u5xq*CVD+IEER0TF3xPko@!tq~6#gD;d}svkm=&K-3K
zTj)<Vbrv1vh)iulR-^afr(7HJ9Kst~M#9Z|#A^+K2_{*_`Oks!hxu!`x=^l8Lbvn|
z8Q?)d4^|fj*5ap!V*#XRuvx4Px-8cW<M^9AoHm~!(_Ekj7sHFP7R&-F`~Vbxqj)aa
z64-rl?=LR=t;2<dlvKl~AdTZ5*__Ja3Sd`LcLwX7<NyXBpb`1Aix{(!<}$`#f3*V5
zv^%8NGyd%7{`@oWGhk<$j*<aeSM-e+^kvRKi^C^&7y!?gsNaw7!F7d!s(>D7TfG35
zCJWD-<K&2>^pc8~8?^~AY!}Ed*tjZuD6kN%=h}{(!@vdz4C81QDIRHL79C8Ha&kbS
zijAR#`cICReq$hxK9fsZ!hzbpMf1>hR+Z0R4&D&nxXT6Lrzftw@JC7;1dl?(e8$>O
zH|YxP2Lf1jSnV7;c-hby1Vulox$u$zD=~WTc2VE{^qE%%lP=zG`Ugeyr-SjcE-h6m
z)J5havk42`^HWuUj@I?ahK~&O3uBMiWf<w?wViDtcKX;=Kng{q*JKQ=jH%BMhvsf|
zV1SlBwByb#^tvZPqIS2quLVswBEGA|+?Y}v53YR^tb<Xj72#lNy!Un%%Oz*ZeEIkI
zpt{xk6f1agzhZcUhI9w?^gSQ<{F#R*+<1C7@~~8Oc>36Ft*)?p4(7~_#x714!W^Ox
zu9^>rzVaJAw0sB@2ah1L9`R8mhdR&+<Ohn;ObSB!S7uuW$lmC&VXv-!)G#PMUld=o
zTW9>UgoPe4SV>$F4jR`W+uYf<rMY0|u`x#K(RMI}37Bkp|3qV#*roZ{rWU{-e(Mr)
z9r9`SC}Cvn-6tR`BL%uI;0CW+6O%J%^tyfR%(H#N)tn2}C?i2)9vPNP?pd~bmUr5;
zD-uliT6fiq7pHM(N%tH<G1Lj4<0m*9T{Hfx8<RX&@4ak+<sc!cs_VPJes+F;fpg^U
z=luQ20Yr~yEdBzcmKYTcs5<&uj7AYAlnBkRyBBHPR%t(nz}AwCXWG$^^Q~`)IxGeG
zD%u~dH7F4sZFK1jO1T#6a0TvN3PZ<_o0v8agMOoyqfe(F5`RQ(pBc&ZQ;Y1+ea1%b
zeIG{U1d6t#3%g&GqRNpacV1O;E@>Qna*+9RM$26UnxWx;5K7tidaWdSzhbX`YkG2U
zs?7td;{EI!&H)X|?yem6c&T6O+s)9BK)uzDKVdg$YHpMg!sGngR^<r4*ah=H&I<ZD
zDT++6DpP&AdTseeMyeEGiDlhr{G<4=?YY+FwbjlwM>G)*IqMi!8t)lqDr+t6b~@n?
zubjvPcW0=kH6`?ArbMB%uQBSu<3Gzn_}97g;azVcuYn@=P-y?bKXG{L?m2(0xFaCH
zS#r&mV(lz`5%=kX<-6pH-K=qGbdFuS;^i9gBuO%fc17ZZAk7nplY&HUyjPP1U)&zK
z)A=;9YH^NtWo$6;wdFuZ##~i^k%scX<X5xmIm_J6@rA&FjO8)jDsvU=ef+=ty7j;8
zY;l;Cvrinm>fs8?zU0EkMSN+20F@LXN*o#%)(6$A2PAM;J5ick4x=cv)YSa(YRCGp
zaB*>`N<)t(sSYoAn6b;b_^OY4&sjWs`#}0`R_VFIo%}nu{-TZFTO*sV+t>tby&jGy
z(OGmQAEdYzX&(N9wTpTi|Fwp9QvSVgpn3mmS_*L6?tpt-6&8M;st1#$kdKir0H8dR
z3Pfa2N*+_;S4RZDaZ8Ic?>ihseYom8?fDgh(|rLVO=_BN#iukqv*`n0H{UyRWAfL}
zzy5SMnEz{cmgH0`8#}Nm=lA?RG=^(<U&`5LTaUYox+@xMa&j`seEM#N$*|X1y_UM8
z_E3YLg6Y-kZi&mw%i+)uG6cAg3p6!uTY`E@-$0P97lTyP!6ACcsF;~Abod$Z+c>8C
zCpjP23*h>EgCti%;L(|-w-28Euh07-3tojE#|@Ah39)BDA~{n~17=vCV4y%m2v10l
zx!9yNH6t5Rp=K3LdHYJw&Zp@C)2G|hfsCo-^SAU0%4p7vWYjQMw&`-b&NAL-eDP;1
zL<dd12dBc+$q6>#S+S-k>pY@p`x3^oJ-gU7?~o7t`v}NCl2i-v3OpV^e%!abFzrS!
zaa<sW@#&K%*b7~c_uiSG_?-~)DzoQc?Ii$?(84gbX=}oVM1JGz*~^_V)=AcDQpHZz
zk%gALWmhXF+L$Yw{^yVWD1T<SF#<Yygoh4k`NZYr<)NkbQ-BX%ZHSTE)7nPA4C5m;
zZ*LrK&a=(C?W0=gPzp0Wv@q^ZD8%s_Ik&j*k^6MqqL-~QZkSSKdfoA3k?AcVqwZW)
zF~OafGa5WakrOZ3v<v^w&$Nkj-O5LT7$x*+N<P%RwJ+V~#eBFE8#=;_!1oH^69<EU
zUOIVAijGWa08g6o_@F05)iUHl7#OYsyX>YpP)T`n9Cved#PTFg^$)yC8jzLD<XI(2
z5s0u=XzZjzhTR@OO|a`vt|fIlzRxHb`2_;LMUeFVAasMG_VS-u)!+QJx7a|&r-t#v
z&#mlQHJ5;Y!w67n?Xc`5r5$;intI9m9K){OJ)N(O@$Pcx>YKk4J12u@anR3q^te&a
z$cY~gr%TBWYQIv*ftU;o4BQxHnBPf7d)+Dbs?2h~rN7}Tx{R>yW))p_eO#UMbT!}<
zJh?@;5Br~gjK6e-oJ8q{eGX5aJ-d4S`pbq?!@e@e2S`f&xsnTV&z=g(pYc0ARg|;O
z{^(nMUwuO=pPRW|*Uy+eU^6fddtjJlSjRy_Lz6M4EAYp$P`B^WZO@d_Zj96841^R@
z6lS5-BE!CWW7-YxM;JtNnEpU-*HJnLts~$6?q5FWkAH*T0jIyxk8bD!WjWY3J?O)v
z5Sh==*x81L_BmD0mURx?TlgF!!CtRHPU05G<nSg&BD4X0{4d{5{P3%~*zBOabp>=k
zRGU;lZd84;yYLw|-li+&SN%0vN$qF}&E%n}S0KyFXBg^pe4ED`E*lJpdD5*9Sg|Va
z#3OoPJdvL2F@YA#{nawcLuYu!MU}Dx`rUC$J+Ck;If{CkU$=~PBVzHXj7_XGb<a@P
zDWLH~)N9*EUGibixy<!<aQ(3E-*z6h79i~H3Zrw@I=a1cd?@fMamW3ypJ`&1BCdMg
zfijF>>2tm!7Q?6CXWm)%ws~f0@g>6T=_{mjEbG#;RPUTT;ZikVw|rr0fn2JMsuuY$
zYyRCo{MLt#SlC?YyM#D49(y0=2tNpQdSG+Se|@3WvU$4yFhX~U!D#G`q}v5MVwHgC
z2Z{bsaoDHhw-Tq4#|$&muTW;4Yqb!5)sumcj7&S3$+KfWr_o?4N%#Fp?e}J;J{8K`
z?93<IsVw<*45m$Eb}#C>P;i^DzNVsYts6}|Qj=&PTgKh2sh8?kp?Cci!{%}4zsE|`
z@AP8Ts1k;o(Oc$u4Ug!=t(J^4r;FhYIoU&THcX*u!=816=xQ4Ydv7gDe8S%UA8FB_
zc$`13OunEUL}z7a!ZPNV0aBq3c(Zu8@c!5_u$Lr|f0s+T^J1heL!!I8`&+7=on1|q
znf@V;d;Zwi*k5O29td@PT2MHnIYgB~NZb=Zc|4g^fd;=(zp!N3eXb<YD9vqfaQRJh
zLDW^db8`OUx86oTdXmN}67~HtYjG*gJrINJfR>@+LlU3|LLf%0OFuQF2JbrGgsYah
z{@Hwl*ORSNU;K_9lK`IjG5Pm`f4s*(`jZMF5T7$aD82lYI&qm;$VD73nCwu@v8Yd|
zln#Lg)fvQ}^ozW9%uY>1zl=O+)H9l+hVN5Iqq|^)-s6kxF5iM8HO=+PsJP63&$D0A
zWj7X+o3*1jlc*xNGuFRV7d~WwA6x8ZYHIhMbRH%>!<BO^p1n8+xHyXB0+cwN`{6Q~
zRS~nUzd*k3hBs~Qpb@RIr`|~yqWSC}8tJC}^U>`iC8cB2Y4CuCQ$cJ-yh55N3QTE5
z!O2EgJ8yKj($vmgts9jc;0{-fwz)y<O*c#=Y`WJwuHCco`1mVU?{V5z<_ss6G<xRP
zq2qZ$y^9L^tgRyb4Mu8|n-+8`8J0%h<>QH~bfVtb*!K15#zP)i4+bu|jJSpqUk((%
z|NeD1P5SRYPd@#x#NELkDI_BwhY7X^508(tBqkQ^vjaBR)k~M2+^O<!{<2TkOn`do
zfyZvYT$y_v23B!xO-==Ia~(UThFFh<n^!Iqv_|b9UCm(pjZR22UeqK(yLW&tscV#s
z(4Y5nRKtu^a{HaJ#aEvXY`)Ay9K1}&%xqD|F2jjuXO`CQ4Yf7iW-R9W|BA)bkMR}L
z?_X$$S7CYe>XkUm%D<C%>rBKXU-zaAnt>+*c+9+_W#Y+Fm)b2Iav@c&$u!Z?>d4Mu
zuS(Hryff022$*Hi!Ipa+*>Wh5MTT{8*|iJBmHGt<l?8l#P`zj9qp;1~MT#9}Cy%*E
zm->ZvXs~p?vJ1Lg(l`8ten=+PIN_yvqN2khZ6iwb=s09jzlZm`gkX5rH40dbAS+^Z
ze)$OV0r@Wzt>;Tt-;{)1AG-u{zH!=?`SI@{Zi2zwY`_(gKuPB@KR^G&-G})UJFKQf
zZVtcjj*L9901ZXtK|}9-ij{olRu?UE<I&^C2RQ1>peO4Jyp`eBx<weGiT0M8^v^%M
zGp_78%iI-X-IiyZ^PSfW9wZylzmqxrDdI7Y`R`uY+~zKdIs1Ez*EqO6JdW2-akY2c
z6hI$XGgns#V{6n$cSW9pTFUu~tgO$IC+V|X1u_AIgoM2?*={)5EjY?*0j`@?CQk9q
zISPFcReD+s55cxJ(wwX_e!<q(Hf<RSFh)92f7ki>2c<#h^V-KH;>Ku?AHOWA8MJ5=
zwwfJjou0X`w=0TbW#hiDJE(~-bF}|o3DF_3Tg&s4XEZ-kkA~1#+INDZeSJpVA*WKb
zF*^7;G|h%ai=8Q>M-}346&}TEJrZBSlRSIvy5b|DA|n%5{<Wn$mm1HAM=RcwvH!eC
z$zv&@?3W}Idasb1f==RL&`B~msIju<P6f<3S3rE9X_MVNDcrVfa|!%I3Yo_5NV;!v
z%>u@84MKsSb!vq>ol69_QVz_^NW+YO#M8*prbGc%HK!oQM~@zbLwKx-l43LmX!sal
zPNh_9!0+n=X2dJtFlgE7f<BH1usW~8II-KPMl+{juQ2pVoZ51&74Y4~2D+fMZR}^e
zmH*4l57^pVHp}Ue3-T-s%fawN4N@(cixm-z4euoAN>e`(2hJwEjK?BVwwe^=vR-^I
zEfpz7^ZK;fv&e+q>U^Nei9%AY5}SB2JUICR5JqWFo%(#ngO<ZfJNyyn-75cg0gW&_
z?T8niaFXF*JmhUP@ha!v$)`j`v<%7+w<J86RF%uK$<b{~zgZQ)qjB&&M0jza)Q_Gw
zVXiyPLw@;!>*LzDcF#7F2-l5pBy9Z1DgrJ|ygz;ai|t*SA`$?<d|)AU5&aXGf|MGn
zj~<Xzuyi7zI(H)Cns)<-EqH>OJX_nG3Wp^uax)dLn??k}ip#cpA7z}^E)x>7!#kSg
z?HLvESfZo(vP;3+dnLmXb0w8L2JES;GE#Zs&v%b6cUq3*F~!BdhCIK~<w@ZB)6d7{
z1AG|)vKG*K+n=>_KN(pRFOfbYZT>yF!%e}D7=f5g<ISRCs-5b|DC4D{w<3Mj<HYSW
zYoOhz14!{~l;O!O;F*c1>dFaNcRwOqg5p_3G4`oW<oIreeIM%upNd7^f86U;H<2(>
zEiG&p6)l(i=Dkj%sC1m7mX>cX+mRy|%e-0Q8(YbvfslI!#GjbLe=qZ7w6IiQWN=0E
z(xN+cDR@k;L#0lGPs&_Za8}pJFFo$!Q@k?<-Ithl(@v&-QdcOh2j|JedBi}WY%{rg
znDVeWE^&YFEo)ON%>e41hq>jh6L0$bf0P^9?PdtJd71`NPS3(*_U6>bDJfV^FrGP6
zlMoBQ?-|QkT>+)5_(P<+_3>r$UeKDnny6ZMQc$^_NOo;NaG`7Xv0Kq00=HxCLD`M0
zic}V-y#0jkjO30(><Ubkx!0L;xPL-4?ab7D^XgT?G{}cIqQurw1nL$1Nc;Bs-=;uk
z$>c!-faSFxcfo|y1$({9jxI1uv<$TzMqnh))kCST{p&BOSfq8gU{BI3nERnUlm-`a
z3N97hyEI9~JR2pLTXyeNL=(p-lUsn5fM`S~@$8LB9pHGR?88!^l&CF8IuBEt4B5AV
zWSV5qbNN~_ZM=<*jUnXb^srGox3WPN;*}S8N1ol$pLZ^FmO)yK;?#@iJ$*ou7L<Rl
zNyDMvaklPlcyLt+w|6O1s<R`|4?i_+CNU)I)yxOOyH+Ef@JV=XUnkwxC;9a-QPX=Z
zcgc6+g5Y<=&n*?$@4c2`*hkCFEk7Jh^-Y{+Ppo`OO&J36fZ9<8z2$F-2y^8ndMgut
zUyxT!GRMNCIZZ#PVHz6xZXhtg_%!TuTVQ@vXLb3!{@(H&3GX-8Qctd)S$ujQm&KP=
zs(-)1H{aa>RP<(Q-R>`gcuQ1rSMtaR4|aLq@#7(eimoj{Eo1}J$vyEo*V<Ap;g(&k
z1SDQ@%kY`HzpfPoP@URCR#a4^{i_a-=}1D>9CU3Q@Hn2k+*r&gsTqVh+_XbY!Ku;q
z<KQ8+wW9K(#&H-vRW?k)4jDRJFR{*JbsEbc!-Gj2V8AUrHcqtuU6;e&^UJ5MDCj?k
z3~0flKcSEsvr2iq^ivE204~Rdc-++!*`#46u@4ARUdQPz+Oy4*X9E)&=_71Kb)6)j
zZ#pWQe5X`ZCs|r!IDMC4LV4kijXW)Z@cy?2MJ4yy{)?`UIZ}AjSsqcGdtNHGdAa_f
zgS)j)jN|?6aN~`o6onb;dYEe2-5Nk2ng0w1Q!nT*LI>`-#YC~i%@6osRgJb^Qp6*!
zCK(Ng4j-qNA8$$wi(HZQ0)k9wbbPKhR07ZPlSDMsyXS=34}M@(GQ!3wS;{H9@-AT{
z@X+U~6Sh>@3<Y`5J$z()D%)KeD_DMO^-QQOpV8i$@O2(Yw^_-=-np|AKl;q-T5j0U
z<}9=LC-(L{+IL$<0cG>#vtFcqtpbI(IwUIM(b7x^)ae698R~2T4(SZYfzgh7UVwyj
ze&N`jOm6e3@Vmx6MSBGDtQMw9L4)C}+3+41>}Z_z<K*lHzNRPe;I&2Stq?Q=!htT*
z(mk{Uj$%DPVVG_Z&>3D`T}9OtE<XOXq)Kus?YF5H@d$5Rwq^%LQwV!CFsO$+UjYtz
z>?s~`jUNxI7k57bH2^dS%_&}+@7epBO~(tW7pcR=Ni`7mZhoMc^#Z_O5xnd>{nQEx
z>%ELBx!I9(&Ul&-d!I*&<-;<I5^(9?X1zRti+2NmCfVaaUJEz#BML&so137M!bUj~
zo80@G2^3-ZS5zIgX94q(`Jjic4tfPGBlW+@ff$`aQeK7c{Hb`Qtm7c8`6@k~Q_tNZ
zK?BV6g`ZYm#cS4U#>VecSD~9x1|X+i>a{p=-h*$bN_=_D;yPl$7?LbzwEgya3_8gg
zT)nm61+zz?M@3UaDlcaw;gmFakMUAVpHk2syL@<Q&U&q+m38ilZkz316PvoM>~oSw
zygP~{7}2GbC@$pO2!fCDft%R}yI`(NO4e&s!$C)+2OZ+RBq(_01>vz?$by~?0<Gpc
z{^cytYuGd6^%1UOIAk{R>5n5sb8KenWe`9=%wVnblo<$S{45I6bpvxifRuG;F|&to
zJ{*GMwTD6M;!~_9JO?qoIx|lg6O{ASiFe_h^?|<UM=@{34}Mhe*<%KzfwwT4<vwF6
zj>8$*<$y=bAj@HL5HF-9+DICO**pMqUYy~EBq>Ds5Gd`dT=hDs6`4i^ySA7CDIg*y
z_yX|c%}#PhV#d48{=HH;c|#4cj3DoKMpt5BVXo7bqfU-g_Yg!zrcd0P0K$O{pfrHq
zz41Wrt*HZ_zHJt-|I9&8-2Qc@JPVrC^)A7R8HQ5;CR~M$ao;j~rR(C;`5^RI9GH11
zhWZvs>Vl&Jqvc7!F>mgV$5Ke8vvG2As;ByF0W8ivdw0IG52~{QTDxC5ejgE@Rq~jJ
zRmR8phcbBf6T;JNTv+NOIs2aOoYI&nc~C`=MBknC(mzL~yGQV%p}L!SL3F$2^9eqU
zES~l%sCsyK4ahd2Dgtca0iL|UJ0t1fa;GWbw#B@h1J7Fndtf$S1ECt;f|C#?kM7^U
zpK#y4Tah;uVF>Cul6QEfJ#%X9)ls2O4H{&`W75fYMHT?R-fb<ZnconrAjn%6?*%UT
z60iz2HN*Nt`5Ng6P9@yyeL=lb-=YbRK7qmouRx+Wtm-N2HA6Kvy%tHBKceL_>`maR
zg8{c*sE0(8^Z;$C0T_vPCh8*2)fjo0PB^P|Wbe6&hn4Iv7)hBGPk;mktEBytCncb<
zzsIBjN1<5I^>NC+;f?!v>qdUVz|G-QT--IBRQblY4Iqm8?wDU$4`HusS^jR9f9SB+
zQ3U=SB=Zn?BH(N8u!9Dt5F}r>Ba>zRA8lV5mSvlDEeL|Nffz_gNh2zuG)hZ%ql6&c
zDJ7s-NQ%;el!SDHVo-u2-6*Iuk_rOfKH<iBd=6*keSE)ne#{}p%za<idG1(yt+f#k
z!Wft=J)hpbRpj?rdh+Hmbg>5(WnWp;N3uVJN<_l`Q<HgqmeVMh;n1RL0oAk3@xnHh
z(vYu>p=4ApceWB23<xJ|(st7QlqDE3cZX{F9xq>*rz_`+L`A8PAW3KI5(8)b@DuYS
z4eu8T8*VD}Y`ir3ERo;e057yRj-hbtJ0~mn!Ww6()ukyFPg`j4BB1C=DKXG?xgH$6
zck$2~yqv)FG!eqZ+=@q|P*NuO9Ra}3`f`R)BgJvW8bveaG_F7}u)I#-l&hbip2Y%I
z7v)QW380d&hiE?WCP|DAUdyF4SzTM+x@W!UdbXWNH@t~z9k|p{-^ivSr*5mI${$tg
z&4WBzoYRxgTn~AI*h=CKB{NIcbCsL!n#uvv&{n<YO16hxXyu8`%DGOuIQ*K9t{BkD
zcLZl^L^RSId7~I^<%h7^wtsl{?64MdF95-E1a4i&U3zH2=765dzc`UQPT{aT+5KqI
z3V?+tx~0mHAy9%<BDL1KiaP}fSZ*$}CbH;;fouc|1H=d!U3McQBltFZ0V<XWwNmBd
zgzp4AnaF_Z7`rMJmO`R8v06f96dyu_UFJv=Q(C7NjXEu>X<ibVM(ho+6t;ZK>uC29
zfwHa#9k(F)8YWy4O1_~sv%VL2#m-aO=0H08VDUc6WwMJFCR@Bs`Jg<>0~NKno_zpZ
z2?YNWUEs0qI}W74(-5rm!9DmpSOsQf7A%6!u^uD~=W(SqIpHdFR;i>V6pb8!b;1=>
zMy$yqM+b+V`HzC5w!*Yxm}UOyNP=#~*WqVfrs=2lX{4<YA3f8im?icdjG8Rc9E!=P
zsNzP}$oK4_1qZ}p{(`KJ$LgLjl1E@})&IV^^RXeI6!O$$QGaqApxh2p5yAV5U=9}i
z=*UHC_b;^Fl^(?%3X4jlPFkGfQO?M(Ly2@IEd_jI(Ny6MRcqd@6DAG&$9ju%QqtDk
zr+|}611~yMz-<mt*Am9rv1Li)vBwVCPD_S4Vs81DX-Q~Hxjk%E9Y{vMw1hLvzKF>;
zQ^zE_A`wr|ijkIh1MqkqS@2B@$%ll&2lwN>zkB*Hq+8Sp23CU2oddU7GbMT*SUY#s
z!NO6Z3()m5z&r@Ojy9Jaiy|99x68_$$_DPT_5&U(3zu>e|8l_5mU79J6Z}r03KT$#
zJjzIwoT67U(nHSxPzj4UN#6R@aZEXNX~yyq%qg!w$Xv{`>g7pSdKBmsQ8hhnH@5TX
z78bPw`;h!Lj#5nXM>vWQJw$J65EhCPom8hK0KzLz8v{q$Y#_KlnzutO6$;hf*aZNz
zN5C_#gPMc@AD?XT$}8f-r{4DRX)BQyk{;oa^XQTm`~ezBGgJj0b-Gl(6Kizk>vOt1
z`(8IZb`|gWo!|XOY6nI~*D0Aj?z1nNMV|5C0=zWcA^cYGX4v&{!?hG_itn>*|2%@U
z&83Ww`<H-W-vIJMqIp&*+tUnBnNchztOb*^61^F8pI|@4Z=amVVF$V4Y6zILGGxmu
zPq(FqAw8RPWfsBaX&<f+!FGAA4LAxECrvdLtk+xK)`$+nVp991Nv#{f!L2$O%aH^b
zx%yDrcNlJiw*m}?MOQ3w_+7`F+}$YsC@YAXr~g^Fsq?{p`t+$XG*aK;C@(x}i9VTn
zPC_D>;Bs#9Cb$jea8~*8@$oT2UfN>f;^)_To)YMNgI;Jc`{V1oQBWa|o2a)+&~ONe
z3O(?KbwwX9_Q=&(<d7a{Vx(SRVJgkT({6t}gOJ*|szUxVTTO{2A?U1?K;w6$_ehon
zwcE%ZWV?3Wt9kJypd=jV#NJh?gl?VNZO%C8V}@ceEV>`CE(gI*<NDev187;Cdo^vg
zV)=9+rzUlI`K2kCXk@6SUldFR#7Z(e{bVB=ctV6vS^6v*z?-&V-Cs5;rvWG_Gvi?l
zd@4eK8S3wk6UjdXbPqN&KexZm2LF4q9qPpf1JTH1=YyqeZFAr3?0$MFs4YWH77bUX
zz-}oLLfgdX3sxxoqd=)Q-abxo)4*fuG(y;jtDp2{>F`cHqpnQ69RZlgb%OBYnz;$L
zvsB&05UWC><t$D0blK#M;l9I<95tQYHuSHHFD2|Ke)YAm8P&d>Kkj21);Kc7#7h9?
zgrOp6dZ6;5u@re97a)(w&=2Ji>7rGc^H6)3z-*E#ej4tQF?d*#yjG*N0ggnw(8pej
zC(X}Le|9D8&6%xpY5mRbv|st9^cg>}N)K5Kk;ml+LYgY!l;Wy_loZ#3OA(k6<SQ3o
z2DZR_+d|rz`8^mdWJCOK$nR1S{m+Afzx`<q9&Vez1r){+;CdqmLoync+-LflegsAb
zg|Ai&A4dh0Rc{dlF9MK?g@VlYm2x;5DuCu~6i8?9CqNLR70qiC_ic4`!l_6-&h^7?
z)E1sQcTVaKCBNnA#Aaed0C96!!s^aj4nFe8_3@SO7x|4<TWsqeFkA}YJ1&UY%;=L1
z$1%b7aSe!XjlmKLmBI}mh+|-5lVj1yW&;lq`MI~{oK!}xt|gN|*?gp;5pUAw{~dCn
z3*Yo`ZBPp2!@vZ-33=FX+*<WWvy-8C+B(F)Sg?=SAh*cYg-yec5;Nw{jSG~WfNAFo
z2_{VIZcDRuzxnu#O#9_R{Iyz$=6}mg`P(!6*N-*4I0$V6p+mNi7J0=iOs>XFCL*BW
zkp?coD42V@7ZP}1zjZ5oXvna#d9zvqz~$+8*;g4C;GCSt6~)1!->6y7sa|IR{zqq^
z7bvkz!1FLtZ+0BZ_se}(H;+$3-F8D6coGga*JT*&>&t<Y5-C@Y=TTBjF~7tE@$*x}
zMN>=~9?BL06`3(?<qR}}T*Cro<jJ;_WJPGrnY9bjRP>;VqX(u}aj&`;2h^uwZP|Lg
z)<-6Yr+4wY_SW{#;!a$0ad{FBg5din#8ux)h>7_vo+?;`$80XjH13hKqTr=a;Q49|
zir!Jw?g`=WALKiZxzavCIBfC4lf-5d4!0_c;?m&4?+Z5P+9gV2?$gZi&n8?_lGrHk
zK-&ExnQU#Z=1UA?)prw5G}>zQ$7z`+QULHvqWbi-5LV;vgCSus{X@h)fhxKXXV)&q
zZd1KI`}RFrOPT>Ry<mL0zq}C&ZKM8cPP{nQgK&$hx|quE=sH~5-{P31enxX(OkJHD
zcg4+Agqjl7WUDl^D%;IH6E0vGbcoelpzdqV!pB~SB2Dm$?#yK+T6<VUn_4?VWI_4!
zN*BhKUrK>>YcLH_2B!9nyLXG@{6?B%b-*e!Ikpp^k3&UT1TfhufjYBQxlMn!J(sQG
zHqc7Cd+Ux<oq0My6$q`-&3xhi*-UWzTzYB@L1~=_jDA!ttjm^;Q&@v_a|X34g1BH_
zTrb4d4sAGmi8}*;5Um3KDGUA<i)R{x3G5V&NIGQUGM*RA{WvpYqi5yd1>nvcIKoZz
zR&A_%fDim_ebbjtz7Jh4j=ni`FK>C~cqiUh#g?8?`$io-8CgS-SHyOT;V68Hr$!lJ
zGG@&+Y!p4A*zV`b2dl;gAc7J(#c{+4d+R^1twn8zJh7l@)4@Z~Ox70tJdan%Twcpa
zhxog<9A}+i0MCxRt84D*fQ>sjg*t)uKSa?NlSI<#x2|(?>kauSBd}SpGJ3T&ui1Vu
z|1Ejbc;N%wD6#6KO5Umn;DQqAL?1{a^Pmnh?=LHK4grui&vsbM4Gbhh1;0QlH77OD
zt8i7a(a2C`?k>C%1rB59bMl{nS&x)Q?rUZBRVC7uE`qDM1xh&<))pL*1|DjpyR5qN
zUJlnzcU;P~)}a|!e?k;pr9l!`VN~u*ab-d{Tr$v^#f!7+qY&qGdrUbuy5Oo8F@-73
z&~9Hycue2>3?9oJx^e54Tzg9|klCZ4zdJ~h2xLdp&PW5YI|9Xhpp~eoXr5UM6+)#G
zh3uZGhlqo-5%Hk--sPtp9$WXlwEV2Pj2~tA81OzXR(c!)y(h5wdlbzwA@{J8k7FJh
z&1p>Y)yCJTe=v*&kCP^!@E@K_T{OOmg3<n%FaEeKaKRX%KySSxx0W;xpPC&=!f~i7
zl0TYdF$hB}aUBoRSwy5jEhmIc_tlUPvTg`kX*=+B8lh<&1?^b`z=u~Js)@XS1=YzA
z28tf2L}B}0;k{@)3k0r_rZ@W+WMUuR&ioW1Iwcz%PfbfAgJ!&`&01!vn2*op?~@g@
z++LxY;LGsk&HH(<rYS7|^uInAaoG;AJVl)t1k2DVh^t|0nd%Jb#o3C3)M(bM#{lh6
z)b;E5;<1%}TWLp##N~CK(<FFP?jgT(&y^XhheC6S$#=oLxA(<`uy8n;Vq)8Tl;qH@
z*-MBF)0x5`hC159fJ4;(GaP9<U<@8m9L8H5Zx6eF|KOU)+^09J)ZRbrLTPqY`i8OI
zpnkENh$tM0uw6ttuAmJznH#AOyf;5m{|Hx#{;DQiD6OQP!HVw&Z)-rrl<YDy5V*cy
znZtI}y#SGKV)!;}L?jv{XLawszVaywt3}`oS4EA3$Egh|bisdIY$C&aT3zxI*dnl>
zO;a&jxAfvcG2GnrmW4Qd>%J1&;o)KXxe@7Fkbv_>0hT)M1a;Cra-tAdB~g1AW=rX2
zMJdGa(ION_U!4^YpAfiP(g2DPEfx_BEZOpUK3h51ca)Vul!Qf-(~j>Iv`8s1S?W-c
z$;<ZRgLgv*(v%aHDuWu}_zD2b{yssDM!pIReN5yP2DI~B5d5QQy$=OWo46Ky@&k*C
zVJ2f8Sr6szbBTJS)hnA#s0m&MsOpr`6nq}5%o~)#S9Kwlnc?dyk`4nvavtctfPjUZ
zgNtVZlh#1@_G~ypNn*4~NlC^~RV7h%J=IrDs|CpaNv5WY=+KAua>Nee_p1Lk8%`)*
z+MlA<k3DDq@PqZ;uz_8uVH$fA^>MrGdsG7#-2`S!uNpj3AnZ4SOq3DkJTiH2)L+8<
z3zeg-N#_XxGS6>6x3WCL35JuWEE|C{cz{3=r1X{RkP(z2nkWrQK8b?XcC!WG<LfHZ
zHqaR4ZF5C8FIF3onwGq370pwJE|4k&?l0Ml8mfl1&@>SHL8TT6Oqwp@%OFUYDIgOi
zkG$s!jH9RcCYVb$du{8W58cRru<H`pNXV`)lY9T!L8sn>)=?M_)%dPhsu}vZht$8b
z|2}5|Yn5)K;pKrt7(dEr{Tw%z6A#~^79Pkoko^oVMY_<U{hVOT=w#AT&eh@ecW;Wp
zr&Osow%lHl(2mcAVXRhVfX7^mNmbu%T-0W!gK6pW7q7T)&(6j39cSDgP(LQ;gW(0t
zKA3i2P~67>wG@;@=OF^b(hG=sF{0SNG+X7KCJbZRv#+3@Z2-ITFv=Wv%?)TxBY_Z6
zTLkrLBb3lr(o_P#UQY_Df$-#YpxC!P{}gKZ&EPz6W+596<nlxRh0;tq%*9^7D9iyO
zX5oS9gsOQ>&Vf0_RY$lE?gP_zB0_0`po-8$pwC4Al?mXohQMf88m{!B)wnM5-RE{1
zSFPmuN+3#d<NR?9-_VC?E0W_W#I&C}=nsr*5wj>v{kYAI`Y%6zyY91cz7HbbmHfr|
z`7r3Lq~ZkJD4X>mk^}2v!5K_LLIDREGnpV?<cf4JnE)`oGc(_$kr9^ryrWugW!)&Y
zhfV6h;WrzB?;urCuSf!M3jW&QCvmaB!2Og2F=ehg1K%v~>Q)?{%C)lKb<LK(?`Wk@
zU+D5N|KmfL&c_EBf7q^j3w#KyCSbOZHwE_3$4wf-_W&V<{kSMD6|km-rB}dMP6MEI
z6>%@%2cn_q>Q%!^QQY&;jms!1hBy@#Wn)tyN}b2@^l3|#_=rtVZM^M7n+^f=7NOvK
zq!S#dvA8D2ZXsmxltMGAhpMM}X6<hIFkz%!Tjo~92e!mLrKKLc<|4Yj*ti8XM2oTW
zFK+DXsYrShwm|{|!o_Ig|9|MhEUhmkzx|i$Y@H{XCy;bNC7+a#FbqL8R_)^IEcZom
zPjzXx`3ple_ntaf+3EqW4F{`6c0s#to47C(eEF*(5f3I1ngvQZo>i5xikh$U#l~;j
z1m_gms!5;l55Avg_Yhe3DX>xPn%|lFXb;d_-OeP9DsC4Ahmb129tKyV5NhfR>?_7@
z^A3c7vq#NaTQth*3T*HQfCW2Ja3J<-OpFpA;gJHyIt(^aVDy3T)u%WoH31?fxXr4T
z5uQeaFWCs5bZj8%&+BG8Ey#8j9d_<J?oQB@;P9udG4n2ZairWN*r_B+)<Gyr(sB67
z_OVXgKGv~XoOi`T{G*<AAI1JJ2l-blLWrIWXdP=8vmfLUIF)#Q&wGOxx0kvV4kf38
z@LEBlJA@wYt3O=DW1WTp5Dte{RY#*90w9tc1fH3A4v=g?(497MZA9OJ`bGvy-oj3(
zDYbb``f+{f%&Z%RT9)bxAKsxhb};(3-^ndq?S$LPJwN90usuiMOwZqPCE+vO3Ypv)
zz$e8^hS36Ih(+Op78VO}!cdjt?!9=(Am9Z9$#qvS88d-q)BUUmD~uRAs#NF)gf6`z
z8Ux>*4;EQJU_IgW6;4{(XP=y9O`s!ztHB(3F|-qSCIdI%@WxTHrihYQ402Y?`gsCo
z((yjGoQjMr!Arvvs@&vrH}(cP|E<T{1mW>WV>N9^;V~D#?56%_kGYCYF-KqUJ2cA+
zySGV~l`(z6?scXxN_O7|Hy7sI(BpS*wtHF#8$pcM9FpC;@1dEVyUEw=8s9J_ZsQ*P
zQoFOw<!6hY7GG7Mh+?AVFuX*lb^}nSFt-oix5u--3+!an)SO{V3LV%aOEZHrvxYbK
zGanm2i%e3IK`6AT5Q&GyB~p+%N1$1B_X4V0!3;UV%_k<aP?NH{ZU;3?<&BjJw9**C
z%=%HTCUA>dD?MC*iRJp?g0@zGeA>(zs1u4x`<2JcNgGSX)dHB(X4K!kd*|*{eRa+F
z`L*5d-QDCauVEi*I22qGKw3<n$pbWb4os;hOn}RC&FF1{D590<JUkzp399*XFk7Jm
zreh@B-OJCYpp|RwF>8*|YFm8*=r_-5xrZ(fUdnbd<=#%Tk8IBRQF<=CIAQhA%Jd4F
zkZ&5wYS;Y<9CG!W;{?G}V=lO2ePsd-_`Tjkj-?k6(1o`8ac6o$AXJvuSI(>HAaz)>
zQEgzLLA4jin1O)Gk4kYFY>&HoQAQ_^jdt`G5Zje3_eEB8WM4a#c~3av&Z#xxE5QGK
zAI~Lu4b#Lm6Qs@ThLS!CjGP)E!JNk}#>K<CLTiQOTtGyW^~JtZTh=DN3iNkW@YaVi
z+}zKhtEnm#lYf&yg%5~sX?CjB0Q%R#WMf9vg<KJA->}#%<);i5)t>Dg_qhG2Mel~o
z5Hsn7UHgf2yN@3}AWCDXaP>s|!~UbPu?CAr75PW&?uio|V7n~XZrrIh^a?WfDD0U@
z6600%c8Y6XmR+={)ch135<-)hm>4SP;XLt1x`Ta6&?wy4mSsCO_7m`}JV_P0>57=6
z4|}NkC`3u9UKk($XK$l5x&|}*<e_YbaZfUUiQ4@)_A#MFNj>7c<DHJAz<wLoP-1U{
z1@0p$QTloN3p*h+HNq!``DXy&9!lU%YNW@2$;yYuk0;-$$*%1*{zh@np6QH`lk?C>
z+SRMa#Lq;A21d-E>Gn63sevZRLH)NK^XzF1IjM6%bmJlCA59$e_ky9q(<eZ=9@GIY
z7(J9tt7^LBA8dbov-9c&Wd3|%G23`l{di{%=Mi4(0#)<sq;ijCYi6yyWPP(r`^9m!
z-K3<QSG78nZMx0Cyp&C?;;~=Not;2rFqn?|(glZYcDY<3(1n@ZiFHIe;E~**I7oNq
z7qZ^G`$_KoOLj<>#C$2qL6bicKe$R8LW$p+AiA@50JMpRP@{_Gyg8guETTpL@cqRz
zg?Zi^SJQ#c+gItKk)!c3%b;3+vY%`fWYW?=Fk4)*FxgHrM{!$rTXuV#A6T=fe({3A
z)mo84f?%OqVPINa!Xom`l$cFf*Xl`=drHGBY=8M8e@SVE23vEXT<>hT$^aTXm3S(~
zp#KiJ|D_1@VaCxevSbEdsQj~000TEbbw|&})@1q+abdtKo4vLORE$zEuti4=D0Ox4
zw-#6+gm;6WUNuwwLe8a^C$Rx!r+@UBD!k-dfmfZdLqhzmNqzN<aDt^KUy~Xo6s>m&
z;8Jq`CXe{s{i*Z{go)!nS9;al#0FJz=<KX540>epIzcp5580d^Zmx8Z7R$wp7m*qT
zp(L^qz;$W_2GZoxSrkW}0`=zcvKtt68v~0NxDE-OPd~)flnHaxb=i<eA2#EYij=#`
zoncm@>d6)5ffxSF1Ep}Ht7rH;e|O2Lw>jgp7m^<X-u%y5<33Tiu+^3YrZrVYIgp#8
z0IHV;>1*PEx)Jnw$}nIDw6T*H8SWtdVr7i$C@{psz*y}_2DJ9<<SeROZA!Km3`dLR
z=e|C#4GkI6_EuPIh<s@vQLQQeoVso}o%ZA;AN2MDwQ=|)l)vfFYRrBrnOOJyiIQpP
zyf2Kar6C%P0Cgp>au+V{4FdJ!N=ere2lNxcpBV8~Yd=BRc=^&1mJJQ5k|-|4>Z@Me
zC3cV0-vezBjN@%APp5DMV9R=+WsT<$dhlF7&L)oTijU{p%$RVtnErWxvzpCf^|!}L
zK)^TfrxyT?&;S02$zlVJ^3LzTbYJNuhr7_+rKKg{nMFe2+AJub1#q7S9<H>paq>u0
zv`S9Cem96XRkL-qz?Pl{P=T<T8j*=NPs(X>E#GTJfAi`I;=?f`4S&iR5q1$buhe$G
z<5L($g68SXJ<N>nZ;`o)Z43Oo|Ghu7sSgC!yMiwKJ}CU<cK;d*XaJN6gMB08d*N!>
z!v?a-C+e`HQ5L}+iOu>(fK|+a_vS$!v@{eomZ%JdNR0%K6D%Ei`|lSoB{yj=8iWMi
ztS#whS7?^gzAk?6hgLt;O{{f5U@m;xcysIY?`{T;r#NPUeS7&rs1`7Bu3aPe?O*)c
z*Z#|oxqD&RP!%{giDsI2=cT{^Si_<{g8>vJ?wM$?3HO@bpvu!-&z==(mT>TQldws$
zz6C@ogA^EDx*uhYqqtu>SK_($p1FlI=J}p^T(eM8kEkd=T2banX$MK5_|^Bl^X$I!
z>tow`o<7}}?jyfaJSK)zltf<a9no(t&%cMFeTmrK0>8c#D3@sVg1vL62{5~qp;AUy
zS`8MF+TfeL<?YfewH@{YAfgefqs1&J=kEbER~BZvParz+`?g%=V;r>N`&f#lN6CE3
z#q>~Ts^3U?{QTrGqr83;X!F8M=zbGtcVIJO!js%LMmtj(P|ZC$cjO<w11Yvy7z`o0
zHE}!a6Nmz4WuKy14i^x>$9f7gjpo62fdR_wQxJm`XlQ9g8NIw!Bdvr!4~S5G(c`lT
z@$4#(xi)g_#j~9;JVEHh@Gwa;{QlEognV*8Or@<Yz%D<vmggUS+FT|wa&iU)PL8)R
zS4son&KMdbRL)g$ltzHxRs)>wMMhBYA%+QfB)px_w9}&z5)ysIDZrNLT|KlyLp82-
z>Qv?}uk@I{c)GA43FCu^3R>USL7Dy+mF>UYdQB6CDH=V<Uf%?AcVpYG?b`Q$dSB}a
z&@NvjBN2ZH<N8K$+`kB%Gd@57RMQki;V`G1-i?nR0z0au##&RRoDv75eM>xONiJ3M
z(@$(vd@;V;c11l&k>`7Y27i4tJJ+J~^STE<K&$^;@R#c0Ux8xi3g$!SRQj*;8vIdL
zA$M#RJP7N+&UT^u+xPFlg40O`)B#;b%v?xMwHi<sS@4t#z0%CXgmdxZ2a|<9|2(^8
z6DI9c>oD%~d5!le8l{&KeA@c=JVKwTKIZ*42j-WD=aYjOqE#)CtPFbok;IPcMh!Sh
zB_oa-6q{ZWu&5Wo=u(D;3HoL>3>+Nt;H!m;MnR|~UwWW4lPuzBSy8dtezp7X_q$<&
z$~jsv_cl>&$qPMANK2BQYDw7*MfDLg$=$yV4D*;m=aY9!R}8Hv36ECVxqls;|H-d>
z_v51$Ui+ZGQxv8F$ja1IF!3!vVK=uFwK{s&e<Q#md5&X>#Gure<4||y6XW_3n_}W4
z0-Pwj2zp<;b^UkIO=KJ+@k{rpQn7OMqg1{f>H81oxhH+$>D2*Hj5JXC(c&;!K4saZ
z2+Zw-8BWk+<ZV~?uVuz}_f;KXLJW0K@ud#P$556^M6ZrceQuO`EF#S0@wQ22^OF6G
znf$k7>ne*$CmqqtvsXc^$aTE-cSZl-{@GuC40(kOoQ`_{1>7W1Mg#oZT&~(%$P)$R
zAol>^Zv^uN<E~Hg!vMddDQ_tZKZI@(xM+eRO5c>h3CDWiMg;V|40AJSZQY^-5?Lxj
z#Y8j8+SosK+eu+PeEk3!3EVw;aOZD7*boDz9g$@kcwY_;<ol#s|EzQu>Dg`BQy`5R
zC~zg{p+}k%sN>%=Ieu<FijJ)@%v_VKCDx!%P61KninEIbTu27ciSQE5178T;L?WmC
z0!(C1tm=Vsy{vc&{j+@u9Cz9hKqob(DFZ749PDt)N?Fm_=@+D0`K_RR$O`jKdS`Z!
zPTi%!BNgQ)`fW)6<qH4x7&4OG=d(*%m>C@83PVIzKy6Uv*qUb|iv(y@Ucg-AhS725
z<TNryP|ea-J9g~YLLzL~yNVPnh{(6uhG^Klr~`@eJ&-EHq@P#9T$w$mV%<{u>pb0y
zXT%c4<>;LhlU~RAOI@)&X2T(y&BKf^MtOZwoh3v3Q|S-k{q`#VyF>))<`S>fA{3>8
zE0d%kfqNXeF8~%q|B%<FW5OSpzc9%|Nx!mrA1gG3VZdwC{P6MP;;?GSHG*(MP<n!9
zb?#N0p#6vER-kPCAf?dOz`52dbo`FBt+SnLS&B%_JANtn&$G*?eyjGT`G^UQLMv4C
zf7DUksev{y8Ia$=EEo&j1Ip+ZCx9EHpdx9cyDN~?>1Lm@UB-vNnF|9R_rdYwC5*Pi
z@#WlK^sm=OH*UnLbV}G)YdmzvO;(urFirXbpK1@v_*DGIcWn>1etC}2$Zd(HDHN)h
zT!7QfbovnCKYILF4*0Mb6A?5`9Jrr7dzJ_EJ47XTGZmLsVfQgGoK-_!K87!ml;Xev
zMm!SAkc0%9$~zI^gfW#A0|{c3&4YvvQWIO_xFceett$eWGlLM)by{va{qj+_gSDq9
zCJqI-+`i2Zb^|*Dep(i6|DE3+wV#L1Q+Ei04N?uFvTqnB$6|=xqg-llBJO3eKa$gc
z0~Wh*5eWhSd5F(8A@PpiJa#HaU!*7;bJ@<M*qLw{_OraiVH|IQlUu(KO=O@v@3P4E
z<+ss9L>qH9%opWt9jbvYwOh*Nw_okwer&npvjCVQKFKKbJ%Qbe7=MB^**9XqKsR~j
zhr7aoh-Uz7N3dz-eQ|oi$27ZKs<Y`_;+=FcW+Egw%U$TJkHG5XVb#8!;^izTz~~8*
z{N5@mM1AXAh-+8w>+1tJ${M?Md&u%{Kjp7~5Vo5VrqGBR#iS4?$DQxvv!lDpt6UXW
z9rZ<RX^IZV^~ws62ABHkhZi?sZ%ED;6Lm^LN+B=tT?)Odoi?w|SjRN5uM}`FT8)#-
zjaG71WNv1t_Kyj_2gl7Z+|uOsv(MhFfguPZP5(bjs9ez^0Dg|b?8Vtkt<XZP8|37`
z$?LWHSWgVi&0`>OXj8-F;4zW`RW`}YVdKf4s}C9T7=1NWGIGS-Xu_w6v|(SNcx5h=
z^YH)uv0zk8ks<Fcbx7sOggISfn1T>9wb#NuqexW#!P<p(#VG9npnFLTKM!Dl)}Z%1
z$k1c>?2Z*WM`ge2cbF`z;HOiz?!VAW7PLCO?@(J+@~QMlPQCAsU;CdxL-BX_Fa3_a
zCpCx(8+WS)p%VD#UJJQ&xrhJ)vH%nxuC9zR0ipYD+Y>F7br6(QN&@(34EUEas01rH
zPHUhcA|$9*7l@C<N)CokIcGbEGv_)eR!8I=E7m0RyFPqo;x4$vY_7<kL$UOVqmLYg
zy=f}aHOv91>BBy8;sm$%x+fyNL+>hyt_xI83q8w34im|1lYGMF-X!<2Fj@55V@zIp
zI70`O!as+|hP<(1>=^-rKv@8jtP#%~SStm9hFx)j;2sVaL(0<f$-rY!nWLCl8S{V@
zmBT=Lv7&(AivSLsBsa9856Z#Xx_D+oav-i$;o`-8tqXkEKW4&)V`XJy3EWy8sV4i}
zun_z6N0a&XWWDM*7l_>5U>5?h3L0M`bugGAB*ytI<MIg)@S5Gl)CD0zm=>Gb%ixc2
zCtC$%y^Soi%-hoLJ<+c`qZJ08=#4<rB(_5SPqAm5xZB*yY(e1-;BwTLFPE*pAdj#I
zr>_W9K6E@e@JTRgw7(mVo`=IqoDzHIbl8SR)swd49^Pe7-{9k2geY~=kN@>n2`wUV
zKrw#Gdu1$@9*3na5mvWIkEer!#^=}N1Czv<yw!Ra^XeYWu>V0@<x9kw3vNYYZxj9L
z>3^)&ATI{Qi*eSo9;`xO9~t=*LoGxJg2)>NLlMmkft8FUD4M)bd=r|Mr1qlEsa$%X
zZ#pXMYp`BMf9cavaBtWcZ@tJ4C~ZaWqbI5yhhh$to5foL_n6Uvmn0w?^FD@lVk7eu
zV1f5@sQ}>14+M$*NG=o~iMv<{IM>O&`}WbnK#jY-Px<{zjb;Ul!(2ZduZDs#-@HNV
zlqj_yrJU{F``G`iEabq4tqZ#9Ir>@4M7wvt6FbzFS&6jvttEDsYCnN;qYlD~N!t85
zRHL^7Ziu3fkS@?ci~({(Ge1b~-nbppB{bA;gz3L-e6mkIB&%rZeW?GWZUaS2w5fyP
z>ul)C{%Qn3Gen5Zu2U2ab_fj7<AA|3l+G|ngF{TfioSBXGJ}#4Nn{F!S}%K|ziQ?A
zcj@A9ln1x(YQgpkw_*D4FT8FR`_B3^uG*bg|7ylJf0aN;Pk%5B3!`LgE6$97`5@zN
z*W6Qzqyo6OxLU#T)$%}5g2f*R#iS%89b!j=53px+KAp>{nalb-5*8rY8bhaZz}u}_
zgP&Q2Uh+Xp!0r;pp)U`VK2($H_7YZZCrEIJy!zv-q&YBMXz8~Nf8LXUf2$2DQ)@k`
zJSI<oA#p$=hu{Iv;UCG8E_b&3TTjgV0sKAo|Mw}8u`8D3hT;8T-yJ`zpREt^7I&|3
ziPvZYIWq`AkW-9}3b;(%xPCnZChOx)7Nd|)l!3_<eabvk02DcaR(A-%Y9S0FQXOjH
zMSQX6(0m2(&7)3@w!Kk5oYfgQg&ur*9Se~0txAdyib=Lhr{W3s+3MB2JN@W|H+4xB
zlRJhywgY>EAgyH$RVxq>m_a^pEg&E|n2`>TKW(4U()^_iUc#`S1?xJ~77qLEiw<xk
zsRixij{JubmAk^{kvCV<tjLbPw4Ue%f59%I>eB~xJgm&rw|S)QzBI`U%UrnH7J<hC
zytrd<twKQ|awR8l7YPCM4v`@2XaK$d!>Lnkq7%1@-NwKMPT>OC2eKGJMWIPO{ougE
z{sP<shrX`SIXk|o>lnnV@@@lbZp>viG@yBe$KMrY^e=xvA3;pxHW1D2CMZh6C@d;R
z`X9awSuWks7iyv@$47^M{aZ9U3A#LN%K;_H^Zah_Uk4(7PJcX7M2bQN3#e6Cac*CN
zT~wWoPfD1$%--ydu3Sm>@OMpciHnIIc?8f^z&x&-;i*8WD_D}9w<g!TLOSM6H?|u_
z5sTeGDAUfnb#uFGOg8vA&Aam7r+Fd6xRh)DMuB+Pe=EvGC~}vFme8;jNZWhK$y18e
zSb>)i1q}K|;ASkWK!fldz77KdmXES#6M8=VQPO1K-LQxy&~|cgExM2_2-_{s_42tu
z>EC?vWHVn-el~#T4$Rc#ptNtDPjp}C0(A-N_}QBT(T(7t14+32R28ouWt-U(fK(e%
z!~ql5)d46*IHEpHSf;}Mohz%2pqN}5m|I?UDaw@W0r}Yl`FJ7raX!R`x+4^%J#z|d
zcwmh>GcXYzsKJ~fi#Pu#keyuW^QeVqpbG2>#UE5N=spCSr|i;q|FOA3Zvt*>$NZ(A
zuL1M$Kkbp4e%y+nJW#w*$~yxOh#SmKgQhMIX&R1)KAKy5m4D~RWBZ$a%g>LXqp5m-
zr|ileTgNxYORGs+?_x`0JeGP0k;@J&h~xk^XuB9C+esGPl3cJ)Reg2mC@W~<pQ^4-
zW)g6~YRDW!yS#YmQeiHx+Eu_dEUCf>s9j+VLg9`b?FnF8237<_f?V>qc^y8<ArF8_
zkb^%FE>EKg0w7Wbm>J11EZ}*q5n*u)gnZWt{1k6K=mna3fm7o~+N?jcj9S4fnJa9Y
z_BxU|`Nc>%W)%mUYy5et&Rm^1A259S@>>zWQ5ax*WrcC`WZ&oMky~v-$Wa$6-1o%o
zLhWB+Krd1C?E@iDa@$ge$L1eBrvP~+i-Ff-1R50&-aS(UT2k{tc91-AjCbur$PJp~
zAyXp+=BRe(rEy&c=5r(LMq`Bho~t^)rJRFjpB%lFT0%^1aF?B7zUXmvVOzrR+ZL`>
zrskdS!N=_tU4JuL_4$tTg!*dU>2i{zXX>CTw6ba)Zxh!9n%{n5Vc~MCt|3j4*q|GO
zM<b|<V;uVUB6O~e61Iv_&W$vMH+WohwISpyGadOdv(u&}YHbVN3ZcR+vKizUKe+_@
z@JT9t;5r__He{6p`<#8&-#}Cd^jF5f=ybd}3Qm1XAn;liYB=3Kd^!|(ALg{gz|eSl
zs`z_`*iKa%Js?+ETE`1|6gf`yw146;V^RC!M-V1ut%P+Hwv2|rEn(`E8&roD`8#B(
z6reD<$5xViNSwa1F6^<4_4!=%hTV_Pd5x`SThG>U1ktpXui!oPt)s<0mgYlnj7HK%
z#>W1-AO0~K{Oh)-&Iilv+0Wd1KvR8xvaIBp$-EzKpEyp0<w-lMn1;4g=<Axou-`4t
zO}O;Fv~yKD>c&HW091BH6%=YHY;>$O_sC5BNUOOG`)74jrKDwU5`5MKEh1dtX6-2;
zY0S0Ca-xy6F}p)iC@eUPhB?+L>%OjT>xQ3M*k|{_sgvMBjLoxHHFcc^Z_vH<@OJF}
zTWzrUBiV5*7lVVlEGQTJw`1Z~iZS%Wa@wf{oueMf0?>t9A4!r%3%KQF-zi%Hg;fp_
z=r|#rruM!LbiuB#cNrZtSZB;Ej{Q0<BHYHIL=$$)>rur>-P|(g;!5I?2SlMh<t?}H
z?eSY*yk$q9N2CisjSSm~ZVfEl`r`7-%)OSU`<0=OE_S-D{ft5s3j$t;-)LU|Q`Zr2
z6_kaWe8m9K{X#)k*O=J%-VcDiE27TL<AZSbA?=OXPK`Xt_R@kbt6)bkg}&>vt5Mdy
z(dh&>2SNFXyr(oeoyBoyjsii<W$N=2b{l_6qyvrjUN2p0DXI?dfGHX)IQ%_AY%9Vz
z#m}CAdxn!domTk_o5{|>faGT3E%NOPLh}||V)YixWkiH`dbUwR)#8_YX!{TUmK`>K
zq^W^_@SLkARl#LD9c?~vkvf93hfmsb=3q-maSwmSLoayROfX(`c59x_-qX%Gl3{Sp
zsW{Ex%^I@a5)#t4XiKBD>M6Jfh8Lx0Z>?G5O!pDm9XvUsp;bH|^YFp?px$F!(?e@3
z!t^?fVw68}j>b^+RZ@|Skj1zqYM_tkJG<vHOg`B(b89~C%LWTQG$#5mT|G?2dvGYY
z3ZFoccMYx$j~d+ux^NKF$}Yj|@CvBRpJq2bBqnd;vhL#x5)TaxEvSOqI-T3190FCc
zk`ZZreSPGypx2Om8etXUXGLpfThqfwj%4bkYK@n;t3IPav1r=4BeGbO1WK+*$VocA
z3y!1k{}TyOR0gz!=K}ED^OF_g^#S=j1|~MkpSZ?<mRLgjY?s#s-o}Fq)lhyqv86Z;
zKTgNPf$ih}YdwK^4F+%}2*nLzDW-lco`yLRu0GS~uQT@ec(-h~IdCI6%=`4{GIk2D
zPM$RHEy`--=^Cc?O+H)CeNo=eGPrywo}}XP=^{<_L+wvvjGGURy;M>?{m`ep(_duc
zUTOD_gP8DWz6WtaK@25cu{M)MU=4IUUF|ZGo~Q6=bE!Sc9LEY2`N^eH*}T7xQfwN{
zlsVQ{WgNJk;J3#z3djI;0Ei)rA#FPtez55jH5COTeG&35nrLVX94`kNDRL4G@=|#$
zcMlk~XTzgli6T%wG6IGpSkz=xH!S2+FFa(7_nf;S?sd8o(!<_8dxmor$oWnd=A;eG
z3&Q$>9zO6fpo6}>EdymmuthP>n%=w@Pj0W-oDWbt9t63r9zcxqxMpcr1@x#uAUxHh
z8gSaOx99>QNte1WX4DR&5)C$L^M$MY3@|uiHJ8ylBD=1&EKUuD6;=e}e?qN%ci_W2
zs@)O8t@rCYGNMB(x#}V4JATf*+ODw{nx6l%|8&wBiM3!0klr8IJN$&W!IRVWEVEW%
z^BEeQ4;vmkqh~y3D_#;X<k!!VjM86nJkvs&^q3hJE`h7ZrF7;!2TyH9MA5xnz!U+$
z#FmKxFb;BXgF2v6s2Lyc+nM7-pmX2vG_?=T?pZ-6_IVH#G(bjj9AzOIhVmKdAoNP7
zUcvA;8Rn=(=e2bPJ?C#{v!&dIw~eN$$atsB0%ayR!?M6~bQ+B$cH%%B$ph1d%qn-8
zai{oC777-@hCgnXY2AK@3&@>@VAQOf?+PosC<(Nmt)C=sKg&y7j)G=244kRnDvvCj
z5Dz5MzAO$b9;a2c`~DAl;R;POsCB(UdRXx6(rDhb9q-BTu^xG$T8+lEpVSNd6!BCa
z{KrWlHjQBcC(%yZ2FGo!+*YW%cH5^+It_AE&jr@Vu;wVvcvKbvLCS<VPpGV1qCi!Z
zHz@t#3(X&#w)0Cz4++reJ5&7t=)QDpco;9Q@vX)shBLurc)Y5#PkZ|(utZ5{PrQl2
zL7&q{6c-GDj{#IY@9qdLf>B_j<mwf(hh)}7u?9?GIFPy++P5=;g&+#LNw7E9K<N9(
zK7?d=J$32@swB|FLxFdx)2qq~Xc!o_GXa&O%zpJY5IK-rnd3{ewE)luy_iZD2U81v
zP27d_c_lB8Jb2E9q9ibB>=2`bvCx--3D^QB1)Gqj91WHxyGKEa9|<EDzJ6igC7{N(
zwh9VnY2ddUZD9V{d;VY_n)G^L74N$a=>Sj1YhM<){Up_2O#3Cw|C(HVd~W-^tKup+
zbU5>Ge@lo=v6^HRx}6Daccj7so9#00!r|$|@0NT(4s>IsUGu=k?!FQy$z5`ToOVGR
zt2z5fb+Yw2Qi-f7=*H;i1O)snOet)a-O0WPbMuxnecX~Z)*RWic7Y*FCI7ihuuRh&
z1HYBd*N2987~tZ9x>rY42Nac3UkftSG8ti@^uGNNDE!hAa5ZGL{V7%7>aYL!J_<G=
zWgM!uK?LV*i&2BEYFj(nxH91njA<DVGEkDZm}an?KX<MU{vsmFki0HovF%W|fSnfY
znM}Zax(BVXhF(4JJ=g$5v+v5eqXn9kwj6^&qxh5|FpWYs(4~%Jp(#=xs>7s9oiED>
zEPsZCt9!Y6%)n^jn|ab&Kbt7Y`hmw@|0(Cz2&0_)R@hAX*T*d4N(k%XwLi7Q=ZJqp
zr5{Y1%eY-SuNOrCotOh*oN#-459CkNfkH4ed9c81ulGX6U2Qq9lTK7k3_JB%SnrMD
z+euX>eZS+l><$9R@1Mf=Y>qYWOCr7sumvlqa`TtJ>GSSQ-s!n#e}WhZs%xPgxA!;z
z&}HP+wUx2)O0d2M^kKhV^ppfV<+d(RFk18`^R&Pkh_bwJ9%NaBivH{dcr+<sqlL~w
zlWN2=%u3q6d*t(7Y3!v50w$%!d>BCD`^(jz?FTiqeH#2gi3GUfrCy>S^gb7Gsfz=W
zMntE+$D5+-KZj5>G?*&JN3MM4EMUURj|+Y~*xx?BD6m7h_#tL&=P$VypVt>zb_4i}
zlu7R6<BSi$hpZk<5(}FkTo-=WP)Ej>fJ78?nx4Il9hJi*Gnl?QGW7L|xh3;dkgz)Q
z-39A4pX8e*oKv^o`K9A*mMn%DW5&6gH-vGYl)QW-grIsT8w9A_z|t)3$dKUf6SF1n
zcBn6$R}_(XhP{-EEF{rQ&DIQ{hr}tTVOoB#gC+=8%!onDWzw5`KrR8BKtK-xVxVtp
zQ;UZlueLOepjl2$qz<O|3|=b>WjuXW5E~e5K*_w66Spz`JtjQdSceVBonfG5pJ<4&
zf}OqR!A$!}?`hXrL!?B`zD2x!5^2b`OT(rz>J@MWCQUoF4PyW8wr~B2BeMBWhr-9O
ze|@O3SWXU#a2K6EujhELZCYwwi7vnL0?kDLY#4z<(Q^WFO&9=;Qc$gCS>=ZE_}<j8
zmW{u=Tt6(FlbS_KP1&C|zSjEay!j297tin>e7f>{GXj5ls3LI-y;egGEHmD<RL5Ol
z!gmh{psA`a->z0vejV&xSz@X-eRCLr;MpZ`_>k5Bb5fBemqoG0S;^M}<_!hxGIn~g
z%e;f0k1|h!)dh+i*>YCN?R7<>HobF2RaOQTl^{XTw)M9vO%)}fleiTEut*P8#+tYI
zn)fx_?XQhr;PXxhne)GG*+IPslk>YY=DW84@MpyROSOlei{dEJ?AZHXB|L`OSmbHB
zpvMN*>)YmGR9+(YKIO7z57?%QX6GFa>Y_xD_4AR)i&b_^Q!Noh*~kS{HeSf<iPbGi
zer1SaX8ooNi`bvn%NQzW0{AG;&JQNsYAX(3qC3*rz-e=uHTuIm^Hk+qzk8^>AR93p
z>AqyS`BoE;y#)hR)S~siPuIwa0}A`J{tk_6y=4+vLsPdm20Ds`XF!P4P)jQOgcVd0
zMQORawqk8=96{IoVk;QrppHvC-~4o*)!W{piY%MnC~XZa3Ya**CHFWy0$bQ|#@;;&
zn$~n8KI^`4S?|!1?eir<gL7-?5bVA1z3(Kv0RFnvisswGkH0*=e?3xspUnqZ!~A|*
zag1pLEqtLuyj=ZZ1q|KlK|${HImzz;l83^tJ0o;>&}@qttPpj}o!=<aqiGNPfh-VM
zdmk>LU<w9>4-viF8LX+0KnDyMMM0imYIQfFZFfwmtE{_xH*0G1pq$=LA%P67O25{z
z?!7{m_Z~MLuQ8J_jXC{l^OTZVK$lsML{*1QT@L9hb7~x3P_-2Y<j5FE9ea-+oyr}Q
z!yOp-He==GaT6GAmpjFZM%PK~(_n6YAla-pm9qn88>}ECx4+DiB~e7|-WyrGXtk;g
zULRdIbGKc&P06=X%>9Mvb0PlHOm~caOIF#cbVa%dQQ4XsPqyQihw8f@2Vrwa#SEQL
zs-kQ%#NIimC@&8aEs=xq^>J{a=_J2<@PrtlkFCq?G;0AE|HLC!A-C6xBV3>3v?OQh
z+pN-NcWt`2`&FLG)Dbyzn*GqE(c5Y)6dcn!=?-8b4*sng7*CmidQba)4@gRrLEeu>
z4CZeWe8mrn8cK|;>SHc69SM|!kc6O*9JqSiyhnYe8ye<gP+2%#X8ON>Dlq|EB^d*`
zM&jS$o^6I&mT?WQEA1I6{gaov6UWi&nmK4jb%JArs%D=P&VlNcuV2z=9Oq7KDM#hw
z$*}n-o78M9YO3~ZB_;ED*FDrlQYxkghECF*Y@xb@DRn%tG3Vf(Ar3#XSOB5xqazKd
znndL{w6IW?TS$S1ES4%%ovv4+(yFm{rN}ECd?F^#cWq-~`2<@~irk7E2nv$-z0Um)
zEey>h?yaQ&qb>#f9ou(sGiUk``hpTo8iMmEz;PXi#{-Wp+5pl}W-l9nRM?=YW`Ef6
z8l1E&5{YlzxY3aZ8K&{;$JbfC&FTPgFo8vbebLQ4ReVYzrwi`h6$%`BYaUd}?`|oO
zjD3@{Gyi&-Hoo9IT_P>qmc$cxV=&j(bkFS4D{n|(XL0E25@FPb%8Y(RWKaUtM@PhW
z3p#k+A7_i$QIbpP*lnLnVYEGhXksxeXo2lpTw^D}b1_VwvUdBKZ0ESYARNBA6kI(N
zfkwi#e;ZD4-~juFk=k?C*J&o#sP<g{!aktUB8>&IK!XA+Yn7c9{MiyAnZC?3_wodT
zB;zk{6FaOF)U)*~SwKLxl+5DPhyTO*N5(>2nL@U%g1yT7jq*Xy?Br8fDm5EN-}$HG
zd~2#xegD#M>)miw`V{l7d*}>O9X}{af~>K?Uwu3fEdq!JyVL9@jUy0Ej|LT6x#<iR
zaTdyl_*dfhV&0gc`c}W>FC8z9!++oL!@-|Z$e*3<&sr4!TIFWMT-}ZJmxlql%Jx`D
zObi<UI33tW+NP1KESO|YDms8gAF`y?bAxw)1eTjEq}#=dNvb*edd|~*>8uLoAE5$a
zo-&OO!u~3mEmdv4I@0szEEUyoyv1@qQ){Igvs;G5)eT3H%*W`+t0++WV~)Jz_K_zu
zz~x7~3G;x8u=*+1no}$99E_zO_fo)&b#-}$!eyq?5znpU9-MpYihpzNOFZH!zO?j#
z-OT^$0LbH@pbsEt#}ZgvDFHhz1*Y5b7QqKb60nb2M!`l%?JB^PdXHJH0p|JsIVFc~
z$xv>KmT1BTa3K#i=^t?U!g+{G5X>P6tgnl|fKioxR)ds(!}7z}@UN@VJ}tL?;K=4<
zUf+;vP4AJ7$}F2fb7(lOu75W}oUQD1qz2U0jmd6m56l-*(<7gKTVSrD5tLFU05{;M
z$Nym(*x3DJb#NmR#LRVo=4`t61I&NfqwOPPK;1J8OiVWktThWQBC?yor*0%YHLf)x
z{WcVHbs)xJO;w0@?lLog=A!btOYhmMu*1vjMPt?r;~A0#yQ%B=_tvc%WY&V+Jo~uc
z1cdI0KNQevY0t6sFb*1Eo(1O+n`$Ch2O??kYDUKtT>djS<0fw&r~z!m1jCo&4u?Yf
zi4Jx&u<XJcOMhDh`MdlKTl*+nl&P-mvI8ud^sj9TM7c7z%Cu+6E%Y#vKVwI<9XWnU
zgP6kYxjrgy1J|zY`1Ka`?Ng*B*;whUK6Mk~oh)?RDIjm{afXqs39zS_&m09*s{HFH
zw2mF_fNF(8rKwqLr2))4S}5p~4p18DSXZMPeEoB;&PUFM^861BC&scr<9$ywH7;8<
zR_x7F&h)evE|3pPq(|ZL)WaCe*zy$xCOi@*GuF|bIH3fL-IwcSm!X)}=8N;9ZZ2Ad
zNC_6GPF*0TltosVg;#`wV&w_orZS_lQDDC5B*NgBPM>SFFz!wJKa40POd$-i2TGl%
z9+=((n00Khx*B%9EB3~58TmQR4D@G(Wpn~OuUCmdC!9=*j_A_$SbVcydtEZm?J4g^
z-@hB~kLrjQ_eg|h`r)wmE=-cLw5&A~qIL4@Oj};^KD4;oRe$UQ8oYK1)bK-kD*J?Z
zAu!Vk2P{NNSs9@*5{OI@a2D~hHI{)EqnxeF|CG2tU-dW`2%+77%HZi+e~&IGMHAZi
zJus!}=vjx{j@a<5Go3l{sJhnZ%!4Q-dv^TY|LGySJXlo;LYZ2NMC$`9;PFsswaA(U
zK5Au7@tNdKJC%?j^|!AdJ6KqDh~Wy9XqQMhGK8(nEuL*@NOP-}v9}-Zoqc)lP^)BJ
z?$XmaMe{XcqpHgfdBjwFI{n|qXfzXS9rH&7t{xw&ix;O_&H|`3t_@>2^QAZDy-G?-
zJ(nQLFLi;9IPx74*Hh~#veF5%@AU)IYs5oNmW`4pyrYOk1|zM4tfZ`J(5-QUc09;l
zE#!CQiEkS|Ah2%^hd+4mv3B_2V)Ubx8Dcot-xRh`*P-gtr7Mj9lSD+l`VS?`4y@`;
zWV|}|IPtGVF5_NB-Y;oKG$y^r)MB3_Ee^`!l4d6@+K!{sSD3w$(gh{0Iqyy_LM!zW
ziGpXd&&D5EnT=H9;*Ad6{k+Smet)dIgd-c%sH`#JINz5(UF-^~`&X(CZVrETV4H5=
zztv=(x?$iZ&eiNBy-v{VP>ecjp&zs2sqL@sQgiOaBMArA!kPXVb7Yfd0xwcNGZp$u
zrR}~F_E322DcU1Ka~c~qPYWA^wMIJsDa$)I?%at4?u|4Q{Leqd@GIe9vu_nO_zNd%
zB-dbmKqvL!CmrK=aSNMm6AkrTar~76`%RCApI-}z;}z3pKvG5Q@}Q3h8`mnfjukJ>
z_X9*0<mb3A=5FR@+$sB-=KA$Ra|U7Ob$KmiwUUeT@pJ1*gDOr>I7u5qPsHB3Iz_Eq
zbXnbiM^?v(25#g*IquXg493S72XiA+AN0VDYzD8$0&kao`Xca*k``L|LbsZo)eE*e
zh_mz^Xc1W^15fs|s>>m^FAQ32-8HpuTOL{bdyS_|2RX&e0$_``&nc<E1FD52;Q%(i
zE5p_UnzdNDr4mhp_pu3=?ms@Dian_C8@LsV*&~LjjeOqWw!L9R5gL)Gp;esW(slaS
z0%08CY#@;u%iV}Ji@OKC;+F=fOxZ!A|2`t6Knd-U05Nc312WP;KJHGv&H-1>>D+G1
z;p}!*7SeX-TO|JWrOq;%&Sdh)Xo~8yCLxP^rq89}CfYu*YTnvK;yXl`W*7h{W&<R$
zEPtdYlZ)YdS7ufI!OE-mQu_)d8wMabK+A2~4?PnrGO2Vrn}V284Pd*;B6)|>PWwUt
z8*q-#0ZDVju{uQ&cjH2wfEJ-ApY4!OO4sP6cS+|FzKINz+yGFZ2h@mT;xU{{qwRC4
z*Dv=2oGY|%f;@!T`^VQHW7zYk{mH@v{^^6eR&QUyOk)JhK^&d%NAYRMl3*p9tI$)H
zCeeVMr!BkQ&ikhUW<-UV<_bLE`KSg>h^xc%>%TjyKYvdF+Wvq<rb~}Wwub<lIoEKP
zFLG3YyD$P>2z1McI1mX%h;~r6br%BDn_~D?^xel8qG4R_6mK#Lu0%=eUw*(#i##^<
z7401Y^_a58CyPgbj+Q-)M<9Hbf!SiIgyZ86deKT^l=h~#L+xOBR`6)cNdIodBfcBE
zXWeU5wwL8%v)it%IPrs<1tO|~jG6-mZnAdxfP>gfLZgF>zog~l!XdlJz^$6fEa_MG
zC4<8~2Qa+FRN9cB6R`&9XY7`*HdNU;O(V?%K#mPCT{(Z76W&=UlLA%!L*ATlP6K`5
z!5VsN2KOT5q;EUBXcVzr77tKEE)Za;HJMq`w+{c)rzFWgC455G4jdDC0BVY9_|V|Z
zeBuE#O#8k+L5vApfx+D+9txL_2#uEO>lFhmaOH^j41lF?-`5WPeDc5PjIg3M;pb72
z?qA0#_vI1*O+zbNwI%g{8;7(h&L<ThwZgV*bwH(U`n}PL7!W!;_2wbmKnlD(*dF#=
z0<wHzkz)V{51PK|+18>>5H&tn@>d*ge+hs2@p*UV-G-RcuLZJaa)RCRQlso*8)Szo
z4z0E0P6hhz$iXw>s*gGL%gG>njt-u1DNEBZoc+*^)#{D;b;COV45e<Nn0Tr0jem&w
zQQi`kytQrN_S-LQ%ZS2d?CpEpzdVYXTiDxhcc2i6f!$!LF&aRj9|A-)?Pd5cdEh_L
z^YC<Nn`%h|#0$+z1DsLz^F1HoR3O8lGH$aY`xxays07_LUD$O?6v3Xzy1z(o`0zXR
z+VaIQgAH23j@a308sWC)sFT`+cE>}``crGC*4yqKTBM0FedzG3yhTjGhbE})O*(-C
zGEg3U`kG2sZzm-z+&hD5|E%4c@&Ph)`xJcd=ivF5rXGWa7j3>?XUlD8A<S=^9*~le
z2}RN`=slmL!qku?YZ5J_i9CP)JaPdV1<_P-WOq&)5BMY>LruS4i@^M?B|^6jY>G|M
zF>;YwUPshPU_?w5O%ZfnFFG8T<6W>2!q)BDFu(JMYQAC|5y++oDBoQj_cg8kH3niT
zZ!&yAEYRQyrvfji6+*q&!n-!!=lAew5-@Q^Zsi3id;SdCAXCN$|Jn$6Ur;0Ey`K+;
zgbQ6SvNDRDE@r3XGW3BeueC@Dh*Qw@oO&10ZJqWu5RckZ5qa5=gD<Nd+9{r$jjI=-
z(^CPuIFpFi9@q_+J|qpAylhGJpH@0$AaQ6}!mU1n-Rte^=R3s2QO$K<viWZ~xdwd8
zR-Sp5rcLd69o^mJEZRE~G)(}>APkhzOAI76tv4=p{q8MdevfPky-lu3CY&9AJ-_(;
zzR;65zPMq~c9Cq`3t->TX4$n93jo8XfRhc$Y79B-J%IUn9l&r|<%Hxc^w4QXfg)6^
zk&RVaT6zvmk6i}6%$P%=6|093_oR1;6-xA*B?I#ERjf1Ye8&f^g(u3CeU`%6LZn{-
zp`-2|1B-}Y>jggiS$8s${3iZik76RLNQijPXCBBd+V{mFF1#N&$WBA{7;|aEg!Z8B
z{ly8(zkZp2!M&Tm!uNm!Z**{JA2#OXZas=?F*R^zlK~A6jfr`3wis1WK)i7@ndX6u
z$%GPfwl?H!<Pcew2R+9P7U1CI7Ye}i8MYG80OW~Ix9#oAoF6<_n`6tlEe3M(j5_&0
zlVfYHhzzrFuPk~?b;#YrnvBSPE{c-(eUs)r+h=y`qv^pb^*mV&VF~3Ppt^j2EUGO^
z(}aPw8NIOo{FO5MzE4$<1rru%N1BH#^uYx@`ixVCcxkyGg5+Rr+>`b;Cq3u|QvgW~
zJ#k3_c~H@VH&)qT9$Kho-kuR*@_vjCm2{2)H>B%PkM*Nq*)Ln*;E~Z_vxs{~8>K%H
zUE{;46>eJ{S)^$*yFdS{%v(&S<{?cGWMG*<qC;ZXBR~e+zZjj&2|SLx%SI&Nd$DMe
zUl1Uo3B(A*=aO3~fSs9tcvHEtk<S=lo_xe=icIO*uR%;pGq8@efJmYcdOr%+-2v$;
zlwAgBuVitNF5P1}$FnE224cxMdr5R<N-yqMQa>JXA+VT}>CmJ1xCy*v_>x%qq|fa-
z`2V`f7}m2p;#BOm?*cI;;+@jVuLGev0@xwne(0*+uZ5~C8E$}(&BI-L@J)WAZB~Db
zwuwm$0gBYRX5@gnRY({~*i>w?&{IwJ6v{$R&widtkV7+91k7p1imYKYY7B=K*`E|U
zso`et(L32+;U&4N<$6>s-iEy%vxrAV_#3e4-b?pR3Etjy0k-ntMH;r$*sxy#P%#N8
z5(X7g2plxg_&P(aaryT9W7xDmanpT31^6@ERHP0c6-a2k#W=-UFxZd5^4thiy}>RK
z$IpDOvqP$DG6N@D1n6K3+uK4l&}bmDqfOSDO&VB1T3447XYYw6)T@rB-}(4;<3Mdx
ztOmZJ`E7fC6UFj_($t6c)aZ|~P~y4@;=k7V9P>+y==&Be&=qz7T9P_@QzCG`*2jF*
z_u!xS`LY00|3he0Oa?p=^uf-O9s#pxF+GeyQ2q@AOLLu;lTbt>r$1(m?3lu)um~s{
zf>-<(fC#9anMo4_SK>zb6~PHCXV&REDj!LUN5^l|kMUZ?Q}V48JLD3LtJkrw^$O6S
z!90Ek<u#S+3aV@GEgiOHD43DW7odO-Rr516J@YOqpmhHkFJB|L`;X$~eRV^C_mNx(
zEyq+QGJ$uLuxYR`P+>0E33KOqpb1rw7=9b_Cq+xHI-J`Y&^FJ)PW|^0Ze`C-C+UyR
z6`mUtbc+9R^}~a!l4M0x9(*|q)@l;1=ldwG4UBRBr<b*+3w!W_36d_n31!#e+2m8C
zDbbw{av;JsqT2@YRX|YvX9v=Ud6S`o_^Qdt^h$|DKu|-9uBgx47&yRs1B^V7oCr7+
zM_@{)?F|b)pQ^tBsyX5bA%8UXdB=b&FRD{G^h?<*g$J0tf4Hs9Jg3PLRGPZ#w)(ww
zIZg%cn>oSrJROKC_C5UOzvL|%V;p>YTj7YAiupUpAid0aYP>2h>@Z|}Byxquk;#J@
zM17bJmTPQ2^pC&p7!F!O4$VmzI(e_r22{8CH@sPpu~}vlf%Jqdpe8sLgIZX85`i(^
z!L6XKdPFTl_1=pY=dLWms}3fJnspvN{I&Yx=G+*jRv*o5TxGw^uUxSB9L3QR;kb%d
zEi`j~SFJuIzg%yEzX7B7wIRAJzfbj}9PnD!iv~$MLiyIP<So(^wa7myW@(?^`nsF!
zE|^w?{qdEQV?x6UG_{2Zl=kavu|pX4X%S>_MDs&r({b0vIkY_v>K*onRKg^%j}5jl
zrDs+H<NW();f-`whhjLybz_9QDWAhD=RODMOqcwqs<zAe(YhF)@?(JXD9w<ZPv=fE
zmhius7~DUPzUoPn-S&#7fuiMw<bD0rBguhKoFifm5D?{;l&=9{XFCmg{&B%-DC3KG
z5-?D!ATd$&rp#KG>T$Lsud#(*3+4Y?hfm7@s$)rd4-h9A$Q8&My<-lk9nq4Xx_h<S
zpR5`Ba`q_1mD5&E2k^y|I`$ro7+fy4R@BsY%ez0gnz$23clWi`313!<hF`)e7ETi_
zfG(?UdBpaLg-o!u5E$sT-U(rc$4^SZD+S3bdZjd;KTB4QKdx`MvP6JOjUe>0ubf-$
zQZ{RTe+l5nX~HbRu&uC=4|FJqDF>wa!pQUpDC-*GYa(~()=Uk~exbsr*8OMPCZE1`
z90d?13fhs02IU_7&sSbNO3jg-0dmRr#%H8<uHSDY&0VZL#(aT%m29?Lgs+<F+tDbJ
zzqN-yIN+OmTwJBWb9w<lD_Y1fP3o*H9>DLp-0y-E>YHC(;OAGT`d@tY8j#(j08OY&
zF%TP<pbl(r&adss&qTv<RE3W_XNv4nqXLCa1la9K#d)v0S*@+F&O|{)I5DongKW0q
zudbw<sCWROAq!2?%D6R-47=ZU{nmDx{l#!=`#FtMSghWac6R3LQi(o`l9Ko(d+|Fv
zcWO|D|B9mG-<L&)dw!IyPW0#^&?oqbL!Abbq<{-r$Q1;B4P9a5ZK5<IKfiYatSbAn
zE`_VG2KvoWG(d_K2bz)zfX`*ztGAKQ9#j)fFCoE3d9O_vzyb^cwA-1Uah=x~>0SOG
zYwsP8b^rH`N8~^_l$nuLLLowAiwYSbWY3b3StNTVGLnR>C@Ykal@(={NJa=D*_D-f
zzupJ>c3$`6cVFjy-M@dX^ZcGyaU7q|`~4cv*Yo*&If9R(fkE?|txE7B?YtT=TY)r#
zz0u{BMRvmlcnsFxxy_j=n5(!wBnLRP?KlY669%qcyb$-17d!93zl$JR`eY?D|F8ej
zTcse7akbQ$dL;QSY#YvjNy1wXc9p9mP4|^Zoif1L@ZKvbR;=Nl1ibk_Pr!zQz!Qju
z!>s#W&fo^JN(9{M^05LV(Hk=f8B)AbPGh0#yg}2znNo!55>wMLLL=BmQ~?ZwnDa<1
z$avDey6jQAVu=C!jFhANit1<1F1vwqx9ewDXC&m)-u_A-y*=@JZGaX6S*COe5Q!&b
zCH|42c7bOE6c(O|7f2k{&V<toDy%F{Pgb&W3PIV6@Jz&g!F^-NKi_!i|K%Hlf_EPS
z1JAejNiP$@(6k#A;HS@=(QBU4zL=xK3Z&K;fcSc6U)YHN0RpXf0v%+2jHsUa^ewP5
zi{dkqJa(Gx@ZmSQLP0O=%M|*4tr`g!uXvWao$9G!jUc21B=6=+ZuU*>-y71`m5eYY
z)R?DTyI<02{+A|_{}G75W}dbfMEG4rdydiW@1-!&OmO?Kg8~w&ysZHKr;n-lANMiF
zM|>c~LR)qRl@+KL1*R`^*fqFN?J!X>=m$<&6EF>&EfyJ&QUo>bz_6OS(TJ*LK;=<$
zb#khJ?<B1H+<Q<`f0ca_K3~u5^-?C@uHgSnSO)`7rj<V!<{wNaO0x?P=l$l-gdtF3
z(Kn><9aXCOF9&^t8VcrDd7LujS@)rDZt}+4-6t}RGMElT5U2xd8Vy6R7>_>%wA24M
zTkmC0uB8IN<uFK>y50b@<p{DDe*4uLh(3#VoMLvPRRPeMr)~s`r{gS#a}7?3g5B4}
z^XtzGRyxZ8LJKapE{X14Fi$F9gn`QBIl;i(0*g&TdVxuAjsfDC($neXs(W7Ei9E2`
zU~z?S4{as>5C9OB1<eot8QoI&(ES!&JkcOibi%X_qb^gFsSC~3;{Z`{D0DML6U?yu
zMLTn1RS+@%59h>zl(E3v07cvzvBbUf0u2DFrGcmZrx(!|k@q@K`8u@FpdabpAbmTJ
zz7yWJZ{G-5vi)q6qj1ywqRTK46R;vQ{6^xNMO^|x={?3iY|wlprNBrhUFHShr!O^5
zdl29s#!loB;rXSp|MOb=B%^Dc*dGt`mtHtq=CE^Vh5?tokSQ%}TAwxNlgIey?tIht
zLjMonv>O&=iGvy^DmL~CQwOrhS-+W3Zq3z<2Wo&Q;FIYI7rg}`fEs8>Qh5ZM0?U7u
zxr>zGVXPc3m4D3%eYly;!TB`(e&NA*RuBKZvuY~bF-?h!Z?53sNs;*fob)-^bLGE?
zZmU@U04_<wfE-Hm4$R-FUG%TUz|NhkNze9;MI4a0d;I^8VSnI0Ixw9H1;U%i6W}GE
zgB;~zkt>V_Rj^+y$L<DDv@BqBcLBEP7?e^9%Pt&n=={|2bC%ERp7A#|^AHuz$A&zd
zfUOADIt%V0xIl&J?DC^hjeGRhZH(0)gsA8^Kdycb09bG_I1BQ%`hIu!dr1PX&s3*F
zpyIX4Idx(GNHRGyaCR$T2gRrB6}qLcgSd#FNJ#1^#^2Aoe3P6=OdRH4_arC(#{rs7
zaS-AY^812ZntzuE<`b8&axfIkvB<sn@t_F}OzA-H^d!4I?c9FQtUb(a!7!w_hAnrh
zJ{|GfWF!&0_9-+{U}C}XLh;#8$>9NbEQ7bDk5n?dIu3mNYbaNpgO4O@AD`(h83qp-
zOmd--+W-{Olfpb0|9dDjUq<ujEDM(@P!RUt7{&!A7*rxvAucc-`xBUq{^x;-GatCY
znMPHWGYY|g88mur{5lx^2C5K^rzvD8*NVepZssD&48@e|VarJeRhr&5lM{u!w~*=)
zJk!)I(xlYr`_SN%2@$@+eoRr<?#I6cnE$$nw~iq$uN(M<JGT~J`5;SkR419oY`2d+
z0xQ;)pB{oBB#DT}lDA4$?<Ra~0Er6C7?y!y5_dW8{<)lBp7B2oLsa9iF7g(*=$Adu
zWFev+AGC8=m~@8m(*w|_Bd$aFVf<pL=v>Opf+7D~6>H$wQ95x%`_mEn&zTie%?}#2
zsGBJH)1;5QVy4>{EAv+)G-!sv&b^9GPiWECp?V_JAhW=4KPXLq+|+;dUBX9LwFOG@
z2Vi54tCUat0YMG_t{6RO(cT{9<NM29%Gp@b2TDwx%GH@ujj&DfH-KE*9Z9t{-T>{*
z^EARlbCvuyPvZ8Y@yK47j`Qs$3#<RfM2p3Skk_d4D8L){fGw{>hWup!+P=Y1myCZ9
z#xS`|v_R=Y6O66O8RTXzr@DDk9K%jpcTzrFo&I6$q&UF%-J%buTe5BfdpwTr-Gz6J
zP<C{o`tz?Rl7Ee&EKwTG5qw*+eh9S-x^NgX3=z*5F?PSCK(9Y_F+eo>nBFIJYWb`F
zyi%^eUTO6#eqa8#8>SjJs?WPWNtE2PwWZlx8^3~YK7?nA`&T9Z)RTn$2+iBeiTlPx
zb>9EzS72<3gafb?FYDKM_w&cf@xte|Evue`ITD)m77oL{Q8`4+<ffB7HW4=k3#sly
zV{HAoW+O?hL?L0Z^WY2n^)EgQH3F?kJTQ%FNaB&p50tLLq@81-GNfoHsIK`tW+_9D
zaPK_b>xFW3S#Ntly?hhA_>$0WEsv+r@>ll(@<ll|SQp6l>rWS)@HZoMjklW2-IG(;
zgQ&0Y@(S0;_Qjby>%A1fd_2JA&;V`V!o1T@__fk)&}{^mw8_x@6Y>wJhCZ*z=;rD~
zcAhy>;sUb?(T5jRpOs%ij)1`bDV(|3rHQ~uST#_0$+9Zdm)X#3-R0RbT5*BKI-1FT
zmq7S}Ot2AXZhNYG(`?bY?wk^N<BRi8eOr$N{IrZ%61)FEyyvl>B+VXg>8pe-8#g0*
ze~q2|>$10S!vWB}eAi(W*FU|e-0P_J3i(;UdV{Z(X1IzPnE9hVM9lUt9<UL#@a%wG
zPJgpi=gN19O#Oazl0$6p$*YK11(mI#kq;H2t5GZ)#i>os>Q&i965P9}0bvnsV2xw4
z(O;hf%U@Xy<V6k;+<~RT{U<eW74(sCB#7eR^<EVqvW|cy;sa=GPEARH4>+vgX@WND
z*(zFkj@H+_u{>6&OTRS-yTF4~B6|;<IE#Exo{>>=L|(A>KQD-<D>NxIjBwEPfU0H@
zJr$xlu5^`<(dHvx0bv+vnOI&s_x<D|tMLiEf=Xt(*2o|KVW6^rcpH@7ht5k`ZpFdh
zVh}i>hPXyAVwV7umyc+PDa}~2<TVQ!7R^6V!}6a|V=z#nYJZS293EN$jmFzsKYvgY
z4v&iN5ob1dN}NZiaN%m%05SPi*8RW#gYv4(m5fh;c5CHTu4KW&D^{&<2vq_S{`~kd
zbYjgwX?_%V6j2Nqf~Hw^gh0n$<aWV=R^e*{Iku{g#I(>HrW9Nc(W;%(r~Qi(DKECT
zhPiV7&aGIgm-#*Opg@N6s1|=gsN4F%C<&$BS_NNEw!OQ2-0>}MNf;0Rb&D~V;kneh
zEsEZ_48_e!jSmlgv%`~V18l<$mdDLMax2lwP-BJm1lUH*%rZg*q1)sj{>5MQqvL=z
zo_kG>j)D(A?O@<5elL7ix0SxaZ-?kQw}L2t<gQuL^Cxuo&~eX_lJOJoMKlMDAf!?g
zI?J^L-oFAH#+3!<3kE70jzfoTuLu>x?#WKS$BfzA8noGx*Yf9JjfgpNW(RYypaZZj
zIRb6ky_B2#46U{YjI$snle#l$qU%LneLdP4dMrNB{lsUXVd!rxKK2px<geapJJw9o
z$o3dyD&~!J^z+Z5HNVAbT->$WC_&KnuTx`e2~Vd*+&#4aAS(T|pMmxS;OMs=^_q}r
zN?>Eix%x;pfL38b2J`1kO{VuU7`k&?+%n^|%R#heKz<-SnuVK4`rn6Z7SQRuI$~aG
z^`0~I#Fi<C_Yi@r(UlUe<7?jV&t)!CZxnsbBEs40h}u!}zuuN~H77pIha=(iH1&Q0
zlan-%#3B+nvc$OdY9S`~DhzaJu#sT6tXtvwe5$`(`=yaBazzJxg9Fi+rcXJKN1(RD
z5jZB+rCEX?14D~_i(bGGK@kwY5wu;Yn8aR9hFrn0#W^fCHu~smOU{Nd4_aT9OO>ZD
z(tk)vFmI9$mTTkm!&_FaR?Z`mJY{v@FQd5XI(S;Ku0QF@i~x+DBekwRpp~=<SaBkM
zkoWD5$g1u8wGn3}^%8dOCiz$~qnT!PA>^dXy_uvF)umQ^8Eej4b}HUR@PNi?KSf-N
zUJ789s!w2nGh<ALPS6!Px;_7}D-NW@Wuhd>elce1m^rY4(0$qeYz^e#my&By;Lf6?
zq-=kA&BzS^wL{$8O*fa(b~i+FL`s=UPcEXBr&p8Kceo@|chX^3*o87@E8P<N6xgs-
zU*?7Q=>?ZnCyObyTO7k&(e&>jJ>Td3yo8T_q_p=~jAuWOIxc2_=J43LbZ=clid$~)
z3V}8OFBT|YV57wWff2WW8FqYyO6pgI%>u_5@p`6jP9lsAkf=N5OT3v(EdVHvLa7eI
zMVeL#gR+mWW~<zOJz2~idBA5B19Nk$Uk|u6o_r*4o^}AdrXv(k_d^p&dAE?N1`v-f
zWj1zp`QO?0r;YfU*5SuS0mlMyf50|CvejVu@p+*P>4!Qi>Q#1miKiyICTyiT)Zoy{
zaOJbiy@WjLr5Ee+42H@-fe#w*vK-2QcfPuRm4D&OvNSI^3S9Uc<N*XhIpsLq`^RwM
z*>#g<e&0LZXy0KdUHX07u)T_OuCi^mj`cI%EUDb}tD?uz+g1a3r?yS)<lcYi*7maU
zgT@;vGF^^N2zCJv^wLX>OX?!c8B)EHaCVF0NCFU16&P=5`QK0A8#e=nV{^v>R4PCF
zoZf$L`gl-Ba^a%}^3^jaEg)Ms^ES)Y<?)?JT0wpG0i*|pDW*ZyI<<2Xw}n$fgtqA+
z56#nCZ>z|g9^RJ6yTUCZqc<htN3!3DS_z|lD^~KD<5@<T7rS_Y?%e6S_vomlhd9u*
zDi(XKQ<*zgBapnb%OfU#H??<iF4^UlSXQttjDlIvWM)?U8U!t#iA&GnCovn`$c$gn
z@&bKRTd%rEz1Po1&P)70S64yd77e63y`^kL;I|pjf;Hk(a*;MYqF+_iLId)5Xd1zP
zEqy(%gk&h+Nb_tB^u27L&}n~@$SSvIm7Dq6HQ0)BF;|}px!l0;W%|HLAYh&{?4Qcp
z*ciFZnHT|oUkiN8CLGGMQseG{(_kZz{w+Z7CFWiv1x&79H}Tb}f?hX3$g)lIVnAnZ
zJ%F6<{QU`@FI*U+hXKgP#97!dIk)_vXZ^#I(KLLM)RhZKR-jTj1I-2t%)F?Sq|<w!
z%5uXx3T`ALT2x)%gxc-UYv1ZG5E<5NO+FRDIlaAdZ8s^tCC!eEtmkhT*|LGO2g~@B
zw&>cxaVCHNT_V!~btfG6;c5`f6(~|iZb#5Sb^b&?+zgj!#BmhM#L$OgTYqLNj00*q
z@!(?Td7&62sQ3xWMjc^|TWi2(N(VE^Pht~K7^s^f+EeW!)I>JBc_+dHJ6rh|t14H}
z&R5t{SLji^VB6K%DhfLAh<%3?yJ;94`^US2yX+1(daEurS4!m)^yiI9IW{I2TO8t!
z!UiR)?%6$e$NPg3#hPRLi7hK3?-~NwRC*YqYZUVK8*TXc?R{Rqd0y?VmI{RIw}Z~B
zpR4ZeTQrZS2~3uy1q!%ozvpEJ#0Ijvd4o1&Fh@%3iR>~Im?;5lC#{X{XXg9En=9{q
zw(jVPEDmZIczHIjodI1j&aiyYx4f))mExe}QJ|eryG&0urrIrfeGL;qE7XVIoh^Eb
ze_9he3{9DYl_4_NM6G8uuR-tQm8||718!%)MQH+y9m>I@zI*c4vOUD1@PhtJ52bnw
zr(3=pj;Zuq9wJh&p(dhWE~F~Gr74n~U9Ci|Y<f^OGx7a|6FCh*tvbB(U+=|a$j6f8
z-boimyohGhG<Ig8g+eTI`gJ8_2G+ypan+K4G|G%0o%f8Imss%~nR4EKZ114}9d~Na
zw~kl}FcBF|;D-o`Rm0mjB9+|dyv4=eF3vNl)I^F9Hhr_OEb`pCY<77%+)*SEd=^)>
zp`L=BP~g0wgm}>qyvwdmJi#ou<*g$ddSqV$U}v4@10?OYe4jY|L5GQ)T-GNtk1|hR
zHOjMNW?gLZdA3Z%ha0+8rgq<#OncmGn3ula$PbXl5Yzo$VDXO?-FCatCqY_&Qqq$U
z6`I9jm>_z^K*g$*c_IC%{1bO)nw(@$pVi0Vl2PCl;_f1eh7DlxtOUwJedcdDRZv7X
zHAJ#o05gm?{7!Iin_bS0>k@HLB0_fl<Vy@bGWx|sSI=l_Mgzcinv_dzy#yx6J#5@1
z!*Bs-vc#seyRsOBZR{~Tg_?j(J(-{C>9BTpsc=jZwvEHV_+tr@1K_(0^TI<_Yd`OF
ztml$+{xUEKI;FSv;YoUubga(Gg&IaEr0xNPtQ7hAEf^6T^GDmq%E|~t$4_=c@dj0q
z)DYx&FLKe{zKpKq%V*(TACzgwxsXdY5Zk{tY&dVFrV<>#=v%Aubn8Ygqe5A7bnoXk
zu10jl8t-o-nADSYF#PAbiZapPw{~#159GsUAZp|s^MKgc%cpUPG$AJ={tBY2zRHvj
zkO5r#%enJ!Qz6_0XL*6W=%k8^#>Z<_>z2TGRIiFF0u%&gY;Xh8dYffbHw}=4tK;sw
zwt6-CA`?QxM%w$+-+%P@x!P!5GxA}@4xD4>ZFGS}0RDdSjmEn?sre@+eT-D&wsisg
z5z^oYfvlF9!!r;k6+(5%dc9FN!(My&%QZl0(?r`L(ynq$TssPd`cb$rH*QuGY<zUl
zhnZ-{N^t=93AwP=>*^wXQfR+AsLnnQYd*_Pt_E1|XuTN)lNbk*?f_u@REID9QPWc(
zbdqanV&vgfpa4z2dXx~@iTkK+l5@)qj)7u01|)OuI_;1>MFlL6Qd!F8$1Kbue$?d@
zx~5YC0_{UOUWJ)r#L>4WctB4AoT4Ycv!)9nai?iTd~H1n-Q&Z{vsGpwSz2>ypvR6>
zJ5`^72)&Gnvc#30vMSyTi?Vi#e%EE8=Q(jxY>V6jWhkHo#Oz^x`v*N(Rd=KMAI}Rj
zcjotoTKak1|JwdPb4As#pJD{jIE6oNOAB2-_cTGx&I^pz;+YFcop7=`%BJn=D^<0O
z*4c7ab`Pr=ghD$r^wakH`VT`<#}$RPo}{cq)iNogrb3aogiH~5v5M8fc|=d}w>=Lq
zX)@CN39xRET;0cE;HYJ#p#x2hvDeJRdLw{5*#-<*EIv>qG{F<l-?IqluqyI6_!3$?
zZRLe_3sG@hI87Se4Q8g-ry&2P!?>^PI_FMN-SXlC1n5jC{x!h|A2dGMMMW0PH0ov$
z81vKP7Sta-B~e%NVjs3qPr>Idkt;H(LOY4jGK);ZS7p~Fa^7V5;sMxq2)wXAY8jO}
zrqf8K@Kn4^k)_g#O<hM$o5%=&W+T~`Tx(bA&3ESkO4002T~Ez@r1j-pEm@}y2`%r`
z93!l;Ek)y)rkG>dLN{RR0NRp^GVhMZ&}rb!LYX^V-tq2zm+d7nV*lIed<zN45bG$2
zXMwO6a#J#^tpQTml|}F(p%Hq^bqtO}+UrU+1gWhFg3I0SZxMnl4L;wz!%_K8$<d0-
zrTz&hNcTnWS@4G-{iiyYj)Uqv_OK**F`R`ICQLl;B3seGweC<pyZi5Jjosw~)R#tp
z0<OJkd<-6<<RU5E6msjbnPsJQFDk$U9C->hKRFW)_B2ru5ysC}XT3J{20T}7C$DEb
zqnkePV=7&*eG=U6Zx@L8sjS;P!fzifjgmP?v|E}5%1L(#%tyBlimjGX0Wt+4%shVG
zJc69dDwc=u7#_Urzq^l+Cj;vP{97?HksY9t{Ao%pd31U~y@3}j8w?=oedkwZvH?oO
zpFl*Kh^ufM4WWs%A8VmEaD#cm2;6-tA<4^kNDf`;EzG>NxxPeRuC@Fc=1q)ES`Z>g
z)1b&BUq$|~gn9sWh)tD4C`M{80l+GcdCO=m!;L7aCi>iz`%XT9s3l3tNK?(fKtAl8
zfm60U(Psor-h+~%m&R%m!cND#{_Agn1fLuBrE^tnG~1UHmJ=NX9!jyq$7{A(fx9iB
z1gl|2J9?>&P0GPL+W@8KOtlnW+vm{NC?OLHu##xI|Je?l7qV+^mH@ZQc!U1GzxbZ$
z;?Hv#Z)>&F6d{IZntn~DLT??Pyc`!MDroE%-m1j}R;(9sHam#_lYd)DlcDoGb+I`;
z=K9yKU*!ppm0Y^ZEWr)lKm{`wMG`ZNtQRcRs|dsZl%8bGw0eCrpZH>^bJ+5?fcSRn
zCd1#qYM&)HQ$kdR4<Uo)Z=aOGswDAlfeV<dYN)x>po}#PM{CZw2HK`w{P4tC(v1TE
z_-ckXcnI_gDreH{ZX(}k3pfwtt6=8!6<d+ao&M1XY&%gG!DV=@_xe2S*=_>ikj)(q
zZf?cOl?fj8t_ZGU6AEkKes??0X?67TwW|9-%Il`J9wG&HTIYRQZq2i(vg4lTPe)2+
z<oze=zL-Pf7zquVfs4y}cCMc-txG@Cmf2EqJ^3Vv-0vSZdo$jo6n%uXYICE!yXqzE
zV$G+KHr8_+sBnu5uze^pxp75f;oy<|o!Y=hEzBn=veN~QW|11jXEn-Z$>zr-Jy*cH
zh2!wyOCdSo1fn`zI>d6gto;4I%gULoC@b$hPPTWO8lH!`Yiip25>#-1gVZ!}>|1*t
z<INzAMItyQe6|13b`vNK0>w852K8ThU#%Q^2Z>tm<+L_3LY=UrX}$nH!qZ*Hj(tuP
z&tgzfe86k_EXz(-6dE^|Qy(veLc2WNM+WSWsr1h>oOT`g!G-RrSK8jz?81VhiMk;L
zzQ@4)PVYs4`UxN+&EGr%#oT@9!`+bIl|6Ni8hAGJ<eT`R#SQf)%%`p^htRm(`Z7>t
zt&K=vub=0xK=l*><&_c`f-&87xM;tz=BC~ir=4fmIqms9{ed}%Xc1q9TT{lMy!VU$
z7jU@Nt-Mt_lMk4a7K*Pt8Vmra@1stOhyhfeCrZl86KxX9rKcA70Mi<ShXr8(^#+$&
z0wDFm?q!+qONv-ElYOASXzPTkBmpWf4ehbkK&jrnX^rqh(?$DB9OQ2FLOpmRzg=tj
z%NuCN<t1jPsy-iTsCh_ZF~kP>`1fT{l>U0pdRaPR_j|m9h=RGJzQc&3=DL(isD;>}
zBhloEug^9Bi4Zmk)6@&x=uVQHAjeI;?6DAmna&>h2MNKCB*RG17Ph;)IB%|flR^rs
zPqwSnF8f1Q4r#f34Y25WVQ~jMNLpWf|6m4m>(lFsS(iSB$-q$V$M{LXivqPm!;u|b
z=}XbC9EU=_^4;FkJKy;<<XrY9597GqcM3c?4BgN|H5R*lxO`YJ92^{Hp{wbV@B+db
z+U1!SE?A5A&X*byc9Ap*C`fK6>YxQ_;Hz3o&CV8&zOz0I-42MlB?Z8%Wp#r0r8d}4
zs!IM&d%XcDUXvV!#E9BVRx;37ekN_Moq4H8C=9i~v^GeQ_F4fpkgIG^z>Fl(8=cd0
zj%H9iU4V5C{pu5LbO#US@hDlz85Cq&-}sE`Y(#PX%!YGKhM8`eKxY?BjalGFciwj|
z8wwZRZS}%s@*>LPufJz~D_mJ+QH??U5wZ}>R*dkw>lenqAJfG`wD9HR!uyP7FkLF#
zBzF~%^ko!RPrdsAah5)NEL}*NIlvV5LzDzw=oMMwtrVpQ)s-vrvW9p3;##KCwz<Oc
zMF_E#3u!{H;I%F=+u%ARS3Wd?c_p<Q>5cvrSTNO7t5V7IrH3%~_>m)kNysBUs@)}Q
zrLzrEknU~$^#j>OzW44!lxza)w;~D1do&A5c`0RS=9%w$zIYA0jgJVSGVAx`r3-Dh
zzql6ruN4^9B!~Oqv0#t-7*{b9OjdlS-TIyE7!FQO>mkAkg)^$i%OCmE152s7LvF9u
zd)<8BvP8WX=C|K|WrS%Q(p_tR(Zc!^Y-skQ9tg}FTx`<{k0X}o<g=H~S`(k1WfSfj
zC+WAz(J#G_cyofw;F!=|xH@MMroGmwZFC=p{&a@q*?ruE3MLKASIUY%1AGozjt`)N
z`~Z8R(S6Tt0-;D6G`eLkK>3dRom?!gJ_f}4C#2a^WGC!1Y_tc_t4#IfhFZX_U$-76
z5WskIvEZ^jYEY(`=sqKAKy4%;<sA0H$NL;Xtx!YFc*kSIu8)1LFW`;c-scj^yFEKu
z5qwj>47fL*+;Q|4!Py47oo*?@HFB7wd2&A4s)h+}p>aSGfBRVysFjuRf(&q#kR2K-
zrF_QwBlg4P&)Q4LP>=TVy|{!0&i^66SPE^##8n;&$cx>Wu{riWPAD5e7AORPs}=PE
zPsRjY5!F#0H4yj@&0srbD8__LYoGJ9Uz_kOb!1;=B{$gAy>Ot5+ql=wd=RrN%MzPM
zR7oebjcprW2Oq15r_L&W`&cCq^0+dL4CaJEGu!}uDC4_+U4v_@8ew9jylOS%NY*Tu
z#6_Lab{E=8F-=h*M^YWT^zqULh=>}XzR3A}71@rb${uYmLzcIDD427s6$pu)r@Ga(
zW7-29*I~t|vlimC!#(S~55iJd603#we3UshD?q44d)WaAc}?}2;n+gwV_e+aW<eG|
z<=`&)+Mz1d32$8{LQYX^iu4TRyHHW^B%*UNq`SYlgDJ;bR~17IYvg8y+q4Q;G+@fZ
zje;NA0CYcnuU-vTIn<&{W^!r)Pr_k<c29KtJ9rO15xChFYd~}3Uo!x2@Arg2P~(}{
zbEsLsq?Ta%4SjR*`qzVAYpm&t_xCN_DUOj}-<Po>ucjIU)~`yAAYz5Ru$@`pjamTP
zFfpBnbh?69$&F$*lBOO^zMLVrzgW3?dZRP7@I+P&U&hyDCxTEGY);p1_JmVrTerLV
z2>vs@moWg&lryOS3S|yS47rjP=c@B8v+2kXt^<j>afqtZ#Q>Ur=h4b>RiS0H3lO|9
zoNdDyMBWVr-ZCKMWSEE(gqfV<>Y{DGp->)o1z2RV$gR%#fl8?n)IwI7Q4?>WerW=$
z><b<h*}Jt$hJm2Ybp0)iPJjoBjCfGn3A4HfXqbzJm^iP3?$O72pZBJn1F<C|zBH^v
zV1XTHH{2>0PhpTd&Z#~^;Dd|_0w`eij`Z=*r0o&3m(z4t46F|1;S0l2q+)fdyF{ik
zoz)l4a2@rby#iLd36dUSb4>2t>>|@{7<r+sQ0L20y;{psvznqeK)bSIkm?H;l5ejc
zIe_ec#l-41d?2I8a3;FeioF5Xw`KszkU<a{x<T_%G;^U#X2LQSibK6-{bxIYXL1Hs
zE;qnhcGG_HqWi}B>@yqFeDurTf#kP3n;&LelI1?IQg~CE*}eORw!PnR=9SYd-S{;i
z3t1HHklOAy_N)0gBe@1hw?$v|7!ZmeH)cqpY+w+tz)-3E^@n_ZEponxp-q`5J3IR*
zJnJZtybX^^z7-9o`no#UMdh3O9$E0A-qGM8Cxd(lRV4h($*e4fUs}n@7F@9T8aSdY
zFES-~kzWnW-g%T-6`%|n{g&jV(wf4>{CH#00}SF-CUY(oDQb2PLkvxrc}Zp27$py4
z0B-Q0y}qiyWD)OkLfxsvrbnQe%Q}49TSy)*>K++#uJ8GLjJP+L0|uQfZYG+rYj(>{
z@>=eCuYe4>=V-bE?$zukU0w;4xaLV+S7#H*Sf=PK0dax#ttCB`2{{ch^Y#cZ5i(hr
z2xle}<t(WN&szsvC{zI8mo`wL*mTPUZo5&$-tvLL`8mJ@4uRJQgEDd#1;TX{uYuSD
zKQz0#?i4mkr={LsSAcfvT+E4W{`xV`_aa!+Ezb03cdCn=%RIQN0oDmMo3&K3w{-=>
zT&C%iAbG=d8X_y+DN)^1p?MJ`Ft&?a=4+T`WtKJ=U@@E|+oG9ZeenU(q;hdu0n*WF
zgD)8j8$5uHqB$d6uE09_RW_X=m;oa9O(F{FTgC_H2JdWWo|)+V3FTaw<n(J==TjQp
zu;3<J|J6+`Z8K;Gk@nsCVHvYlS9S3n5|%<T@xuOZtv9#L#p31qK55Nzz1%hz(_I1p
zCmQI74G)Ywxo*t#Iq0fRyUx`R72V{PW>2rzw%yZ?jC7GrO~2K7$PojsU&#z~6&beL
zczY^n<6{fJ?B)2k0QKMhUq%<oM}Y}(I8nSjC4MMSLdd+9aq95hZRln%N?m@+{uF^V
ztFi}K+PCayhakcRqzrk}VYklv)(fLksooe1glP4|RU=po!FG4b@M;0&5{F^wMPz)d
zXJeo(5(UHVnzCn}Od9L$R!8`JDir!EW_|arP63s&ltHsS!+ELk6yMvk3y$xSh-8=;
zAF*Rk`$$(Z6Z|;weTRm(mu>KlaR8WQqP}(}6b4&J#zsEm1>w&nEtdz{y2q1F6CX2w
z_FCtCZxZim08>qXd*!rUz((xCV+AtnhXwIbFtz2HxOT6-Y{&1+dJGI99%vi7$)j@p
zK>3OugK7-Kvu02t^MP<a26iQsiX$c^(p)trh!!Ad*UDBf;_!rE5Gu0*b0xqB|1$nH
zjfIWPZ$Z|HjpA!$#Off#Jd$>S%HnKbWz`-zrzj5Dvl$V(nt_@ZZigsv%lYjac5jt?
zDEm;hB)97FX65D2t9?#05=#TG-uJBbE!})Ik+YCfl(dku)Z71w`O{?zKztz+Sr?>V
zlx|=Z68mT&kDsz(8Td6HBnaw1XrfjLu#Wag$0c0vcz4Y_XB`=Q_KSV^3f3{0j+%^h
z7am5M<qoCsYr{3=5+mU$E{8hC;cpdY99ZvN1T0u70v*~n@AY5sh9;GoZ~Zs>MDH$U
zy2S8MMgkV}w|C*={^%~BHUbp!!A)3z2j-Z(suBy)y5#po0Swc38@-RbFZEm?wTFi7
zTQR+XwrnXQA_Ew36UU<EuV^m!Aus8TQ^+Tw;g{JI4Q(PKO%F%9Lr#t;HXQbwdB(b#
zb@624)BDe5SmX#&7nWG$Y8<dTpTXfwHL*zmV5Z6dw^*8K0C5m4UTS?p$ho2+1?E`g
z<6~~t2Y^F4?qYqj9m+lWpErLYBkCNT;~iRHk__So$>Ys&^w4w9Pj<bxY(C2xy~`BM
z2D<Z2=p7ymnQ6M3LEK60S^>Tm>$>=So1B1MUrB^ZN+|sT{rSUUPl)#)?ymI$hMO(^
zz2YV;p+BRnbWL<RG0Vtc>-RGaHSxb>HD@U2dL@MY8T;a`wI$d`2DuFr<j^DZ4kykx
zdH?nvEmc{P@RZefuqeTK7Wu8mD;-Ll&l~!K;AFSNq%4uiFbVFFi_2A8Q|y)49Vq>e
z_kj0)ACl#xL6+sMm4Fu5ckuCOQ1keni0s@lMZj3DT1N?9Rxv9GU->{L32ctUaqUyP
zxqxqz>LCnhi*d;3FIV*0Jr{mS(*IaIeEWHz-^ja%=M$@{p6FjlSL=Jnx!X_L5_9_N
ztc>OGSDl>$Fnq^*e=P~DoQ?*5ac?zpHmY1-)Qdk)sxtr7GtqVUHWkD4DcHXW&bl8b
zf}v<z=g`)72zcfJ>!3$q(*B|cEMOw3@b^Gfa}O$!o?eSI+>&c}FIJ}tp*j?>{&Y0{
z*rkAx5pyIrLOfr`m3b?@moO`N@Wper)$&eSgkjPL^Pjb(d<{TZH)9ASxp!2*q|H+4
znRkTTRp*T65V#C;V>$5PTisZ>moTo=%nY*RGp$A0iIk4^)O_!7H=_D2L3jSQuZ^T6
z{nr>4Pu!|w^gSHOwsqE60@0noR=zeU;B}@uGJ|{g)%yu57OUI#t6}iBufx{Om^<Nv
ztes)go_tkCCfI!5a-18K0ZQvx!5L^K8<}qOv6i{9w<7{L0QTBxtsedFrmgE#y_V$^
z-t`>4{QeD%MrB`*#g(o*>6Z2Q&&m7;SO-n;?KUOm=GutKcJ3Y>7IP@44XCicfq%w!
z_X%VIG@Ae18?;{w(`f`FS953vi@$bohcbw$XzJI*!K{%F-Z|xZ*olCBSJX^`pmhB(
z#ulv;P}%&#?U$uGWo-)wQ@J6Zo(7;}@*;gtii8W9zd9S{eRZ+(?SL*R9c+1O#)6w7
zS3(cpLxob+=K3|gHE!|Dudo>*xs^FL*_W9@e9rq>8T4StSU5fmpWp=vo<Cy-)AFkw
zFMtZs4veU)lVNa^1isa?$L|z=TZ;Z4-$*#aqB90v4~{?A0|VEC(v`N%!eAU<Q(qR{
zC+QYq*J&S<79T9cvA^OJY!w3-y~uW4x6C;ce6yueyQ^SP%!j-X=9Bg;or}=HL;wz#
z)=dyOAtRaR{Of?j#tR97MZagqIW<{S$>|5yO-u9jN8+<gnNQ$Jv+N=0TXV<y4YeKG
z`8CX(SVbKfIZrUdA8q7JQWk`&vP%i(@fQjifXfCQFn@rArWZY>qYmtyg3W5;0h=)Q
zQv#xfMz<WjW^6uSAC6Pq$MqHf6X$~f`MgJ#nDmODsMFZlcwt)`+YUXn)<pm2_jE4u
z4oNv;uCy!TsVR_&0#7O=lNw~(0h1qiNBj8GS4H-r`v3$(<Xv+i^eY1CK%w|FpNIiB
zs{L^f>d4PoSjYUK<cN9<id4k^K2jNpOnv9aI$`3JL$;Fu9oE=@ZudB^OJiBvq5P5n
z^zeOR_s#Y0ejQtz?6eatwl0kb0n(3s2eKypY_g2sNYCXYp=K)!SmYsMcCQ%<D_P;r
zRvsYEMg!F(ASYb=v0hj3tj8;*!(N3FW8Ea*R3j+Aoo3{||Bm*T4rUj=A?M(yFMuS)
zFYmnZU@{-IT9N~|bI)nE3<8yxlp)r$UPq*?3YhD~!mvtLM0iyjnkLIG*^CVVwh^?E
zFjK$u#Tp<4#gbP)<!}Mcjy$VBPo<C`h!)&od$T|uaayDd;NMQmFCZKG@C9C<?IPH}
zTmnA6-P{W;0_u&8)oUL%G`)u*dPTxe_2IFPPeT~i7P^d{oEL-=u5=A7>V@?RNt;aI
zHx%35nC`Wxvle{v7<>Eu6#;nDq2uo7z&N-tY&a`BcOSJIG~W+e?T!Jq+XS*_uzBCd
zO2otUQ#WjvEcrl!gFj=1DBtN(X1u(9GdYmclwR)?lqJs{t{24}%jMWz42!qK-DFZu
zy)xnR+E_~!k$Zd}>l935wc@jNLHrL_m+_yVmAa%$yW;REm>w5e+Svw+hytzZl_Yjf
zH|{+Ts^XVDR}kqu0h^cr6pjBbgZK)LtwLR~o(DwijzT&WXpS_8x%BDvTW}A40%~&(
zK<uzEin8A_Uq7Cg(mht{5VkxsY#=cDb^7=j-5g^Txw;hD#){+N0lQcRiSX^#B<5b1
z+@{=#aNQ`v8Hc3-n1JX!y5}t`K?f-S8!ZbM8=S}6WMGg+eWce@Zs7qXE0If<hxOi?
zOn^;(f-i6&H3Kp~M+QtVJ*B>tftYS!g}h94T?i=@1;A#L$_4e-5XR`bin+Mo(`)Js
zgW~W$j0X>NLduGU;cibH^b&dsp>!q%Rx$wsh+PLFS4DN&rnzRmss<!dLn3czEkDPN
z7rDZP+Z%aD(qrW;xQo~>)DSK_?%KR*#2TS#FB`y!-$<5(_vM4}zHudaZlb+(*duXO
znU9*0O<?&<nG*RyNt@!AmT)iIS&}+&_mckK-^&h$#rZuGKoGM4v6qZQsrKoX8i``p
zz^yl(ChzTHa>;*9<9%71l#_hgR;ee7mnXp=D(-j*7e4<4O;6Tj25zzIQ&%veun_Dc
zRt0m{0}-t1_N#(#wA=`A>Bw?b)GsDMPkNki9-8T6s8d=$WRc+A(G`5{Xdmk}539|%
zD{me*FGn6C`4&xkFyoL#$hXFl{dfJPY1jwnzxW*u3?JJ0wGk@#DgM;H2Kp|YS^{7W
zFGny44RIidgJ-L*1=A9|?EDEO-syTO*>tsa3(W=@v&@Un;?_Xy%vILDq^yaNHtW)H
z9BE*M?1n7T&@LFVl%PSsc{2wfMMT1klR%r^LZT9ICP&ctQZ0ZP93}M`<op18h|j_L
zierEk2(Q4w2)CIAwj3X9Ov-Lugx{|O9d)a<t#$fci&}qTw3boMzvj|YHqzer&}Z|a
zZNLbTQz&{P;5@Eot%2OwD7~sPOes(vJYoGw84OEk`0tPyTxl^VbB+WaRs8kVLYtlo
z$g!_uEbzVeuPgk-^XX1{DYAqg+?i{>f7#r4$!dv&FpsaoISL$q1AJ++i#5}`E&Lhr
zIhI@huPd7GWbW50egvWwX?B&rEups2(eU4T-*xPGp8|deX6KN8z06BbjJxv*oG1i*
zV`nG&tJufG)RFqY<fT!G@0B_EWq?sq8L;?PP`ah!Q=s(^Fj5b(Q$Sj?WU;B)V=w#I
zRz@48fu|G+3%5?Hjfm_fNT^Uq5ci6&SZbZwdwOV!H9cf3>cP9)C!>#sGz1lx|Dc&A
zNPyHjn?rYSC-^)>sg*TBn_BK9M92(R>5{bJT{X3YpNdmfnlmE&RytefS3K3Oa;YYY
zr?QdNq!QG;&6x8cVM&WYiyfglRepAxH6+;-Gt`c-VJIrOvQXy--a<D>$~%#E7h(Di
z(;L7E3l3q*C^5CaIi3<^4h^I5FTgWH&1~N|5mH-g%7nr_M?TZ?v(zB_W-od8;Tt%q
zT<tr9(JUJh9Mtpbw=|tr3gG{8bMjbrV+<b{dnSmvD8f=E(}Ex#mJ#gTy^A^0yI6^?
zy%RtooHo@le7iv#&@Z2$XC>19G0cbyf47vq$=oFqE0Wv#yYruZdkbu_uvj^)^-j`t
z<$%!bX?fqpWUb{8J1(C4{6Yok+%5Bwom7bWxm`Z4toGAIT}=S|w+tbtL)aWU>PyNS
z0X#2xyNVwQqTtV`y9w2K<LCn;yO`G&nvAmYZ1|u|@_f^!yW7IEL$&(0_33_PA-kE2
z536jht?ag64tAs+iKIPPq7y>fG@5_8PljcK_~efzmJP1svO5_+fepPn04`Ef$yRc0
zZ$=rKy-SqP0|Eo%Vm@4Cg$5bI{hHoGrQZ_orKP3iuRpmM2^w2^P(DMrm&g<noIo)#
zm>+6Ome%5w0~uFpL82&6NH3cX`BZoQz$!F##9B<xr$NPy7TSF7Jt*effVIu1?*JOc
zL|0tjoYg{}o5(|*RWXNQe`FPhO!%@zTPt$E00)8{?e~BhJbs$ekzYNc(HGY$r2YF=
z;Tf5GFPSO1Og=bi;NIJC#IDKO<%gQ+PCW=E!2P#(`%x|QD>BmWPl!eFcQ~LLLPHu*
z#*6dPC$yywf*qzDEI-jFXg~BopAE2v`yj_BP&V;`!L`-?TP0(0)~8q5EE1$G^C<7#
z1FKch_|cAoOBUjopu1~c=|7u-XK0lo@kN<_1~XvHkKsRDcaJE#a{SWH`w&oKNywM|
zQt2`J?$m5F<UWOu^=9D`mNq2-y}M^kzg`f5m!%o5lzkz981-#HA3K|!`K%HW;ZOK0
zXF*VDv-pvOJM}4k5V)@?m({W7R{T^%Rv)9#e<`=VIvxmB7T@cx9%PRJsP!GgZ$zAr
zIH{S=sw$p68v!ekKhim6eC?Z~?+bsIt*4yhkxAF^YH5H6@@-+<>#^qhJe)WRw*6}|
z#~zCZtRvVMD(e61il@BPM73PDg%xvhr(Z~*gd$aHe-m7qv8YHKf>G^;Z|H99Dk<Q&
zRmg6~I%#oA1hOG7h+@|7U0TnKmO!<dduy@hGEy4omD>}p4_g7ncoeKI6!!OjOmSNe
z;o>qQ5O4}}9ZLv)9sE+HG4=kc;HOf*Gg@5nstV5KIwWIC_H8`5ED=Q*^RZc(Pi*y9
za9t-H8hc;qtF<Jc|G5VZ7H;<>Kq=%s29S5sVSL~SRaQ_Ag>Wu72<MgzlOQ)p-E=lQ
zM5on#cV!Gxoq%!mF1gqtC`ME}!`}3Ru_^mvf~eqO*(GQul;GXUdo75GYQpIHf&Ptt
zm%wCGDAr_!h^;id5K0XY2HHI&SN!|{k7kB3Mzq+vO9k@n;moje@{FBJP(!i7K&Ghd
z`RKwWBY85-eA)G<G3XU0#iJLf_Vx~6<rV%F=reM(EGxkSZBP8u#G8t6&uzM0Jnvhv
zWLDRMH|B16Pu7TZTwU{B;a?*opBWYlPeH;;{*$mMaS7|;ze!m8eIWjyg+2`Fvsi<W
z?~^VxU@$P_H}^P+IwM{Lc<a8hACfgI2)1Im$BR$@N<mrd-ch+PeMgi&tv$@$=)0o-
zO4idYWwJg>N|jus@9Tbi-9TxDO=fd?>Ah6j#VXzz;<Upf*zvQ3`;htsk%WOPPS+I<
zRGdf{>D?w_OA|rn+yE!BL;8u74IsH~7ji=kre96q`5)>EVo%i%LKneWIn?XK47{vU
zpV9>o5dmp87H7Zk!9@NZsOFRaElz_x{E$XM^CH+;@!y)%(Tr8E->qr<1HwB0n9Qou
zwF?{2|3{%pwi4)s<`{8qhV?UFI@MVl<~c-e1x|c9cKB1JX4??R9V38OLARs{v<<5k
zD;Q*r#98N3QkUf)wygq?^TV%lpl79ShDESMXyH(;hyS`Jy?Fvqtq0Jy`#i;LU9RB$
z`0(1y+1``+{hP$ap6pA?U4czNE1kfK;C<VZD3H2@cnS$c%=1J8FiS!8`#h&ls6Kn5
zqGC)<M_2RGrfBTDfrM*@*%xHoHZ~4KiHNX~zw4+syT~39Sw*B9M)?)jpTftf7R8Um
zh@R|XNW`T}79OlL9Lj>w2tSc;0U3SlgGKH(7|{}uuH2h&TD2-p2H~Rj!TrXs6T><A
zj?7=N?e4Inbi(+mZg(tLA_NoFaBSSb{F}qc-~%OycJIq;d@C+ztuKK}2XN?yt05<=
zfDwu~hCP*l>AazpJ-_>*Gv(P?q2b1k%jVUSlcV3lg4#R78Nw#@b<XbT*OPPn)-n_5
zeoL068U7fQ)<nU0IDip8@Zb@^UKs0d3*wT%$up?<NZj)WeK9P58wko#r|GQA{*3zd
zcec9qx#M`kQ@>q*`pSln1xidGI3AQtoV@)C=S1Q9NhV1vE$5IP<8RmkUiz>o!<FDQ
z{})`6!3W5%%^t9b0gn9ho+7pEi^p_<^axatv<#)@<WwvixF~3{>CXea&g-@D`zaN-
z=8<P|nG`ZC5?#TZDv6PvdFy9B*@*6|)o?zqeUXoC_r}Yz5#<J&{=47F@Ew#{B)lkw
z>2}NQ<W?3%$l<~Ju9TsgC?WtCZW{k}xVc#<GG(=dkS6@?%z;2!-uWLRgaPl$_%WER
z{~An+QU@Ka8a=^Ior=pL#_YJZ5b7vi?lLSd*%nUEkSnI;4%8D1xcipz2m`cz*>4%s
zcv^*}-MY?flcP~kV}pIO$Mb~aH{PNgqwI|N>06N;Bhqy>hw4gNQwi80k<0(!x7kVL
zjEB7={9a2tY?bG}aAH;Q|1}!$*7S8^pm(fX+iA7!7GODg*Q9KcihgjuE^VN<FY#6Z
zs*0dy4~l5KKX5<p4s&4uxR>Xxn3L6L7((qbAYEv*+FA^8DN4xt6RHxF_0k?9eF+3M
zOJRhLTfiLB@}cqw0b8kp_)YP1iHQ{|!ZT~99N+R5**NPuxRuEXH72fm=}+Lj#5R->
zu>E$AaQV*fxbuw@&^PWl{bTDJ!}0yah;Hk@0@rWmKp3J!?{v@3JEcJHbSN!=_@Rpl
z9e?Z=0NjPcbG!H-)2#vIGXmW}G+LiH2J9oI`hi!SS}>E1KB6v*3}^u#y$4f1cOL1~
zZ7c;nC4P+fT=i~bL+`dc6oib|!Ia_%^qP@SXFl}LEuRrXV!Sk$N2bhkUs}NS%myF(
zK;Q9pj^Qh@68cdLCar#B@PkIXd;6uHvTv>g1AVrRa{Z(uFcFKk>A^bKKR7W^!0l7G
z!JGiQX0uaG`gTf}64X}ECJ*m}mnTib2~SHzb=SW?6OM2ueBn&^|2Y$(=uDi~;U0Ry
zt4v6)h5J2&dg$1X>qV$z#JRZ64Y++{La03OL{3NDyZ5*)yQ)l$G6?7dy3Y3o$wc}F
z^9FCz>?BA`9VK3WUpdT~k>>$dQ}SdFkn}G?Y&X5@JXr06AfAU7tpI7t&s1%C#vhB4
z1<j@fShFZ-yn5FI?wC}3hT>6jR%Waby+Wq%Zq=u0Y`!0TDW#%wH~4(<@6JP`&xN&G
ze(?yaxA71R$4g)JqHrF;h}JEw35so<Bk4lOU|8GXd321RVL{NR=)aE*IS?C8qH{#D
zbB@9Rg`WRlo=(X<S<7aHO+A47vTJA2)md-gKXBX&s>U}D?uc9)U<{iXP7Q|(f!uaR
z05J%^^8@0^&gB}U#Rp1wG_tydA$JFr1rm<vU+=L%wkG6xKon3Yg{^B(jzzT~%pLhr
z^*F<9yd5?q6eu4Sz#qz`bQ<Mjeelvl{wEFB!H4X$di?|Wh2q^=FYZVT!Ut2sAKG9h
z7>5VIQp2|swN$V=P}7F2{=U)Y5L5x49x5eoiE#f<a~XV;CE6)rc2a@8?@my0J)De{
zXlblh>pi5!5ZZ5BN5Tx>ru!EPk23*e?g#ckFSK87J0A|BZ0{kUPU+uRHmAgOf#7%$
zMuj>KedRf71U=1DR~#XT2Tcna99DofQ+@oHWI3W@BMLZF+aH!0SswfT>6)0?lZT0Y
z!h|D(1(4ss0b!dft1sMvIyefBspq>I;6fc<>T?QrYuQa9&I0^P{$_JFcI?D~<2P#B
z0X2SeHc>X6-D%h~`xJl<;@-i2cl;g5pl_C4mpEy}Q(X1=;wiU3uOxyJA7ptE(C;?`
z1Aqe#JZL*OfKxA0w<$0jZnoWJ|KA0pGHnP(I_MCQ?;Iiubckv=<muQZYrPsD7_p6E
zhrY|=5_Tmk0%9uhXal6R>L4hQ!$@W7$UMy#gYm8eq)LF=Ep*mLVEcT-gTk8OgXgIL
zL1B)ORru-XAuvLZ2Bt}GY7%VK>$^GmOjz$KVgJ<QrCIz;K-l4Q(H;58!0Iz-T~`s%
z++5XL$yQ1l9o**U-YWJ*$8e@H)7PD85Rl<~O3a44akjX2{>C$k+{b&j3OXFZu=TIO
zLQrX#O6*QqJ=4H?f;Q)eQnmF`H;mrCmqLl5%6T_`E6lpUfpKDT_$W5l9w@<d!jxP+
z{-)UWJ3cAv?UKSyCo$yjlG0GM5fJOgsiNSEe#syUIxx=mS1NBT07%ya3sgL$$oa^m
z1CE2m0XuNWe_hb2!~gTChC^_MeqgRtd_4V76At7D!y~Y`Z5GdkVap4clyj*ids`J)
z&Raio`#JJFE)6hygw=HvyD!;s8I%?39=<1xHz{OVnY#HN>(FqYZ*lV5YMgfT;`yd@
ziCvfA9N(qP-;ltJQn3PCI$vfB0r&&Hcx$8L<Nm|^w>}YmkBHKj0aZgK<??(2!1Yh}
zpXuJZL1IvBmEJvhmC4oibs{<A@t}<JQcT-FOyE4=wofQ;zQLQU-Ha!xV-$B%+=oFI
z^bE`~eNx!{LUwZk0DJ$`ObP6%8-B$a4NI*LE646KnmtcOx$}{*oCG)|MIa+smIWS1
z@-%Ye02Lw$qv;ypkgr+mTL%ENLT(+n#0Mum_AOc~0^+}`F_%HG!aXJ*k{~^J;P@t?
z)~J!Kg3>$j0CuVF^cafbERL5VN54kh=!0T#2t%;J?4J!sfRTP`CprBIa<FJk)YXLM
z!W$UPDjwDuj^9SVI(|WDNm5}n*_w+_qEW2aaE8lZJ(;_hYaL_ib&@v9t(e840Y$bY
zNzzktO9NL=F|5<c_r5q<LKQ+;<f?u-vw5z+m;<-`E2{^d@Av#+!f6jn(wa&fzNW$S
za*;&wuTS;xFjg`4Xu=V{q5?#IbDw>6Id1+Y(slt>DA;iwjQ%pUvSW~`Cyq8Y)^dCj
ziBRJ0UqexQdH;2OhA3hs>4&sObqyv8nYoBdNds+ZpM|{rtQk4BpeUbx3XNKh5|msv
z_mTNeCft4=w{X?C3jM!^0$-JxsJ=ek!g}S!k!`q1{487;BF;m8tj~E1{p4C|-udlA
z&#4NH-{h(sodsrKve^JA&sdt$tpHZF)YHWDQl1sh@%WYZxBT@a=sD2p&EFt$X?a{=
z(R>)qz+k%7TS#~rb*xa=MML;`s0FOy7(`10$}W%B`oe?p^-mz1YRGsGN^(R;vjGP^
zL)_)b1-n9<{<6-a_s+#j550JB_R(jfD8|yC>HXdHG~ebqa77J6-jN|wz7m*HTwt%3
zc>Hl(ZL7lj>ysJcikZt8gqo77m!8>*D)!fM4HlqR%XW1j$6b2K!_|Qj2UQK##tjGZ
zpQb<%Jz+w%jj=GZSC_S!?T@-`uS5I9KPVQRy%*$o{h`Q!xCgXeP)%LrEuRvA)uWO|
z*WZghmF^4u2s!-g9-G5^q7rEp1wN1kUBZk%NYZK>f+jm(r*ZoQpxbU9D1Agd5Ssi(
z=-Lmlva;q(W@<bI#e;#(0*v*dVQNGtiv<WpXwt6i`(eZ|Z+&5@iD5f_{6pb7+E`F|
z%MFsaxQ|vmMdN-m&!{P4w_1@9*;dh3itimMaVN4~uZWYBe8<N(A+~>i<mdrz^lNu+
z75dj|B7*i`eI5YDzwwYixVL$Um2G5jmCJnBcaE)~ljel~s-2OQD6fUU0rwZZyD&gL
zu@;Q_;aX8?qIxDUR_@KW2khHV%x=H2!MO0y`0vXGba6ta<SC`bVb}11{2vKB5xXl_
zGT8w!EH1R};;kflKR$dH@q^wl4LXxwfkDv)2rSOb@=lx>sMz|jPJ^>0`%nD9`}2u6
z!mB~hnFj1WVMZV!qP5OFHZ_FQSC(Pjm0xSIW3GOwIYjge<=1WF*8!st{^Wx9I7AGV
z{z<?XVXqJMBt?_A$i7kM_8$IYsb=pY*S*q=zTHXloHqUNJ;D$$?|is&R5WDNf1M-T
zzYgxkRW(tF5%TY<njolZ1W;9TW~Zv5gsMhDfRfz5ri@?wbRA=DLCRrC;#*MhJA$y`
zDP13Q4o8u)(B$ndJU(1h;wuPIuz@FY&wZwp^MJvJGjupwFCcGc0q?0Tto~0reZlJk
z9QkIT&y6<m<b`5tVPcJlJ!qgGJFVqy0<DMR2YK3u#e*$~dJINKLMtS>9WM;uDkn)A
z08Kn0b;XZD5^K<5)n2p!+gcJ+$@ZM`8MWSwt`QVI+x`&fRU9MB2kDoqghrYU$WP&2
zfb453>+P;^yhi2e|MtYN2tdI9p3eaP*b{Q<J$xZVW)k;m@+Zf#2DRzuZ(oJ(d|Ih*
z-v0+enX(5jrPJ9*6h%MerA3w{flO5_^IeX%zYT5dh=1hkg-JN+AiJ{_3i!X5zy|vW
zsMYv|DSlt3`W}w1WQUaRvREooguqtZ<pa?aXEcD$g_{|;bQws7CPr)@Vc=JtB0#>L
zFZ&$tj2GQA6w!h|n0gl(AU-g>nF(i2%5rrww;b)7w803(zDaHYngp3lAaLG$uZF-L
zUO~&22<T>dEGekSw?w%Dr<uXVnF|;r7|kCHzSSaIq_wk-j=@bw;v5f|-;-JT6yc@C
z!&q<l9XMNO!*(r&UU)3Ww3WR6Qz5uM!VU3R|8<MDY6UIYU1-q=F6>P1Vqw%kr|Rw3
zU%;!(aM+aGN-$Oa7B&`EGZkL#Y!Z%z&)fhbFB@T!4Yqva8hk6zj2;^Yo)uzi+Z4O<
zAht}!kMpyx^S9g=I<-68>81_XlD#GuW~IRe&;a$~I|W|#L7Z{H%`a!<_nm-jpiIN7
zPv<ZPl&(UMq4s?;?0N;$&F4J;*lqBG$tJRO--~w}Dsn}DUef_Hs5x&tI^vW?05E1j
z&yh5axq8*rk@OqF=YIA?yL>1dX}oU$gtNepx%x-q`rBy%yxD^{L!XWpeDxoZ4mA%a
z;7C@8N}&D($07RfP^_iTff8f~0mV9(wgW`60*ZCqf7ZS$hm?JHNUWTB!j(}ke6%-y
z{B=E8Tp&h>J|ws(FwohgI7%9B3?`QCRbE~UK8KWI&id^k`^N)JFg*3-rC0A$4mD_}
zfKd<%Vu#}vhiLwM^ay!O#<*;LLqEuJ(g6*iq(r`zX!xud%NhYt`B7jPBIh9Q+h30C
z)0ta0q+cJP5oIu=47ycX&`fH;7juA`^t8jgGa(-4G`g&jYN=%Z+o`&bJ5}`m?Wywa
z;4|1ndp5yd<#%kbm)z3VK#BCQSU<1#qHH*z(`T6tO&G`SxtNSN-Hw$T$+<#|hj9(U
ziQi#ygr)S#!Y72TgXm3JPuvnlkNnK^PcJL0<S})0h}ddy+JClv*$5lE(x8sa6iNF}
z`BTU019Jx@$P(tzyN!**RkOn|8rs(oJ~>vTjC{CJ*?!ozw=fdYXSWSiALraqyM%z}
z%*CZsijRZv-JBYN?P!%c9e;ORIVS{74=H}1U<jeh%J$9pPi6$efssL~bC=iQA4gRy
z+>*8n&UL55=KdH_|2p68sLrBN<s23vH>JW|W!J4?8Z25jQ;~9c>-9-{!$sRgytxTb
z$QMSou?R`SLa10wRg=L5=mq3J!3Y?aAQZpFu>$uX3`Ng~@jmUk8@Z@d-he@xaznRK
z>hFLUnG>~bQ8`ftucP!qwG<k{ehx-9+69)aFM4XJk!t|0fpa8itUkxJKqk>VafgZt
zf0lRPkcQ*{^P_hczNg!75YT^EH_g-(tE}IzAtOyqAg&6hA~A^#|Gz#3A5C})5hMw8
z7h#+xx9rTcA0KyP<djwXT5g0y$c@nt<Kd8t;d|&{u#aHNf7QKbSd2bkO8)%RjqxpF
z5KA(S8SDnfVjZ1`m4^<eKXKR(*AeT2X)g@Ao4P-y8VQAQJQ8<n#~whEc(f;;TQdXO
zX!8B)<Kt-5@!|2Au)I779w!anKK--|v%V;_>J_$wvZF}}a0nEk-qbk7A?E_E?(vmI
zaHT}ce!=b`qW1m3EBACbO~R|t?;ivE@s~aA8fbA8%rXxQV|y=yyT<8ywu%@;%(T9h
z$`syV{r1Y+B&`+e52*n^gsj#-%l`*W(7i34u`mG|?Gex;-Nnc~5_2&JsY`M3%|~f`
zvw>w73xv`ltndqHr=v5IrWy*9TEvcA<bPVRTwQ;m^7KGU*{~}VUh;7C94UEo|74wQ
zw4&MtDvgr6C*_#n*|XX(j+kvdbU#(q!~Tr7#2oy|a0OBBe$2u?B}*~+23Em&fg3qr
zfL}LG>D16CTK!=w`uem0!7i5m=73m?i^-d@-N%DM9ljo!-KRF>1B2N{s1(BTf{{7u
z*t<v87=D5?&*GOslZ@8Y(Rdu`sBm*~gZY1`<=gOs4^T4z(tmT1JAyok{?{B(V;W%e
ze+-KFD<mensV~L~ScFupOk}@Nprt0oa8P@f04ikBmMjA@d@O&^T4(qCou6pDh4ujf
z-wDZQ74+cmf%>lE#~X$hp6fTkcGh_>cX6_<5p2hUCAr3DzYK+P_wfUB3k|d=SD|n*
zv5tdfH-2Vt<kNs@KxsRy9!b28Z^Qo;ZK#8+%LVhZ6LnPhOrs5vAYGzo1NZ+30AiY;
zQwz-k88E#WSt03~q37>WpPb*H!KbeD*Sk}#FN7X{s|WS~Cv-6T*L!nb6dk5^YWpF9
zjLzz*wRN&hW0nvz(P!@LTL@JW0y!mn<pOMhBz8#bGRJ@7wH9XF#Q;qXVO~BKp`~dL
zwKbC6b<ptcw|lh;Hpz|9Qr0pB_T%B_J=$;Y`Yj<sE=S`wTj@wT(baR`O-s#~NY<gB
zwZaZ2svq#q*k)q=CpM|>M-X<0-+w!B!Yu7mR0#5qkW*;1*eG!YJ_r@osbQ_un`k>e
zDFA}hf)TpL{JM<)>o34@-}E@7En!e%#_HxPEW(^q5sc}c9cIsqS3~txhauuI)kzcX
z(cEhce0NKOj24(l90qlSV7BDOvf1k|Pzs&b%TR`IZ2`4F)TtZG=P7HRcY-qx*bcM9
zTZ@9r*bECHWlM!^QF{m{=q$H(u1#;P1)3$=G$1oMG$S91>zM+Fj~H5`TOY7``M{J!
zCIqm!=J>OwWn^yvtjb3422m~-dTdGy>axesSG>D=_hcU1erOMhi==?MapmZNg0T)o
zGy_Fm&UUXBJ{8858O{PcmP^t7mFeWG>+{R@U)d`3>`LpgU0_geqaPY1^XF;%6k;+k
z#xR7jZJqDct|#eG<euq*lSb7#f4~YDbE#d<9N&nQ6JDN|T)zJD*WHfq>h%nobu#sg
zXQ7Z>g0`z+)#<eVpL%G!lEWrK!X4_Vd$ET#NT>bjtPTvqP<3qTKy$K<Q<j`ux{eT+
zr0WjhJvAy*Xkfkm^jw8aw}vmuFyg6O83qO(NORFwz}wQN5BWvZlX02}zwEac0KAZA
z^X03~W*J#hsMKsyjDLC-k4QWqIzG^M!p03K-hoGIfQyL3P$6Li*;+qqYdsE^B|pyQ
z_h2{h_>O=CBMSCOb@!%UY|G1^(AL&)A1s97t`+FN!2+C1NZVvmb6(br^a1j`L;3(+
zk1m57jUf6Pg$XggZOy~C%<jqv{)EDZBsU~O969X*v-z1VSbFZI+#7#9tQJQ6@@BVP
z6m2(76vLB!BUX2`(`(KvUia#pBy-Nj`43g);7yYkenw{e&tqh9#OGFgYU=0nzZiS(
zc&^*_e>_6=NT^7Z5f#~Fla;-atw<RmMP?Zp$tHV8HkrvL3E4Yalvyf7=I=P)T-WEm
zf8WRB`}zKHU-uu^eP6w=*ZF!K=W#rb=SY_#K7<^unh^f2f_=O?<0WPG3!5;&#C>1;
zu!z`7K*Cj`jcupsUV2w_*7{{gR3-;g>^q}#!j*2%MYz+pKWJ2wpCb*Uctr)5GhBv6
z$w$F~;=h|D&p`O>7-yEpM1V~zu<&K{#UZlInZ!)d1Rohr=)D6NQ8)ab%dqc463Ap^
zn&@sn=IG(U{PWC|VG5kNJE2Qs{sg9)77=NYrX&e?(8Q@85=?qPe0`X5Z+k$Q4W?Ya
z`;|Q#5Ibou6xyTA4*Ne6-ApN5G=qTZgtL2WrvsdjXS<S(p<y>0!H$LJ+yaFT`@@O4
zhN7?F9|L9nWtt%rT4tmsfK-(e4wjoBUTIerY#>4LaYII~`9KrK1CnD<4FD0o<i5?P
zuU|=l`xn(?X(z_W2*4&jLFtsz_9?3Yl8*12Yf#zVmlNouu9lj2obGWZOFu*{uw_<t
zj0q9lC>^rg(!8jVBiQoj+pfH)aFK`<EsVk(rYxb$JF_;z_2PEm=Qp+hYk(tB0OW{@
znR9h>{rkvU@I6;*w^dM<!pv3<P(#vgJGj(Pb1d+wAf+O_GUk*uczUoBo?pLH!3%-`
zq}o>>hid++3&|6#3x<XUmoArhDOD0Mu_nI^Td|4ix{**vs{EGHk2Fh*y^vU(^^01%
z#OA~IsRedNvG4`u;p9mho2dT%5r?n9H8gUm%h9X*;R{`R_FVSCwSNHWR-gX{s4=~u
z_mU=cxwq5ko6Ms!?TpKhK&Xu=w{{7z8T%O4+|IZpb+`Ncw<%(|>L%ZMh%wB#*OOM>
zIa+4LT1mkyh{zsjeud39K>_zCRM#-_6-bf*`3cCD^1!&ODUZx-?bB-rfbvZ1&(%={
z(F{S)Oj!uDNs1x#qU|mn^U@4Kh(pI;MGfzSIsW-{YJIqN+lSH20Ia|oppy2oTuZ54
zQ_NxiMxOA9M-?u2qyZ%@2p_enQie*w&S%+AekT4kQY9l&Bp?wyZG%tOf!YkE*qrZ_
zv*>Fu#baB(Aq{^^W+fY~ImI@oj(R^Qyl~;{xKp~qQ{aN#6IIfKzJjx6EgVEvR5NO~
zd?gUAj>3BnFMNw0oGwu25?%6OB8=`@*Q7qCougs@9d=^}E7@kx1<>kv({*oKXp5hm
zd|3S<>nU*f{@y`G@-z4E1s*%x6M70Vy#c>M=#@*{U?ydg9Ri8VOeJ#KVq5IL0=jEw
zAZJ=TAS*_!qj<yCgN^<a=dTt1;uQs_N7+f2Lixhzq$gVz&eJ|Lc^$2HL+fK7o+CX!
zF>>L>6QNK^vd%M(x8AvN=Z${QxMJ(|P~{g63{*Zarl)zus{9|g8Z+wRaQGu^DC}F9
zBuF>T9J1{nXLj%70{-})dyW5mZN}bYdZX7S^5546E!pKBH20rclA8MouMM4|h)lGH
zjkWr%BUY!ugT3f)lV#ryY|=9T4{I<l$up)OAlb!YJesfe>nK$Kw7J})xOc(Moj#V=
zBK$j1CVNZD7wkdK)WrvIHu!&W2TE_d!m2%}<0?Q#Zym&+?zsUVY?%k1JJyr6<CfgN
z*?GBOG><o|Ys)!DxOHd9ZsDKxP>#0QG<e#SeJ=e8<b^ia-G(;OUzx{a7{`yi0jA{`
zL}5_LREN|e19XlGX#VZc(lUiw#2+6`gto$WZuIzfER>n@-hLg@jkC7W2in*SU^%7I
zn*2k#IfBi_l9J}AW9aa5d7?(x8bZRpA!mm+&_so$b->RiQ5UJ<AY`ifp^r*}VsC0M
zZ1PMD`Ya8nDy_gHI;|Fz%cmE~trwU=AF=2rCBKu5+S}5(8uF;tIf(G@SDykJ#aa=e
zgvY^h(3?A;-uw*LnSZ;0LkQC21=2@S*svI!pA98&qFYov?9Dk<&r|$&b8felVFk>X
z<l;AHnwDT+b>%kct;|HigT~9l6z=V-zTDq};lG4&{e=WSQQGz}{sipfHTiutE>F0;
z#AJ>j%Nk%l^7<H(Ah%~2V3^GkZPHNsxb)yZ9FgK5uj6gPLGe3w1JIePD4nC@wa+1V
zAJtfZH~z?N(tX1xPqV>aBjb9j=h3;k`Pk4<23Q-PSdz{Jab=4PjiK3aey~<bD)6yy
zXa-c<F!a{F-QpXRV289r;E@Ta@jlHrw&x4l@5maKfj`HqgYgqE(xahzjV?xA(hZkN
zX&9WU!-Y>ijcPDGKBnN^x6`?;ea=f`K7?8VT48B&^>D^(=Q1<ZaFDb8)_$=3KE2Ox
z&wq4jSojr^DC-?GA#@tNoE73V<|jp69<^AFA)?5gXhZX=y*^hlc7_;e|KoGv52wBD
zjaHV+-~{(X-V=XR@<gx9l$J&@aa_646M2!Cf*D}oRoNpxS6n!2v>{V(Q0-5^A}XYz
z(EqUeaPj{r^nOt2d!`d6jL7ZfIk`r*6aQH=@euy^nknJQ!fD21Lz$PXkNv8?rFsk~
z*2CGEhkCv#e87l;P1c@K4*GPe9=%Q4$@Ae=6%Hb7L3>l~@EbXUf|Aji|MIi#?a6l!
z@4{AIHsBAIG}7Kd2jH3ZH&==TMPuobc4loh!rNfOvoxd)jgA$_NmjU?1u%xYk~K(C
z0-i`d9;gXH9wo_DE48DWw{tB7qZi-{B8)Gah${rO1N3RFmf+bN>~G^={#tmYS`!7`
z<+irK9k82b{3wW*&vwE9e3DG~yFi}P1DyikG8KXAR>hwDh~^1y^H<c?b_xNhccG8@
zgyp+EI&J^9;&H48;a-+nZ~?2rzg<~95zb+YUgTr{^`%Z5GzMrihC68Oss(^QX+x?F
z3Npu%-T&yu>lO|nL%Yu<$&*cmLzaa#f8u&LfQ*kUR?{mwON3HGIe^=c9JW`;+SV52
z*XNKx8?>*TA1d-#m83Osb!VrlhfEB{=f|kYHmr3z;Mx5<iJ`v%=ncnYpbKmjU(E`3
z79MtN$w>;ZO?~tQpB8g>7_d&QDGr5MzqWH%(=GLOntYSl2gn|^=(|(=m{$a<{8D3<
zN$_>nJfiW2z!+~~gI|lJo<yA0YJj*{Uo8a84}i)Km-Km1bPOQty2{A!`91aZO~!7u
zI<rdHm`n6*UIwY-1Au<MT02@zH`Cfi0*KSd%gn^og0ulX)Lxod!&*OYeIF^~m-L;Q
zDRpaQj&1>@!}7+$>jWUXDS+7$$tWZ(q0qH7<HsaSL6SnqT;1gQ`dfIGi1Jb1Qv_Nb
zIS|B|kML^3&pEN9_>%n&ofTEon=>cUQa;Y(rAhSedCzgE(d=x)rl18};cR#=enN^t
zFbc7DETl##NNz$JRWw3DJ}8S@P_`tC*P%ce<j8Na`#8E>0Bx}kVTM3TK`<<P;Qso{
z)41!A7jF4;Aob6Mr^uEzUteIfEcphI^W>3+@ZEgci$F)ur|6<bn?S3<Jc5|ON%-xp
zU|O010a3rfX!{drtvjqffZ_PbA?5O;7tCVnpGVwHpX}dniOZMcUVfi_v@l2g+Sb0U
zLu>5$cukI9(7rYKOXv+1Z0%EG_L%?i7C=4E#nAKH;cfU|l)RYl+HW|KSohZZq+Ym(
z9p_GpVs^_CZm`)pDZcutvb*NpfketNxE`n##6{RkL~Zv+%U=R3Q$U6l2-!UR6<~+b
zd;?hV1}MI`!~8P1(VsFtzYE<}o_VZ@bMd3^BLn6{9I(!6CKtvZwu0oxt)D!C7S`;<
zx2W{G`M2&2=-bxtRX7I}+jRZ~g`sklm~}5_IUdgWwWXY`m)cgA(C=2m67Ys*&R?s7
zce0OWLiwOYG`>w$kI#2@<Oy{By^u(?l$!kw1cS^C*r&kw3W?UzT?|q*WCXO_&I@MO
z%RsX@3(C_-P<i0#xaKUwT%6=Z>1_hQbh8QmqgXsoKscW7N&lwCq9W0tV4x!CEey+)
zBX7|^&v^RaDNBstL~$6h102Uo+Zo5F2!NzX6yHOTM}o+w%nfo<__wun4DT2e;u_1%
zmxb+#YZa1tN0gTtDK4#tH`_tCc1$260So)13X?)I+kRNn895%i(h@)3EJLj=$69%N
zGr6V)-C`$i32S_u{d}_E#;v!a6~Wy9|N1S#Fo-yu>%`A~oW(4dnmG@<uFSd9Qlw&u
zf{h1Y$`FCzZZ98b9#wURNsHTjMR9@GINaxPcs)z_0?n`e%GItjEtTElY?AE_P+hu2
zJi)|95wgy#gHe$~-ie<}-vBKQWp{KKm0Ron1=_skE}`oM^&`7p+qLf7Oi<4FpAI&{
z&hwfNU7mu#^t^e()Cu--G^qhlq8W5~O73G3T)GlH<T`3$J`%k_V&Ib`nR0LE32I`8
zKcB{-0&J57w7lH;h$w0D-~}@AhJ&^(?@>ZTH+bPFg3jBe5Z;82o?DIgi7`ON&fpYz
zzuXSj{g&bW69CNF%_tM!Lgi3|+$Yg0wqY2=q`lguWTC0n#XlqM104xaSj??$v~QiG
zI#1|NB{YZ?eAv`-v2??_yJtyP8D{%LlxQa_wrRXIrsnhK$?m$2{s66QHDjd>i3+<^
z){wYG2J{4qbx(GCEoI$taDwT?FRQ8=vBo>nzep*oQa(Mz#!(}z6|wwz@8fvzPuRg<
zL#wVd-?)GNxc|9L5bLAj$N*L}(5;V9(A0v!6vKmAToXEbNs4q%H}sIad-L!-Z&dbl
z+|3|yy-v;LH=ks@hKK3feyIW%W(OD;QX6x}xsGj|>XU9Js9x(;MERTX-q1Q3*i=m8
zf&Bn-vjV`P-TWWwDZjqmG(75Pn0vd-Nm*M=?yo`f^(p`Dge4!QV!?8rKb~4EUBa~J
zHJoxlFlT=0A(0_p7oK~&vM6NIBjbpfPx+p*XsZu&hz#~?6!&uLU=ad+?&oK;TKA-I
zPv^1qyIs3Vnwt#MNef7v^H^p+y@B56?PraPS>4T`<h_Xc_0c?Fqaati1)wqw^B3hH
z2KCJcKd;vTDp|wL@x{Z#m;kxVf^`wA!1$vGV?T+1*q$?|^FbYyHYCKZm)VsF?((-n
zdnN$~_RC05d+Jj64L~>Zcx3bj|KALg2CvtbL)AU~;YGe~Z*O03wmo1q23>amA!``7
z+!W%QZi8&4T=XN)T#M%`+sd?OB0U9*@0?|t8n6;7JUs5=!^O;_Dar8fksjU|p;&OT
zW=Hf+kV>{N5{51`Y98Lqs)_Vr$Gk%imqu{K=)U*|OIiK6K;lzz{>ndI0Irpv(?OkI
zM5~4)dREj$3n5St+!T^h#-lu3ItRlw=7#dvKrFe51T)@qgFdI2qibh?$`L}7|4MEO
zXtm+s!}oD;8q{EH@KwPSV={h%GRpw4W*DkD)rkLM@Y-htl-3SN?}0{UsM;dH6+||y
z@qYns=4!&VZhq*4y02kg)3SX?6Tslb%8aWZJy4~1VT0!apK{P2WYn28@5Q$;q#u1*
z-<&TaZm1kgf!>r6tosbBtx_xq=5^{r`cE<hJ6W_ro^w$v8@Tl%<|ay3t+UKEeogZB
zmcrR~<zz9@0cDM_wXct85xLF9?aAug+DMP>466~o(@T#4+@O<OUdow9i40#bYYU(?
zs3ePBhR5|1QD^Nzc#CAK&yI=ju4&GkEEBLEzOwvFkIP|EtH(;*$^`+iv3HY-;@Yc?
zOf9y-l{%&zgzcZeOHIayPCl0P=UkX(uhkd5zwVRI+l4=+`{@_5F^7z)>-JWbx0?yj
z!Drp%A)&W#m<)7dJoTP`4G;c(6BxMR6}Z;SHj)n&oBzSq3%-L=%{X}X8XjT-dQS-w
zAMBQr{&@v7n?a9SMEiWh)x0lhW6mERvpZqfg<<M-a;#1WT+lVKw+v5;4z8UL@&r~M
z$^vgcx@Ox5_*Hu`ALY&!u&XLgJ9o@NH0B-bDlj0WAy|T_YF#=9V!79swZ4^)+C(5`
zQeH1-JpI*2^2je7ur*lJU5C7-lnWqGJF(kn$)r#>1`z2@ea+o)ON3>P!b&?*ioB71
z%|!T$5?}~M@<pdE^y8&Y{gv+$%rgX5j~-~kt>Pa27%VWfit39;f(8_oPz+yMsNXl(
zRv?5NAQybyvqYwPbIhDy<sRRFi&$2`#l;&MkP6sSe1+Nr0=3%A*d9dE^B2ZuAGQM>
zUHptjYV|pp0RI3y%U-JiuuY0xPvv$|#z;6p{nN3CX;*E=?nQa?u-j!yxw|B04W*)y
zIxFTV+bYhVGa?rLg3EiW$(REbF1UQ2tTZoDffF63(@Ut))mM3%V2O;r%2*Gv{l2*f
zWTpZJxpx&#K*99Os&|1FpM;dXHz_II%MUPfo}>)+_fSt>&)JP&$Q~Xc_&9Y{5qk5G
zH^B45^!1?c)w^iZ3SaA{-{G*8{O}_+n-nQO-HF5dhvfhPZ@5OgPE7kJ?L~sJYtL6@
zTg<C8B?hzB51!x~WO+odWVZIV;P?$E)}pQ=0oO@dl|0%Jw9*1-+9=>?$QfZ4Tq5cd
z<E6*tr$_J>*9F+vTFH$z=aN@xxz=*CwjBCP!wF>oVi>i)0XgUKH>Fd)z;qdfs`PEE
z!-mU~<<`^njMu<tYJm(?>-$<bQZgra7j$&s<Gq!;0x1W?XRh{}6X~ecmRSu^yGz(8
zvj?kzM`+)p&7ZeSXTluB&YkgFm^F}%WfQitU))3P6|<m-AYmg!FpD!_G5O=oyl6$i
z<ueV2SVqDwCi2gYcvhVIiL4T1xJ<uTDiPxyd>!%iXf;X`3Q(T~X(^(5rz9H9z+2j$
zM-;72ajZtRjHPU#(0J1F+zPj+=v)(RQbNTIt#-!2>~)7b>D%}H1)E&x=g(r>9G^-G
zi+}X%V7*B0pWq?px(t3But)A<;Vw*bLr5eH0q4?|DqCCwl@+P(&5xAjB4#YUMDjH9
z5kuX-h@5uOUqI+S=MBlSf3pK-6GodzKtRBFRHw%m%uU;1X9I{XbMyL}^;541m*B(u
zyg3Ib2KQ}wMLH=E#~W4O1laD*nm(BHqHg&Y9lsel^$dba;WT}__uv+~wp|D2^?|^c
zYBV1zH0FdN#oKfV`{TySq%oM<o&IwYLv}OMhC1W)@Z0}RM+X${d%rC9+4DiV^VY*J
z#9a8l%Onel9=x}%P-7}T@{c6x(#XbF;4d}K^JISP#zFYB1%yrIYkVMBV@{k(H>SIf
zx<R^v{IVV>8OZXWEH1EPuW~`W?BB)!C31cRT2|@j(iBeC1s$#Y1(c}_ddXf(!q_32
zgdY_%r26AXRqbZTX99$#%K80`yagBoqVFa-H(L_MlkI8b@%NL+vR9b6XRSkvq6pya
z{U>Y8WuWl;Y;8d$wl#nL-S_fQ>yAec3v7BF&LsjxYHjKJNC;mU++sUh`2+O1)i3X7
zoy_Ab@yYr4?GhIBf}OH{0{9XV1a2T?ZT{5IeTaZ>`gqv<fk{TY#cth2VDl=M+53mG
zHF#R7H4?of{FUlHKr*OX5$332DTojcNeEE!%U=7v63;H5N&6|Y*NYuQ6ZFsvT%b1(
z<2tyNCke9bHVq$-GsC`(K=_1v6byU3>J-@GnPB0of!eWA@_xqwZ5;*3zcB+(r1Gm)
z(<uHAmK3_mzVkkWbjY+`B`MH=SXE|j-TBtfl3TeCse177RdAO7LCe&Lw?U5Ff8zV~
zs7uJTsz{WKWNr$Gfl)AUkLLW_DDx9yiU|W!M0Ne|il~3?qX<v9j&6ai3$l>CZGAOU
z=U^c6@J8?+K2LK2Z0F3e;V-_zH6sLpbV9wG#m@(}z`N#@@w=iC-a^hck-g0s{)ZLi
zgR;nIVGNkWHzH7Y+f|1}vl3hyq&l|%4njxX7qSMfd)|lmzV-d}C7y;f{BpcGLg+7>
zea}$EzhR-WsendRxevL{0SN{i%6g)GBNg=609Hd9q8`ply*{+!-U@%o=MzWf9<~H}
zlDGVdJmM(N6RqVE0}OG99t@vOFG=AXHUuxW@>gpw${J=$`TXr0U}$(H;@0JWdX0xi
znHcb!u&{QIkHIC22Av02PWv9POLcuDkhTKk*@|3aKtXJkCKHWpq;Dg48$W&N2hHM1
z>rV3F^0^ohK8r-Ut?g1CQl9Ay#TNm+1rOdSIZdg-#Lfvk%-QJJb1Il8>aT;1IKzx8
zDYQF%*J0?10PQ`%8rr><i=VgDq)6E(QoR0d?Qo!2gZKw^<#7F(Q&Ra1&1Ir-{Qg46
zJRgL$=1bEL!hQ`YdGk3m7#~x4B8=kxN6Tq&W8hBu_&N)pdtUJE{ZaE0@EaH<uX=Ne
z4fDDJEJ!bU47i{3P2^ElhrR01>k-3`Fra**Svk0hjIl95zVvb@F8}feU(q=?>x#Yu
zkWmmukknNr)30F;x2#)AI08u?yOdGRL*!-k`k+4l%%MdV8YPA$a8Ks38R%QZs~M~6
zLU?R{o^IdBB^j~ElMUcacM-;HrJi!2?*9h%$;fh;TPky;ofNP^XW{Qg0J;>rr!_wf
zb%zg*zi{A^QT4Dnk+u`)iSB?2OF8@QRy%O`nZehG-!<!Rcl2j)(Y|dJeS<MF8AjTG
zKTBF8Kx~fWwB5g+Z4=(xp<$$eP<kqkqJZ~+;N~?;){*-^#tz%r=b1QlMGSL3KWVVW
zH(w1!0~}k7JAo|Jp=ftd7yLv%DBye3ZdrrZXT=p>g~PwHqn{L4trASHYX9~XZ#DFX
zH~25nU#{B^*C1yKlFo(!-R(Nyv;Pk3o)wJUpiJ)}*$QK!VjsHu%wktuMB{0X4bAnI
zRKO+x^olXRz<&DyfMqe?dp~uGbcgqBNSCevX-YqDKROYX9!OosUnSNB+h>pOpR@BD
z-gyy@E<KdLPm{0~TPVh<d>Q4~M?U96OVc&oN5JFHH~)<}RlDGNsLhwyPm?PI%<d-0
z?{u@=hZ`lm!=L(>3`73<y9w72bEy`_xY9GrM2yG&b=><BfT~1^Gn-@PG1@;MoqQ<9
zKej-Hxw8{kfmZJPy6AUH<rV%}-NW*L*h>7?!zQOen|)#Qt*yYP!9bh9qFS{`J^iHz
zNH~aM3V9czAA&E)xh<;L{_rQ`cMzkF^8VA!pVKG492=1^UhbiB%Tw?<xn{)=D(>>Q
zjveGS11c@K0M0>R62k7b!!b?>LfLz4e;$1eJDFC#Ap=t@tIJsx&j?L$&eHt&^rm4|
z9ogf9BhA2_Yk%SUZUa%~9jjcjKyPD@Vm?6~U;``}Ea*5vk_oNB{flJuA~zvE*0uyD
zlkt;<9Vg#=AznAR4{N2P%lG-`k`y>NUqU_h_Kl3^&mC$@cCMMAy-Xv?U^=!Bdv7$W
zDt)eyb?qTDu4E5pQ4~;JqSUO5{aiTsda}+?YthqA!6m`OiOwYDs3uO?mLFwjGb<cb
zGcU#gMkjWzUiAD~W$IMSjV&=o(S7%sOFq@k@}+#wEoVy03ygkKzZSQyC0pe9R}!}G
z1JEc-Np*u$zdY~rPL#e4SJG2FV$T^wr$Z<%ti5TmN#J^-a1YMz`YlC3<?BM9Beq_7
z(-c4Htb2jWdlCVhO66Zw_Fu2R0|}bu@Bl)xJ^>0yQ_NvP%v_7Ivf3aUkosQZfo|?5
zw7Mpq)2V=6(8m!=JtsUIdD#1YxUmAp(RCsO^Ujf<#Q`8Pu0yd%m=nm+w0{y-i>6fy
z52}m^G`yWWK`-S7eTo>n-~}36SQqg`(Tk~4KGvLH#=sKoH>~FpY}Y2o0OP-3I#8I6
z6CS{TRC=DbTP7h)ZssBzDAcYj!Mw}~fB@g8Y3_#jds@#J`e|?<(QP$G^Obva0htv=
zJpcL{Q?GLejgJVS7x~?CKJAxoM1aaXjXN+3Zt>CnMgBdtYi7z@8=4=83Tpo7<$iWz
zyuMq9Dq9tGX@2dSj3QmltiNiKK5!%K(@4hiSw(&Mz_LW7+qMWa5%S%(yZ||z@LGmH
z(8#<9Ry2;$;0iKnAQ5#0@k}#IHh6s4R`fW6uX`kz{Ns#$*cpPW$dr_+84Kt;fxqnL
z>9HRuOYA2ZJ5BpM5S7I{(1IYgcnp%pS0t4LTZ8e=@1O6<^p$U|3QW3{xIvbOc})PW
z(W4$2^Vq<LID<B#=em@3LlOj?_u_hmR>EAb#=FT5y7O!OVcDB(PMHe29_E?`gY1+5
zYzKdj22vaxY7-?Q_NF@(P`7SsiPXT(E{u2GAine&-CFECVo@7t)l1^y0iEMMZ5MJZ
zLH9h$S6OrShK22O!uNiPFe*iWt@+8@KL+`c(d7<+2sa{P>p;wS2lNc)x;!DGe_{e5
zrP#cLdd+zmh<Eq-Vx8T6H<9%aA&0MG>a|5yr5F>D8Is4uj;JQ<#=@mFk>LLS#OOt3
ziuLIKOPNYyP4U777;xPA{ksH*lDaY}9~va5<ZA_1JSTX&1a?pxY{D^q4OiqSiaqPT
zrVdSnj&=EzS8cl{qU*x|(1}8N0B%cdrWl(?jzS2(MkEEdUoua<BLXCm?=<_3ypslZ
zBEhTkeox_N99H)W8MWHvlE&BcYFx^GWh8*+pS2?U7jql{H>bW(H@r-3<^m%v@C{J4
z>*M`ICbpTPlpx3@BTV(GTX6mB#XDng_5Xzj8|27t7bak96kp-9w=Y>&K)LDC1z4=i
z%#bsQ3hbxD;gjVbrt2$4CG3Rnv8#`vgn2SjVB>s=@~Fkn+wTZJ988N?-hkrobKTRr
zq6f@3%{T<h5hXPv+K3%?tGu~uTQ<OpWBjnyE(>eZ`6*b@2Za-^WT)H$dX{&*dMA_T
zOIk98{UOS)vkR+IVX=>5Q{P8KM3{Z|nlqbh{_yP=zEI4C772zQu;9AzBQpl;U>5Ns
zXipT@(Yhp7qmZnySfU{Qy8js+TKM2k>H7>p<_Muk7JjN&9nppo&*moD>p%}}ZBC`#
zz4Tp8K`2+^8Q%g=Gk`(5%H`F~uV#p&H7LJyFWG*zTkL0BwKFW-g#b|vz_a^g4`$3x
zU_MQY5OIEm4yf0fT#@(C3biN3fo2Q1h`4ZdAhwie1q0foSsMk6TQvCAfEF(y1R9-#
zE&d!}>W`8v@<hNOn8X>17gV%4g22;!*OB4ts_(;hAQ8I`{6JWX&ePlgzu#LWepXiZ
zVVA@UMo+GJEU_>Lje&d)K-pVxL;JRP3>0w94;jJQ>;cYOh1@)Byp`(JU?f(}=0_1F
zmyFz=M9Ad?d9%Zw$--Y{{)K$1M%IJ*7dHVH)y(>qJpu}=6FoGRKdM(%Ove_NyHlkh
zj5|zg*EhHSq(f)#$W5vrPH&e{|1&7K`(<dM?3Hn|asv6!^H{s~BLJvOee8aIKyv=_
ztR70)1$OMLcz)iOpr!m#xx|4Cf`OkUzI;JD;mS=Dl|sdJdL_#Rm|T2grT{z$G><lP
zj78b9*=M8-@$h$IoYe4C6eIf-?!p4(p&wJS&}A#JMb=~G93?_BEe*cj@Dj!W6!XQ!
z<rGG|i0_n85Bt6h*bXiP+c)`B0H$5|=<J-&(MAHH`G;k&J-q4@Q`^V2RHovIjZ56%
zn!5sbg>Q8;DRY0}$83+=aM$H=1Xc%(F7Y(0o!jtiW{(nw$0gD8M;PozG3zTJ78pv$
zOqi!1wD&BaqEI@K)CRas_5wr9g+uCM+LZ;HN1#}x)%pcXb|^u=B<dHU3qDt$s4DS7
zgHo5!S=t|KgqXkAiF9HjeK(!W3{TSj&Lc8?pJ>!YIS=HW`61Oj;BvX!H{jCTRF*aJ
zp)v1Fu)w?uYY6=S@@yUpB96~Xf9(&q>0FqstIG#8K`|X%^Pz!8yhRjbY%vs_w04JM
z{8?ONq}j;(I~;X8_A?(+UvN}jFre(KHT_767(|-8X@Q?N>rPQUcY>6_py*E;qjr%=
zfAN2uugEh_$=A#b#?a$Fx3Sazg4z*ZURoPv%D-dF0+O~~XmoozG)_@@;{RDN`wUl7
zB3=AhcOhVuo<Dlq*TWcpe)134eQq-wd^z|&1ka@yPxd>^0AuEiy_U=6>~myW%-w1W
zVAI!J20F^6_C>_222&-?8gl|_PA$nsn0d$7Bw1E*qIS13iDTbMhIcmIXqr~)RnX<x
zKlS(FHW?e9PpC8@(@6&Mw@AG9R`XLmrapnRPm4GOHqQ&eYhbh#3T@o$Z9z^ta+Q`;
z@P|S&sH|sMUaJjVB-(FMflJG{jag)mmUr9`fUEsP<NgT4*k<0b-0m7NeNtw^Nm%AO
zLyME&5p*{MUbwQ)7ckQi_+jA-CET6PppzmsiX*xG9PhBNIbHHY>MUB}P(>f~H8lVB
zHJAV2`Wmd=-_IQ$w!`+28POH~;zb#_jOdZ_)TSWpwJ$yX{-dMHu;~QGak~fBQ8S>M
zn`;9r_YM@^v^4w%1>--x(Y4qLpSI5RLYqR2<pZ~c!cTKzcLn*;GH`WUV@`SY>{<OZ
z#cnNdOw%}Z?vGdUS-S?kgZ~|Z_#qitY1q*-Ads7s?(9x(HL@WGdSQ`n#@M=INt{;c
zH0=BYYU?TQp%iwc`$DpRSSclV>8B)iz6X@&5-fa3{(eoAsJgiiWVTy*&P|n10c$Cu
zC)<@~NFtbjS_mzHtO!|H5vDcvkRuR*c*iuahuIl0$F>4(nf%YyC0H=<TsyM9#S*GP
zX+8{o<p%bx%xT8wNYW8nU6Di-`L~14v+(I>PR{Bw^-G%JLVl-2w%Qjcix=N#&Q7h=
zJzFDpTVXIe9|%St28dg3pmg5*0N_bJ-fdNe_C8RP&e2OE86~J`kdK0Y$S;leYEs9x
z5+Lbx0L;KUv!0#4-fmc!4?bdTp}R{wvQ(m_uMuA<eYi%9M>}j+f*sKu3~K?^_GsNS
zJlLglnfC#RN;L~!V{boeQpUdF!xR&S7snBooigNq<+39_6+h>PA7)@b4?X~$9|2O4
zs_(*CYXEBZr&-vAg#(B5-Pb66?sS!vyLe4M5Ye_qi1XC*0Eb($S$FnM4*JC2_4fl7
zFinQzk1@8b51QjK$#KIH@J7ArRVIf^G8~!;InWTDXSX3=RJ=H7itX&qxCC2|G|syg
z)~QWr(c}f!R5<}D3Z<EFTkY_mo%wK<in+UJQIMl%O#@8=Xtp+w4>U&!ygi&XId)Cp
zFc?JBvi5?1E2Q~lQlxsb#d@k2>FF6TSO_dj?|dFy2WyFJ`W#dJqT4>mwu`rL4%y)Z
zK7BN!rAIcS;Xt3!R(jhkLM4*rOWinXdTrL$@I3Rtoz-$M&Q_KPifa8u9FWIc00HS+
z!%3eP-@)BG3h48`)hhj_^iMW#NvCwW?7$qy42i&8%8-r4-}JpS-)I#_1DYO$$9kTt
zi~Ub^ozm5|*=gs8ccN`yKYsbX>M6<6bE~E|_p&@-wu0AZOquQIfhZ=K!U>e}vj8y#
z2rk0E>nBZBQK|>7g?z-!v)wt`7h{DV7hFtT<oN^`?Jy2e)uV;m_;NRem6Tg<fgN7C
z)jd%a8CES1l;?Ou;CJEET}9AlRGSz9--$$W6Smf)Bd`Zmg*+hZl57+2fQx#Xh%oK>
zG&e{D!vPouqL0bk^vmN=fMN;0_){MiAU&QHJ8>ASh)+B|C_n<30u(J}`q=W#El&Pr
zp*eK%Fiz=UZ*g0|{1h1U*pmzO7?AtBCJ|<g-vo9xjWMZ@t4)E@bv~$tajDV&N)JAE
z_VG=CU((fI$(Hj2Vp%+|W1xWXvdR-X4|~{>tE$apSONrZR2(8WY7f>_be^9$s@li2
z(xjfibBII~+uFS?t5f{eF*{I1%B7PbLJHkx9n=@-L|yvHhczYZBs3mWB5f03UL-{g
zkY4o}hpsd+fNW8g`o46Vh^K1p=*3r)qv6hSv1_5?s*a}y*KcPV6`ABD0fmi<>?OJX
zMI>>{T`;?HFg|KxL-KYgKt;kG^DxlqsjGX+`+VtpbkZBohENFjo`xu*$2%-QU|yU(
z@~pyj<rYS8Faczr5xja!-)(`?e-@<Y0{7XMQ@-a7X~Mw6S1<>d#l7;rKti#s&00R#
zK1#W4{J76|s>^xZne)IGsu9uWMdEJ`f{_d)d7S|#EaV|3eW?@s*w(d{1}y&yOfmAT
zU_wn~^@-C}eb4^0ZKyl0_QBpg^p5gTJmOF&RC>y@jO55E#?)*4rHU?1r5B%aW*$-B
zwQEQ5(-lmBVcPtS-b&g%_TtU420c}=rGxX47F&~tjf(FNT!7}LaIglmRn(detZUN!
zByQhMW2`0@l*K)*mBkpkMwAHmZw({`i<@xQ;^b^}p+<5fFBUmeKEdVfP5(mT5=eMG
zlv~y0NBB7R^!xj<k0TYxa96*EPLFd$<=j~%8uE{Q);dk@e;#{B)c;kh*@c61(8mVu
zc3S*8qQXH4npwbuEIa8!qnbdg_W*%pWNkz=kTCio&o8ueb=Q|U(fgj6kMY=gAMgd8
zVc~b5W2I_E4B04I#3M<)?c~HID|4{@MIvtn;E>&c|ClEhG$4~=a7*L9zon$(vz@Q-
ziG51)U@QC+*fRKCEKz)5b`&n={?Y4T14sU%q~22uA`Xt6<G^$h3>+AK1*cE*58hk<
zS$+#~Ahcuyi&Z8}Y;XVFoyV<0><sdF>g19#SL{5n1N=R_p6!!q^|Hy4lVhoe<AlvT
zJ4$TF;Lju(#uSitE{j24#cTG0ZewxPPIj#=SzlF26_4mR&p1!=lN!6q+}xbpNo|#`
zKm8*4wL30g4NlUb6QKYofVI5oQBOPkpYN{FE&id4=<@kfVXk#CnTY53ZE@^VMCopN
zAzNPZqNiA3Cw7X4MXrfR@H^3a|6^TKIQe_aF%8qhK}WD9@qY^O;fn--&;uJk=J-i&
zGykk95=(W}Wi8J$#pWHBSyrGl7$Fnt*OXlU>UoSl6c_-E9cSm?-s_P91U2*S#@{$L
zO%{v6yx2cSW598Bvt9bO3HEB8@VN*vqNn73)j~67T61k0#q#{$aW-kW-aKIvzz(}q
zzb(nM!7+ykAD`mKi1yJ9hL#%$fF&$`44||kQ1Q~ib=pSwO!D`#tD*0a4GHjs#6qa#
zZ_9epfn@I@nzmEhfF`C5`j6Sa!aJ~Y4ux)^J3?-K5<Za9@4(z>%Tj+OvfU3*n`S!n
zY<d9h&RyF*^JDjM?TDD;a!gj;fG#Hes+XD}RC<JedQ6(Z&u;PIvK_u4<Xw2uHF6PN
z5@hkDyCjiLhyt5D{ZG7c>E(-%dM}`~*<j?fd}?+2!CV%8g?0)+Nyx)YdKHDnFTUV{
z1ozAJbB%Tz9u6Pp9;P}(dG<HWHA<AWyzsEaTIY?1s0U}qC^k*m{%4%jIFTNGwO0cF
z7!_@mn3lTQ@8J)#)z4`OJ7m*BbfCWo*7|H0Sl+e*q+&MTo!SclnsqAnFiDSu*!NF5
zAki;u1ueg_s3cqfrrRIgkANT|{h_b>A^}SF<4a$l1-i~lc&e?y7++`lzVZx&N{aB0
zD9iK)<W;S|hKt$?=QXD|!&!Id?zlFutPF`CPN&?@`%4!|2^{tMDR7%S<%A4FBaV}}
z{~OY;j}BtP>?%B7deTii-FFA#ZA4bx2OuQ=`ZY5785sy<Jq3@TUv4_Du`hs0L8BVe
zD+y8>4ufv&%iv{rvJK(RP~Tr?WF%OndGThMX5WD|8-Q#f_m;Mv0)d``UE-OZ)|3pD
zN000DMCn_+B}h$@7{SJ|mmg(GaK)b`S8)$MLyDC-ZhAgcT2R&#_ajrNH?F^yC(5-a
z55vBpTKTP6f74f?Ek0}7O5xF&Sgx~%yB(j;k>^PunIZ`Fk}Uga?!!Ry%eur%LJ)V;
z;G~MXRY=F2l$Gdieo&Xs-t{mttV!%ew#!hD_>{CXRDEKh<g8$)%O+wDt&&v!LF0|5
zj!NpH5Z-K(dL&u%a#K1xye<rcI7T4WK-MviuOVdjCA24s(4hB;t^!>OxiMt`PXh+v
zI)B2~DNlyeqF}T03Ap=HB|ES^pM_Y(p5WBGDA6E`pctC+fe7NlML7;&t@Ebuf)Jg0
zz=qliS8VeOS?q4dEvt)s>iW~wv3DLtx7jQ1UGnYS3A#6Ba?~@6dTN&;rz$x@4ikRK
zFpGTy*M1lP?lCZb2v10`7W4y7tpZ3IGoW4908Z_DG51$+y%$3Yyt}4r$AwVuw)8zC
z<8tfR)G;u#H>>x+(zDxz2L03@ci7m&cyaOaWe6N2!bz>|8;FMO!zW-rGx77^uv!5=
z_PXK75|Rvq;WT%85$140^*kg<vy~G<r5KK}u3_KD&+zE`QAE+g7+GUGU4%bt!pNe{
zh^o`INRNStg4h$MYo{B?{xq6)#gK&v%gNLSVOwhCxUdhyBNOdFl(nxg=l%S@A0pss
z`S5c-z5#w_7_zaEK26(maSC$qUcF`Ad;ASfG0;)9^DLGy?q-5zCWixx*$ZP%d2&tT
zo&BBn3J%{#w819IM1FPRCRh@pNhK+r;=F7NS~H06ZN7?nokXq&(PmwHkrzRoU4qtE
zKi^ZWpC?;4B{P2Eay{DxdmcdYxzC9xZ2f*6EU85-@^DyxCi}SFY1okS&TaDQl_4?<
zQ%}68m`d9~r@^rk6r=I%K2DmB$A<U(ch^o#F5?ln0&Y^ZOuy7I3In;rNfkx_>x&T@
z56Us(1vY_pJbNTA&Z=9jiH1d#F=E%SZDe=dL_0y82oPQQAZ|ld1%eytfpvMAVW#Jm
zz<#h+1EY56F$a}Pn1HecBJvhS_hh^8%%Rd1vqCl%I0=p+K@z0XrwJtKqa{csu{AKi
zhv7DEUecqE^zNS|R8QW}(7QIfIgJUiW80j7P$+K{iQByfMcxpm24#5Y_|FtQx_qX?
z-|khoag<K<C!hXnFduT<nOt69LR4}cQfV_8-vg?LEPH)<`O|djEsRgf+N$oXw8JHK
z@_SN(IJP6<PoV)7!$nzmsn02zz=f};hPzVYf^w>1F*KpzyACj)#YRm;czEd=*exPU
z<5_qXL*<mmk27&_bgHp@l|{x0V-*j4%^Z{*tc{U_473aEzA3GX*=dSVmwvrWEIopw
zl@j=1^-1&xtB_-AZAp1=soL0IRcb};y!+x#?@>qLUL5K#p+)%f3Hi(MP6bRfBLg`<
zRxY+s){zbN?wXBWh|fXmgva4s4key+aAT$8N@_jbUQmQ+UpkPy!9afxKPPMN$3(@0
zS4qmQpsLS-td54z!MC=?&al4J?09Z9R#8Mf-xC^3f{6}HhAQ~8_4uD_*h0^0i0(=F
z;~tz0BmI#ZXM!b2ZzyPBlH4WI32urm#3`gs(=zUuaEkm1{QI-wp%%{A1)w>cZ9y*7
z_qG7_WkCVKj6ozgjm+rUYSn*gAV#BSyQmuE#<3>%Tm(P;GEzBRhc;~%zVG8_&vGKW
za=1)ZBcn=N(`-V}4fcO2hK;ydS9}ro7iwQAPX~m{T{5mU03v++kJyccJ~`PC`tLkl
z&|ILwUu<LHISr@L2|nGwz=wAUAG7~-x!8T%u{BZnEo+DC9k>J$Lb1YxJRM7I#z#V%
zefOcO9QykGbJd*>l@WV1#cX%8^^03uI$tuY2A6t!{1ix2-qd>GL4@lhZg(rQ#|-lr
zaSSe*Xfybsi~^8l6CX#maPS*wd)@BGkj$nmYBbOc<?G+gx`BHSbeDeMc{LbOQ2W*$
zs6C>imS3TU+d~>}z+(n~q%I+vy@G7TC!i6zqTPK=8jl}yONyYgQWLcFksu9TBt1=u
zeRyNxUmXvYCU0pV)n2@YO0z!>RxYa{4;!!BD=C>z%1TPa#zzh8XCYgUd7SkauFL7b
zzY-r>LL}A*k;@$5w`NDwR>t9C;Df3vQGo-O7l-pw$fr4}eI@vyk}2PzKRy$h86xS-
zif6fJpuiCiIolcFq)B>gNB4ApN;nS}rAcD_;y`W&d~13>ux71=#0!@9ZJdUcOpAWB
z(03`neEh9F7FmsETvv^KW7c{c(X>E@0x=B^s)k*5-$(eqTO>mT>WN5)IacrIvGum&
zF%lZR4yS(o;`r-0lU+|Z5?W<YM8}=hcP5-i-@3AFRp54|$nrIY9Mn(b{P<=@0Y4Ro
z9KXs#+mjCA$LbAaHPNuqPy{+fO@u~1>?)Oj+XQNP1Fq$5NUK4#F~0P_BdB|=-rHL8
zHC(*Kh`4x+%4g<^KGO4~X9vN**C#*}NfuBBE9l}~dV$9{3nz(gIn~n_F-6V-6*;h{
z`@slJ#JZ59;6Fv~BrY{wgaj!yg9fm%q4T&#c<?Dv2q~RgOh36wF_4%L_#UU9<_aVb
zMFJFUrTV=M8Zn8>)ppZfS+vG5<{&N1r`mSdL<DA?<x7S~2Cp|WEVfWu(y{*XKyKHw
z(xQxc*Pb6oP6^UQkrW*-POzo9Tij`hZ1yGWf5p;3FXnbLGM7EK(;FA-B5sAOUbcdv
z4Y^5sRciD-bbxl6s6I|u>pm*bbw3lNBGY-=vl}=^V!WMnxahrN!g44y7B$oqz0)^|
zmTJGj4f1u{9-t(sg?`CCuH!|xc?DwQx;WMa>Y*o_h1#%|?pz2Br>x~rLB4xtN+u}Z
z^vv61Qf+@P`qh^;nVtI6qoP~$we;+}?;RT1W-6!lZZo$gG4C>v=1@WBVQZ`=jCggs
z^~yCe`Uj_LSBFkltKK*AX<bfz0FtuCJzZ%}7B}vntbuTOW!wc@20tj<3V?BFyxdS}
zfw)MZfa$@I@GUVfKbL`={p2+l<bpE`8wYt#qHW9vR*i!{JLNN-qKz#in8kDNy8Ph@
zGA=kSkwJ&Cq?}Cw&q0EL8|UDfYB@<0=0+X(SN?}#XNQU>_>(FE;DSEqk!TrJkzdac
z32eknWj&)82>!R6Ms8SJ=K-}*3lb#b%}cbVeBl<?>8DX(Z>$Jd&S&vkd_m~123a%<
z6E#(i)}rIdI@(<j$8!heX~K3elReGG4SZ^~D<3VSC3XSRi51DqkfV=*K>{o{^?=*;
zGAH4&*HvG3fjey{l0RnYmE~ly`Z@sC2Ki8>M~Ekk(}5o=55`-|h8edEag&{35|$Cg
zx({ImX-QYU)nu8F0;jsWQyTZVi6p~Ktb(uxLXC<0*n!S(*H3<zIpTSb#tt%LdnV4_
zFqV5Y9<eG{uY&dC^Juo>IC{I+@Uyn3*>BpzdZ|yu!l6T5m24gL=TVCai4)s3L#hda
zjBqUB(Ae}v<kg-ZhPx{3Y}lqAYI46%)Na%Q;%2$gS<`oO<8r9oW!;IL71F=XA5Tr5
zl8SNBSuo&U`+Gf{?;%TFaNDH+6Rt1<&n$B2OxY`ABQWJpqMi$NrY9lFWNrB~&wz8b
zvpqO)@ZhMct$yyOD+a{ybz33pCT)x4*E&%DlsUV&e2y#xl3N(KkG)nJEQe^gi(sp<
z_D)q*YxM`R4|0#tL3Q4f7~YUjEVUR%$`(eYpLG1z*eX>YWf}b5koZiIDlzy??OcGl
z7>0&-XlO9kxHlX8QeKFs*)4sIX^z6u6S_Qk&~=FTX&bkodkgZ`vw7<(kY0Kg0IZG3
zUq-P)4_~i=I=nxkW=9vfF@IMufbDqum8zT)7{StN1wY+WSKZWTpzk*WZP;6v;#}5e
z*{nVpWpwa;_w%g)kgRg?1*@Ux<vd95%4P`$mjX2a{+3%a3xSre8~JuiLoxquX;V{E
z3ltAzaChp?9dYahLc1WPXum{?3-vJ}^(tQ@;YClwP59$;>ze8I1A92^T|p)obnDZP
zY%D7`suQD6q5ixbc-dHtlx9<_Nsh~LLNfB*WSGF$Qn0xaz>7#&7umo;7qoW?<n5H}
zx_z$ds@q!PH-D97dGsvCuwN;zMT&Pw&^`xB%<c%Mbw)d8Mg^nfVnhCAu{l%Qdl5g_
zqU9!*)g-w;{-tS>EDNrqkRFwzBf(-N`uSvPb|UpYru^bxOF~t_ae0_=kx_Dj47az;
z=G|9nWaf|Ycc2qY>F`vPm+yYO?M4Fg9U(OrL?b1yXCY{r_8Wl<*k`FAn(`^HD=sTK
zO9;z`dhuGx91YJB3?x|CLg`qP-^`#>dUsUNt>vFi=`8jjU8n@<pr8g3ro7J{c1mFa
zvgf`Bo?HzZe3hV@g#%$sjIdH!r7HjkvKRy|>c-d^Wjm|M`;kv~myCVK7Zvo7r5lQN
z^L_M`KS!sK1BGaBZf^3H<PxO9QLFH&^LAhhIKmYwnw^AS4p*qf@SD`G+m^<NEC?$w
zv5oEcEH6%8fDp9*03Xz!@DW8zTEa^KBJ2iyyC7oOkYbw?Q1Od~?(x7-RDu~jztyVE
z*{nP8j))39#aTWvl~j{h*^gf1YqcI*lFtcm?7(b!k2Tjzm=lEoQR-}z4~W{V$lyLu
z+xW3-xHwd3i{GDDOQ|*~mrf<|1?rTlLt%~WWUXGk+0&3y_y2MjDn#y`wT`~L-RZa6
zzfu^fgSYpTib;&0C8Zny0SN;$S$Nan0-yi!brw^J$?X!*VfH^qb{Xg`_vk8uqe;V-
zbmH{HvA+rMUVpy+P*ue(qz8;T&1ZSCw_2b#e)I+;cd3iJ-H}9Hz)xeC4)^qZxLG~v
zq_R7T5*VgRU9EBK?HaSL^x$C&#Ynuu#~hg!{=to`&Sh=j9v<kd^Z0BK*((XKv_cj{
zSY;*vaHL^N%tJ(uPHCo1wh1v?G~8)zh<yw4mrN8B2z+fe000zXm61y=N*B#}^b)c6
zJ|V-bMFNA`C;+(p15}>uxiwQ#hMW~IkRemm4GB0|fGuuN?Zow!<G%hG>BPHw+ciXI
zM4Z%MaQT+V1X4E7>dh}Ne0j@KZTiK4g%=LQi|?(s7Zn>Is?q7UJD6Ygrb-{{;=hO=
zLos#QLN<uoh8oWJ-Nrp(IKx+%Gj109DRwuv@YRGvN$?TuK|;H(2|ZOyX?RL$$|j(R
z{SB5Tr&aWE-E9DiW9)E!1)Jbvx4-r)Oo%_}XgeMn16{;OO6|ppM+@OmufUvSq)+4_
zG<32AVL+&s1vG8x^FCxgw}QvI$Y|c!i3%qn^{tw>+KzNTcSS#TB>X89b<%m_oUbhO
zyyL`8@?!(sd$Of<PX)2DZ7AS+zbq;vd~m({6JuRGDs+K@PgSXdX7%p#prDuViCU4V
zvunN%OqFMV_hPIX-iB7_k%%3#f&g`}AX5u?uCTpQ8X~@T(-llD8wHL-!<JdrGbGlF
zd86E*J8O8WSfNQUOa6m?YBQDE4%~v^TXw;@`A7s?=q8Q=AEu5_u$ru=Te0}aj*&mF
zpuF8KXAo^K#YisbzEz5N2;=9oSuL}zJhsB;fI>9Dax-TN`@w?;dhdV;hAw{@+{ZtS
z`e~FbrM|*8Rm6GV$$DfV@(P1?xVi*U%w<FkFMds{H9>b`mjBd*;%lW?v~~TxT+($N
z^oga0i>=0Bv2kJ@H>ywuS4T~fT7x4;8TmW`H|C^cHfI3}8%MN)EY8;ccN~V5TEHCS
z1XP!Ic0d4mlfZP|Cr~!@DVo!&b5Yy4UpWj`s)}Ie|6!X5O!EpXqQKcR6s!<iqb}=n
zFQ2bbF`|%H^sqA1<4%{$IbtfN;Tp9nb;MJi3f=T!i_0s{&=Ib0RH(E?Gb>W7i?e8X
z67ZZ(R<RVg|6I%yCLJqJu2(DVLZ;}J#JY$BeKhCgEU(8YbKnUj<63E+B;xEw{{>Db
z6i9(D*mF4nnTi6Cp>w568ORh-klT1~qegpb4xp?B?Z0l?8>0i7;?R;HvTh`zwYa%e
zcUAW!IPDc$^*;W6<MYE!^T}tQW>y9WuP5KQ!>rjpsl><ls%vVo>NGX<s-##GWNELG
zg%mNr2q)p(8yH8j)nEDkxx!eZ42WF>+<Oc0Samr<)_@sDA!6uV-F4@=R6qJ~NFY><
zuX_FMHx-!ZQXb)g_q-U`4s77$(Dl3a_<&m_SUTNX9jo4`0}0f~gX%twf?U4~*~b{a
ze=hm>Msb`KCB|xHsWpUl;$bSx7D5{*Sq)s__2Y03rsX)C-6~|ILO%j366$dR1FQMp
zM6(-o7q|VcHzk}j6}w#-PDzmJQ7Hdq!cWKecsA#QUVX4Zjb#rZBrQUL4v?SMl?KyH
z&lB>Q8cke{O;A_XM3g#E@k+|jZ5Y87#)9N<)H@BrhWV^aO!p`{U~!pb9}SnpwfNas
zdpR_VCqACwVuFNEL*ZshRgqirI<)Kz(m?k{e%#rA4jV>md1D}sWwqu_AcGu@Gkvb&
zOZYe0>iK%FD=*r6A2FREskufAAF*dVw#s(nPxDV{4?7l@vihjB>p`ixWJSpY_6+O*
zjhu@~>Jtw7gMsE4JUs4T!>byN=zEK<hpmu6y0dMT>%wAE<p%eIFV`;LD|)P_w**A}
z5i+@cQP`f<iRN4TK62oE0+*`Z&wEtY;yzrk`zGbJ7F(*lC+OH(ah_P8<i}+uQN74_
z6!jwpP;kulWQnCeK9I+uW3S)#los|iKllOU4cH(p3y1eGU~7~QHXdp8^QIr{!GZj;
zU?c2K%lyc*O?nr%pCbs&bzRJN0*|Vx07@WuLXfVEKgl<)c2Wo9wQNy41<|~0*xHAK
z!3H?JDCN|!XGz=J7iUck!F)}%N52st5(s6+Lk?~xdP2E;C`#(%lBQCJHxs^O=I8@_
zY@y-Q>xr0Z6vlw65w_5^#w`EkhbHZEj{Ck_%TKm<6O@b{C7O3;I@QizP5(&CXBoNn
zeROD!3ndf4MEx6jR}t_drv9ZbN}y=me|Fp0?UzrM2cE@clz;^#^OU3bpJ7uNvRAHA
z-7#>t$iQZG;sml`7!;~z;np_=wMyWuv0osH+wSTYz%K@f#0t}Q#fwy}J1`FWu7P_g
z5|$8KOwh8|5q7=-<CGP%Q;_&hOfNDV$`9O<h0i6KjLNu=ou)jG5nEU@m_#2LlCLt$
z9?FJUBJ&UNcPY4j0<jBKKXZ_*T)af4)oD8xcKbzU{_7(}&VBISbae1(yMvjB%=nU)
zIK9%vBdOT|Y|l`CjUe3$HHq<KIvZo_GUGG*ut}Z;QOY2Kw)eJC>JilAjr82J-e?~S
zmUTB^Oi!f{boi08mQ<lln#lVk<apdG+K}I7`R9IX^l6;JyL~0<I?SO88QJ>sgxL7c
zt^-o4rtpyNpnx2-B~aXiVFWPov;66ByoJ$8+R`ri0C8?GocY+Nhb`oE;U=c{kf=XA
zxsy(x?NZ>=@J#&D$g+S%$Nyclu<Xf)7(EUVqW$<CDEe*Sbf7`iGXGrP1&iv930wtR
zum6m^k|R@BoxN_Q9#&c9u*8BAHOx*4*{4I7_c4?oNHkg&AtJy1`JeZP_IzxH4d++D
z)%cfni_M-x8hm#sKQK8jZ+bnx4R3?l*S41v3SvV8LP)di>yBb?ku!QP%#r8VulR1H
zwZT+5ApG0`$C?sa2-?e3!df0&E8W4Vy6~_i2t99jKMm8IJ3hPfE3x=-#|fdb#|{B>
z`=iiDt5T01&#w(08C0XB%4NM<^Jm2BH~2d%B1%!e1jRp3kKs2C+=9$VaN<S3bQcPi
z&|mOIgl=llVxkaK*`YAa>OelGXe&x2f!N9Vl8%5~?2oF(VG|MMZ9U?J!%yE32Y2sG
z>u2;uyuBOxscu{)UTe`fx7Ftewj=(}CnkRE4wzLW=Jmk*1rAYyG&P)bSrEq_zx~D7
z4;Wq1tzYp-iK3+;<BVJhrsc4vLHjsQI_ni~Qu1`T|K#g0a3*nkcz04Pc+$b*?>tf-
zKdBtC0?(OJ*1KO@BR+;E+0VfEWRIM70lpNi`Z6<vEEh45P+V;qwjd-6>7icmFFcNQ
zv#uo|Iox(Sp3`OwSk7Ot3=whe4~S!ciCtSdt}qBWM#ENtq$>EVxXnFqG8;+<MhTF}
zD5S-nhV&Gf$~o`vZjjs=_)7!A#89x(7DOlkV?KcJ+~AUdT#FbJNpP0Q;Q--DT-AjG
zJe-c|m9a!ay_3RkNF2i!nwLs-F59|#I_^g1)5wYn!G-avx71CZM@Sd+i)hfOgPKB3
z6KL&Fz|H&C4d9LVlAX1Ib;bPX+d`n5wh1f1mCp*hU7qx!TOsvfn$Z~>O_%MzSbK>c
z!%rEc4mt5MI7kss?c8o298ug8tIsdshydC)IoyUEe8HF_8on&SaIDIwV?dSR*h8C0
z`@+NTDZ})+1S!XN!8pu7ZBm8yHw3dLuF(%X7;sOnN@B}e{3^v3oTm$;gDg{%7BH(#
z-%;-e_zdZ;dkA{nv?;lfo^;q4Rw;jhw=RQwtHSMXu|ee{1$fv@;L#cd>RZ~R`LBj*
z%H+EMW>rVb!t@*kguhe0T1SpinKdM1@;V{v#i=-Fijcj<Ya?x>KSlEB3e#ozXLTR5
zz#S-6ba9~o?L+!A`THY2x#1?a04lvU-y@BX1NA$)<Zqz0N;fR-U(&*AuGVz*bwt;W
ztY@Sr-C7OX(W~h@@iJW>b49UX#QtuPv0s+c$^|6e<Ld(FFcT%F;K9BUbwVIEaprDn
zZf=9?^eeE1egnFewQ{8Y?xhuz9?Y^-;ms-BfIkI(>-D&#bahPbEcaiEcln9?zu&jA
zi!UuO7Wo+=pZTnh<>l+XmWy5={_z6Pf*<A4fn9xem{@83KaK~TQ8^)8Q#XwIUQ)I+
z@SednWIr6E79`GvdMgSCn`4$0!yE@!)$qg}ItF>HX17oe^&29uj*woJV><TF$fhH%
z{k;s_7mA<mKWUvOb9{;6vIBWT+L?zIIDgfFXT^<*&023?5JP%8ApbJnhb*t&lDn$t
zBLO=AD@k?$vP7!lA(#;A*qU|jOOSqLx}*)GO(@{ZzHYVr{Y(RBdk7zqNY%i#A;3?Z
z%GasZ-}6wd-GUa}C43s%>Abi(IPFdx9dBaV8w7SirPsq<#?IBpJXTcgQGpiH(X$l3
z77iB)zlgg16xLs-v9Vx+V!US`=cgulbkcQ2RQ>70qW>v2>6|IGv}%+u_M!PzZHAe&
z1rl??c)2;^TMZ_X9hfCG5woC$>KF4qb^P-iC6mdXx>5$xyPOj{HjTw?NAt2wXTp2G
zgr#9cwF$b)PO4KZbDB!qGmx{v!CbqpYES=PCnNEQ5a)OcaIOfmzp2Y*#_itC|MnC_
zTYZH`wZYxI+-7`H$B3J8O6#E6^wyNSj`#gss5G9$GFo?$`|BMxo0fjzRw@ZxD{%#c
zR?@p9Qz-)6;YWp#HBN8UW7}OVFwUrPd47H4E=m^xa@|mJp4h4ggb1LO6!LEA+GTHf
zM#w*1lSO+M?wVDV-|xggpkOgto_{NukdO<OEYc&8jBr7G$7Dx+7=#Mx(1M_o`UZjD
zu+SD*J`LJQ<jNTO+7?9-^b2J^UM++*0EWy4-iP@#GgMfLH~?EW$}@8uhXFUM<Q_W=
zxMdPR=6ICAEmNXc_0c9G?(uA2gj1P%Jv0zO%F1sSL31sw1;K3p+(bvP_Hp}n7wAP@
zP7+pw9YA((HBb7^P8JK!6F@?dz{T~h={UCMxF5&$$yWWZjTzU7EQtyNqLVq8RR=4X
zE3Wrm_zrD?1#bi_fABqztQ=#M_(uL;Cj;LM0US+3xvbg)HE8|u140@}DGu%)K{FGa
z5n^m19Cq0!=r?)ZqmUP}U>ROz36*~175P@*gUj~qo2n%;UJvgl<xt&H0Wbn)b@F^>
zP7M$uyV8>0yzH@Gn^%7a@*tkJZ=aK)4Q12LiPXxwsjlULxBjdEh^Q`qG7S`zlp;l(
zf1ga+c`5wZ1}zWF9`Q&H3{*TY#e@V5AelJ4sIv+HjD<oQ65ujG<IoQJk8OuQ$Wm>o
zc3xde`;V==vn_yn8bn6O=0ZV5<<G=dU7a}rAdYBxUJyBTQ?mXUw0M;FPCO`sj%Q$8
zZiYgFApmCwR$vk2-ICF7HU{8SYVy|&w}Dfmfy=R$7@_q)=VFWHFj)ykR%x{tyQ86y
z!#J(qTb;QN5Rc`?-oUP{yZR)L(@<*Ws-=<1=pNDU5m)`Mg%=gRT`WqO4Jy2nF86}F
z_h)S*b^o`jfWX>gl}~mnSG>~Cc}O%7z*_ZzVH?5W#Bta|sx!mgv+-EKlWuI%FTvAn
zv&vs>@*yh}H92NaI)jl1ycq1S-tpi&Y52o2b3~Hx$~SME!0s>RM-MKaRHU~95zDSt
z25euD^_o+8pFY4QhB+Q&U7{PvnbYthaCBH;+KxBth||_0_k*ChOdZ1<caVFAgZ%?-
z?^1vE`HMOQys(K<YoJGhQvB|e-@wsOBBZGmDuM(AZ`5j>%E{6xOafoY#?Wt8sQ=J2
zjOK*4PYk}W97kI2oylg<E3hMnPAGbtBbSScii%<K{r?zy4{)yA{(YPe@<GWgWQOcS
z%E*?L?2+t3D1=Bxc1FpbnHd=s3T3Bllobh)%HEmT|JU`Q=eh6azQ4cY|9yJy<M|%P
z-F@eJU)O7#uk$=#4)YIN_i`Mj$i(Yzz<b#+;|ZB)vO0<yt{*ogRwOF0MP8QJ^i}#f
zh;Ykg!qZ0UOOsr%<yG?(f!BXZU&0Prt(6A=N*)D-d|~JcJbZrBs?O3hmtiPp5`YQh
z6$32+FVw0)p`c2wfAFnUb$kD_#(2VS_*sDu`?ouUL<7VY?eq;M26P*?<&U_`zjlD@
zBF8F^;J>zr;j;h!%h1!{reuu@Tt9VBmV9Pbp#^HWaQr){{-)&U!o}UwsXJu>g|h<3
z4~<YE7H3n2P7<O+^>Q*da`w20K?gMSG38thJ2wE#I&igeGd!z`?;I3N;7Y&oDryv<
zxp*)GZj8Eg-4bR5us!F5g7eb$Z&ZlcK^hrqi<{;)c$Ac1zi$9d0LvTm2kc<XlL(~t
z8(ga<=RDGVw>)0gAv6jgmVI{g4oHfE@VIHaOFRn@=sr`^u@6f>qr~a~zLI%y$h{h#
z#<Y~mIy?;{LLgYQ-NcS%su`|dUUKW~{hVuV3>%`Y0B{_)xhS~X>Oct|i~+kwO<Kmu
z>`3VtKz4J^KLaWkc;*3JEGG=9nST8E(ea^~e4EIp-^~Ew=S5~@S1Cfyie`z`e14EB
zsh8`weOEL&mc=n8D4PxgF#oz7;dS1;|1?xPCmM~1UPBL{q`oXJdk9@D_8M})E4jm3
zum^{O|3I@Zb=Hv3Lx(Bd<-=D)Cv|kWtn~4SHwU8`8jzD|JP7Z{@(OArK$wI$tSV|(
z6o~Y+@Ub<C;YXZ3M?czb0-zj+T%w{L1Lh!n=>wgDQPW+xZfcM`hT-K0n_m7g2Df{k
z-DUZLdxjIWzpsVIa%tYKsw_0CO$<+08A<E*+JtdC0?m*aMonyaV-ZiEeM9{oaF7P1
z8}H5|N;Vi|ggOp!;5bwP5QL~HQ4ktILFmNeriugrM{bjTNd!uM`0-eR-P)JJDXAXj
zytzPhn+K-u8Mx6H$9{-}R)B{H+qPx;HLvkH4L8vOJbW8jca)+v&4^Z@U0ip-&DRHr
z##z#u@7}!m-6U`1hmmnOC2)1iJ;X$<irGT-O6hy+wUoUZ+q^0tPtBh<c4jF5^?F*w
zLPd#{6up608l!PLFeq-24Q~KTg?j@qLH_^$2HHr@Gwr?s#4FE%q3D|3<zlpB1>xpV
z37><DWyztOB@ME8n-#zyxjGaZT!^d*5f?cp_v;N9%_H-9o>*%LJ%}WsP*rRRmU}qn
zQbVFkHt^Vlfu$eUn3b*ty=I%<_!bAc`z~M<!|?QTp}rb&D?}`vBwFZ53?f_{X(4Hs
zkJ^et(oR9}zN;ukVvOo#t<g=UeMKj>;tgbI^n7KyfbH}GGlakaTrgndaN64ND1fk+
zzy1bhL{VTAs0yU_O~?DB9bbXj(s=L0Ln!JSXI<#6O<?|cfMRWJDMgMDt&QW&KYnKu
z^X=-IZt%@A-I$oGYkj_*oywt+ccVp5=|xQpFFY8q(<PPp`ivyuGNsdfNajbUMDB(E
z0U36xyFnC&LSB{g!K-lSZg&zy1=V`rUBUEM>mNf7l0V!+=&~=L3$Fb<owys601~)s
z)12gPSxofpQZ@7{0(UxP$zht2fM9AF_30&O=3_*QeW`!dZEtxYbLG9mu-}0juKLNG
z<bFd~!d1vJh=4kfB_v&n=f!)iCRjUkZ5vWXE->qYZ+`P$xVy1A>noab!ttx2^!COz
zg?lUSe7D!lk(sx9jLjyX)bXHO@_a^cUK}8S!m!Hi-|uCtHh%x|*HbVf?4TKBNItUf
z)M4oNS?)76+p(c+X~@(dM|?Ol@aTmZ^jInl2n}w($iOTdhHr;E5-~N|{_bV+V;gU%
zHA6{HPqhle*xv&jlFJ@JGn@z!c1`!gtR!F$F(3+Rjq(MF$@~|~NcxU1B?cD5#az}y
zkAO62?@;UiIe}AJDOlQEwJ@%c$+kUn{*;4`ofenE1t_i%29x=l`+q(M#U+@<!4QsL
zOcF<`y<GzAWE{~}n&ZO>$Py!8xd9V^G3+WD!$Ms(6iNR)gAK)9I>Ys8^1;mKM|23p
z0icxw8B0ON1ElyyGr895N_=Qn2e?4dJ%yp_G*~6;SFy!r99mzQO@yk7!R29I81jMf
zI-I-`F?8rSJ@H;TRwR~le|ljPIuIndSGafi3IO?g?I!48A{31(unzsN?*TJzKt6wV
zZ~gQf>JVpdSZ|Y;RS|kU4q&~cyPESA&LFW`$rZS|$a(Vrem$n}dX&-Y@yETMsWJai
z1K$^bJ}b|UwzAy#PQwXMiZ#I8z297T2M1tM;<$0+sJP<Mla!%*{h{lh9%x2NY6Qar
zsqOT;Fy{mCWr5bzS$n}P@{1P)1kE9ESy_To-Uon(W|&RpO!GmNxdGBv>4ju<NpLL9
z4=aQ)xukc0rvECerq&z89+)Ib*!DJg_o3(7&uZ-CBzev|`{d(|a|ihvnVyf5<|J`?
zq@672jET*nBBVIR2{<gzNz%VpWmvJ(@D0hwLs{?1v4*hDaZk|gHZYdKN*4QoUSnZ4
zw9~*a?=~==?;?y_<Ai2Er;|PEzfTI^mzRn_+@!)u4vg9yuqv4W`_L{Q8HX2UU6Iy6
z&$DR&7bMfwS>J6wAhI0KnLLCLV1NUfp)#|EX=<-0Am2;<>D{2=pH>#*$zWdY)YY_`
z{s6sw!nv+?h&!l+mbBI_q7g9C>)h|8bCj1g&_#=RKn?Ol`L)oyce`Tr>+hgl-0*tG
z(ms|Hwa5`A&zS+4gB@{5m$O4*c&H3pm|+{C(7^dJ7DAq4s2I-Z=zxH_={amSLj4xk
zU0T8b80^u9Gh6q=Y#zc%`z2`Jl<=zw#K#^Gy0&p{cmbd@@lE^Vs3){Ax<v2D%1$n=
zd<B=Q(?p%R{$%GVLQncbdEa!rRgxV|J0No^_l;kxgHJB|h3Kex9lv|OTjGJ#Ggre&
z&(<J@9PE@#A6h7fiq>VY9^b)ymn$=uESU5Lu7Vz!;gl(Sda2X>E0>-aShPnf4}V)&
zF)}U)>vg$wH~zVVe$4#aYxlUByeae4dG~FF=i}oOen1cqWP<<cl{6?Iyi$9zRh$F{
zm6*(#QwsHVD_?+%6Yfmu#0UKBWAsG0RFHMLenPdI&?Pj+|L2cSRv)S>7B*}>^Pvz|
zzx_s*96Y+ykk9`b&*fq$CKACLfWAYxmkxE7hfebAKVuYkQ2`o4E-bpv3Xsa@N-AV|
z^$6(Wz+7iVC2gGcOH#1Ene16MLnioV(<OUOsTvFF<b<9q&?j=9qs?Tq_6LtC9cl2K
zi)-w?4z2^`nfrL$3%L?u!<{x*8BZk>rU*x_EL%oOp+%kx90k`tK-r(4S2+R1-SYqe
zD0V$gcUgrtp#er=<(6X}i{0SLg$#R_)XgZdMo=cU1?de6LxtWf5k3OlODAa%Z&V=p
zS*XwtZ=wo)oJF9bc7msEBIFJf8^=xC0shSevWE*K%1>@u6#godp^=e^FjNbz8jVHr
zb#Z|fRNpw=@~b-c2*c?mh$Ku%=x~astBUkpHt}gk*iZ*$q^LeK9~JWOM;1SMg=H9S
zFX0lZlKPF=h=!cjWg?T72~1__`z9bn<}N<&d0f;xyOs(MdNw!~xKaEEVUp{j!@EPo
z`IEsZu=k;;ln-_IbRTMWYW}WZy`2wC(J}LU#)8nLq$TDa?gJlI+x@_geo=<myiSUI
zV@6AI4v*OMI6UWYrc(y6LZKidYq>ChBHx10pC9TPK$Hs`;6)Sgj5LBPhJiQ=ODrV6
zjnJ||pjSOFHnn?Dz@?1LV2$drQo5d+Hpt|67<hm2VB{g?6<mzc6j)B%`SUD(m{67h
z`g)gW`9jl}s$KdA9s@lY6d{4}r!#ed8h-&}!v3*OyCV&T81Bo59xLJbk5OaYScy%O
z{K%_<X3_Vli@d0Q86d%%d+-+XmLY;5AOs}q?=z-`<bbQ#xk=|%oDW1VZMT6`ew3&j
z@o6hAKwt>Il#jO`s7OMfwBXGwSpmx?bD#5Dkl`n*1#U{Y429;CJcL2+r-`(a4)@Tq
zo({{?*Q~F-km8>eSYw|eTy=TZn13`r_w;xB3l4Vx<_o3`3}mMmMgY<(-r(ZjYD@e?
zGE}l5H}%R5{;cZNiSJY}>kD-xY#CVEp*9gGlJfvwUiQ=WBuf2$(>ob8aN6`8egIHO
zh)Kpuetd_FdhPT@R^lavc(6wv%ge7#ha3qjIBZoGY!v#fK%1G1!;PkbC3k19^c^~m
z;4Y!vIrJ$UBaojLaTn-osGHIj{n)wRWziZ+ypDWr_M6AM9`fd;pG<Dg!f3)XKD)Kr
z>U&#a09ay@iqc@diO5Gb#*aIT&CSVpFQ}ZgVF$rLXJnL~EVfjRwWA>2RQqWC3E?JU
zS(<cdOKH+mX}XX+&XET3MFmDEqOSKhYP}4r%Z<?(0><)tOzc3!CIXulp`cpuYt}&+
z4fnaC)O?ox54nm;aTQRI(jZiCMkPVOzVjkCd~lY&r^cfcAp5tm4~p}>6>rnDi&}%H
z5Q4~BpLO9@`&{?+eicNL6`)aK$bQVB2S54AkIWPCMA~XEC4UJQUbE&gICh1i+)%gE
zF80i0alXqiNjeT6E=fCk?{3|-Q;(wY7y4hKyBs9Z^>_MLpf=TaZ#P6mvp)rnVk%3{
zHYG`JeFgxf+aH~8rBc<)g@i$zM4-e6aG91-L_B{h;Jl|#%bwF~3)<col`+B#0BYXx
z{vRN^i3UA=6Re4PxM_O?{;z-I@+fe;AE=Y}!#2jk8eZGiGF!j)3s5KkQ@(tV6k<^<
zpaYjZy4GN7s#yErd4BzY%K?*M@vUR6i_ZuPQXerXsNdGt6mJFwHX@@n8$-PWwc#Ba
zZmoK)li;%UKr`d$Wt~ZcZ^M6sH`kP)b<crbx0l0siJj!Y|1se3cOrIaGe#iB58q+{
z((jfVKX*fQ3s?xA)&Gq+s$h?<9-m8rY3I;-Xdti(Gh{`g3@hBU4tp0_K$H(Q4k+tp
zn4)#uvm6ef^_n&sB)<86SNOH|K%jWEH6s!5o*o@{5regXEE%hr=OM3sCKTtb6WUFA
z)uO!%2!eyMonh+1Ku!=5hKJdv9^CYc{`0~9miT0ycG-i2Ysc+#ff#VolN?v+4m$1P
zP2ZU`7J`IkXXFOU@W`t!aqMx?!r!A;kD>LGBciMTp*BahOq*NeqM3dzC`;o&6KlH-
zp^%Fg8$dds1!E5$^ST<?O5U;fL<4A-NJ|19J_=q8RuDB$6G^MAS_S}y9@_M<MV;vU
z9G78`gdpw41t@=>!iuVo%2LkrEvIHXdzlU%v=Li^qW-pLZ?Y%pz4_)bI^O!XFwi;-
zQI5L2sP{o{(H1hyEDO7@ZO<AN0Wg!R(-+!k3;+aE1{$OwuF$IIfuhw2Be9|R3VMo}
zD@&88PtjiCBQVtiuJ_NyvVdtKBbHUlP*;Cs;0zL>!<!=FGH|2owK0fJi=5a#9dX<5
z4KXBZCd^9^5##`5_lm)}r!$BE?DWF~f9bsIt5-MYd-f~89%#rSDs32pWMz^5;lkBB
z1AgIkr9Vn*j=tr?L<<4}GPELO6Z-dV5BISw{>-5Uww^t{4ikALwYRvhcXW4FXd=y;
zmAMKz;QCWtC_wH-(>2J+h?)Va;|U<OlBW5!2iw#ZAC6O@#O96_6Nm~dIxvmT^r$Xk
zqR60b1RQf=FZ?5-Zn82n+Be^!9@tw(2S2-!|L@;71jcenTN(p!Jq!cr5Xl>0O$!?&
zgu2drZ3R8qLa1QIRiE0nANBxbSgH@WKWAyDUY5qvzQ^XX5^mpHujXGJzwxC{FU!e*
z-VM}ubtb>y$9E?{xw_in5xbwPEx--f$D+!J1cpn#fB6e&1#xP)K(+UD|8F34ud{O=
zu~VBkb|_PA?lvCA@xUO%_{c!^^d&;Ng-e~>f$%c~nbjA-rw_B;H)gTqrngcyU%mtT
zB>6o=4Lfqg!cWY`!?Yi511CuOQ`hawjyr9^am%U@p;a5>An*8|gq&$$toAw<;4cyF
zjQ+)8D0|Sj71+7Nq#T^)iO*$v1k4}H&@X?EuRBp4KIIxR)Ixc1`B{#=9~yYsv-Cl#
zl>latyv@hxzx=qnPzzjApupoQ&kkoLR^r6#j304*Mv3W8SyEElEG8QTcbXMnLGP;o
zKBFr+Oo*sz4*U6~0HMALD0kKLP`A;aXhPxn{y`^Cs0%oHF%pJ1nYx%Bd+M7XgZYdu
z)R>2y;$f<YgfGM<5der(%R;|_mU0E>)hkz%j7Gw_eGx#87|2Nt+B8q{c-V*A1A05e
zmsQi1yRPTQR}Q1VGxG8QRwDG8%^!SU9gd1~H66HDscW#gY-1QPd$Y~uEi7z}a93%u
ze=^V*alX}-5jH;d3pwPnfnuW=;Ev@r9iR#RJ+GjXAKLAyO(aI367Pt}?!o;{&&Uu}
zQkMLgJ3TR02QUMU%#Ux&@yMI^Hvrn(=go2DA>ilb>75|X{f@d8X7>mNX_yB7o)QPY
zYF3I+KVvwvQ5u1rr{H5y`x%yR0ZW+D$qhW8;$VS=os84oiepgn%lfR1b5#O>5RLmn
z*!%S`5y*jwxeI7TTD1>}*s_9Pw(A)&IU_eJ+CBqx$_tSC{E+3_^O$a`SnGzxw5OLQ
z$4f+~B4Wg?zubxE@H$|#$5~4jMd)RXZ1B#10*9debv0GuUrxb~bGHgot2(h558cV)
z2qR7cP|vhm2c24zx@X*&<bG+k`h;I#kt%%e0N{!{mBJuIbcNYLl)XMlG+T)to*3&Q
z#*u6&lT6W5mKch}sucZVm}e$}ai+xJ1lorHs(fq(BmCThz8J*o_ucr(;)MG$L5B6G
z3<G_}ppS0?@^4<P;P$A@c6$t;UOp>yo3IL?bh77$Td3@-aw~cfx9UfI9w(;+3Uwr&
z@Aat5vig|HGv%OKjE{!#X2v}sc$j6>i4_Z$@5s#lo}PgAixW}4%(Vms*#Bz;Z)9RZ
zHdr#{XW#2J@Z(+qIRugtjZFRiiyim1NhWz+ZPiWKnR<mpnoq)`5I3rxExmVzHWxj-
zQPR_wai?IgJP3B&ya-<T@})kVITda^SSwUBiaFIcjc6UjXmhrKkGuycrLred#m}Q;
zK!W*8-7obKwwz|-To+*DDS|B|$3Z)c7TBW=4#Obf`P5`BBHpT-L~BfV)LcELJI_Ax
zu2)NsjxTGxfnbL6fW5i)kDnJYJO>hL=<oIcaVR9$l#vFXP#z_Sv5(@1|K(=LT9NEv
zP9_p{Q|T5TsBf4`9NVptcV5U|-aN8nBWdjlMtC8^q8PCa4Lo9-MeJiJ&lZu4AYG$(
z{kKWdvwdEvZ1tn^*ETIyZW>&BoJ9lwwn1-y6*3eoJDZppMQo$cc31p!Xo(5ho`E^4
z3@2T|Gxv&2umLz}zQ_Gfpb+jKMS|GlNctSQ64{Nm;X!(ig)}Yv@jA^>6GNbC7ZbT@
zy<`FbUEC$J>xP0(<7cuKw=Giad)!~H!2tTn2-B*SMfE(@!O0C)@`+AJ7b3EbPucza
zZb2bx7^<rO&rpGf(dPNX2sl>e6(9r|<8NodG9{FsdbB=i|JD7J^=vg;j7$IpkN{$u
zdMJbJW2H8)E5`e%h>gQvetu?u6%yRfU&xFQ7qE{ozkNV}!6|;11VMy?CMcFNfy9a?
zASon|a3GT55d;<^X3^<dY2dd}9Xga!=(r@k0o@*b2MBiAKo2OAlUsiO$hbaic+it2
zb_<7qqU!DKZPcT>UuW>g@M#Y>;Fo%|z1Sj|cN#z7TXDu`yt}H#s%i1L>w>sl9;6b9
z#^7hH6b`7VGLFQ-DE}xS40hw^qnYm%T{bI(mH=7ivy0;jdB0&4Jw}MV65J_ijri-6
zp3`-<PGAQIxoCReen-Q^_*l09oXMzCK^w_jWm{@_|H}D~<+pX<^!?TezcRPIgc3?$
z>I+EtR~J8N5)?Ut?iv}+!Bm#Kf*-|@2eQCjw+25mQEEe6j>j5VKxGFL4A%2yw$UcF
zp85}qK~W-%#<y-6r`$EbFOVC-c~f`p+=)dM7_`q^U>a2mI{O>T)@5iRJF@CVLL+c9
ziPpZpOw8&u^W((07klXIS>-L#%s8p@ll{6{W_ITdgx^+hg$Bkr_2%e!d4?EK%GR?$
z=H1Xfj1!dp<Gk#4&#PmGPkHtU?iMljZ;GIpBFoBz3gU&pTGx`EVJsvQ+iKt|-Cr89
zg(MaU*usa(c{rp)x$<1-C4IjB&E`Zm?-_I>W@w$R>;#&nK}6997o7bSRfujZqqVn!
z6%MMN6(?Z@*}_#Y#X$eM`qX;(+mP39?+Z}L=*2$*?FC@l>3ZY5`T79Y?j7H4VjE#l
zG>j-iFl@Em23C<(L*nOmFSa+Ydce+}Prd4%ei)mh94fP6##Y<En(-6Lq`k7&Nt<O-
z5)<eWK|nzTA^GynSuN1t{pJHO^o7pQztv;1_0M->vb&79bk_{YJLvW{C^G$Vuh9-C
zUtpQa#aDl731!#)aZ-#}(7WVpNaMC>T-uV{dJ8adS_ypVgZE0GtxG?wG|+~coC1_|
zeYoPJJd3y)Xl^j;YMLW9<%!HqKVw>R!&w}tw@s$@P%Zjl0HTgQHB-S}uVaOyDy#6I
z>IBzIrc0d~KH9px+b5+v%%X`y6U^_VQ~}wXhv9G&P%=kVb@@_6uPzh1X|3NR3eN|^
z+9Q)n9UR-IIau6ZceJR959Y&+(jIS3hz=xN)_EAZk!@GtCnID0B`CAdr8lP|QzQ5l
z6cKTdv*mti)1j~RzvZj{-WPmY_zen`kgnr;4=^|5POGG(G#~xMvaC6~Z;x$Sn*5SH
z^u`zgzr%!1SeO_>;AMc(LeeBJoYS>zy)n?j+P3u0tUnqtQshCu*@f^c74GfPN{PV&
z8Dz|fDZ2bFoW2A0WB4^|!yhIbgYp+bJCYg>pv>XN<Bvd`cabbtxgQSL07!KN!R|1%
zJOj1eX(HRXasVbz1CGT3$j-tPc!xdR09SFUg9k@d7tFLhUTn;(36#8sX(cl0(t-Wa
z$*>l3s3ooM6yM|#qKLG*!?jlloyD%N6*e+F;`<B)qK+rds*}8X71TGJPVF6Iq{0t@
zgg>~;=lsXpE83-|R9qUPJ9HTrtLSQba1%JENHok^eVOjHCT(Qgl=M7rlw{B%lCK<U
z-s9+Q6|tZ*x~p@adhptP@hkt$@%|0ePCJ#(%jHZ@`9(khIACCie{QLsV1EO&<vVHY
z7%7;#>#Qw~Tb{Kd!+jtNfA9uU16sG#aOay!$Ns8Dv}N?>$S-%aeG@dxZ=-I)+Y|;y
zGH}B7E$Jca#xPRm==L{K=EJIL0#3g>tp!3~H<pqVxCe{n4u8ATv=EuN_m}o#851ST
z<DHtsrCWQElyid?;7W6z`Tv<e>=eU;#=E^hbp~U`&Y_6;VB2KTmGc+DDM0rxclaIY
z5dtLeN_7Ap`Dag^Ll~@>E}B5C-TFVCT&M|t-X7NyAYj_>hdW~ccn7o73EbnFtvNpE
zLIW2%TNdX3!+;3KfTi<K@@Ke9j-gTkk`82UTRED=fZb+D;oIW6{$}VVQZ0bF&P&}R
zpjDWMA-pPpF;aGqz6h3V|K=P&?tCY;&gtV;ptVyF`Q0<2Ro|FM5ICq5z%%OSXAGU7
z)0NhC$A5WEINdjzt)&VJW9&ovslx%Nun38Q6b^wBl$N+W7BOzb5EBf<*?mcPu_S{O
ztaLpP@4Ik$?B@TG$54kMkQvz3-EQGE2kmbhT!}p<>hJ`t!9Nby4`S(v;9{_|D@TYN
zdAc*vE2#(6JQprSy354EUD7uc)YLy!$*imat8szarx_V}yhTnF=b^E*5@BoA^^4CW
zTxMQXTRYo(cy#N`-a=+RQg@%HIU-&a`Lz1tSrOsyN=Prgf@Zs&4eCAD#p5=B+x(-<
zvB=nsz;$)c$}d%O(E%Cq^Gvo2?#vG1d@J@Z+fVTqK06Iv5hFW3W57bT2+-7bt^;_$
z{A&abWya4V#9fE8`wa?o6alrSbE1VJB@Vd~9W|+-a5_)Q>Q}FhV1enzUlZG0I3}C_
zAR%Zv`N4N?UhnCJ*Lw;;=WEmV;Y^c+J`?D|XlUDRzcuquBNkM7(bzZiYIu&=d2HF0
z$IfK(fS0D&n{qT{4b{we4Xn>Z2&QIqkN$@`z4Hhx<aZu{MHY3M)|;KYBQN-A5ALTv
zzWenrdHM_A2P`<{(-FTwaZ$*-vR2ryy3&z*XSB3`SbmgRZ4n?fVejwIE&IfaIsw(c
zL<s=T-Ej%}<>FwDYHSV^bol3l(5~U}+{8a20x96;Q3*%A6P1o}&H-Jw2_$l5PkFUk
zU~UII%L=a;SdoG_Jg7^~Q%AerfU9#w?y+RpBSYSj>Bxoo-u1pAV2mYvK4fd{yPl%b
zYSW|==tV|s)cI=_%_V<Xh*0=rk<(4UsYKm>OU2l*oqNE>(eJn^2c`d65QeXZGXSve
z??h_f=PNgI)5~zJjF<2HkK?yWd7iQK;bvBUumg{4euu7S#E&=cZ%Vj1`cYpN<(2wj
z%<8qbY7eX4=6v70#eOoApj#zzzs^0D5_nS<!jJv_TmH!|lEeYH0Zd+#%ioeUh1Z0;
z$eEy>ywsO;+(rZV5uxBp!#>Dcs8?_pR^E~BG?WkgJh!!Fp#7=a_;~s=E}onv%f|!o
z&!nvhc4-*^|KGam=xhMcVxs|3Qv=Zo*D5&e#K5+n8_&)^(U2ulW|@0=vNFc$Q(sl@
zX0hs9yz#!m)QwsrxoxsaZW|?EVpBjC6m1lA0Ry9hQZJxA8_hQU-dv*pH|G1->_0ra
zTwFvM?C_VvQSk-wm+t?_bFi`c+@$byI+85wh=S&XeixszRgn=6cZe3(I6VG{ZsFz9
zNO5&bld?Hg8=IKo8E8;ah)9RF%w<FNtSFkv5*I}~GAEbyhXdIv{9fo^j}IGk$}jM!
z0{~IQqHQqst8RkjWEg=XsL(~6)<2O4!qE{(LdSO_kA+bt2zy+>-}!M(Jh%_sf?eD?
zg{|rUK*WP`DalcgNS6%yZlyiyJ!~Xl-}n89;u*Qx@xsBwW3FhU;1E3<59DHbiI`8a
z^%yvvKW1S6Q^J7F(jSR}Sg+K}Bxt~y=x{W|3O1+dcO18ieWiG=)rs>HYnm)4N&5V>
zYwkCW`(>{@aT`aFrzl|ObZNOGbX6Tr<|TtNnjx%^j^Q&g3}DA74to7;zeHK$bT`~j
zV~7R=EL!3>!;SXPc(wZP=vIzl#W?)}2e{!zdDi;Z!Bp@(cx5o!3L(9rH-L?Qjt4ey
z@j<}N5Q-MS1%po*;2cLIubknoph}=rsmy$>RPP=&or5}12X-L<`y`6%Jkj$)P-NSr
z#1A67c!9vY>8LBJ-GL9e8#<hmw~hpEnG;nO>(sGcak=gcgJZX62OcZ1K1Ix^sjcCu
ze^z_=gj2hkjxXOvpE17NDfV*v@}YmHDA6A*ZVr2TV_GQr#Y8A%K!L6&o>In23}21K
z^gVMpcN+ERlk$gWz3CP{FY^E<wUl!b>f#JP(y=)c05VuH7zwAluHSxVzkNw^Y~<r=
z&&`Z3lZQBd$>}1$Pr&NwfN{f;CxM5)jD((y^j=qO&B~*|8H5JwL|fJn)XIYXz=%WB
z-dY_)MO3b04KjDhdzvS}_Av?s2OiG?O;KBhm{>+9&=y57xY=XOIlQeoxUQaO^+maw
zgGanZj#2pi+lG{Bz9y;Rh11^O#2}2n3A`476*Kf(As}`qSO2$|1!LbGM2GhE5jXZ}
z6thA)M{>|x)SyM0DE{AD`QE>N3RSwE;Q4r|?%TQ}&v_2xd+Y4hAv?E<hn?M}HBjiP
z<~O)i%;zq{kFkQK`9LlBcxz_Hj~s&g@d4X{6Hp0Vi}*|T70or+36qPmu5Rtx0a-JU
zp7B7#V8W_cR)aV*H6RxHbDt6f&5wc&Z~pFm^3NoMb>qB{$`(7@L5H#m@m{iT>rtwV
zuNmU00gFEt7!412=V1$Q1IQP7K#rnT9!Yo&V_q!7mVIeD;n;hpDE?nU!c(`-xSSjR
zI->E$Iv2s;*4`oq{X@Z0gf*)wHg-p?_4g(M8@L<e_XgJAR@ixl28&c~=)jgqr(~|H
zjuc7on?F94{5|$qCr3zYaKE6CkZ<X8ZSs83@dHD_Nj9$|$Q5)d<MV{Iv3}%o)H<-R
zvBG4qsH4*AZ1*Wsj{_#`racZK641`ph)ez@AvRDtdEyb9_CkYiR{WwE;xF2F2>?aS
z3q{yx3|=s3)(HF)+2IBl7b*Ey&s3=ae@WExurV&$%8_})tXt=W&rMA!kS6zzqE^GR
zL)Zwpzvs_hLEGHGX1Rg;)OsL|;qoVE51Se0N*}3)+X4LPQff!&2Y!(k=pA;Oi(q|t
z-5&P}`DykkfDawK!T^mCn!nGGaj&1y9;CuZS@os2e_7uV*Ft>hcJD;<EJD5#>u(;(
z3m-3>FI_78rmrs$Ypmp6Fy-zeYEy>OU44~g$jY7$Uu|1Fj?X0j^Mf{IL=Q}@aYDTE
z6&e?_*BiE!-6}Fs4OH6bg7w?|1t{{te@^JsmJ15BiKVr5SP*VbFLW{JMUT{K-R~b{
zcHnkplR^tjBQE4iV+UjC%2#oyvB}aK1f~ztcml^}@oosoo8Nj`a(TB83LXW{)~~5q
z7&(L=5^X6>Yd-_jAghiXRikG>$5ugv3*J{gv6zfvY2I0%s%|U0$=kyp82H#F`q_qW
zkCqha9PAtdeZW!H>@e)<M^Ic*_LZm7{Yxg44gF&<GjZeLeoNe;8TMOt0zQi+<cdo5
zHvQ~)N+84grphyx3OXlifo-$1>Ap3m>xaJ@?hGWHesYHvojg9G(D#iu^L02avGu~h
z!hNY@xS`G~)+jFJ)DMhWrayjqYrY{@h{Z^7bT{}_G%(_$b4;`S%Q>3i)7IbrnEf`_
zVAwmRJuyo|LJd-Ye`?Cw_n#1nkYDZooTP*mP#1#)Jb~~J{sseYM1BX6er<9C1FqfG
z#HX$KZoa*ixV&l!I7v)g90yEzV`0%yBRFvL!DL)jzX%XqG+VsUaooCz10?D3AS(6h
z=r97wLMKL!5qR&-!pJ-y9On=&CvP$EJj8lFOC})y<-M$l^u_!Z_lD``R?i>EAMA>0
z|E3B1>)hLW+CWW!RRYqW#UeLI&i+|^_*seB@N}+xC4rHKwjJ)oD_kww?14`u<{$LN
zYSnX76<iS6#J%Tlk1u$Y-X!wKbn*E#+^^TR^R9Jgm2Hf{uXEfTBN6mIEgR@LwnU!K
z`S~J0bJO$&8r={niau~rC!P;*n31Z+|3U}#H!+@gA-(QZ3(Pst^T>LFPeSm603ifz
zf>#_T{Nuqa&%(?C7y`%^e}Ptu3Q|QVl|Rf?RRFj{(=O>&F=(OZ0p2_VOzC~=;5)~M
z%yeK5>e&I>>O!Jx$m@vfV|NpBN(6hW<*8m++EKVBb?Us3z%j33wpEK?@bf(8O_DiR
zbKuSQXRmr56?W(^&AoOw=XUU}V!Z*>CicU*JAKX0+y0Yv{kbXr)Fy2`f2~apD<Rl)
z_2NgN)x~OdmyuWHnPBrf*_A6hv&i*|@Fqow-i|g*EXUGL=^hNwMC|0B2Kte+T^L~n
zIA;$yla?kEJXG_xKT8+k_>rwr$>^twzziky6q9`ajg63V;*Q=k?d=pi>If<&HDj2n
zO)iZt0d|)N`*zo3yP`(B&Y8zYvnw-ne1U?zo)C7Avch;M3hX-e7({)A8zCD<ayCsu
zA1yfu3kw@h_W)CMD`15I++x$Pf*dTM^RiZfJDGMi<>bexky_{ZA=Ls<`Z2;TQ#B}|
zYhZ}}`V`sZ0qj}RDs$<r3Ey`I^lYkb%X|F5xn^GRAFDkA4OYQ;OAaeMWdUd_fztKQ
zLKuGZ;s4NVRCapGT!KK*)A>yjX{1*2lGVv%5jj`HSbU7_?92&CCb&{D@a`M+j#_mi
zp3A%fuG#k(QRYNHa;5!%6Vz{FZW_hMuXT}CuAPR7WYxND4GClzlCndJQ+2JEi}Z2P
zFIA9&2d0;HxGImncHnxS!_%wZq{G5U`2CDvo=78TW&(_X=ytX*Fo+z-;ykNXW-~O%
zaoR;JzfqYAa!~_4KJ#8$ygXzFGD@fUWBW03;X?E=g?WHR6^Sm2K|{i}2?mqx7~R|r
z{+x&<)VM({B((~&3BLYn_Z5(e)YK2Pt~A!@zmq1Z`&@s8J4Udy26@JQ_dbsiWR&AX
zV~7A-aauwXE&xd<tHSsE+u%y<>@H$cg8$>{wVl??*DJsqKDyrt%zDHYC*n!*`ATuB
zHg-QQSlqtsk$=XA{!39nuA4-qFwS`4Wn-;+UmSD|=YRV7k+)3jH-ZWF2;uxlcq~0Z
zAzV-knIp&DMb^(a-z{rRxR=HQ<GpUdzmz;Jw4sQsu~YJdn|<60_1=-vri2&T$Y8KJ
z>`?%ti^CeVAuv5iGnv9=BV$>j)Iz}hkT^9^D|W!lYZ|s37iZ+Jw4>*bw&UV60UPHC
z)IDlMD?rS(!)Gfp7xg(f03500V;2JAU||Yh_CN}n`amR@@6-l~HKHWFdH>$4!%#>t
zbIoESpmu(B!V>gS`BqN?&ckIrl_NXz;Xv2ez-3e!)y9-D;;$fEr7h{9y?>tlUptgn
z;B$&Bbg;8Gj@@+Kxv{R@S`?_LRk8RqJ8IP;8Zw+WZG+)Gs4@Z~;4^oWZ|~80)8+9P
zsYZy;?GMSfm9!j4?yqrfJ)c~+k!e5RHITgTL#F<QR1`i%H6=E%p39siW`E+;{>JEa
zXrq@N9wcpiG9`)aRskpDzA#8#T+5^m=-70~${058q)H@4d4{~!Jq!cb7BDh`zOJ?#
zD9++H<}zXGC01)Mz0y@=YHUz!HhqmH3_#utS#TDy23_fv^^t>+B`9(+(742d&4_1<
z8Ij2t>|_N-8Db2BYtwhtE2#AWK?DM}5mgg5i`T?TP6AUd3J8QfWp=S(0OZ+WX=z#R
zM~OuFK-pcuG>S6w3SFOJKEJ#`vJh{m_tYxdUbt|9DMpkFUxKCp&^M#bk9u&sjzC>?
z%wOq0Qu)8tqBXlEf?3+n2jELrQSjWB5B|g!G+MFj23N%eO3yLJ0LsJQ-dl2FnZ^Cg
zuui3YeowOgPPg1LWn<)~>nG&<=~dX37c&xk#|<h3GYJ6A6T4@tD?-L`gj22W!lc4{
z;0t86{myTKisjHxPnV+TqFH$jR#_Oe_*vlK{#P7>&t%D?nd{|P_Tc)RPhK}il5MO&
z=09%ztMV#Lp?hPW+??S+uqu_7zaEgr-FeJvD$G|Kk-`Xkh&(g+!N6y<Gvme&64dJ0
z7@I>Gn*j^S&U}jlfoUVameWAJtpwQ~tO?hFXY(^Hb!H6=@Nz+xGCp!4&ljUAI>Hm4
zUT$2swUpyaWd9*|tA(y>oV+n!s^wjF#ly>}XG~;%tjPKlRjZG-hV1|C8p{jrUgJ%=
zlA%x_paw;sfxpi`ij>1dErE6t90RYCH%H492Mv7@@})j6Hs#ZJPh3|eVl*glr)ks}
z)9;N+2Xsb?X>_WuWc`u%Vy+iP0*0F6mLUnOnS_^a-c)(O-@`%`F8#Ra%;v-`c^Cfb
z{jW}RI9`}=zXt)vPD&mpt^0;?1V|OgB#UNG*CYzNQwAnsyBhnYbpHNawSxe8>v-yy
z?#Cw-rmzJ`2M~Wa15I%H-7I5^A8*1?t+0<xTH1`9S(5t}X>Q$BSh6E?2Kd^NhRFB@
zt;bHEj^l$+-2tpUu#m87G7X*vMrI=zq7Q-L`LUeR3jptsF-uz}dcPRgyhdwiSoq38
zeCRRoMj{?PIq-R6s}O~U&8$>>7^neeUVBry`QN&LRJPqg@Vg3Z1ucY$q;OP($$lX`
zYH}t{QHD95=o?FX>A~V#WlWb}E#;)^D|mizV)=lpt+4KqE*s&H(CA6Tue>nE9p?F;
z!Jkh*_q%7&cm0c}xZBZ}*F*n%Q7zY>$UEZ|Mc>yNq0;~?<5YpDEmF)!J@flSAfCS-
z<}i6M(a3#7bDiU`QzR5!;h|{63AI!+w{hVn0$HImYJzEl?le2l(>JKKs80fin;&*P
zhWHXSAjJ7}fh^Pp*04Uex2GGpS3X;-HW^eC)t%redSxA!H}Zw-XGwtSV(r6Z8Wd;V
z+Wu~U+75!v<T-$IL2}4BahzIh4nq>wDD1kX0WFa0Ts+Zvh2g$yjU@+p&phKNbXHLv
zhlYWt?>%sAgEXLgZ99LB<L|W>E(rTK(EKUDQvc&$0eoUqDj?H*wa?74d?bw*#8<d3
zABgrnDae#y88Vr<qk*aLjQP37SD&<UuIc=0Ds!8^EO`rqVA+u>B6L+Xv7#1#yZ~_5
zHfZPC7GxA2j0x;>q9}?!nW>YT4J~cN3^N02QEm{XGBmIm0r`!|5*KwJpIdSr2WURG
z{M1L+SuV@G<S-`q8bz|eS&)=Dt0Bsc9KQUy52&0u69f8hj{xOeFYKr@R*=|S02@$o
zncuD>n}yW3mql7qzpji*FRBnEtwxC92kQQK<iW%2ImR|^jz;!xRrbpcPLy+aSWb|i
zf4*Rkmpj$y>E82IxI#g(kt(AxouHz|0L27xu8+zvkf>p)fRacOT}t=~{`wa`g55|<
zPDgX;9w;7<UyeiAfINH2N`txZXv*vPkMCkdk30?<NwQ-aj<;!Uuv|HyamD@g17jRA
z`Nz(6W<~=BJfLW(eoy;2lF|Q7j+Nah`&!6S7VE0RAL+S{@6k*A{e`YClD3cll#1~p
z@11W$Jm8U%JB4D<B2nsr2DV-})>><HtliyLfB}m@-53yYn^%8Qe+XFr1~g<45xJnC
z5CqA(j0!yXj@CD3aBEnbhyc|*GT=`GHluvt^mQyhUlQ3{!ffQqIP#1|My0TaXYcA4
zY)1X?Lsj9y*A7p|R~Eisd9%$_sdP(n{@wON?qAN)*S32u>%sAruWjT*2--mO{y#Fe
zzaAbs%R#l>>g!>mNffZ>4!TmQpT_ggr~Xng`sjn1#}`V?(zahRW7nR?ai^+(3%`AC
z$LWeU`0hi^7@^sTnFWynJA5W!n~gIVP+%X{zc*)!D^S}Xl(TQQroA;}cwYkB-GFR)
z@~}Xz>7Xt{6MGeN#4W`RT1;_4Mf<MzuCZXvT)@WmR2s22&D-^D7LuR=whsAOCV6GT
zRZIY@I!#dh?0(XY99~mxthxl2o161~pgCb(oZn%cI9_#6O?j{co^LJnd_U7~)>GXN
z|7oo;b2`<k;V~i=+t=5h+zJLYvX@;Czq`{q+4XSr^cfC(e2Nl9zwKZIq)&=WvEpIi
z*I^78^=CsZ^0EUK@EU<_;TXR*Fg5JJ)#y?m@S6+3&|^f5yhd%X=gw({CxbL;Rr7DE
zBXT;;^1R6zx+-aMf3Z<s5XhW>ZsLH$>16n4woulu_+LwnA|;f?`}be=lx=;-Ph!Qt
zOz@=ju?=M|kOz3dZ}d@M<yqyN7Izj@tY-^uc`x}%zxq?&zRZl*;DN7KJl-VuI8EED
z<3tc+*^E%!>`{Cmw~|coy)YwWVjapXcE-QD5i5I9N7#+oJ!wYzb~mGJ#0NX>i+U$W
z7oUWcRVZDScGdE4!CHKne55c)j`hRCeqkMSU?awgHMocOc5AO1Z9M@G;-o@@$Cm}e
zrYgY?s2zkz)~zqG7}h~X|7}n?bV9Q_O2tl#e)n&2f1p~ogI<Y{bffQ<@9U&jz>l<p
zvh4x)J|>AcSW{se@w2{~FkMnJ=fZ+!N99=FfE!2d@>=}0_PGd79@hF7Bm5N@v#YUZ
zOzjBf;UAf=fcK4f!s#^6&gc2(U$@j`ueN@HRY+}5ux!ME*BQU=!Kav`$2gOymQB|?
z7}YK?^w(^+YH)a7F^Q)-Kc134yfl(9vXw72&LB^I&z+vvDHiWtO$wgdz5c!catfEW
z@p?aFA4Q8*!S*s+wK(No$udil%QRHLt(QDcQWTw@bGK?y7SGabe=Qyh`k?hf_*Y59
zW$R!1e9OPdRIYCfe)G25Tbs```pRHWdLlHm76Y_>m*i9N;Kr$D!V$UtctheMa^dwQ
zd@+1<>wJ_l4WC?uK{@BOhhSN#37uG8Ny)J_(5rn|2H&SPaPQGQFhQK;AU4PBAEHTZ
zJLc~DDHnFDj=p(uLbNC$WzTa#IeedIzk(hPGHoh(tj^~?RAgtt`uV-Y4zAZ*Fzs=d
z`#<iUtMbkPW>&NPNL8S4af+bU@yCPT*-OYQqn5eP>uN`<SJ9_zo3P6VK`Fqpm51l%
zl^0cL&2ax%IhUzI!%J-K%s1(aAJ(sLd#SbI%ahkh6}3rhQmdRj^PsKdwd<^N%}C2g
zLuznBM4^(MPF_dNQO$GUuag&+Eq#?5(%bd@nLZS7Hck<KdAAr@2M0}F;0|&Uay^LE
z8E%A3f}JGa$KXXcs>2vcN=h0A2HvNG6zDO9tFiL%gn<Z82~0P-9#JN|h(yfH&=Uv?
zG_e7p5Vx>!jIec+)w4z>P(AyDTto>L5G8-L`=WQ)NoLq%ew61hx%Tsm`n>H+7Y7QK
zB=*^!+lx6vN`am*raPL3P*&C&^IKYrp8GZdKKg_H^Vi%djHlPTJ#YcZ<*?#(Wj_Z`
z!U<j=`C79b;Ip<V;n3rI=Lc2HzV?XX>VyXp!`B#IojZ|MmZ~@8*aJtN$U1*T9j?@s
zS8LqcW4A(RnCaEIj!G?u!uOcNUed#0PLl9%bKX1P5cbr=>?5_psf0`=Ho40O3R0dr
z-R)FAS5FB<y`&WiTRj4FpB$;?;-Ux$8ROjI;2Br;mfEyxz~C$!^e@e6&+LAtxqGZ!
z`4#cyB=t{1_;d364-Q&X-S~AlS2b2?Suz~GG7An2hy}A?MPfTREo#7S?mWN)?S5=Q
zPn!>d?vbYi7)&S;$5}QGjv$}SrE^Un>Tcv%uLKcfq4^8xT#dAQXo2!ZQM5WX+n8`h
zGCaF=bW^>1rq1ah;|byCKJ0=^cDj0Zvo;F%)~fR^@%^^t$Bu>L)$Mov%;rS=?}1Xb
z-idH=IeecydGI0mZLz)odGW^>{mFzh?LM|#DO(sbJ>z4?zjE$wk%stFLWX8L+8TXa
zl++uep}EfkQBwWaM3K`0*H9HoSFxfenFK+MuH8#c4Ut$5jw8CG&LQJwO*3{zA&kkN
z+B&qmxYZ~0jLr?m&pY2;B)zUTp@^YdFbj$-L@!DfAOG<!M#K91YOxkCXm04Y?K4pA
zs*5nvkGe=F=EMq3s`^ZvX&=1b0?^-I|1`(}=Y10c)3zKCJN9S>mEsN?4{gTkEV-Z{
z_r||pz^IHH6w3j^Rt<x3E1(m}0i8%25ZrFLY=Qf*mPO{q`Vi<^bb)uoee&ez!YT0J
z)V8wX7;cDuJ(2pQehb9>cPh-t!4@VDSb?FjgN8&H(V1F`(~OQ%p}O})R&L)Y_K}|G
zc^c|<>$jaFI{r_NvZZpe2MOnHX5Ml1qeg$KfBUuPNak}JdkU9hV-^8XfpVwF9Ru;O
zbiwuJqs$!>--_D5$7!+$XHMDKoo~vL^xb<|vnn_$P$uHAWjVd&fZ=9t!gC`s3E2j@
z$LaFoteL6xPp8|<Zd{3*4?j9Y*ZYL<1SyFNTb|HU@E-{G2}qoPOMNbq^9%;PIRO$|
z&LKQ};uTNz>B;<D(T#xLpwNRZUA0-4{3je*CY-kT*$_hbgU=J*h<4V~qXEc3TN?in
z?u=P6z*NcB$vvYg$t-!}n0m6X;>FT%NuLcv3N1tEZin<>{%Hr(0#IzQhxOBpe|f_G
z7<^l3Xr%#KREHr<F4`tC<wAMj$;Iun-+>~=b|6w#D^m7DnxWjA?=$SvxBQtr`o$Hs
zikka~m^fd193otVE-X0Z7#z|KKe*bGc=Y;^)&BL@ZxQ@jDa$}O)sRC$jC*^XoeshM
z>L!edG4Cj{bhp#L?f3fal~&j3hwP&h=bBWrtKzQ5ec)<+h|6Ts;%B`Jwdoe5yV7nS
z%8o~On#P?-*w2rbkcuj{V4(s6@ss6P$7`2Jazdlo;*PlS>U(!vvQ;QeJUKTQn?QJM
zq&(-&Q{0`F!dxbUgeI3(r@bz5d7M_zaOl>ZGX#RtqiN4zt4iD@MNbMme_s+bwGshR
z-5f3;X0WwAwN~o|clb&ve0+RZ&L&-Gpl&v_4U4ArN-wEbxeVVHkNoYopqU|^hID{W
z&WqJKeaC>TB7usE>H~6;RRoW>2jJzB;^0CI4D&qjb-i=Z((=})DYV?Hq5X|`<x`7T
zE~llDYik?TFUDWSp4(*4(D9m7{z#@B5=>N9+z@ibY4Q25`C&B{N2z;zFs!Uc{ALM+
z_A$aI_@86H5nB&Svj|@<zIk&R@Xf@zncnuB4@EbZ^!kxVD!p>SYXka3DjCs`{F;-C
z88W^apn}PTS4I7zbkLm|9&~P06h)!olPu%-ESCOPdFvk{>K;mNeqfR(uVN9NCzo%K
z1ARNXeS@U_2Y;!SEkQ+64Hbze#poHzs^*2p>RSb-fg4o|&D7)txbK)gcB1n;IjZEw
zRyEAZiJ+1jJY`8BAauarx-Z-A*JtiXF{wwEBx{-~QOXnZ2Ama2o_8cCmJYygxf-op
zi~ePq;eZK~Q#fh8>({<qUYwn<v<Z~s&SwtBYNfA#12@kWMi%6s*zh41N&16_R~s}M
zx5BLFc`S&~OZ!O~8Wu+Rh88@4FJ7;M7Xx{djnA1VJ?Jx3d%Yn36My^_U-Oc2owv9f
ztgD%Dg>nZCeyF^_E8iAiAgUCxSFt=`k)$Bmm`HF(5#$caNgsGMLM*#UOlr2mgL6W~
zBz+RbJlS}Mr@9KU<u{LKoXK6iD&z?UkuRkD69WBKNT6_8zlpd1pG9-%@=U)PtY&5Z
zb<xJhOcH#{Uc7t#stIv4k74)i%MA_1hV7$V!2*!_30|oxxN|%h8nWtn^}8QSrU^6c
z^UF0->}&1JOy9)vcEef4f5X|>?fUFIk}{v!eQO@A`#k5ru2VC11|6@xO2li<ap#^Q
z!vp@E!snRj?DGgHd>p!)PnrnG5o5(0e%zuY6yc1DN;;eNp6P*C<j)h2L&#c_Iin6M
z1#8{wED$MfP62n7arL$QzC-ZlYUP;R0#LQ(f|(%|r)kZImJlUD1p;p|qqsel*1805
z5x04r_bYi9<!XaF3v)uV-b{>)XJIMFQ^{Lm@Ec<Z83jbHKc(Zq)h7LvKq?|%n(jOI
zqZ`noP{qtQ-(iNAjqKpm<A9u4b#wh$*Bj*lfhj1>^?)&~x(06!=pMOa;L1m}4jjhu
zl&f`5s3KtyfCa}1FW1$1>uYOIEZ>8;H~WJ#V}<gdeJ3@;$!!j)yXr2*th6pGSI>nT
zb+%;7t8&Xj>E6x`mvDd`p55#d<&Wq>{{2@mb^}b&I)QO~E;ue5me%`0mBIOB)p!U#
z)_oWYsojGq<Q1@xOM4R>I!S1lpOHL&icH!6)+69Y@qQ=<H+fDb8J|?Mnj3QH5kmE+
z19*@19a)*Ci@$iX2hyL&HN4DxO(sviXIiXcAG`=Ml>aizUzle<UcajA-4q~hPFA4v
z`R8Mnb9Xw3_WU;blzKU`H@1jvXprVQ$NXV<O@gnwox5g8QMP)YM%L|}@F!U*_lo}a
z0bW86FjjhEc@KJkTFPm-2bk*j5cdFEj8w_t0sg20mcBYN^Z7E=`3p#JL19E1oJxn`
zt?^;_(~%26O6;{q!u7a=+H<(RTJx8NE!t_*GYMS*c!|Exhq>*gL$gY6V-JWAeMpW4
z_Jsy4wY{H|L^si%P5g^c**PCJcb9nS^ZI<}Cq?xdKv~*4F?wQ|P-z6*AoO)f2Cf*T
z&&r#(Z<DXR0H6Lh)II{uB3$+UhcUk9K4;kvre=*gchqU=g_4Gud)*x?UFOVJb^GSq
z@(6P@?{-^Xj!s@G?hmaFO@%D=BW@^-|EH^vg`gtdKoE%MCwcJ1W2nBV25iK1M*!Xk
zTHyqe3|o2%8+ncMI}n3EA}kQ!x2AozNi=c^SJLfavVp?F_Wer-IG187v4fj&2D3Fv
z)sg+VLy2Z?il3T1bXIEQ;!iwQXuY+Q^*m=Be!qifBt$+%>7<XEkLCOyjM#ilRyIYG
zN4TWtT(fq|I4<e98mnUX-uog2xv@_KVpdPnbgN{HDT&DA!)GHZiNgIPH7c?NsBD(}
zbpEorvbq^DAtA-=foNFI6Wci63K*}hp~|@H9u7i$iYSWz1{(O&Qp|B;VC0hplyEKY
zjfXb3H`XsewecqQ6Yzo+tefJ5M3I3+4q$D`Klz*Ls|y^yR<zsy3<No$E~q{@PeuSv
zd<)qD@trzkD9I%x)IPzuPEAj*$}j9cD2?wky6`qKZ(Om)UG$+vTI2;LwGLx_m+Eia
zaTZXKspejK?*w;e&>8xaIWZSK=+FQD=2%%3@Z`ybwRFx>;6;sQB`g8QOAl;q`7YQU
zec|-|6$%X!Ud#MaH|AvXU0^n_2ym4yKzz6Ao-795VnqISc1kivl)t0`cb}K#-v2zW
zL0U{mi*#n!3)5`&Q5KI}AI)=Xj%kvev$Yq~iws1Uc2Zl_xRKb!Kxnr29XqCmWvreH
zeZ`(V_B;1IRrO<R5TkvxKNDdYTZz4wh`<0tuKM=_x#Y8YF$@MvktEN+MbcFoOu7Je
z8?C0>2VY>064OG8+U`=OqGh+vAt}L#4RD9Ka$91vk?c4uei(K2L|>}pNPTt3xkrZW
zYlQshPzDtU>((at)_||iDX<3CI0f)F!8(9{0@e^a6@j$K75r1=2IW$J17aLNN2nPt
zatjL0Pq5WK%)RjGUjCu_0P>a~sA@jJ$Ud}L!6~Eu<?;Pv4;5{=r4?zuq_DZwUlDnF
zjmI-vUgK)j_gJ?{Xe2e|m9jy(0`;{;R760^|9+mqpM>7}4QwY*CKQ>!&;V9!95`b%
zfu~DWdQep0VUyd`z#K?HCccfx&>vzW(+Sw#SyHOC1krB{m<FBTxZ0E?#13v;x;OS8
zZUN__)5^-t%R>Lh6KkXQhphhz%)`gv5|+X0h(fLV0WFfQ{b@NliHqw`s>?1b?=`LZ
z^>ry=Z_>g7ImCh`RszqWh~}NDUScyqr;OcCsF=yvoWJKhN`JpE&StB7Li!W_^$hp;
zemzWED4RcFlOTJWj$R&`fBWPSVtk+mjai9k%R|Q-KtPF}Pn!U~vOIa@)wEnS-=jIF
z4z7$n+gCT11kEdTHZ+?gr(}|TwD>HKe+ofsE9_lzAla3$wm~(r_HBF1=7J*b{dz~r
zdV5Ri7nvkO=jq^R0*L7txXOxvTP=9@#8_lVd7XsR*qy56T>^Htg}Qm<oLn!cuwckc
z1n*Hh^@DfK^=Tv6O?ODjWB!djHA{|KV#Fojz=f)YPK#Ynt7q~3mi^?r4UdpDd61~W
zz0TtHvF&sBE+;g8R~PkuhHIqGx7pHR=}@u1w@uZF@_!@bhM%7xsrX#!HJ=A97s(&g
zzrqLpzmOu6k33#I+u`R_NH|yq##XBI8%35NnxIOt&H>2r>Ff#UP85J4{q(HHm@*%S
zw9O7=Naq@gavc-#b2(WqKx}-@)`+F~8m7oUx%f8JeU)_YUXkRztMrEwPa7|P%gVTJ
zn%B4g$KIYj?rB@jDzv=6J@>fZ5wiE$DQHR*US4H<pR|%vWKe!T0*4~<I@0XHTqoi0
zV^+WOT&YJr`*Y*>v@<%d3oE;LHCBcP84A{q1_bf~j&tr0K5`fPh05g^iP+cL#a6QM
zM8Vo>6}9Y<$!CU;3BR^cah_>|3-|bX-E~6$TZs!`Y_!41f)6;_2@I&2vVdAynLuBO
zL)I(^Z^*Ozj)wWlEYE(+00Y?5U}<o`sMKfE^M`8qvsX~8*nyNM<`Wok=D>qgvwwiq
zybh!<{^_rkv0wm0c)V5A76fptblP>^9=bZ-RjPiX#YrWLjGUZaD_g7%_w+;_7}FOT
z38$n8$X_?c^=k<kCD1I`Y69}udt7PX|2|bbOa#gAtf<#OvU)VYcJj*SaO5IH*YE-`
z%0Us{L@H$cR%BMo+vo+JgS>kAQEx8MUx?m@LPz#+$Azl`bZhM~nX_;IBt6SNNv{px
zA4XF;U~4%Jvjc;HIY-zM&{^{I(A)YS`<kCAev18aI=kXgRR;YhqPD7%d$+lWAI^jx
zB-8WiD6eIiHM2ibxCj3@<g??Xj1{u=38)7&@eAW;9mF7})P&^Bv3@VcS$Rk6^Dh7G
z0<G!yr80D@5V$j&xQ9V-R{Ih~f-bG3M!inMOP%_cLs2)_(i?*{r?G&_S!+U?I|CvH
zovFo@Z@!v<Oz5TVr-#3Piw&7{tO2vfp8);-joY#r(l>44){dN97}*Dw!Fk}=L-Op!
z*4iDQx?hyO>QOJR1Ei?fvk~6l?=ueI6c}P$Zyt*I4b=fRkdxoU@=<nj&NYAU6Y+TK
zG5<7}AXoE{>`&>V3beN`=|A`9R+pEBdXuw-4Fhpq8|6N3M?L8I{wMcY1mXoT4!xAc
zlTE+FV35uQegsBmR8)=~d_Q~Y$B)#$p=6?>)igyQY?gQ^IXRV7GBy(|W^+=0yE8dW
zk9u*5cL-pP2<&6cB*5_hm_u?MNw87I-$ySTWT^U~;rqMVgCd^oc2(X#E2C4sC(P=r
zUCLF>_`O2TwG%9j6{wP>05u*Uh}z5~Rl@~KxxDPKD%KJa#UOiCt2kwE`hfzD6OZoo
zvAd1uQkk$Xbx_Z})`j9mN|9j+wXS|%glFZ+pAj;i%>~r;^sD0;P^EmzRZ~XL9thI7
z9q|O8N_sBl|L|^r6PVkNpRNZ;5Z@rKJ%4d{?)NWCxllInxBlvQ;GBwv#(^K8-_HS+
z+P&6BCYUeZ8t@opYAk?atpmyf&XcDl6)YC&Qba7L70)C;pXjPTq<VHtJEDoEX-Z#a
z<Gu^Rjfh8V*|B5s5bhRxkI#|)Bi!9VN56gugEw!|Su#-Nb_b+F>^!%BiaLIc`qCMi
z$<mdXKP$Y0>8&vfo~#7`L|p{I;uazGP$}mFkzj--S=c%YNQ_!A3I2F8o24dHGmGcp
z(Y9ITS%zAkqpRPZXyZ@xK8jNFxXBs2Q^oVKzjp8Q7n-fUIe9lBnULwha>3x;32iRy
zZ(8&iNcWvOy^>L*6VXK~+rsZ(=BkIyC0EJSr>){Y*sWrEoh-E@3LZnSleP$g4f0h6
zn{VY6*xh9<G-gZe*I^s7I2bv>wdBII3A6irDC-KJ?<WfIdqKf%S|tKy?`=l)rn!U6
zwp{Spd@0c~Ty?e-K&ab^S-b}+5Ix8ikfs<{cny5NE>U9B+2<D04~f|HbGGW+W<U5q
zuk%SXBQE;~k#LU+e(60z#!a7>Qd#%Pv=3m%QqW=d%gmAg<FId`!}em>Ic&2!?_`i*
z^YRoz>)^UPbv{h}E_O|O-|e-d3b>I;=IpPFpH9Kl>2x5%`%YpdeYaA9*{JxyR17qp
z^3l=Ju)$a0ka)O*StWb$0mr(DB_>`;b?39sh4rda=S>GBr}Gk)#;rTb6C6(WAf!b|
z@PWT2O6bSP{d4=W?9jbtWd85HE<pFX^U&N|YILvV=-qJl+7mF{S6a0R^rak-B0w0H
z4N?x?;P%REHbzyh1%IxyKWt>7+?E0&864^hDvKw<;0bSd!WR_NBCuS6uR8ZTc<JZ?
z*&6uf>8pC+M>P(GQJnNm&(eLwy_ANVeJ70OUhwtm4_Y<$2<P6v>Y0#z*yFfFGC|{4
zPCu{H?j(n0u)oRiy;**1m}hbREsSJxVSs{^4PFDfP*w!_M$Aa1{@MfwW!IG%Q)CY^
z+>(}pAs|R=yp%5j^_%PZvV-UEIW`Dzcg%b@-6DUp96YG3t*spg8r;st@UFf53{9>6
zl@9d6R+&I-)`bd%Lw`)J;X4d}0wQ7C*by^=_c!aesu+N=qX1|}tku=ekLVph-*K_@
zp!Q+B`YmN;<tv{fr}7E}S66+&i;RbxJM`(LkC%%{95a5txLFh_nn+Gg{=m-<?|JhO
zW+;g92GbB2Qt0uRn}o*r(2SP4O%?deh0K~7hsXCanT|>dqywU!E0rI+2y-x1<@$XT
z$0+=-)~yBylQUgjz5&1vn0Cor`+9IcdD>0cpX@E|kdn;p)9q6XeD%|w+AP-|vgmo&
zd#q4ud~JJccH0&{!O6>k{~uv*0hQ&tt$|7jh|=BNr4oX4cT0nSQc{9QNefa+H%JR8
zN-7~Kf^-T3Dk-TbEsb}6sB7<a?z#7j!PsN1wfP0V|NZAX=QE$6KyN@A_rG9{6i3h=
zTu0O%;XfpmwbHM-9%LG!+OHJjtI))%xSHqNLO#*5IM%NuLTD;f&aWVtBrFZ#j11Nb
zG+eJw9~*)t&M@qRNQ5KVy1|Gl(CZQ=DjW(oRIxX!!%tKtM3l8eE&y1Jzixmk>@tPa
zxM7HaWyXSZd5vF?lEW-y61Tp7du}&>u4U^yEGF57W)0C$x%`i9tIl1B)YoH{)Xn-_
zLXCl0C&6H>9je5Q-y+Tk(NKYb5%BR-V_wJp_cb$v8w7sO@4@b5xh;YySc|J!)>8Wn
zZi7Ykv_070B*NN60h9sm&!Fqn7y^iM1z1wZr&9WJOJW?Eo53@^2si>VbOA>ffk4>n
zuCjMi@+>a3UH@C07Fk)@(3kNcxQqauJ^?8_)=o!92hLjPRKPDkN{C#zpV|M|#_~Sg
zuR|~m*ukjKHelKw<p-hxrPCOrMo0!2fq<GKvWhKKM~wh*V|hd2U-Qc14XWj=IF*;B
zSXA=XM-uZ|-qMs%YU650j^`*i*x0V67fGM)iMz|mp#xh%#g1z0a;V-@9c?&`PXQBe
zl_XtQe-Mj?BlO&?J{g3NN|1*u5JQ%bebmKGDPSknn?1c%q?)4PYwXsm2EMCO+z;Q(
z0&g6_6O}1xqfC>nXAbZ5BhH>(9EzxCcGsqh-~q&e!>P=XaK49R<z=`h@ZWX8N-yLm
zGpov2Zw3G5%K#K7wXlN=LcGv;<TzEnX$Rnop=)Q}lP|iZUI&aQM93sEX#82nP2pql
zs=@P^S(o1&^WuKSgHeJcBjgK+dmyvtey_(5hZE-+37hNQnYT@O`E1=rVitqHQ`?Q=
z&2didPm!d2iRt*W){B<;yVmoHJM)Wwj&7J5HsNeXS1e!gT*pJOr6C;6_X`;VevMDJ
zeSW^=h}wR(?|YR-FxKVWUZLWPGYpuol=7%wz_Q&zo^;_9^7BAXc8)4!iPQdgjh@rn
zt0;b?{FZZEdjKx%nX;P{RDUr9UW|!5?PcCE=G2fqc$ZhyRCCK*(Yx}3riKnBcYA8m
zqM{G}x+kibtCpBi$Fz6OF%+#6)j+mz;}g=^i05M7ah`K-tm`+w0fQtXVY49^J_ShA
z)h9$<tZFg4SkfkgJB2IFaR;|>n^P0O-K?WkCpy1Z{-2o+ZVlv_!Hx!wd*WqTT%@Z1
z=PSqn-AnP_)cUBArGa_*G|ZmJgw=<!(NGIXM?>3?jf#U)aD9|DplyKYZEGtt)zAsB
zUPizzO8Kn6b5|=i6r$wu@A7+Q+))hSZvmrJZOEB?-14~zwDZOg#^ycLe0kwou~rbm
z$l^}-Of+|%z48fJfWexfC&d2-!I^tOWJtnaHiRNs@uiK{fj0*Bz$5rCE_eHfwSY}4
ztrbl6ZA(<Zbq{>O(HXER)L(o7>A5z*`w@q&6IyUpAQue18e@v0My(<5V;-sjCJWaK
zuCd=bptR$l(sReZ2?!7?{f=R97}sOk>@V&I7B#`kL5-qnOVGv20FS>Ayqz=Dgd`U}
zzN?0Ni*Q#w>h1VB?hiUJahMcYe}|qWfn+FF-O=)<HVJ7kKZeM(`mvP?(+ODnWTwQt
zTE*nO{%FmNw<G&6vXW2m7g;f9)_SJL0XBS@GbkmYPLr=kqo%y7zmC^3r@#IxcI>~J
zWy|&Rw$l#RQl9o>RqtB;<7-8W1rE+MA=KMr9qUReFk*M3T|6DJul`rn9sE3MYJ|pN
zb@R0A@iy~$4(><dKhOCN$3xeX?5+*#Yv_73rCj*VkGkYmsOWtYg}<A=B_>IhKo0Xd
zFExuwVh#*roLvuI!jbz$I74$SFE5W+eH%^e^!}#|8Xq@2K<s)~<=(9wdk6DJ2^l(1
zwKQ{i&8@jKAG6xe!H^T$^`0>Q5!PCN$AV<Q$*n5%sf9<L8($vp;E%BIyBu2|td)NI
zO8QmCo_nf=5tn;0aK;<8R&iVG&IlH&4+|j7jX%#Hoo)yJIh)5sGfSS})a70qZd%F$
zCGiR9L?^%Q1N%&Dqkk76|Ml3sx2fwq0BeRynTm$NCJ28Q>ZuSLf|QLpjPbcaAZQxb
z-p(B_SoQ)CZzNA12V6|k@_gsopy)jTNL<sgQZ(7*3N65MFnPWt#3}D~wA=$wJ3{!}
zZPxN)MR?CS7~brQlyD2Q;OsInGBRJYW0(i~4K+mS-6jKG&n!D4M9K-Q7yMxM3BIlL
zOmDcZ5=@49{0>%%`iYq`3gI*d*M{vCm`ZL=n?Yub9rOWn1-LOu+h5?pFN4ctJ{ZMb
z;U2pr4fPToB^Cx?ldSkQIEQl=z>TU1sQRkIkxadh<z<jQQ0S_dIY8?CLNu%|;}Ft>
znjh~{J*f<C)-vX+CQm1eEvl+N(4H*bKL?PpMY0w)UYbWtp|Ze4i*N~X#L(A3Ha`dg
z4~?%purbM;IrkRZIxVfcaY=D%OrIpyaY<14RCBAXjSj3OIOz_vzhX7Z_zFl?1iHiy
zawyBh(VCGFUn)zgt(yH@%)?~Kr<HgA)W9TFK@Hbh1<InbnWv)AD2~pC>mc9UqI=o|
z=zjm;Ab!hC(VGx!tu{|!eLf6vWwsY8z(m2{O9_l`?+QGf0Tq~A>b*A5`v4-3I0WoD
zm#=A0#_9`}X6n2MidgO-ai#45Z}`;B?!>N}m4_mbzCEX0FjsCnKs$#$iS-rVV(8Jt
zFUUvyRr^#U$T^XaIx(wh#VFd%IyC$~75QX$<UP5Xb%x7Wd&Ee`sOOn?@n0R|%tci0
ze&Z)9BtA!b6+L%PzzYg#r<MKv6*>pbEuiw@E(Jhu!d!>3*AT7DEeDPgMxqc}TEOnF
z_!<&GBKq{_;Vn(%yDqWynq60PYCc)G{;nXxVcDa2el-_RTbWPL?ATV3*M%-ZEK?!e
zzT>?&*4p3-m4;<lbxH_GLZlJayBJq>#+x_FuR^K49tuD=0-?*ZIW`?d_3kS&s>$4O
z%r^<5kYsZ!UBcO=P1u(u-;DdFsEis)V0#evtEsjO51ewUwqaF>b==!ZUX2<hUjwOw
zAMB->L)&%Qasj;752%^jD!gHy-?_$;7y{SKxi7lPw7MT>W-MkOhjH+Z@xfBr>lxk1
z(UjoEMu+B~UAH2`-(7cuQlN{W^XFR2pMlrDAQ{EcTPR%18!>84aTg;b!imb$$x9k3
zyi*?Cx1}B2dt$p{zbctWI`@;}*h^vhoL;FCX}<7Vr(dJ#?oMv#GPtaSdm~%`hWt|@
z98fnRx<*7+@h1c38Y(O9a`YRySzkC^hKou)<6kv(o>mHVC~(<kGM*KK7cLnwupm`%
zIVf<-ocwE0gs`91H&GwZlg~w*S3z4FtZ@CJ@MI#E7F~)y^o(YBz0VFO1_nkDCi0>^
zYLqPED;6zhGOjurY3LE7&Cv8hCL6C1_SKh0k(ZSS!x_b{2eR#*fU$A#E4boXqfHv>
zDS{8t;}fnAzZ4{6-)wQ(#ko@z!svZzuz;8#Y&8vSk{f$sk^vc1mKpf4_DGRqryr0f
z`0rzfje11A@kn?o2l`QyYaOU%aP_nKHyOaLPntWKY?sfxAr(MoUgd$gt%l+$;yvFW
zR16&<ounTH-QOZxTU(2v#gozTPxgYuqR4O<(V-*DlJHYGqK@-SA`62U)*;vgwy`qM
z;79|@Rn{sc3UM(T0oS*#r{GNd*o(=;^TU;jRP>rj;=QI-u#3w>R&+4%J_V??^{^P~
z*YH&1y%k6mDS$P9Mtp{38g^97r*m#NeLU~ouFA;DMv%R~ybGS-IDR|pv$8z}t--$L
zsV;oJ5T4vh$=@T;pv_+shrthf2^jz}h_hr(BC$;tbxlqKO<1P5rFr?$NJ#;P6RnuR
zaMA0%ktqTL@dPK%E)@=q-V5VVUtR#)0ozr^qZPw<L>BwNu*n$MLTSD+zmJ+6bQs)0
zB^gGB>9YL=qC|>4*QTt5LvGM!aPRe^aziC#a^~AzfWgfkY#{YljMPGZo#;M^&o)?9
z`=k~?*Bg3>5P)UG#XP)->{)`E>%H>NPX(r<1(1b=g{ekjoj*@Jm7w&HBMz(xw7{XN
z*G6P|l+<lqPO|v8(c$cT80KF>*?)JQ({f0X(5f^re?sU!!W~RcdKZ#wFH{6)<+l*Y
zgEh?$_O#N0>&;ZE*Lg%`Q(Pp*US+FfewiMkdN-E$(jGPO=2q8BUn1+L1xp&Tu<XPY
zpWi`;!>*~NK!Y5k9NeIACvD)Lz0k14RiPj-j-PrK|4Ev2TH+CWG<NxnZGopYm5zP7
zO3ASW)|+Lqt$x<J`lD~`n#6mcb0n^(qM_kYi;g2Q&^ILq0ty}BjIt_#+#7F`^Sj_@
zoJi$B!~YVjP)iCIyAs0o!~+_AcEZ^tdoZ%j7=ancoLRXV1H7P>S)(VSsBkMeTMQ+k
zo!TpY7si0hbTucVvvVKN<8|byn~eOUJ*ev({D$8q6JlH%Zqw}6PCk#nyw2}a^7VF%
zT`2ll{B;uq7yMMJ_!gNX|M_aMMg`GT9yov*E<j3pPI*^JCjK!TOa!qEE=T1%Su6(z
z%dAn(Q+o6Dew^I)T)3@e(yD)$IZrq7G}Ls#t_9#fxy);Zz5`KxJfRF;Z;%}yw=x%*
z<t&Q%Au}#~0E@s~J6t_7Mt#82dsd>1#<BvNRJfd^_7QicD{Ot02}!{>aVD^hHM)~W
zfGoRAD;L%Ey7A93SUrVeeB*ZH#cd$}1>2|ADyxb~_g*W^BkGAwad+Ta*aPaP2jS_u
zI<uE~d86M&(>QFUU1rnK?VN>8Rd6lV7BS|JFhJT(B0N%u%eAm#daef?kKVi#P`URO
ze2jWsd+q~zi#0IX?m>z+4!X1Po4e!hkL{v}qHeEKN|QRo;fN(8lK$|k;R9R7B@kO3
zuD*Cf(dmX|3Tz~XfF)qx{}vUOJMqnzvB{Pj$?Y6%K^Ql@GOwO<YbxQGpLy!9b$D-#
zK=`_Z%QsQ$TU86MI&Y4BC6d84jvK8t#S*bQPu_k%5RcHp`l_w7bEdthCX6gE6mY=_
zOTc>6`{Ui!thfxzpO4uLdua?OsCSppXfuEhxkNA%-~>yGW(mb8cuOBD(@R`TlX;6f
zbe|17U6|hB!|AT+DVokm)6o^6T68>KD^iE=e;=(_1>b)j2*Z5mFW~5~bd1(UZ^~h1
zAw5JXH*`z}=@$^NK?iGHj(18VjO`d~reNSNBS^jC%6NfJ!gQEBE4Axqchci~e7*HC
z8yco_5w8*A*&`}pmF`z$vuj3EPh*<z5PW8fO-)q$!q`8QMKqONq*Z9anm(9MZLOC5
z@%TY5dJXmr?T}Ll`x$iK2QXdv7IF>Y|I@!B|FBMuLTdGj7{4=HQWBEoV?njzFct%m
zrWsHz<RI@j7kWEgh#~J?L9E6sEt@A|-O6EwcK|b1JF9#LcVIcPL&^fm<`DD{EJ8h<
z0Ki9lLck}HcBh-tCZ4E#a?`Kxa#dhSL2sGK87Vyc%2&8I<NGB=6KUt%Iu0J3o5)me
z_|SlNy|59~2o<8%n&|m1(@pwT%aA2Z5`x=?v%!O7?RtjvrIJ#4*0*Pcv2P(fLmrOV
ztAYav!ND3R+c&ChY{?@M!yS8JVw?1^gD{4pF6UM3ebVE-cCzbQhK8vHB7pW6LR+l$
zb>wIEcU&r=hroDd?&@J=0lg1%-4fo<cXqHvkJxveG{z7=?IB+7O^rTl-Zn4&Oip$P
zkQLCoJ~6|Hondzy723qJC?^Dck_tv5eG{>u!0<Ua5`mMh79UDdx)+wg7ednfZZMC>
zKv(bh;bgk!xA|QEHOoB4-8{;K()Roy7a91u+(NVD=<f2+sTxe8*q%0@tbZz_(^f+|
zn?gu(LCv#y{VFBdWM~=~>Mv;2OG9P6;MI8!0fx-!k%8S@DyM{VvN3wIbR3YsHX_wT
zbbsf8tF!BEOz7}FFQ9vQa%8E2@h-fgNnr*^Xsx9V4P03qYCw_m@)S{QW>Q~K&xU5I
z5T1BiUTaYas$RSdzP4ImyMG6_i0q-JxK^gY_w%q1$#tbmd9rQKFqkW3V$gGEyXO0f
z@{`hPw0-#HKeNH8v`dkP{Lue=t+OGFNT#WZsY`09{6Ss>HCKZ$X!q(0u0W1P8YE_1
zu23G@fVNN(>T~?OCBX!OerjpOw(&;$g1zfDsMRCYTZ1=YK2UYcid?UvBsD8Hb=aBj
zB8nh?FivL@H4jVzRbQwa?sN0{9ngv5w~4f|!OBjyq2==>*i|UuZ$R!!QYI26c`aLo
z@QBK&+L7Y!kIy33kCWVUlYqwtXjVR~Bi_SLVHv&icy@M{FyjSQludbU*przy1UNua
z0Qw4_Ag8MHOT-f6`fCYS3zR6B;G@(lI{;K-BX#zICVXD(S7TR1!#HI=(4F#t`=<kc
zG3JG0nxT@|`)r5OQEO(e4!hVFR=!hwa^SdLJnW*EAG>`=67I<*v@?B}C}h~{c-QnZ
zzF<G8Mx9ne%yn90et^LnK3CF4v_*2l)x7-tr;sgH9eFYFcqR=~z<<Tahce?Pdvkj~
z?tVGCV5tol|7U&}i<x}v?$m`TgPSaeKSqUgsXJIlAV@ib;(#h9EiAB#FYaaaLd)^-
zM{(63jzPo{j3GLh1wgi<>x6q4&z&(nCmaTKMpLEpr>BkW^t55lSz~@~jD{Gu?pwl?
zw&`tA?A{FEk{lT5fPMlQr_Uh&%pooAbV=i`7C1cyP|5Ua?$tFeZ^nzwyGz)yh3i0z
zAkw_upCw-36{7hFyN!i;C<Tx$;P={_CVZJIN$ga)@j5BjJqX|Rwpz~}j@vQIA|Hye
z(YG~VUFx&Ou!s3CT=amk{PwKO0Z7Q1H7e+()YQ%|^rq=r8Bp$nFU1P%PLgkOL1a#O
z9k&bRH<cNTqlUoGZX4n_g0Kzju!(C3%p{uyIHlr7#PoX~u0-&?^6KX-{RI%K9>6OV
zUB28kN?BwRbfWIzki2ZEVfzh{SPJ;=-R@d}1?@1jZp=c_aK{GT`taP|TGDrqA6#Td
zyLoNN#eD=Sv_h^Fv&@&@{iYB9cUrC8>>f02dUbBrT=hJ%u#H{@ujlwo8_NX59?7vV
zTpup49j2H^fMI5QCX?8A(TWIEhSV5)V9LZ_@>rWvcWKyo(Dsg??qc^(@J}R+k$iFA
zJd|2&0?*iY?bf7=56|`AyO*)JxfVVwkt=6H-ouAq2)lgF94_<~$Qg?*nC+tSMe88a
zhqjPYq$6DtcwPkliD?>;-Ixg2<;_USldXeEL(0bG^JaA%*Elb$yd}kbA@fF>;^ACr
zq5tcvD<6kNJXsZtl7S&;2W80|=NUT@()^6D;L}pb6MV`W<rz`1XOdObxzAbJ$Xuaa
zVA%%)clEcK+9(*f&CtUf;kTlnVL=+aWtQ`=pMH4!jeQOOEz!B)`}av|ihxqqD#zAN
z5eJL=KqNDJPT3sm^d(W~;4~`@Hf8wf@A7b9(dBlFv5bclmPsSvecNT9MRlrva}2kW
zr5ECjP&mFF8GSwehfkrp(@)hGMyyx;_J679THL}y`NX!cU-f^WJ${7Cz}A2}SZOZP
zIO@BR)>UR^IUONIu!YfR$Wqc^Z{e(Xh~%-p1k1fH3esESuN+d^Y$8qTJ=DF(9O_c`
z;r0n98@$H_D<s3pcag{A&5&pZy3r3Tb@H1K{Z<GNQdeP(z=N<bqB17}O!ETZxcXEm
zUoo%u;A#AIa!7a&wv!>YeT4asJ0TER4RGnW0xbu}irXD<dN?^gO_9ek6U&1nPzLdU
z?35uUCs%@$&qU6ooy%G;La0A~S3JqEm`b(;Ty}dlj83IDH#e7<fPkT5BI-Oje;`6Y
zQLgmbuqvr&!=8V~j-!aUt;}$8$lVYBbGH1}ZVj5f=F|CowhkGX*63jqSl4n6^+Y4k
zMGln4O5nLnpf`MujMtF}NxqC2ow{V(ulD@PyI-pqs@IC3;UxN%v-PBfh@G`eOtU^_
z`m$ctr9zyXyOx4oLow%N1-mcUPNuo@81TTKT-;K~1}X@juNU@U^?nBR1Nt0M0>^3D
zins97vHnldGL#I|uj7;@4uebbHMYhxB{+@psqCj(I5DY~p(?Q+Wp>)zU1cS=mqc$l
zxIJXbX;3}8HVp2L;O!a&T#i9-_gFBtnLma3Xoza>ej`QV+*cViZ5|6cB#Zp`Dh+-=
z&sjZBfU<@}eH%o{Nw8swMX49e;%E%5lxSuuy>N;b=A6%od#pOUUs0WE#!Iss&YLf*
zfWIz+Ob2zm^7sEU8X&NgF6xo*8W0>(MlM>lVNnV@XSOY>Ff%Zm%7+?$D)E*t|AZAn
zIK~C)$)517GNY;ohWJv{R{}tEK*Fvi<wzQB+nXwl#JY|ncVGkiwC?OszM}h=cfcr~
zAHCBAfh#yTpW7wuIOfZpJ_$)Lo0?8T*qk)5aY<oE@l>~H7@}KPE2@?;%W&=p!MD+~
z@y>Ktts`{ZAD=%^Wt%nnBq51Cqk7_4BJ3+22t^kdv>f)BIBx6XcZABXyamD-OII`a
zDp6jxc?e}mgaMO<kwhPZQ(*a8=bMD%oYCFV*3$Ke`RG#BtgI9aM8L%Lchg}fX8~oc
zb^Z}FbV+Jtg!*{jKoX&zC1!MK=<PK!Zjyp(muU@9yScr(Vz!jAyTx*__Jiv6PwiDx
zr7lN-jRZHVgzI$mmsgB&ssGnA1N1+3E0dQy*cZ5em!c1NPXvp*@#W_z?Zcge6F89t
zAOUOS={t{yKllkp7Ei!v>{{&|8XyRljD}FyQuYvOdli8b!)Z4)OplZCdp*oYxCh-y
zF<_{Rz%v%=yA@x#Ul5FzHP}rgw*Q_ON)&@yt40d%W4Pr=u*Lkdq~113Dlgs0dTRL0
zLB8=C3)zL^a%K$lhDBqP&L<)$pP6n&laqj?3z5_Dn_(#aYnTMx3@|@$vb=0rqd(kU
z$&U<Z@OFeH83!+Kzm0bTiW97kR}wTOa)D$qp=K*eKt;v%+!etEyO@0|_s-zR8a_al
zQrQ6~92W<`#iCf(N{_=0=tF=*GgRR7MhUMsgW?CG4U=VX!S0Gc=!oR1mFi#~WQggb
zz5&100?-S(denCqpA&)6Iv7@t%uln&nv5bfl?XfKypE|Qr=uZZEb|qM)pJZZ1;^}c
zgWUGTQE`C7ya!*l=Weo{xq?vd`Xi%?PFtZ8h5aDuKiW!Zc*$>VWfC@6XAu4hG={XX
zzIc$*vVE39M@Ju5`_dZ$RvSc`gcmWUu$kG}JBfGF20-&L67a$GwfoyTDjsox2Hh{H
z+rQFDhwpQ}UVrmgdCcuXVC;QrlJRXhC<~g6DA$osp_B@}1Vfr=C`}$H9Shh5!Wa5U
zm1db_$}`f#c}n9L$**kVK=JV(>a#=CRCpks#gV0xfmmD9#%NM5{fGV8l4W*b6dvm^
z{U!o?ckZzXKK#LnM3B$PgHYPLFEgH^bmW08&}XnDBpJLgwtjW>aY>&cbf2LQ9;j5}
z(Me>20W@C)!<<AArskZzU1a!MYntAxUiCis7Y=f+JW;_Qz}&s3Ur%^DrkKVg@jR*m
zLV3^MAwN15QT)?quMx5vmeb2CstTvig?7Rd*6+JWTD>GlQsXjRZDIpcTrnhkY%io-
z*MStt?}gw0TxjuyFuk?OV9nu^kjQfrTYg+m9>6$z_T1b#@C?OgyL;#b*>tUQ9_GEg
zA1-uH{`CSV)l_~1hkzQ;dBa*+YDa95+rRGIYqNyy!3dbGRdoHRfHQ>5T%Zb6`MT5>
z23&Ez3K1e~pb7v~uB`L35$;PuD#dLO5f{)Q+9#l7RNV@ze*gwpL*TvnJTfTuenag2
zqQju83eYr{;O@Hq-Cd#K{#_k#63$SeWtmda(*7vdd(v&6jIkMGe3ILnpell7M}4&!
zxD=!P1%~9|{`@izhKn*+TO~MU1$&;|Bf1sh@o95tRC%nr8|Xkhm&(6|Lbq$~gX(`e
zkP8N_cd|#0v-&ApJyxOV=^Tt7$x(5>du}UE=k}xRxM;b{nLtHZ6W~Pnzh_X)PAi9*
z)J7grskvhF{vy=@tZGQ^Nr3r-r6T2@CdO-b4^icDE2v{oo;ETG=w6jy!|43&oIin*
z!VXv_i?#KbU|LATdM3Ra@)r3J`&8ogqI2j!#G{KCwoVF*KUHuhE}9eY+**Q}lH?k<
z*rNH1&hO`-e7u&q;4Rf6jEGu<4@qA9duR0mYaVDKa3ipB_(L~|G)fVDlgP-9f$|Yp
zy1EcBIXMSZLaEH?G5U#f2x|z2sm`z4@IDrUmy^vQ!YF{1(B+Q(8~|tZ88PC~dF2pz
z6EaF}GZYC#nFIYp)`FA=KMW)j+E9?%Sx>74xRPi5m`W=BRLRn0G1irs0E^keBF$`-
z_z6NKq4Zc;u!2LldGXbCRI^LCVj8gWbiPIm1s&SjNxZ^nZ2)=nN}GMZ{^81P{==1H
z3<4zTJ)}Z0vj8$_uH}DkV=lC<KViZ!BHD9LhcZ4PA=K>Ao?-=uF9d%f`~=+*TM2_B
zN?9D?>w>2Dev?XY1v00NMOMESIGCJBb-p2`7~esd!+hc>lK*S9z_7+CSSJ`|r$2M`
zS;WR1B?nNF0_fp9;59xJg2nqDAxdK-|29i?>C5&cXlfbn02`t^?Q+o^Cm@0<P+hfu
zGaADUv7Uf|h197;TSC<W0o=f&@y?;wlE9R-<_jcJniI^$yothK`vNZSjU${>Ydv+x
z^wAQU1g^FInxeLEsVqKz=uvGnCfiXMEq;#y+e>#8%F4jN7(y@wi1FY~$N1O0M2W=K
zQ3HWYMdtvKKTikTA^-z3HDGe!3l;e7t>^Y}8m4AUhuMI#QRqUM?m1lCYQnU~{FbN?
zw9%W1h;-f8XQc0B;u$i}!4^Jc{Bs0ieSkl^)aSt?YhghT#L5Um{FGmCO!~OfU&DBY
zXo8rL1F~1Q2^I$|O`)No1TKWDj)OX#L21A#Hl2$s`8EQ2RH=#z9qmboB})Jf1`9_v
zlLC>O7eE0Jm*^Dl0gz7!0vJ|sMDVA;$lhZEb>irdTyyun^-}Rpqk~atS;t@6tLtCd
zt5(FBX}b-%0We*-6Y=~~hvYLvSsc(}`~$Tb*x1ssZ!+IRpy_vip{j+`p00bR(U)o7
z;VRAX-5(4WvIFAdU#vCuI*omi&Z4%5jkcip>qWtAuU=k=fP8fTW(t%l#03N}Vn1rh
zGenB=?F1>Wo5b>tQ65#qOO-LJrje9-{R^g6k}3e?(A^bE=frPG?Eq7E_~sj@cgu|-
z-Dwkp(tJP+NLkrAc5Fbgp$Bm_f0sW}Ae0LyqpylR2eURDFwP*bX~x%d6T+C%=3`}5
z01O4nK05$*l|)ax^8^RjQZHGb+gY2Q)+3RaWWLQL-!NN6On^yxXWhfAS=#^WhO?zh
zeb503aE|g&4VBKJN+8p<#MB!*TI2`(^Hr>t1N0uC@UtpE1)k&+NSIc3yrt;1HGxnK
z!JK;;KDd?byTF@oI{s-gamR69Pz(4+NtsI+Cm?$uacuCav(tX6g^2eeJfCVpL&Th2
z5(f)F9c0-X%F@OFpNH1VBe3Yz2IOZiaeoi2TXSK*p8CZ%hl~N!W`MD2jvp3X>EKlg
z2(qCnLl`t4v0hb{9`hw-O?lAte%k5UoCjBk@a$ZZ2hhHDJ9k<LFoYqU2Qr`cQHi><
zt3$G+l`a=51Ex0vCLEX-CwgD*jp<J`*^QlAW01Qi{=;);K~0u4Uhps9WY(!TgGT51
zb)?H452$~Na3SV*b}A{;p3H(m!Yrb1nIKDR$oAl(4`GH~ZQ**bxOr{ZdSoD1Lh1E|
zY6+112@)A@BPieXV-y#pngn9Eu`%lwmCJ6g+XpX71y^F|?1kUr2zD;BH%}Sj1ZJ9@
z_}PF-7APmY((|xzK7%qqi1Z%=)#d@O>fC5vy^0FU|8RF@IFyl*ks36>@EsJWVidQL
z00di$)4D&+(2~Iecc4`2L&saA5b7#L=eHfl@pXi4t=8lC;I3uZ#VNu5OY1jnNjs&r
z!lZh?Bl_a0FJ5=gx#uVHx?t)qv0_ZJqZl<xpV<wc%83@0mdZwy+XMbV2FTs@&sUBC
zYPSEu(<H-I2*H&>;&v5HZBWqt{SiA_6{Xnzl=E=w7Z<Z<_?KJAf-&Xu{wjVsut33x
z3x~^IxxcEs6(9-riaMe=q6GLWpnV^&xQ{?7HFf~SBCw&tLvn_VW`HO4TZ!V)NeOsD
zsR5+E$-+rMN%?F~`~XmpV&L=>4j?83fW@VZs}ba5i<A#y$7s7Q!$1LNNzGB`JGl(*
zCU6mcU@Fpl?cXiUuk>4K_(Mm?|3}EjiC+s_R^Z1y^g_kNlvt!nf|{eGV!I&P#Bmwp
zgkTcr&boR(K~#|P5~&>&E^s6gQ`t#`$VJ?kp{#W+3O?JA_a)rP@Xq9&_I2)L-FCbd
zGIj5AHiO?-x4i+Y*{3ZGP?0UqWYQwWD7j=aXvpVH2|GklLA$Nm?x}V@d6Z8@Em^Q2
ziYe2Jw3fL~0GRmI2MmeAJ3Y;W0C7!<hDK3|=8C`#QA4g1ePxwfS^V**ohXtnh@ME$
z8P5{MKM~_;46b<ay6bHGw`Zi%FWz`imj-qD2?F&~A3qSVo>BHa^4s5D(LGqqa0z+*
zMQv&QsYDPL$DIoj*}FB0A$~O8(~a^B2Qx2FqvK~AaIvo(94_vQAyCYTG}Mffp#i(o
zE%twEhP(x|nF3zx7Rs*RpH~Q_s=yJKgucRoVs=N7o7aKjx6ouFYuXSXHc{`*0_F_+
z{7*o+G;U0=4hjWLwrc#;W2}@{5ILz#06W?uz!Ov*-3Z`Bz;ctQ>^_vTqlJ?hPv?Gg
zQnZ!Pk`(3@;62kn7fv8I3pXtF{K5!^9C8S?PvS9;jz(ulC=D+46?&79%viJ-6k3D+
zv!3m^(R7j;h*hdyM+Q%oOR^=5e29cOB28WcjA$0N4^fqEFbO<AE#`QI74g9x38gTp
z00Pg6AB)^kuu6zx-!W76fl^|6IPX~Mzt)|!5GC}NbwKf#b)ZLI4#>(^Fc7R`)*rT#
z=jq`GxlmWB+|m69E}g3wb#&r|+px$>Q6s%L@#utt)HT^((r2u^FLmD@gVSWZj*8>0
zkWTIw6*jZ0oU}5y(O9z$HP{xt_v$gVB!8crKNSVs?Wq!ZVp|pURJ~_6(?_t|&{1m@
ztmyX;{5Rm6>q|0NT`Zq6T$aTqcdoD8-3IHdB2c}smwI$3au_N-O{KgP*dmDzY+Mys
zn4q6>^44)s-Ri?R_K)8r=&9GYZ-i(~P2fsXp^{l;4G-?Kt!!LgVWs#uV$(b#@ghp=
zo6k6Us|3q&R=aqf|KZuJpui%eC4Jwf4Q4U_b+eFX1b#-n;XDMSFJy?Ob%N~I{gBYT
zLY4iKMc8_`F1}*TSXhE#0BMW~FZ1b5ms_GGfu*GtpH|M9jd!&GvB!^3MyxY#0KG~%
z@^<p$tWaviWAN5sPEdgvwNZr?1|lBscD6?#H&C<1W=W<vr<NqgBCrKwmytXgAHOsM
z`}rIoksuVTf=nnzsifS7^c8802u~b>4>Iyv*`gvcA-(uFEt=jSnvemr?xiW~ugM<l
zqR!ExwngVpi6bm2`p)Y4v>`~->gMP;VY)x{>RNKy#NU3z@}GVrs0HW^cAslqWt5-C
zlF`xe-D3g`JjqtbLx?)Wj`Zo*n=Vi$`q<=GFJceWLp!)T5nf|ep8;<2FGBUGYB`E(
z5tOIq`4_%8Xj@S3ILlqYk)5v{RvycHUKxh(7Tb=RAB_~3;>$DVkf&aNZ&)c9nd}_%
z6DtZJD((B-)uYeE=R%^n-U#Y+JVqN6048V|wm$LcYAOA+C+VZS6&svns^}dSk8EXN
zr9$#5+H-KBOQGTL0s{G8LWP>ll2w}^>|$<B5l4x@zvtCj+}K`&r@=75eUA@#fi_#j
zKbw>RVex{ZV7?ioygasZSzFU?e3xxyB8s_8H0=^y)<F%ILZ8FJ#xuI&&*De}8$;)I
zKadY`wc^`h{dfuh{&VGgB%0YxZn*DfYMNP431Z42gp3sZ<03*J0hof>+z{>xR;~?n
z<y>4w#ODZB?kYNP#2$I&R=j9wJr8cv4E`ycxZpa|M6bY5jYw>eg`iNKpDC=nOXQDW
z12h8CfO>0}$=69x%VdGk5?gQsC4^G&76Q`ON6jygCcwfm1!mqG&svYtz|h(eotlW~
zM)d{?&Y7xCwB~zNBzq#;u)<dMTr+k#J#?@D4t2e^HYFZGg_L0ES0J8kg4v@EE9MhY
z;55J;=K)k6c`?!j83g9z)KhZf1fGJ72^tz2i-!q;=1-<yW?Rz(P%c=0goK3P<H(fI
ze0IvHKS9v8_s2OQ`CQZWD&4|v#)@N-J>Q}~IEBXE{}3X`NhgC#86;mpKO>C_*JL5|
zOz%Nsqz=!f`cT$zm9!)q;a~Fuy=VwJ$+n`<8$C3ldmSt5vu|GvcTwzh2>q&Eq$(!m
zF&+e$YV@Q_buSgFX(SF~6}Yb`cMTtk%L+cd8Kfp)Yr!JBzlv>}r1sk59UZJ6WU=iV
zlwog~v8^rquWE(#KieT^D{_$SQywJ-dO`27d@Kw2HgS`Z`=!D-j_1jX_-o80oR%y$
z1}?kA_;28N=@*f`#&zKj-v3XF3}!;TIPoax$T2($4g4H^6M#d;Ec#w5iScuQ-{yhv
zgg~=p@$?>qr1Tf_<@HK4YpOB!ZFoiOLjCoOxE%K+2W@Fow7%Su=B-Ab*p`9j$7cn5
z4+V7!*>A4xqKN(XdJ|Ha2$sQ?aP_bUAfFh+M0(i==PT9cFt*?($%mqZy#&Neb^!ss
zD;jEQ1<<r$;e7SxA;dX5j)T+|0OXX=DZEfbaYU}1<EAVJ({#nI@X+%xhR^Unjz{Mn
zFLz(jQ6}}rtq1%zgOObFc$*0J<XtLa_i=`zV$lvS+4PcI_EFMB$pYaRtn96=6861d
zzDGd*;e5rIUZIU3Sa6!MO4fiQO?j=?Z0ccZ_We(UXCM@?a6sblA&kw=Nj1Pd60q&3
z=Bn?Cjg6&$K3Qejatt~<T^Lu@x>o&Pb9v;*#oni(fHvB1t4~;8V&h{g<b+2XqJK=g
zJ>N+%*Q{&KMovx5ozXs7-9`GWrPzQ0(>a*!;7!7G{khle<;s7HpvvDOxR=rbMhuAA
zoeiy+vine;{9<k6ePr#GR0TT(?p3LqX-!5|Hm!ID)m!QY?(PDJ`Ou}b>31%l?@9X=
z(&e*NL_Dw=taPS5c<FRewniFf2|09y{{QQt<DQ_DSFKT0eMH%=aF;`VhihTD*nOIv
zbWCY_S@C77H^6=URWd5jDd7*IA>;c98j;IE&b%mK>B@7Dr4k>&viCjAu1PksK#0GG
z3bzlTM-HD+Wx;gSnKSUZUrdNv*pf}mwFeMdSzy|}7{4hI{pb-1t16wi&xqrDE4Qk;
z+mhl}+)fz$>edAs!!>_$1|_O8Us%(F3Gv+>t1^!em837U+AOi`XVHBTg|BZ8(a`_c
zsq+Uxw6hi>zRkunegLR2Rz|CSiT)#ms0NnzMKGof!O#;vCKv%iPpaif&uJ1hZt@=M
zbf6slfL4j#bYRU{Er@LNjtG!Eih$FZ+S|&;_izbgGrARSUIa(3$aoG{QSh_}QQhtM
z_A^Mooph%q5CxGzMPQwEX*~p5A<pXaOGgfa3D8nSfl!^054YxggTjMPNC5Nf2%tRZ
z6@j3s!{2@SBBwA`*3^`G<FFY<fs(IYFR1x4VWm+>KNTl>_txA`YB4XZnE@HRWCoaF
zTOYh@u&iv11;>Dr3rX-fQU_a!T;VA%*_n13+6Ha3m{~YFs*!gLRuMaeGR*mXY}O5%
zip#zB&{-mLv$VUsZ*)87p3CD;tyU~-8kzVSRs1+Eq&#F$EKSo?9DPoSa9*4D43wM6
z1z-Fb1DLS?LW{o$i@|z(RJ{!vSY~0-YKw@`#l5^%VYH+CA>zFUk=NSPW~2C;@DoSW
zTmyd1(_B0lkkCkR5fc`RWT3OMy-=s2cWK3#xA_SeiT#faJP3xsI*^#sBb8P{eUMsO
z>*q^i2{C5trN+eXAMhOsVFF~O%v{%o6Djer!j2Sf+TR0UHpcgxZ!;UCKa^e1_W1F6
z<>u3}`1ZCoe9b&1mCYZ~ePv>;1X$!@Qm|DHiO_wwL7sIg0`WojViU{C=_N955Aswy
zgv!9~cd%`<1=m3Mw8N(J?ZTN=IeT3nUs2ZEGuvnfw>IomlzdmNR6O9&DH5YPVU%C3
z(b&(V{1BWct_%ecu_u(0nntF<wUZ#vf9_@EA7C2%LnWEG8-qc<$}@Z&;O>noMO=jb
z2vlN|q!du#T5kB=zRfQVVn9RmU+|IFUqMAh4e7H7HS$dmR^0w|;s?$IeC-__CRvv{
z*woYIps-%-7!|cZK{KDW-ant*W1H<|833dmj0zO<dFKJe{(LqD%%4R7Js?H{5{VZ;
z$WHii8w9BnTgeK7q13goFq`q*`H3otX&ZM|H5b1}j9a6H5hEHk>Or|zxT#VC>mv&d
zr9E5P+je&B@^W%Y@krUz>kQ(urGqAnapUc!cwkD;x+MSl>GhZa2T#mvz2y?c1W}Yn
zjgx*7hrqH?XHZ7oT}k9rS&orks{4JaV&8^i{pVQCWRPana^MCW?~qk|e4IZBQ*!!c
z+h~yqcnm9mrjgCT7KW^QMwyb@@M4V1!#1KTAw<7UI?kEEUybF&wd<MH34Uwqz~EZP
zk9|2^vI{R%FoMn1`!^yW7Z$52?hBRZE7TC`5yl9&myFQUkr*X{$m9%aH`bpD$kj7%
z^;MgL$o-o^E3ZTJ`Yh#2IDY>jC5#ByziGb&TfCMr?B(yFxn)}(BNDb{`EUm?W@Y9H
zl+U;UnQGWk2@jP%R%Qm2@p_Fucgs&N#AX%?2Sx`diNyfs>j5w_7m*EsB7GL0#W}?8
z6n=N?J7?8|d|xYCtVP@-h8c%2p@C~ImUYKr`|j8!y+(JBUCNNpuNi@RY<3qj69w#9
zp!+k}$ah0Z@;_hv+(CdUVKbhG)KHJ77q|!4VTDS>cXnwVIJ{OFCDKpVWk8;_yp$b~
z{305I-H<yE*fNz%lY&4v$$;`*R7+0!+JXVgbuc4V*3t2K`IIA=Ia>+S&~7!ZK;Mck
zJqrai?R^Q)d`l}Mcgs4|IbAEOygFJIOki3v|B&@J|Dk445g5;+Dte8sZ2C)O`x~<g
zG=mW8T7UCH;ael@K7c`w)3kSRhqne~yS;%6G3jYrp(k93QOU<MD{E#femB}){wCit
zMduM`gVq<3e*3FhKyTCfNDY1=jI&_Un}bcnn(7U09v0Ei-r2X>dx)0^e2^Az!j6Ec
zX~L?#3@l?3YI$`nu4dPes7Y3Kr|@ZlogZ^n?`)5yQFrU>b`G3XUxu41bW*=)OcJP%
z%hyAzIi2oTC6qqqA(O%FkCzd3L3|xay+=-uNf{RKakTBmEq(U3=}4vwY)2%|b!R2F
zhJ|TKbw@j3{lo_2yllMw16nvQ)8BMoV!?T-!M@^C+QHZW8snMXyNT_w`FVMrFLA=E
zrZz$n;9?jC_F`sVusa;XUxaq!<qXHb0h;;kORCwDHo48s05*8RqVCHU_6ncnQLn~)
z&c*H-c(Q7^Y=y#UtxA>}`Ock7_L|B~-KTQI`D8F5jHqh1;Nh=J072cW8E*!B#{PS>
zOghtiBMSh~4IdVks5ys`VrhtEdjh@?O6gNzcy4*Ixow%gbxp^JIOtf55J{B31<@Xk
z$%q*o2p1UcY4F?tQW(Y%s%^LGW6@Xi5`%yXaDsI`bJSD_^|%Nv4$|QvE?{NVjNGXe
z0CC^?Tze5Q26;*B#=n0X<cF5`=b*%<t7-Uc+H?a@LmWbp*}d+<2xTZBdeM6O>u3~M
zjc{>%e2J^+=s5WrS&eEqR(h{Ki$x>r^p!YRX8Ks~S=P7t{V}^<F}~Ayc@Sb%p3(ce
z!?Ht%0~HMis%y;*!|X<9!S4eli_3&;*FViv(-B(SEeCVPRFc|LI@}g1m&Gf)A(>R)
zcL|c7U`c%U{S6$lU*%=jN|NWdK`^C&1pOTW#o)KAWPJan;itN6#!bF;7Fmep*N@3h
zmm&t*@IaW$M4G9xpZlFk{h2-|mHM)Fzc6LK8&_tU`-=h<U<%}hzMMk#As4<DpOS(K
z&`DhyReGvBl%V};TGol2dY-77KkQdxXv*YQA~{6a0SFh(Wtnrx2Yry}wEZHYa@m`M
z)8tu0e%0f-)h<LesDsYH;Lqj+_{FShglIU(D?pwIR`%OnY}tb_y)WZq{OX5kopGbo
z7&JEH{>wt@M&4QRjbdZ;vLDY-?z$RrUKfZGy&%i+f0i52;5LYGN4?!-E3=QsfpPZ<
zl#urb>E^jpyz1L(AbsMO3BrU(gxI<0Uy?qb-HvJ~7cBoH=>cpT5Fp~_4HI4_xNzGT
z@qP-`UNcU4vE88m4Vr62J=McHdnh>Nn<BjP!mM6svN<3d$WJSvcarh3@mcJp;54l*
zcl0-Jxek>!GA2ei+jb}l`l|FVKXJi(q!^sLwpS-E->C$Icd&uA*D+B90tsOyhU3?j
zLDm@xQVBDB-x=GUiOORp=ALg;wQK4@WN8Db7vB<J+qn$#1cew{9<0k?<MWB99;gGs
zu^WTAvK;*UgU-e#Y(~@}9jmY$v`#2rMv1yyZWwaKtBP&{?<8ZfS(!9-W@C|kefkD4
z9GCCoNzhMaex&6Cu&k3zacSS>8R#%Trpa_?7)EQYn>PzYoWDsuT~m2)@}b09((CKJ
zu}bwO(!a=ih#>81x?6Nr-=xH=(kkURFdk^Z4MJ=pK>FCn0pN%{G;#DNL!HlT0J*%1
zY?#~@+`u05S0FDWNaW1i<h%mMH=d63=t~w!_3HQY?<D7<NVsg8u*oI;#eiD<S+79T
z&%B;W)z6kXzo<f)*2Umwul@2Hq8(2gTqkO^Ut@av7fL%&ymDo1S<Sdy0TEWQV|XD9
zFv`9AqF<3cFGy0sVwYWDm0#TV*@;V&!K<S?#EIfOp?G+?fq`S(nhd<MpmL;O=ie8A
z@|gX@TmbotV4hy5c#QfZ%!*?}pp~DSr1N5o9&WkXw)(kF_f%Ha%+_-C2pl%r)<n*n
zR<dwxBsM42kSA-IOJX2H35-EQysVglMbU+hy%TJnHx_GHIw!_+_yv{^$*u&aO|oMc
z?WiI1>JOT|v#42!ba7x)!RcS^*-Q`mkEupguzkavZ17G%VAzH%4wtD0y&NeN>2P|G
zuxSWqx7)xdtpdbdqV7|F0TAFAy7%Ff88y%Y%~m7DmTMNEi)G+qwCz;z%K&V*$}3iL
zE&A$DP+bf`!@LYbL|ZT_98}WOjRy_A8q|huNXO1W)!)SIFOk8a%erA-azoe0nXU}&
z#s)e|kT6M;|FI|wGc%U|MLL$C<C||H0hu_rP#$=;<|%BS^@=oTN(AjPv&OP2rhBFB
z{ZxF3ip2d61Rjfd$9tXMN}aj4kds3{dmR)&p{{HiYyB5%)&P5I9k(e331a~eU9cUF
z9Vhee)zhheU#Ree1?H?vyvW9$1p%&kekyRdt@MwC%b06F39;Pmda|%f2fJ9J52BmX
zx+h2b3-Mha6AwxS8Xcc-DDWHvecK$_Eq&oZG3oqmZES@1#DD81uuRzOOAN_EsP#f3
z@}#7gP$9)J2SiCKDrJThj9{uT@dGacDhM*0g}8&TCU#y}9)LzEai48|d6w*>eB`6o
z`Y3ijL_Q?xlg&VuQ}vIJuT-@rE>Zc;BSgl70Z|w+PV(KBx@<^%#Krr*8pQd3y!*J7
z={r{|#}3!(mrHt8YPe4(ju;#BmB&;Vz*mstV!<wQ7Bz_!{g9(T&r<I7?V545xB@gP
zKjl+&wryCJ<Yxrj-txBtCn~ixIG+=-#KmTgJS7YNroP|_n$(P5c^zHtJZU`cBx{f6
zg?XOTMgdB%mWBmz3P-@net{zH)Ro6K^NeQqya)#DYl%F3zbtObnUK27Jf88o#07`<
z!}lhQI{{74PY*B4ja%v4$|9Uu(<jWgg`Y|3(;(|eTu~I>z(5Ve0FaFaMHtYW|Md$|
zS^TzT&xM;1p$`3|LIu*J9>|%c{rX0hk?H}|a$AE7&ZEz+jpWK=GT<ytL%K}P18dH=
z+g;CWA?1rv--ifuisT`iqX3wlIaZ4<(maIa6Gw6RhO$1+A;9A=hWCRl9rq{@u}DWo
z{6(`feH4Zfn@cKuBgac!hXSq>B0gD++_?&dhj%Y@r*k}_l;1a{h=vm4#2-SPNGR*0
ziCB;exQ?9imVY-`MHOMoLse}C3Xh`A2eOir&z}|$foE@aBijXscMphpDiJ>m3A)Tv
zrcdK3AltaH@lrCnG#`L3-tUfZSp)!fo?Mi1|3Io`9o=h*b!sSL`rfIc<P26P5!mk(
z0*ym;oaUjYs;pK@%QKpb+iwd?PSws6mx~Z{lA>1cOSQ37Mi1qnawvS|**k~&5$U_D
zeYHjZ)l&VZy$W0-`s)kU`)6Nz{GcU~Lw6bOG~t>W2s=RB#VS1s!ee5Jpc1@?d#-OE
z2##vdv;Vf<ExeB~L?HA)U3>O2KYuK|h9z7;Nh6%mR8vVQT(0(wfu|rG(sWXnM})7Y
zE{BypdHw?Du$X(4J;6-Ja{K!dSRqkFHqmfZa6OvuON(w*3}NqE6kefGf31HG6)JZG
zLt_!LBT2vV6>%+v%1!E2r_vJ;;cY#*z8uB-WVuhJPT)B_?|1*}^Df_gp6cfZCihdO
zKkWln%@m;=@p7|!j@Tv#W=RI{ifsh|Q(}Rm4On{A_)tH1fsEH9&6x?=0H?!>6PA}l
zR*<y=A^`0I`*)1slL@ibO|iKq`Ie#Mvp)#@N}7i#@yiuqg$BDo>i>_YeFp2wA3yBd
zpKV{Op)g<~&OxXsmxDlb^B!i#K37tMVuP|Q5nfMxDk@bZ8Wc<9_G_^47_6E0=Sm5M
zHMKh6_bHh@S>jiZftIuldp?y_YGoCbJot#{$jOE5n*#Q`49sGI4!?yNL{+dErprN;
zB|e=OlR=jE^8!w%{?V#Dqq?zw)`^(+^`8xw-#xhIHncvd)6cxqBk+^etR58vv87tF
zw3@)E++}teIVGL^?}xnr%NwIWMf<$$?Q}@$(a4vPp@Z$i8WT-~4IHz*3#a56s&ygC
zTWg{Kk?MiZN9?1_>ORS%b+o%{Q<EO3FdpK^vfvXF%NdlK=5DW4*j*#VfB5^VODEiO
zeNSRs_a(jX?iJlDl0;y#BDTSbhXvcp-SFSr$}33bL%)^`<$ngt(>Luu|5~p8r@GXa
zx%-_3v*jPOS1?gQ)hQ_IyH}pp1}FzR$f*}=U!altg8v%<++9lXAfpTO***ESI*I8;
zeC=?3HjG#aRY!(T2SY|CZc2_=k2}z8Xk3e<DX{fWMYkXyi?h8sM~R5bLYe>{qkxlz
zG?r6G#z2%|Lh54;QL|&EeSOC2_Ov;f4fx{R>A#ch>gnORIPt4CiGvHSXNb^A@2t4-
zqZySNkXQJPk_ng;YI)tGqN0cvtVZ~nkA_Dz;N#@@sal;<$Z?sos<k!ZGVY4--Fj1x
z0z2CN3y-?Hi;EaHzZ>j73BhZYOh4h9j`Vi){qe)tZTiu*&2bf1e=;*NxiC3g4DB0q
zJQ!CNUyT&<A$c;o7E_Ui`!|ygy}cIofAK!@t!#3)GwpAF^f5=!oXLs5D`c6q<>K1A
zH$YgSx9~#A_p`L~;!na%9s5@7##Fbtf6vRSc7Vxls-f-(hCzw74qU8pGbHG8b6Ayc
z;&IU4X_k~bernx(j&I#;0j_wxr%`)agS7IT)|<56pfEpluNb@RI#wNYG3<5U8L&_M
zS9Z_eD4%&gi<blAO7pS5d?`G*BB;+_6+C}F$UO))IUvMPG^K>1lnl^r{f8|RFgA`x
zvay`61qTP;1bWJfGVN!fD+X}wan?+FWSBO1T|8vv074d6kDDj9;UXCC6>8iXQJB~h
z0CJdm(YwkO&Bd{Kv`yo-rx(1KJ5o%}>C!wXCliI?GrBua-`~|1^D7a3Db>$g?Os!b
z+=tqCDQ1q>gv+8zaU!J{!-Bgjvh`7^w~UqF<H3`Q<+#nlA)g2N&h2~iy)sG3iPy~!
zXPzt)l55F5p?&^|yDaX*V-v<6v%nh?o@6S>Gs{0m{vAg)U({Z(w832Q-p^GI08n?l
z$>Frd0c~clf$l23g=Y{9!vz7&`ruDeuZs^Egph~$0JR(MqT1n6Aor)Eng@-GQ`dIR
z<zDJaYs+=)%BwTT8L32VP<**MLP(A(fdh~HGR4*Z8Tb=;n)<`~`;pHGi7y&4#NGE^
zjEwuJJChqm8xQNa`rw%8=wyg>?tGP1Q4U@y+c?8uU6ibyc(NPt>oB`v=Yxr}HSn-V
zSs3JiTQS+4lmw(vK#`&{q0DwvH=H|*N6QU|zuTe8zNf>#+(2mcO3D6SR2X{q)%)p6
zqi(qpJAT{Q=oEr3Wz@$V;#~^nkI*ql<I1i(G0ERY-F#+sxY@w_nwR~21ob)24u50p
z+r17f>XPsbqzYRaNGdkFp4~{p;QV{4HQM1x3;bVCx&{VPP>HO2lY75fx+A+97}$rN
zxYjY{_2oi_yn5~F;q*CIP*~WN5p@#3R&$)c#6NskdQC))b;pVQi7NUx3DUDUjc3*f
zUI`Z|qCezp*$UFOec%fBIr8rO>wi+^TkNtc;yT}OD{54`#&<GMVU=K!t4ZyLmfB`{
zzY}~0LxvF~3w?I$;VG>gA0OYwks8PpVFH0$3LpZt)lUiwRRG|yhK*6_CF@+^UgaId
z8#p)1y#(-U+b_{WTpVbAEU-cPsn_TJN%8=)z{@wo7r*e%9l6e94=AYgga;qX==oYz
zHk@?%?|Y{#$x-i=@-VEOcg*vCX!CV&noK#N^Nm`S%Rq%&@25Lc5tJzC+eAtY>5MUt
zV&bAUx}M?X>3CgClWeF%KYK;<iuWFOi$lS{%~cCGlxXh+P5QTxlKu%BIm@5arK$t(
z+@o=VsAdEwF;O2VQE6QOeqhAgS+h;^0UjMRlzUj{n3z341uB65{u3*e{dfH!Q&Lt=
zh{!(Ta0rmhKQ#P$jf<gdCn7Q4lG%08?~*$uJnp-}MJDLy+2L`cu!S=Gua8^h&*Rn$
z(_FMro{zPELx`$Cd<BRrBM>49Rx4(2OZ1+`9|1n61B;n$oGS?9xfuv?2S0%|F9Mz$
z>;ac)A$fba)a>d6td<qD52j($`-CkN>)bgvE(85U99a-S+(okeO35NPvFQ_;aUrov
zCfcvhyxeKc+uQr6v$E=YiFk7}hfY~x11`M|`~IMSWi`3nqn<Ta&qI~rUrnO>*D8*y
zKf~ERe-6%een*or7M$%|HM;-5vptEN?X2s0Jsd3;v)_jhLVa)PL{&I!9qiOMe#lcZ
zHU!;TgwBs|U;2cn6Unao4vkF~#{_=I0Kx=^N}fCI@8fC55Sy6ENhI-^EyKg-qypnr
z2a(YhrT6ImTY`4kJLYJ49@8&fs8ObvC~bMku~B`Xgyig=A3Uu+|2o}JiY71~{-mIa
zpjPvdgHv8W)KTullEbV=F65YM(yL<i06>YJ_nMMC2=&atB!ZA-9t3=lSjAzGR+)3O
zsRe?V_X%P*ClPo(P2`Tv2gaHOa0^NU`jg{ylNpkWn8am3x`0{>^lL*fgP8nkY`L&U
z5w@LCX*NfsD}Yxd`=#>IW*1iT&f)w1s?7toLzY>U9~I~QUY33yZU~+}sIro+Gjlgf
zy`ns?MK1nxR3;EaaMlM<fzeM$kaIrt<;J{@Z{=6!q}HgequD+d{+#fyenzj*%$Z(@
zR<$Lioayj2zm$li*1jX={|M$=dc)3o6W0~rzJ2>%vTAo{%bmp%VV7A$_QA&sSy}ky
zfq|QuwGe*;#-KqMc?@74jud<xQ+1W3i(9+h`b#nL%g1)6MFkH7Sasma`6^C)FlH3E
z8!oDOW|GN*9wiI!ya2^{jQ@^rMg9on9Z!UJ{NW=x+(NO3a!T_S!x1;es|n%LgIpR)
zYHANnKEJv(s-ig<t6ki{?+actZb0_4Cayc4ggg%SAD@lD3FE6i^quY$ptLnfqcNcF
z55Yt$bV?a-rb!RGUZs<&7SL2N&G+oerf2ZO^csWc$Nd!V8!Y<@jJ?0Hd|6+bw{Bj3
zUG9|7nOh#bT2_6gqv2!gOUN3=x_PeGgn%pixoEOc&Rl#vx0*8LyZjl{1_jd3(c*n@
zYF+xhZ;U{nas9N<m`>Xh-druqM_l;k!KP$AZ2{ip*iX*&eM`sVC@A$fpYIHaZ$c{c
zKYfM(zE=EBbY{Jx|EB=$`%{~1dESk~mljvBI7rn9!;YS;RwV6)|4g;blWU=ZhuEN<
zhx`93%1Dj&#v%WHS%^?%Ufkfplnf)LtR~)l*%p>^;IY|*rQ8*25tq(^VgV3*Q3yMm
zzPnW#0NWluKoQ+Pk&=_M!{$B#q|CKRpPc0PpFDgNkU1*0N|+n8+26vd^CVMLnFK^|
z)^i;J-HVfH&8TkbcCo%9O|KR!(qFt?g-*@mh}n((IsIV?Ps_!!YX0xY+4^;koor7b
zCMHf5YHotwIX>2{FyWMzK*ndg)(f~^k6v!B5}G(e%Tf1K3v}XG<OVHv*U2djR5IeM
z{b;T&+Q8}40r(~RYBP`=th?e_{xoYdh;kKR*pGjkH9b<L+y0o6R{ga5UuIa6?rnbd
z&FH+e8F@L)Whm#&K+;o79QgEqv`WV&L@7h}l30(7Q2C!X+v!C&&+(^cThgPrhZaHo
zP+amMZqRd7AQC15S?wl|8xP>u*P1|Qv9Uduv&w(76F*JNyVBvhbuKu|RC;gOM1+Um
z=BnxG4x6Tn!{Ca!MO(ThYM!?4Sn%>8d;YCydwctp4ZrQT%@?bSuN#|xqj_>GTi;IX
zoBY^GICQ%wS_Kybq1zoDdE)e6-R@{bFDgljf+*k2_rcodk>ca|_U26AFAm71A(a!k
z6yOY=MbXU0=E}aXIjBMZ?b|Z=mGA-*v|hIb>>pqs2r{#<v@iKhprBmE`Rw)OQKj>w
z=v5;8tEU~<S=1DouFc`+W|d2w))#zwnsFAYrc-E3*rbD)T|2LLo11vY*p=<G8l!I#
z&Fs@l(p3M?>&k*V=sPB_Xc>W0g3XQ!m4n5@vnUyV4rS_gAamS(WwvNb6O|eGy^_J)
zv=hiY1)x9Oe^)r-4L)IaMznd=ba}P57UR;>Id?zbi2+IQ$p1swd&gt_x9{W8B|=$A
z%E*?zrL3atosnJ03fX&A$SQjlB`ceZY*N_?*;_Vcll?ni(!1{Weg8h+e>|=p4_B`1
z`8uEHIL_ld&g`<4Onm_NB0<oy#3nQM{*dkPmN{wpjc4<*uJ+lN@P@QixyKV75+eN*
z7ost*sD(0=-RjBGd+QLfo@DO%L6jQPld;1$4y?><t^+ojUR=r*9q&5hPeDl&Jsz&g
z@aBxJ>=d3An>X6$g*$@SbT#INuYUISJ%%^XKj-lcI)+|+7IBfq3An~fxAi0<Cz_!F
zvl?p%o0Pa57Bmt!4KPs)Y<c(lxJp+)U;X}BbobD*n6R}26vAFDeh;gDTA$e)=qTrC
zrl#vFs2ttqS^2F%l@;vKtnB1jZ#N7l;f83x;D6iWIq6Z)#a*hVHoJ<h+|XB_KI-G=
zm3y@d8_>W>K9eLAh_M?#S8Rkzm4Cqmn$^<J(!~7<4-GINpov#_GP?6e^?}FMTxPD#
zeHh<xaUt{t$>FKqD6Lvx5T`);TVPOsx}=qV$SxciQ6VHFTTrv?iZn<g@9>;+@K^Y{
zAKiL|?-Daj!jmu0`8F0&r8c~(b2?hrSF2jP?{o*|>~T5}3j7oc<@<VOG)7$|x)>54
zZ%>RVwr7zF>Xe<24Zf-n*oeI@h+8vSmxt@VjZ38GfcA82fja9j^c<e+r*36WJJQ_O
zU=a~8Q7S3wWmWz3IY@aR(q2FWO&8H{5v{q74))u${36=tJ<XZJ*mE4~`Ll8+v5_G<
zKY6+t;W)NzrUp;SiqFnDQiOU#40T?|M}?9O)2f%@$J#Kj)OND|+iwbE@9y_tkP6zW
zZ|jC{-<0YsOL4#NJ%@GiWz^9*YjS7Xl+)TMlEzFCG)Rm>k!|z6s7D%_vYBd71{(po
zm$ROQ1CylTP|@9Ewb8O1D9+x6oyW>q8k{976>6~HDFx;gRsiC7r;4zJTjF?|vs(g?
zT8&G;-@$)x;FYse5tC12;q(O=lYG(Eom8<!3@)~^4UcEL=8#}}QWnp#gHt^q<3LB@
z=N12Ly?BMc5?O)jPg4xw5%y40+^GbkG5oE!I2951p@VqTes@hST?y!7FpBw|UXT3-
zXjwE+mcJZ@BsWlfBqS%da6B;%i7yIYQEIg1BQw{B*biefBK>+=AwtQP-<cY{cFY3P
zHqrVv4Tgl#6P})$lrVIHd`Sr0X<oe`5#4{DuGbdEiM<iZg4Hi6Yoh0SE?u(PPYd08
zM2(JOq5kD!^*fFSil3%`aoN}6q&Ffm<9C{T@a`K(W}gc>cVx@nR`PI%-Lht3>o{FC
zHKNdFv(DUslWm5!?qJi3))vgbCPQ~0nItVS1MW^JY|$!<ylr~Jnrx?5x9?}YKx-Wk
z^%=QO=&(mN`<e{%lW)_9Shp8hxlZcDi=TEL$oqH>uKA}J%$wdY=1m+^E~qA~<1QP1
zOhSm5il2umQ%cWa<}qw~Z{gy+M?4zSmQl+EUsK=hFxb0s+DoKYJoVV<Gqzdi<-DYx
z^StfFrN)GwtlQsKJ)jDEfnoMFoeKCC(I^azzF6<x!@}Se;m_$^VrT0qc>!m>Rm>~5
z=9&!NZaQXByO;N@XgtH-I17ebIh{B|&;W(lHMLVwhi^MdmXHU;ky?}MsT%^m#pc4`
z#B|cc_Plg?WIjxlZ)5#R0%mN<&XlPW3Y>`xIf2iZ2b&`*7)hdE2)TB5`Pzg|LmxnU
zo)pL5-ycLm?4d@mb7=<x2s@Cr!jZng_3PJf7OhoR?RWG*`ogFnxcpe_f^--bUAs6i
z90r@R3kvin3AUDYUdQ>VBtL0*Y)q44Bu<K$?g+*^=SD~F4UQ5>qDrTqwbOWc<ztYk
zE7Q8T1wn&9Q}~PY*hj<tB>)me1-B-~NYKD+-%HZa6N1^c8*<(Pv#o(B%#QnA_)^n+
z`brFp7HsS3?Y{-ZDu}I1MkI68US2Bt+iSC2Fq*^i>tL43YXD#(U637Lg8exjRs%gJ
zT<tQL+6GlSp=7WUL&E=qj+3O`f?%%=$9?Y}I+4GA;L>u;95^(57PGdbG1D$juIS4R
z)gomEXYlOE83@;3d*w<OD)#E6=`ru?C*8ItqgTF&;Q^W%fYf5ZrwyCVwaeBnRsK=6
zWBCS<QDhaAGDy?iu5ij1++C^*i;9{$FCi-Gg)Eor09v(rY?=>9JqLyzqn<Z_29ixZ
zxuYR#!H1P7q3MZ80oUjE`u#e~wd9<aQZxD;WuIXXj!vaOt0CMUQUp0cXIyeL6fQf5
z@$D?P;;9(=Z)EGY9uIy=&!TtlOCf>;fvfXFEEOaORJD4@v63B@R%B*#!OG6$-(e*T
zOjEy~mIozfMu3r4okbi(Yd)g`FESF?%#uL4c|rOF{sO+JD26wUCqqMsqhn*A%r=9y
z)7-`!q5FIhlekp4+8MA{V&ljde8w1Xv>3Jq2uydF+5i4gNchue1cQ{%Ts>z<&YIrF
zWJY$T<~UU4#2o#`_i1guY&tPMScC+UgZwJ>D8DzNN&x_K3g|IG|25KWK$Y#L3gKl&
zl2ju>&;nzAeKWX8L+(H;N&IAAp=mfU1ADUtbBXB)KDciVF@WZyRBi`Ux?4av6Wou}
z()~_awA*Xyx0YCTSYc7#t*19K`arZB7dN%!GZPXYjqP2ZlL|qMf3lF%1TL7j*Ei_^
zamZZXJvQu;&?>$Eq~$`lEc)?)^x^0iS4Md|<lo9=1stmvtVcuq0o$QJU=uoc<^X*8
zj&et<#0w6W$!16JwBJ3bJxaNPD1@%mI~qD{Ty!`*p93tiE2N1DAI9EK(jM?=JW>0Z
z|K$2;pPf2bs@?d>3*Op@rDFI{|NAs?X_wv)Ts7Prd#5e&HCg_W^EZ)GwnI0m(6LMy
ze!13{LxVd-3=o?m!$09SsOrN&ESJ@=AdrZH9{c)EX)=_&hfaMOgR;}|M-tnC1tyg+
z4#AZMngSq`KpCs}`sS%V&>*;#u2>7_8u|05)9ja5j{rbQWr%8sW1-G<Z@HB-DB9!M
z?;x;JBDQ*`C*ur#XMz>WtB>fsi}9y%*lJp3FiH#*Y_T7-#M=&^<Avm2Pycg)#@78-
zPPr3Tw%=fxjxci1#?!y(e+h>A|Bu?H(;6;Kx_SkTwY<2hp5Qzc8>G4q*PjGkdVCei
z>XM0R-G!UPzm8~yxYu)`0N33rn%>`emHdgV!GWuJ^AP{7ja$LrvEH3rqG%6(M2x!6
z3V=&?TH@%x{3<RR{@cY0+{B`*BTDkKBa-L4nPRQT`QWdW9Tli^pG(ZL!yB&gYM#G&
zQD+ng>VO}bgZ8QC1OhPvshLGc;1NRtLVf`LYm}}9kn}0oM{lv6hS9RufQGkb&qDX4
z8M<f5z*4ZBbk2sPTnan_>Vziqm{-`s-Im?T0@LZzitFmxyK!GGIvXnzDS42``jw)<
zv1+e=G{J#nK+j1`6S-mqIE;-SlIrVCY}#{essyYwxKH9ThPj{1B2o70X}FGFP(F(S
zN5OA|dkzhCpAGS<7Q*Bhr>o(P19qe`y4Od5b``r~={*h?v^%c)%D4PVUQaGO4GxdK
z_dV?Ahcl5bOF4bipPxBl3OT(PY%fc4ysC)f%>syQ%fnh~bUu>Nt(qy!V`E58_g4~o
zPIJ_G8jpBUz(!z%lP_CK8o<Vw|MNVfv@DUW@Z~&50H8zmba5VnIdmlZCnxW^T8nx<
z_!|FP3jlVRfJ?=j^*?WunW*FdTc?pa7o-6MUo-A40+IOA>U5hU^V@-@hmjH!b&F7m
z72DtqJ=P8FNb*2pRo#1>2dm0sr%Y?LAq$v_W4{@A41Eee<}Hhwu$HZCc<oaUrtAIu
zhgrZpWC7)^mzQl?Y;lo7af%SB38p%@Kbb_LTAPe_1i`(!dIJ$a02B{z<?0UYyh!&y
zJP%P}^sA?}BM59ao`gMDZ}YPWu7%~6($iDKYh)+f<23*yp-&L>ed?76g}@RwW9z>Q
zt5F7T3)i-0nZ?D8j3U$H=cNo=!*sh+pyd0hcZ3ce*k(xuf**YxD}SFU1j;8WDyp!E
zhzUZ)gbHBpt2l$QsE$z|iSYhU1LuYO`oRFG_nDEh3WkizBaQWxJo4v1U9dXiS2tEX
z3%tW<Dly-8k$px+%|C2cng)f!qEg*IxR-TBU|MEmxYw`Ekf2%WuHeuF$4Gh^42y?<
zR3s3sIfLGWQ4QD=EgRMC^<_1-3)HKAyy3?tne-?;NWf_>Ci9p)0EqRG0)CwsV!a5)
zT6e$BpbN(vDtb+rjGDe7mjPI~Zw;ttvEiI$X#fGsUq<Ca_;bAthpWR`t~N6H>-8d)
z-WSJTBraC1X$Y0(OO?Lo`?$}CWKO|A#Uf~+hs=o&U+rHv9G~M!jhE?w(PC(`ZgA@D
zO63)oMBo48C|Nix`guq%KX@50YGRU}rB|)1S!gn*mXIF?^$F-Qry&MNdSlj;hK6Pm
z;eCeDt+pit(~Wwz{k0#r^|gCb54`{nFEI4Co9}MTmdnf~+o`A8d%^H-*}wn5>}LIZ
zbK{fxd94?{9n03MVJT_Rqc)JfxBc?^FY(cBqsXJ$y2;qZafTOvdCl~!&8v!?43IUs
zk0%vi!1ZS3@P7;@S$t@c#qT%8@&MV1hLOf8$<GUZsix>?&De}>u2&rvLVQn^>NggI
z%j&~Dg{F(v#`ADM@1&J&rq+qiJ1ovIUCw(_v10hRFc{l0GT%$|8g#le@Od|`)x@C-
zQfk=h`eP&M!0)&YHb02oWZXe>Z*lq36WAgcb%Hl%*_4tAl0Hw=FC2B3J2$Qb@Tk_t
z*)P86z(xy&$%@h&x-fJGF~)owBhK=|{y`IPcalNNrT~(>l2&LRmm?y4gy+u9s3o|=
zqGL`jSjAg14y~gG1(7{kM3;xTWKR0}-2&X0VzH-}!v#$C(xIUdK{OeXwAB{g{Vhpk
z1o4>s>}K(LV4}OWk|>DOWN7t~Q2Ue6SU)xF!c<K;3hd9TyFZDM!<lwjr>K5|pmo8~
z@*F7`UP!!Gdlg5EMNq^fD+h>ZZ-MB0ydy?2*ytfitO<zawh=17JfBxC`h|nb%m@zM
zcxjv7n(_RZi?3a8N;W^?s`3?eN7-O4QE+0O#K7uR^O`<BATQ+oZ<nd@a^&mxmFf0o
z;o!U*dmSs9i;RgIub}4mDRFeuMOHJMnc=2Ce}3A6F$8A6knSL8S$4B-!}V!~s6)%c
z^EF%00D*328_8gDJR0e5`H<*_fb3FPq>-MKj7&ISSq&h;FvtfR^|*z+#Ub|`YRWE7
z`XJK3l}Re{-L9mQ`m_1<M-wlu$t4%8`d*=!Iy*W=XWK^kdJ{T+!;s-4R<J@C^L8N*
za!)7S_&rVGPtM*}<-SIH{q(`5I7{bef7HgUL!z23QZN1l3amo|?e0+JKR;=^01VbG
zC`7xyKj|jOi}6|+>OaLlsGpyXg?X58ntM}vUS%Z`*-}wfDf-b@Ssjvu#<d?SX=<Z$
zY4cPTm7*wQ^k?y(4&mnkz=wJE`K{|PZTz=waun4}=|C1;>WY1)Zh)O9>d9Qs852E+
z>&Ltw9>G#)+}P}g!o^RQt#p<KqaqA(fs7CaAjG>Hd$3cX2~-~Fav9bEnUI7bE?R-U
zy=b=)*F~Y(q4E+ftv+?K(|!n!27&Zju~g|h3*ccW7n&%Q*f3|iEe#QD&G<oml)Q_z
zl3AIQT*^H_inj_8Cv0mv1j~(RG-k6kupjQBs4hBS0fa(qH^<3-OfNZivCBFW8?WR!
z*TqSB*-$r$@q%5dzD~LG2(e4MyXVd#cIhXoo5*T>OFax@3CcmZ>=WHuR=iWA=ZMiy
z5St*B;{L+Mm#+;p4kPkC0*%sjD`(qwy`;@IHX8Y&)9V<agsPdzw`b0!okZH%52fxt
z!7`zR%;@G9yGQ>K$iY=aqaVbslrM;!T{N=TG-IWaP8-Rc(rr4Aj+M@OlyYtaWquX*
zC%g49<u=NOp!couvqFmjH8$OfnPPiTi$=!9N(!==r8|79+;{G=hQZY9VV}@#wbgR6
zPaOuL6B0y0-jK*LJ|dnA2?c8RA|u}$GBRgf8~MRIx&KrZX8?ESg1A?FG%gnbt{68&
zlGgNDEHo4g8;5b<9Y`&$<&wxcxv%&0vLKcF52SK$s}eGpjwzY0Ac7n;<`7_xQ&-*z
zg}n)TE0fWBW+R+J)cSY5bF23?VY8CuIMz83m4vT6j4tDO<x*$$<@8OCWO0+1vJ0@)
zAa|1Ft6d^);3){&XX6#$Ujr8{NO=bRna;l>&C=T%q~+cK*4H(6aO!@Gagz_Bp)tyh
z<X6eF<bL{8J|>U?I8Wiezlkor;#2;kOh#E*c`#{z;ceRllr9d;&1Ea^ao8I`=i=>I
zDPS!CP4-4k`Hw3lD+|n=oSZh_g82jL#vTmtK<=j0U)1rfSLF1ph(-AahO!+qf^Tv?
z(c#gQxw$vg*ItDl<7YQT$?^Ll=upco4t%I`@_$u1fuWZr3g(n<U4nS+%Ny=>h|g~#
z`1}rme)Ug*sqg&L;(&ur9pZIU-Cr%&q*o@9c~qN%?M3p!z<>mNMfHnZT*@#SbSWWW
z^D8+<_MvBsxjh=V&M+~Qi#JPg{muBb6_*iT<uqyQq5Aj7vKU5ZwJ))<wz^y+B0$HQ
zK(LM+mNV4<nM7WE8CYX$wa4a3{tTwDvoA!1F1Ou0n{mnk7=R;35ymZEr0y$y=%YKS
z550RO5di=$&>JXu0*i%=0IKbP#qA|4LBw>+kHU^!mqLZE6M+06!o7gn(unnQeg#xL
zNW9i^2FA)R=<FpG-tm$6&(NYHAe$;QqH#UogxL`9-Wx`!Z<xkA3aP-mof2Thg{UY?
z&#^O5SFlw7EiTAKwhha)Cl4vZE0HqX1J%p!@zi;zRe%5@V<~hPeO{IF7G2l_t795*
zTMc#~?;t`yi$4QGM=s+wIFGMm9bVfjSZ>AJF)g<#GR4|{==kapv4lUDA~U;%Z97tP
z9^Hcw(w)SMrYF#Fu$J8Py`^Y>b>;pzK^d&w!qqHjc=fR7x19SbKBvY!7F}$DLPH52
zc|fNGZn5r*p^vE2e6ID0zJwJ_T(B*+L=t++)JS>=j-@QpI1A_}O8vza%)serWRyN|
z9k3{AgcU(_92^I<=unM^fjnQ?_kmQo3rIb_s_u4tu&K2;Y4@&|jq=WFj%^&?UR)m3
zGWtYs7ItqKQgj23L1*5|(C%~={NR1@Q}CHDiu@CsPp^9_m6(;{$6hUrT@|d7^nm%A
zM3htT8WC-4%uNVCJSrbW;l_E}A%%BC;bH6M*AIOjZ=e%V<ET+|xob?ykjN>FD`vZ;
z|1h|7teY6l%YHhl0e?vq;;-Jvbfm-g*L%IZ4lp8ih7w%O%gHViEJlCQl86F(6W>l@
zJ%0_fP>!m=gz1(LUg%RcnBGz3!<UqnCZt2FgFJdD38=QENS65iz0JY84}xYxH+T}r
zz?M58h@*!{iIx;BFgG=hfKVGVG~J5^QejxV*$WNMtEHIhpj>0?r=LL_o>r;VkD_6v
zjx1eM1XNoGP)*9+z=b0UK(+j-NA2LBwovyxT{X_`J86e|Kbs|``hGsDE1CTY!5DAY
z>l&n~<`8c^e&ZP0TX@nqVsBl!JpR~)`N@jK-9d-Yn~4Kc4C<FEVq`sXUfsf;#-v`c
zy{k2b0ZAWHVAe1pNuLqDKR`AAk@U43<~asai|giC+?c$f^BIv#=J7nx!z(?qs|jHR
zUv|3CM`LME&@>fC!B${*+j#_309h3(iN~-n^d*b-JH-{3Bt6*MGUzg&okaVnVLJxe
z<bu%qxgBW3l6cioKBs!$MZkq;wTB)uZz>@6HY@e8V2CVUVW64*(Qfq4eKCh_b424t
zbk*!FcPL4eM`9>~=%6i|wWZH<h|K(*@97q4KWQQFmqm}rx+};-Fv_b;bngmh2pp<6
z{i9ciCFo*JtP9wD%8`Dkcf>6?w2xjnQ31S!{D}~jJ}2ePFPTwzp^ferq_;Xg+EL;h
z&#MA1$6=D=RRSmG#C${OKh6vClCK1c!f8KtrWS3M3?J%UHMu~{W{_w(+QVmxi-xi}
ziU|$w3AlRF)R=V&1jh?QYG4*)GE!YP6voW-pu-(!Whh+<_iXC6F`x<^u#lqU`?YJ=
zR$$0h3VIPkO>c7<?SH)cc3M4u<)ICX4!_aJPs(gz!I3zS3fWh2lzh1QNg$h&pP!Y1
zT}VxD-f~t!3{I>AKRyQ+FhZ**ruz7BV#;W&o?RA44ooVgMc}?JO{;)aP}wNh6oTo%
zR};lU?p?Ycn)_ewIz&~p^s>X;G^XL!RQ21l^99<KcTHRf{g%RR*v}@)e`lz16+iTV
zA<>YqIX5ak{rBzTL%-ngJT#wjK_%s_<AG!4*Eh=Ak;SHG&`?}Qqzta<Pq$=fZ8*B@
zI5miHkjY(OY0?7qT32Ej+5yrqJ+P8>FmkZ34$6b$#g(wS1?h+(Sge-wMkEXZG=pTr
z7}Lhr=*%SVvq($3rjXobjnZ%GwX(cJ+G!J7svE}<2&;$)Jk<BB*~Y&*1XDy0E{6vQ
zU9pVUc1Tau&N}i?DAXBF$_Rgn7V~6(hG(xxpdzjky}$4nzTOP*(b#eAAH6jA%OIYk
zUwUJ@^2vC|sEa+#ep4}r*{M<?B7%t^Ye(ap*Z-Sbn9lH$mh8vsr$w8YTMwCD6q=u7
zGjQQkOtCnB<CwQBJef~7sr_0?IlBmf#lowD+ntOKl6O(ChouqB$3`3<D4R&P4d_9A
zpz|@X@#Tx+ZMcMm_xn1aaG;~3y8)RHTXQsf8~FN6T5H8H0Y)kE6_>U79tDtQ8|&W%
zgdnYCl8NKzL!0YY&4zWioO=XXC0yh*LXa@S<l~1lTg;dtO7^>S#_xmWY`3A&K*eB-
zU2*XJ&h52t)^aCAHfB$B@G)tJSIy9>09+-D6SaUC;xlJ@X=rpkMJlhtr!1j=JamlQ
zMLIe;-JJ=A+nYP2Lw>r8{e)ckjuDJKBuk>AGEfniy^+(yI;REh-JlKcH3BmH<+$>S
zYqq}|8xB5n&p5F)e2^2}|M~HI0B%HuejOU-`jlo@l+?saCiY;O5Vj>K?wgS!i3gO=
zjh04hJM!*`oy5hJ9q9&y*BM%-Lvt{8cN>`pgBG7XbDKt~-E3H)d5=7>$|I4g3qiCf
zJ$}|^v&=UMZ4S~%ulNS}g^4G@ni<ke--D}GH?rIcn2hh@7GdV>E-jmX4rGn%V$OC@
zFSI%acay7zmIEaRIR@`!DmNU-6H&-Y5LmkT>6tE?3{7G2sne$PXiZX=@Z>O@g1<(@
zf5yECC+H=mk0cOYc!zku-zyd3zs{lhBrRQ9(@(t3Mun?I2b@n9u8DV5+@xO%h#04W
zR=;Ik*)~9Xx4@C&a|FUoy%&4w(=R0JpZluMi#Ayw=vuv8P9_}r(#(XP2Rm9xfmg8z
z@4Q6KkE1Y2FGMZ%i_DWspcuwcXx7zaQGtj=?@o&xSTn1rg@uKQOV#9o;B<)vs6i^Q
z<KW=f!XXOS%_zE18h<w2&v+b0M$N`14=gF}jM9COdC%>Dg=I?exRDZ&t34NX*95WK
zu?&L=AA|v*w7N0J^<nx4goL4AA;A@KaKq0j)}eXV(~~$1Ak}m9MI>(SLENl;=3eFv
zP)85MO=|?88hXS#kMp2K%k!x}J{l^E4!OM9DKZd9erKik@ZZHnFuk5$47G4&pAjCU
zx0h~wXMA4LB;V8Xi`XU_=dl3<W_D1KufdwQ1SxF-&;7sPl0$nGRrM~aP7FQy`9qzi
z;Jr$x%C8yP+JQMU$I-NaK#OpAV%9j0=o0ZaUxI!rm%%689w#_Ec~~G2Yc90&Ff8a5
zK)5zPm+q9v;R!Bzv4BYUipvV<6p6VIK0y}uEkYD+D_xIo?*)CA&iU^{@v&HUK+cE@
zwonBz`R=<;ETTv`Ec|G-l@3Ury%n=F)0FoWQ4?|HoIi_{`Zx+~uO^wGcf^e21K(dX
z0sW+ZBcYJ+O}49FMfkIS+aB7D6*9aFUx^@=3TbKbr9v#_oc2*dFihu@6YH_oH}`@b
zSCP)|LMrn6F4jn+pPnu{4iP+b<e)6wRfBadI;vfWCRu|+iVVuD3wTgo@g=xjm$(=Y
z2G}R6@z<Mw9pQ2BuU~m)*FRa=Oj!AysPqR0xY|u&(pOS?b`QPC|Do`m3HwFilcPbV
zX!~FfjCqB~{?FHgcAauL>w{;3U`9)R#V%xlq_=9Ni5d7Dy@L)?>`)ve2=1+QB9zJ7
zMP`vm(*)#C4J2+?dsJpt0L>G@sV?ab{_HCmf5G3ho>(UK^hNiSVYknOcDhQ+J0MsS
zkZ2p;$gkz%XYbPwJJ~J=Su4_f0LbihO;(DZ(Q^MciutYA?|aiONDibw@Z`C10Rw-O
zx8Rt`!d){cZB-N~A3;0*!Xc;nf;<}4AFODG_9g9rJFqhQoEhcQ2n-!E-8^=hh}P*Y
zGlEVd9m)<L65Z1?cK%-108-|d`M|;Q`{<7zf(*{?&jC~C_17Z2!OMHKaw+#xIOKMc
zxls%MH0Noa9X03AdxHc=1Lzs*e3O%tBUVrtm7WByT?0(4YwsAbxWP(4MDHc2i_DY%
ze(yyfA~*2*KT_U0Vr+7{SA(2KBj99{fKaryKW)d3NIu(?YjS_sU3c|WBnt&C_X)nx
zfHA7LO*y}9Yye2A_SbSM!H_nl)&Mq|B35L>e~Qp`$S-AJDiygw3fCcvjt;#z67mT)
z;OqQgi_lbu^DVQLF@y|84royP#tP@|%V8w_!G!YSYm6ica9ypU7p}V8-uz;9Us=sp
zq9^>NXxQZ$VN%?hZ@f|!5)&7%)d-`z-7`F-_x}|;;1U~PK?h7Kx3Ri0)U$98?c3AO
zZRrw%1b1*zH@&?+9u=dqE?!5pQ8^mLH_CKdFvkT4Co6k~&ii3nJ;0peQs~yOXRRcI
zii!%MU?E)#t<n<aCXxbR<3)hBjXa<q432}=ekJ8|npTI?4qcqB^)CmR`STSbc(S!L
zSF2_fJrUpLdHS4pF@#qLvljs0(9-Yjs4H;#ZY}rl+@{@e({H=H{1n?Km(IUQ6cU=J
z0T#)#tCdL3h5?;d5?_qqM|*#L$0fDV@kv_JE9TK;mnjy%vt@LZrvUpmUqa_3EkUl?
zC!U3PNOav-llaip&O+0t>8;HDKd$lc)y*>+tMjJL-@l3QQfd-RS6VINSu*i5`I<sL
zK&@|o6lcRJtRab;{lIy@6YH^$kAGcgtdu$6l{Q26J<hxybnWkfVf*5>CCv93K?S7@
zupHSymc|m$VYwjN(7C?TCR8*rd|K6RHeP<61iaPWv=@O1Gkm{n{z@t@FJ}xw!tlv<
zYfi$%m?Rhct|q^x=TvfZUV&b!C7*MF2D{pMC!QFACNF1xfJ9C05ZX{Hrsw-}QT8xC
z*Xu;qayv>djKucbnc>a2s;M5~jrVnIRjZQT%flAR$-4+*p)Uix%=-f0QNjNN$ir~)
zGF;TmEKcz17QevuGvP#?5j%GQ=dI=KD>oEMurvUQI^@wO*Pn+p?TR(-01rAu!Dy%o
zyLg8|exX5zy>ddc&1wv;jxy|f1)f2ZQZ|6+1+eWwiSUNTmNmF=fP4Hd=cIdvYdh4-
zw>C8|&A9Xh-Y!Alo<5U;le_@dCLBit*SjSD+cp-NND8{&f!m2sU%FoW{)Z1SxE<iV
zTJeF~dE0vK8+hCD!>X~o4RiUA+ev45Ff~r>+zz~;pR?lYSC$JV%LmlMJT|+}DolUV
z<&)q=?b46V#>mRZtZM1;z_ETptfY6fZsso|L!x08UEX%Ir{Ormjy+C>BtdB3qj>4{
zfbOa`MSlG&#o50}Zg@A2NN%QxDj*ZgUDyQkGcid6%rvXHS@<PIhGYwP@3z?+HOKJ~
zIGe@wj%dYd^wS-MA*y<L^TgmR`L1y;3sIU&XUSqRg?u*f!BkvZxRve!lxnP*hotBe
zbRnaH67F`|K_3`05>2u+w-f<(cw^MIxyr?NS0Fd^=5yjt$w*v_D=S2vFw%6>JGU)z
zT~qu|NE4k{wDq5rB9nhno^-YT5x;8a@Q~T8@aYp?qLxUbRL+MBwQy;4cB>|VN&H`S
z;^l>negi%Ey9%Y$GFO+X$>0vX!uKs*hn#csd#AsO1J?w{<b`0Y_-UU6F8A{Q)(HPt
zGowVKJNYOqh-%(o?7a^{Los4)vfTpt+R**Io#|3{prmyG&?;CGZ~th8dIMT`1xB0B
ztzwL~1tss{-lrz66XSt0CzuX$^iOekH(yu5H{AZT29`TT5>vDm!lLVXJzVsYNT;Z#
zgs*uA$%_egaJBow_lpwXH`k{ko<Jz;M^9jD>e<E-GdCw9{|8I)qCl^2yKK1B^1^2O
zy{b@|<NRvq=dOv_tLE>3rQ{Wg;LXW>`Yc#8%y1&2Bkk>n9Mr#WPd7pO*O!4j72k-5
zj!VnPg(gmm(4QoEItPO)US2w=!YVd|J0?_cL>EsbcyO6axyVdaz*Cr;>Kf3+4PnF+
zq24^RM%Z|uCXu%0f-EmVI!FL&b@b$xPQ|S`z?$RneDb%dzo}oVcQ8>;Dux-PbE<O+
zr`$VVEh9}NC)@W=zCsGA-V@yoDdaA)Z1WJXCsMhEH`NcLO2(2ZW9ZCXZz|`!dL(mW
z(haHLF5$@01E}D|iV}RmZMNt0iuDW7kAFKDQEu<vY9hlY-2Qc??`T`MTQXlKINpku
zuqx%tZH=JO$AW6+{N3~CQ0Tt+OKWcWPp)xc{eJInyK7-nq!sNKTF-wR$@$5+o?+9C
zy<9Dbx18ylL=6wRre5?7YRpcu^FNLQ5>t+8j+IWuVLdKRD(v=7tFujL^Evd@>atl%
zBD4I-=DM?)0bU7*Vx|dp3r&EmLI^{G6LrautWNRlv$tF>`?@-JXB-t+OVb{$6B^7s
zH1(D*b#6otm&6i=DW;tdH;kY{d4=RQhRgj#apJ=Ju_rpNa3SU0hquRaLlc72@36wD
zAg~{t$~uOiPDJiHiElRv!}UL3EE{z?;Yl<?o5hJ<R~Y|N;wo2KZJ?TcJ&h8D%~Ip~
z0lhPy20e-hIg(PZlYsvJr*9!gF-Z(lDV0=biM5qF#zw_ai?lvFTaWSGQ(It@Kq<hh
z2;;Z$B>XlWLP<%<5C9Syz)A)`K@h$VJQL%=rh;6^ac&|Ds&(Cf_Spm!&k1Ny*o@Yo
z0Ut(eV=h%s-9yd*5QOn}N&C5E#`SZ9Srxc}Ink9d_7{*=PcXws>MPrp0ff(khN>WZ
zq97YJgG8nDedoC%;_F7ALsH*=RMnLWrjLaB4chm(v{1jjY}99ja0)D!gwGd>_TPUA
zq*D2Na^e}fCq*So9bm!ZWQ`Q_89HNIJ7zK^8<AZ<d^ZS|0wGExlsYVp>ouLL-gFOX
z|I2dh&M+#|D)tK_rt>EC8>5YbyUnI!!F>6$0KXy-2|Y-@Y?rYBgUB7b$TUR!bpPef
zz#<H2c)_TXjH04qg7@bbPBBm<nJ%sbB1^&DkJz(c--K&9cSrW<y2|Og%%-d*0fkHE
z{{8#itx=j!9d*6ht$;4%aYbqMT~4N^b}C>kA0HlKEy2vAi}BHx)q4)4FlA$X{GMSk
ziK4EpT8{QCzaMBjD^Im0E|zSVAP@VC%4nt4>Bi}Kc@wV3%WXU%-JxS~Yd08%Pd?C{
zc|lZ^MN{Ma%6=+L61)?|Z|~%q{ue#fD-OLJPcJNbcqCiu?$GWyqq6u!HJ%PbNl8YG
z9&a&~J|lAT%&wfnMK8m4auNPw|G@6QpehV7v!P2$I{TH=ZIWZjDX9$#s}uzFH!N~o
z`IHxeFwnA(%uKT_FL4_f+kn%5azIBK^N<2IA%ek32Ldu<DIi^@z~~znto1@Dr14T_
zW*LL_Jvs=mR~qT}_;CXWk)(jQ6{JJbuq)wfhN=wGEvw1~{r(j|Vy4rLeb6Uw%mZK0
zusLBBD47{ffWz2T-eOddg-r2FFZzDfownX~t}Ccrb|Azb#CK2wHbsM<7{k*3?JiHL
z$<N<D>^+nF)RkOa0rnOdUmH{f(P+h&s|;pHHHSgGWcMQ-?(4sjMnzOH<vT`T07QMr
zd{g!ig?fVPo}6F@T=yR9F9PUQ*ia6ubI#G|>!2yLVRi>UZ}`s(3w&d}my)5PrUR=)
zCz*9r_da8Psq@!qHri}b$H`+Thikxo<jVB!)r;B#rhr_JR05~P%1uMi0t<_YiJ6aS
zj<`gF(6AxqN$cys%=`@NQBojTg2ibBKxu=4ZCa4El2kaR1RdI*VxYoF!m2Cnc_Mn*
zkZ4^l-Bh-oV>tOoOiB$ZO~&4wrjI1y+dL8UJ&{Zr#}|xD(JXu4uB{P1KY>1&mxjir
z1D}i@ebL=kQdlodPOX2rjW4A8)O%k4K=}$1I5$^jGwP7|W79rI(=9(0YC3OhX?ot7
zsctYi`H$ZMtiF1o{322eXEp+901KCqnq5R!;R=t#0UBO;E9P>X3Y5cB&-rf#d!0v5
zNWl9#6VCVVc4J<kXk>ahZ9(&cI^X!XI2MsM*IQ&HMWy?FdHA@vXpu+O$fS_2*$M<U
z*4zC>88xuu$#CxL8(p6sD2{J~*ir<WPQR)rP=jnhX)_Qyi;+ew;x}(fCkngMLf)m8
zr9ov?xgKRG7|X1eg^V0_!19owxu~u3Wrz^>AVQ4KI$BwjPWv{Mt;wZ*A{q4+`qoB3
z!&7{We}UB1;~eA&P7Ft8M=Q+vv78)RDbvMAF;*z~%>yM>Dty)c$kpqS`lj7{=+8;|
z?`50>@YjxT$!<Spne?rE-D|DC+<gUMZ^iucpgsLu+LUBxRlRKKTJIZ%6X66M^Z}2{
z%}6_Q$`ByyFlPUbn19)xq`XvQ0+DB=%5=`T^RH!^u~+UsEAyqGc?7882ugBdG(P%{
z84QwgsMLzeUgQAvNm2ouWCa??BHkdp9}v*J-Q$7i8-SL9G)MwxwsgKfqAe{UqpJoA
zbPZm!S3g+a*t_tp;2X^DRlp3N?#A72f(Iub)Fps$B?_Ci=9DYOT>}Pa$HvsSj*x}5
z^_YeSJM6wam27{U?KEF6RfsUzXPn^6%t4K^a3*o4#pjrz(o^z^w`;j<9z^<&AzjR@
z@g+AleDCKeEuB#%MHYa8*$|U(D%p9jg>t^(ihL9S$0G;~tsF8z1pmD0&$-mJpf~d;
zhFuE8J0}cuqR_3nIGM=;Dql+G<B;brA_wv!e9TOU%OO6)VB5}6Ms*;~8U@T=@IWbm
zMTN!1&CRXEnde9;G<*S4iBhpAkeuOi&h#hbe@R!Ofy)>i!K}{qS`wJj)JS>zT*Px&
zHLB<O^7uz17zsueR#s%h6bKH7vh1Z=Mt?K5iF-q?IAr$G$n4;g@>gf76L`Ql8TGBK
zFl3adxeE7j^|QYwRQ%d#lWx2SzZ^1nyR0L7aw@Coc&FJ-K|1Y&jF_tEd97Yw51xfu
zXn}QfmW+#4ZX|weLFlUZpDy62lzwY&+T^;z{+SoHmn9|JkC7hSPv&`}($pE!_tfhZ
zG{#J3X74=sSJ8X;PkIu>pwe`ARQH8FKb{F49L{*SyEs;sdXD)K9AFzdt<BP5qmv_F
z6clrfV7f8~3pT;CgFagXG}GTTAzLhG2TF3YbSwK+qw>ELnXBI}vorl<wmkw&SA@1_
z2p-OI^0qn%rqjSJMUu69tbBVm8LLb|p|+b1Q9=nsw|`#kRsQ(M@YBorl<To~_buZ#
zSm?pX5V}IUw%3IX9)>hAC<ouxl!LXgcfVM&%`D~T-|suoQFhJg$AwE>I&e@!X<leC
z+VJ9N9J&tYLHJXonfIsRGeNa96jeQ${n68ZxU@b1M&L6msp$&tH45I8=0u+&01OHl
zzyI@we?U~M7zO>lwlqWN70Gfh!8{<BrxW*^GZc5@IJseJr9sQ<a5w`p9$B^}*reNV
z98YjGAPHue3P4^P36hyC2G_Ri-@>BCMwrRr4$$_D<f3aA>-JOj1Y_a?k$WKL>pdr7
z+})WsX(Gui8MFz6UeJxCq@;o$cl{>?nol<yn?E(DkP-E2NUl1yMSQAtvbB$je<2pw
z%0<?wcX!$@+twmUgrm2U#3AMmOxO{}4eMufkhX1vq>j&yOH;<Dlu=?_W?!SBm+Mo8
z)6$Z^dJQ`KzL2-ka0}aZKajNZ`BT<-NqDEk%QvmD<GEy|Irsp*%J6O?;~PhdlSSo8
z?kH$T3MJ9E*g^XBUk0Wx=AAro%h_x^ccpEMIW=Jab2qWk7qOmLvliU5NC)v^s2IRt
zz*WXTfoRD^NDb_7tN=5)8Rj|`HQOxi%zSypW!gboB6sJ`ofXjE_7B6NE~)n?V(2Jg
zprdPqTg(nkp_c5p*=7(mU^WnGSo_NLa(}Zss@y6)%P4xy=q_A1?&S$iJ25QqUR0^x
z*8mnq!x-M13USN-A&$1WVNrt2=VzWUavSjPkHdbTn|_J^#cf467#w+^q<VS1LcNo!
zSM03S|Ko=%@PE2}w|2Igy~A#np_O9iukSnrx(^PK<8uz;^H?T~&^A1kKl?A<;i2EE
zQ3H;S4L#%j^#jz$swB)4XF@U9d84t70=CMS*P&3;hfBxN`<R0a?EP5{LNSD`((xHb
zt>o2=Wj_iR$+PUb3u>FNu{9bLY=HA>O1E+A;aXy`4q<Mp8N~)`Yjg7#30sAvS6lis
zU(P`6Hx4kp8*<_-#J1xKr^UpM!35tp`5`n8<J&9Q)6astn0=cTYUMt5`b6%}1oC0C
z)i~d&{oYH2b_0&$>2Zi*qL2+uhrul9+-WZ;DgSZ9OV>u<cGVkxUeTc3P)sxH3%fD7
zUZte``fkh>@-Iy2SD?9|#Nbr+Z#*1%e_o-evQ38IRzEvZb*p@GR*w(hTB@Cw6>vBK
zh(j!#MMI%JBpxiZFnM{RbotW-9fDvAAx&UgPXf+F1&tQX;`^6C9^6(H=0;=4gj{Zc
zs^=op+vfL1)2FQW(h#3a$ctC1TSUl};z@JqJlRX<5jzuI#4r{IIc*FO!{D6qR)W-M
z5q;6aR<cUN@%<(_PF_7SBBc7L<%2bb$wM6;{Lu=<<A!B<g^(UaCFnkckb4{pJPaDW
zH`O0E;Y7!A*dUvWQMd2Py7kIVDRxa;?2RVdtyW8u3urokBc$V8eeu6MjE1)UY-%{y
zUh|{WU@peICvF@SrH~G{6moJrK;h$4g&voH$qfx<ebA@u$kh{3zfbA5<Sn@Tfhfmb
z0SAVh!Y)|8tIh6*B|l8C1O#a0+tL8$Rd5+UyFcWryQ(M?jTPm1f^W=F#nyS%vEh{K
zRH7L57j^KjYwKx_D42y2t>E73ef{k&?U=N5?fIoUxpTx~MjNBxIFh6(<B~I^A)05P
zWZo7NK(ffEBKHzTjl#XsRvQZod0yPSS@M?;fGKoMFx1>GdT+~nxdPElJ|yA%Q2jCa
z`b`W<o&@F)cn%EuQ2amF>dz}&x(B=hstIoEBm%a}8LvJ~&g+*IWzx2d=+z@)3dhh-
z5CW$3ivZzufYNoIBqkN{sK~4Z5j|;8h1HoKW*PL@`Cec%DXrcG1k)y@g|1mr_kO_&
zulYUYGjj^FUtTmt>V+}_;!h!Y-)b8b!3;A5p6%^UfOpQw>QD9fO*F6fYs>DM<w}~8
zDy8ZlH)4nGCOgu%WMPE54TB*v1YE8n`e3p_`m8?3pCjl)MSQrs=w`22HEhQ;aQ3@;
zW+Yx@W7`d>Qsv$8n2qhV^hn5Uio#MU|4I=5IfHU6x60+@-NaQ}%eNvAj|Vnjf}2=L
z*$Nt5ux<SLof&GPozFs{H2S9AAlcUdtLEkdmVbMM?#kP}Kx!I%P<JXP1C)8fA3sK&
zR4l;l9nHT}PfEH|E389J{y@)ScDSm5&As!@76t?lgl-JIjJgEqWkB&*h4bNR7=lNK
zJ)rxz`NDnhwaZ6Hdf&ZqDtp;cfA39B7+eIV+Kwy>)MfU<)>}l3P@mC`&OTQzwSJ0u
z6b^o>i%N}q&UVlp@R{Xz=TvTJf~H^a$0TYpJq%@)P9&TfevGPiAL;^LjvBV@K9+by
zL}_aJ$Uy2fR1hurQJU5h20&!`uOs3pujzv5a`!#E%|?gnTfZ>jePrTub+ad5oCW0#
z4K;vD>TfgLL*2iK^lN8fn1#6mW-}XWA5|H(OUub2<JqN7%So9pU~=q<&2m5&r5}?1
znhww{n`KPB8tzZ3z;#YenA+=jWsgMHE4EU!+ladWFf#zEOOlwab*e3!J^zRNqjd?n
z8vzkf1V(PXk1|}4Fp?(hu5}H3CvyumC-|073<x6)Z|iywpKulvERm#x)mz18P9FZu
z92|;=V-7I#u-8Em9HVgYW9)}JD``7M>$z$6lx{CFK2EL=D=BA)=OL^xV6ei;6#1A^
z{`a|v*2tsF@I4Q4j3O(LZ9uAv+UC|2?KgL${QX7o^TLn$lP&NJd<gIgM#(1#UIJ-Q
z>p&+Ewy~Lh;aC>wtf@2y0*vofEkH9Fsi?}kv)?`F+Y0=&mwG>%$e4&1FZW-blpLus
zKWh{E3nMk=VXsEC78POJ1E3L@PKy4ipLy!R0k2l}5kGSG(2q1Begs!`H3}&@S5#!Z
zz>mBIKhlKw5edo7UzbQj0g|fW)=Q0K6}AO;vnL{MAx#>iy^XhJek4kjk1M8R#}%o0
zBscRpj-d(>e{!~I4W?rMT*0GP`k3CCQg2p1WL7uy$*8ju{}}qtEOoFmr@=)3@LCNU
zM47y1St15>zB5mrJPCm*^HOT7_9P!<_NnMc>vbq44PizuCj+#-!+@L)XHpz8w`(6v
zVC&FaTuH)%mtDB3&T|jfjCxWgoLtt{Ezlmk>%X#B#>X`+beNOPrx<YBA8wXc^)oAL
z-i%oGz%-kKSo`XSO3g7@<V7Px$Qa~fV~D*hseCwg7-?A;;l3|2L>itR*1DZC1(r`*
zt)I9Hwd4;f-YolIStmZ__tz!IU+x%k9BV~_3@I=9zlU;=4v&WRX!AOQA^0^G5~Bx-
zFc3R=R=L<vsrlngQ7jD|IB>PZJL6(CpU^?#xLELDchwRdW|BxIZ=$RyH-{V(kulJm
zIanhe29x<^a1u-VC4NTj&+2*cm@kSN+Kl*iypz9#FaYO~f)N4T3yHzET2#)SLPKG*
zJ?INX0Gz`Jb79%0?L4vABy@8j?}$L!F~Q(WzGB5$vM|F@7^OwkfN#ABsVQq`M+YvK
zvhT0Q>t&DBb_3c>jbw=MGs1+mx~e>2hprx`Y`9iYN%&S_htjR#X;-1;CDI+c<0mhj
zk8vlO$I{RQSmG#n|CdMkA}P96vEzTTx&grLV3)7O@vfHXEPFq)vn|rrfG6m477|uu
zkQNQ(rXGM#d!t#D0c)80ct0StqynQr5XJnW8Z~kq*rkR*Z<|$xRD>OlP{E^1uF_&}
zgPZH~7S071!G!AEVmoua+s|@lJ0Bw1S*;WutsMAZ1r^IHkepq|YSFe)tQve!fY0hw
zJr6YsGOE1j_%f9Wj)&(14+j<z&I4VO@kWCgC?&DcEIT+q@^DHXex<;ZAq};gC+T7z
zE#A8~P&XzUvD5Tis9>+=#*p7IS5je+M~3&AQx5xtC_L1CdhjFha+k;cS3Z(Xz4URL
zXyr$b-!0Pttl$%v#G%i0Evd%n0?*&M90WCK#Tf{~(TcYq6SBp|J<lTvvJP{6KVi9C
zBkXui)-JU{Sn|lC!`bfa_{?{ptX5#w#ne+4dG_>+py=te#Sp;X;{5Ejnz6awN$9lh
zq=?NU0SF<y77`#v^L!R6Dr|<|MpNq?n&1B{4k-5Bfjxm~fs9=gnBRAhTeEm?!CO=f
z?K93>vpnE^BOR~KEhA0MxPz-HH-0_p`)r6}9QiR?TgI4p(v1MXwWCKySj$h>&k7GZ
zoT@YW@D1n)PiRk<DV#X|<px+vSzhO*|Mn(cUOuQC+LVTGL>{55)>3B!n2Z@W?bBJ>
zhE%RjpO6NF%|mkdg|8eOP#DV9j6HzmaxxR;i&f!>sLAS7D}s!6fcVy6h73%(^H}uX
z+E^S}EUvuv=)|lPDbn9T=C$JD;;Keqw4E!!@Tgx_+MQj1A02XtG=w9*_|0$;ouTx>
zP34F{PM(qy8$Yn6$v!$hRtQ%Lhw@6^@3DW|QeLZ5(ZP`|<<2QLYJd|yZ!UZ;cndMA
zk>{K+U(OMIjDI%9yz1Pc@PrpHTr<TbXWCe?+V4SSX*Mrv@@|(;w7rhDhk0a}16z`q
zm(tfKK)CVvUu59KTpBx%X5aq9345ucg!NpthqYXKnZ_na{cP}tC1~PgV$B1lZu^}A
z6J|jp6gUT;1vX_(&^*{)Ges;0MhEC5vLz_pZvius8rC~C<Xb>%9eRYWu#d-KKrK^^
z^>(32C^8YAz1tMY42lq}dTz@;nc97x06XN^XR;s)a?NiYWm>9m;{0xviRU3?zouxm
zR44*~sPQ{M=z2FlHFSoC>gAl|bPDlS={rcc%Geac{PR6Fh(_tjfAQhrjWd8s`@L^P
zYb3*-HcqZ&*Omw2lphg~7>MjByO9DcP-sqC|JWfp=)}uR$n@+d(`^6C=|CA3jzg;}
z7h2D-OWia*;@~tmk{_`4UAOu4D?(@*;lmZ9OAWn<pI21_g>N#T?kg~=J4uSHGdBXV
z0U-fF=<CnW;xYQUG$zmlt6C>S4z`{#R&(9X)fKv3?vO2;AlS!h`e?=IO_*R3pk4wS
z{I0bu#;N%qKL;6TibMo*(2J1J1)D{8l$meCT|WAZ9Hliu5U(;%G4T42cfkZIS5gqY
zd;7`XXXfQqiB!hAt8#^hXY>g^s@T}Hx8Z3Y{;%aew$V9F`o@HQG@e)2C;|riQj>7q
zUOsiY!r$%w689L|DU{%zXGm03YAxdo`tLvar>2Bgw%9D&%kL2CrS0jeW3SoHl3k>%
z8q{vmRw+E@`W6Q!$5|mJW`CT;0}ny_<*l(otcYh-y*reFj7TvA&9RB}t1!YtD&%7K
z)J6dVRJfx+lc`OGx6?weWG^U4FHa%bTEx@SVzxU6NrV5~bs%GabU7WsbN`n;oB?}q
z{d`sFkZv%8WaI}w!&_=ap1Kr5IZF&8Eoo#T;pq~amhQv0VnzteT*9S>pZ6Aw;k)3@
zhktiy5OY*Khc+2E&Y-(@gH2nWNtQcxV6!$pLvqowNS5F7M9^Pptt>sV0n7AjdMMa<
z*-Z*ULh8oG#@PdHp94>j{me6JK6?oS9ya6<2nAMAnLJ&oz(%;%TR=&qyL|a_E>kLt
zh#i-1LSmH|VeY*+Gh)f;DB_8a6!t}PPd)xAM4<NOFlfBCWmYF8N9~gBm|U2ycM}?<
zFp+WeJ2f6i@l!3p01gBm+#ZIT)C59wxCGmt_8)E3VIhLWO^!eI-9lPw6XW^|?N{z6
z;W|5}J-_03BmVU2?BuKXOzGSl-qkm?^r(e@NY{sql>hwkQ$l|%&VbAHc9!RPZY5?Q
zK2Ov|Ye~)$c9%`Ji>uTq5-e##4x_801bFcp;W;kDCy4oXxkFKC1bu_?2bU+ZONPVX
zDoHDcxprs8Tb4;x9qjLD6LGkKumCcJBjB(Qli50Wmcn_00P4x72$Qb3uh6I|_uOWD
znN9=_Z=uK6>f*;0UDcsyPGNuf*RAogp|}8)!-$qrGZ8aY?C{%<;xmR$9g2(n@+G|*
zfP>!`+<JWY{9nd<vHO!nOkjdeby;v%BKb2??whb@;CmZ==Kyu=qY)`p^Cb|1z)dJs
z`K(=$p-DN+N9JzM>pZt{epoGy0v&2`p(KmcIAplwpGWfh)h^>pZI#`hkI$8!OI)Z|
zYwbMaAKuEX9`tn>NqRSTiG<Ko@n;50-$K`z-aQK@*`q-jH^4?o@{<8L7#rxgH-PdI
zcR<I%#$1;S@UbV=qVfr02V-v4c1Fe4R6JDJQ|$7LdO^TLRElti#kuWBcLf?c^z3cJ
zteiV+CKTaeW9Czy@sg#=9}xn_pAXu?{*P&)JO5A9N<d7j&6&aM(6rp4s2lb43={Y{
z@^ysuQ-X9zE%@A3>mMRd#}+**Oc!IXnoB<W+*R>LnjCc>F{A-G0kl7Kpg%axPXMiW
z0kkU65GgeLOb<g+4HE1nA<Ca!HxPv&YovM{EWHEVyrlyS1iL0-+l_b9-~G9~veL{x
zorh2?k&dU$dd3dBr0k0;s=7{NK4k0M*yH2l6EOY|O@7ru67ml_kX{3c^w|798&6}J
zGo2@o9TU-(62EqfRfsmQC*g}=4uHukpKMR1-1#1cAQ688&J@WL8Of-bWzHnS&#F60
znaVLO=Lc_nw>0@v_SH-w=d*g;Yw7S%?Fb-TTv?#CRjlu#%%-?Zz%KSeiY#bJ#;|Qj
zN9IR)et^jPq*ExCOK>tW^32!%SA~#%XfnNEGG`LA?y`TfDw9adSZUGW4Hi+n`NS(2
zIhk*b40fm%YDMTvN=a{T_E0%24zr#*&(mN(14&*aqN-{UZ`0mTVAPxpyJRdU9kg%n
z?2yJTjS1^q_Z;)S(otkVr?Z`^^4o<RLtDkESnaxYyI?t)-dlEooUU~~x*Isoe@g$n
zTu4oN9e?<fAyzIb0vSWPu>u$7`hA@Kl`0x*uKV3=r3;E80y$}%HkDosRK+wG-@A%G
zMgQ|R{(9v`^wVygGqzl?yW|Cn#?@pKFE_y?p9mI;`?WJzC(+EYAc;P|DtQJ<hL?to
zVT2?k(G=ejJ@+qx&PyxzFvxl;4ZE$%m3#-SmoVVK$$&7pR_P11|43}zT{E#)s{K|I
zxLYyB|9Y^gg#XV63sm}!v2(v}v?wN8Ilav}A(z4+QtnU8%a9W`MX_qpa&oq3Zr^4O
zZK2Bat!gemt%DE*SWljJOO4Y@z7g=GVx~fEv^a`*CG!(-1Rr3BG<etjb_{>qJMwDr
zkb9VavigZlFehFp#fywaZpSV!2+X)l5Y`^a<JQ3lvaC1cSj|8<>A?B@ya!_}F7fiJ
z9?o7wvDywGY{(S72ZR(9Kh!M!B4tiWbyzhy2-yD!m>Rozi=E19!<-gGh5yqFWKk`B
zR>w4I{1hIC;)c-(o?Jc6{|qq0h?OsHF}NH)L3h+bTmyu-m$bG^Q=G7w0k+*|BO;aX
z?-MOv_8tfG4-wRpO=Kk`61WxqHF*;AjjeLc#;o_C_qYN`!CeZDGylsiTnq`UsTU!O
z+!`ToGzt_71Xa)`BvWn7(|1M3=S4pOO^Zjh3{S?W^clP-CJd1dG>dM9hWzu(RN%l@
zdJUd6!VX?T&|jWC16e-tq`K&e>xYjYrC`k*QWIKsA=5ud{|t0DVCp(Id0%AX$8o_7
zroC@JpYfQTms2F5dW9kkU{>uTCU&$ZCb)pj`10b#vmvk$<s!c5E)8`6aV<91Mse8_
zFY0vBnOBTVl|5?*gYU%YsbZzLN?KYR6C}8SNI$aqyp7b|hI_M$pMif->OwO=S)N;5
zK2T4tyerL+h%OQ2Kl#TKql?urN_X%18%1f`Y3Dl7RdN&>>DO@L{JApqNIUIY^iewv
z_41exnIL<HQbR4{w!T}gJSS$GuKfIRUD@9G^`<LEhLgQl3iH0$xa1kr=)~M=?Y0gf
zL-QmYC2LPq3(Wst<!0}t2W#R{F7)g@W8ZE&oUXGjwW?aXzg|_5^X=0|FSzkZA;%Np
zGCz1+1EeN9AUDmyEIvCA#kc0B!NI|b4JfG0z(4zJpSj-)#W>8=zJ^^2<=^K(G%CUM
zhd@cO=&;lN9&23XC|+~gx%0~ML?`H(@^vF7OQrsl0{EBFRfY$$wni*V46hsP@OF)y
z|ELibNc+bv1L%J`@hH+}k-Q9fWV{;gLl%^9#&&68GMOS!Ws}MIlZ(;417`~#9RnOZ
z<}?XcUbXng84jnsE3kg19GiaPQF<R26rTU{62d=gPI$*AT6LNy?wN_*GfvC8C47sU
z@`_$)<ilyS-p{%OH~nh#(9hWXY<}nnMN@bU%tozkRv+lV+|%aehoE_6G5D_CK;(5D
z$N<>SrbIK*(9~rgaJLD*hSIXM2xI({r)Ph5J2!tXS5bLxc5Y^WPca?8)Zb<>`Xyi1
z-t%s(ykX42+B=sb=2;*C9zXu{_Fu8n2R!l5k6gb12k&D<a0b&DLPa5u+T-t@PH27p
z#h^Ao==`@bzKd7uhpA6b)`_h-C8A#Rggh*j|EbM!;sCLO>h8ey|7iy_AqlbBXbRb1
z#td#dALpPTb6;Wq=G+#tKjPcIU2QCVkUNt2jn^U3t+%lK?wUZ~;sobyJ`&k|+vJoT
z$4z<T%%1AGs2}=Ifk)qoVw})F&*}|*H7#$1m4(c3mJ9*cl8KI<-lgjRW)+nH-W%NZ
z_4S<t{T(F$wg&r;n;NWAO4x5sL`t{u9Hf>!v{Ww?5_^>UjJ>x7k>mWGlhP5!tPJ04
zJ#cXKe|pJc+S-w4a(ujF+?wp4-w}NSHB#EG`(y}h7g~?8@PS5SbM>h(zhMOkI61t1
zTOZx*Xu2DJ!N~R$rmCILtohpalhPTK77UoB1WQ_Ip9%IHX3t&z=Rf?XWc3m~b~4#g
zIJvehhs)yTNqZVz57EwwpJ``9InK+oO5eGa#t|TYF1Nyip=1zSVb{4Ov2uTb<gBK=
zQe>S1-!c&zii}#63e%4TRVR8;17qXIP$sP4A{TVxNPKzS2Rq=&6QB$FtqjP!+XCWe
zK3w6P#z(y4(VKTCj^oZltwiX+nZl&HrG<CF9l_F{pQV@!4fM(;RirPc1<()2CNJgT
zH`|88NYas!UwnWH8Zj*w$(Lo|YGIdEt9GCh5-Btq9pxnY{VT`Ns_~(us@9-CaX7q%
zuMf7qRPi#6c|xD;1~?c8hqa$S%gJ3_Hd1%feEK*~&=^l*zdGNU`z8oS(V5Ci6qKVR
zSYBL^_+NR=#Zeg;EEb^c4*Z&Yt<_Dy@GdlWDX|nAYVpqcQ&HnLpo8AKIuNW@(oD{n
z#=x$apu4B7Dwcf_{ruOxVzoWb1#G_-nfE+wO&j{$*ceK%S%)XmGBN^#b(Om-&07QB
zEA#{FF~OLrx!un$>@x-3=^DICQ^#xnz*hAco0Y2c%0GsKg-Wl9*yJ<cFBxEs9eQ^)
zZmz<YGirkRPjP(=odE#g>1341ugjSy?{)ojLUte%9EB7dg@@V($G(TzE_v4vb;A@6
zQcnYm2M)yLe%C!vyP96M;?=ssC;pH`ej6;-Mf{8CnS7LALyT4A;^ppEGOAPg(RPj#
zfz;OXOJEMhIXJJcUo=6Tz&-a8IyIB6J&eOkFb<Ll^Fv#!3NkXCM$tN{tXjnrC8SoL
z1y^N(+<&vZz1{Nr4GS`1Hw|Rv9B=_42VR52VdUd0PXFOCAp9M??pyG>^^{<ML?>u7
z-prFo>;Ab&SP?h=_??Lex*+xX?pz(IHav@2RlO^JJxzFqmL1FgA7$?y&vo1WkMm9;
zWfros_sZUT%SvP<H0%|kY$9c3&&Zw$MVX<Jk&(Suk<5&Y?BDUy)s^ng_5I^_Kl;P{
zc+}<nKF{Mk#`Aa{&m)WR#V@N`K0KA=AjCk_;8Y&yuHF4MKhD4R{<&@@tAOIg7>BGi
z`+V?XB*S6JSkx2xa5SG3*8ZbX0_SVv8}kHrt|sq{BlmE^Pb`cl5aRr7KO}IWaEUA+
z#X#!BYAq;>REw%tRniFThDuqAu-*feYGh-5VNe@Zcsf|K+1&t)N=0404-TLJRy7d!
z_i}A|?seOV3p?HCxKaD9QpjkqFqWf28G1<n909%_N_*O_Ygb;BJt2A}v-}5H!~evb
zB-T1LksPaKs$UPv8u258gW5Ehbf^y$VxozIH~L71*XD9+6BFJ&8|Y1^<@1u#?#<J0
zezxxnsoR(1@G!;#-yAveKgOs>W8y*+n7OUxHrGzimMwz|K{gz|J4Gq!spGyT1WFnj
zS3t=_Mn<LwNK!D^!fHPUC}uZEf_oY~fR)9$2p9CR_&Prj&R}=kZclcqN9}A_My2_x
zN|t%zLC+Vzx&u4+2jAaF)R^!NANdf`dJ`R;7w>3X%O*NEeQ?e2AFuGJu?hXi-})&7
z6|+a0(y3pbLWL15_0zR6xTkc@*0&{D=ftn=|0H>{l``InWt1Mzf1}0P;~-n4#O(ti
zrJ^}Bw&6w8r`rzvuSU$_`TOXg*aRqhI`Z<ev#h6Cr??|zaI?1vNzrpof^W!UdtZ)^
zrHRNp&ZY-*UIL~8Kdk<K1v)WM*t9)agvU|ztgte_DFL=e%~yVnHH6JtAE?JbbLY<Y
zk-_KU@dCJ%Ej`Imx)zNRo3tdlr1VMJ3`-<U;2_z$BL2t}P}d#LZXM0s3U<-sX@OAE
zd++oO_GWoxzrgRL{p3l9D0OLq*S}l>w=v=pLf>#qnaHKb-#6DV-iY7WCvMYU4&;)J
zWz}7@I?HD&l@4C48jRZ&l|;AQ9;tNXSpQf+w62MK;<NDgPK?Bk51r0BlEz;i`iz9J
zpLt*hyUw3U1yK9U;SmvOh+q@*v`Mu>2x%Gs1M@9hwXMTXMO_OEVK$+9Q-04HgmcJX
z1?BK}*tHEbYk8#jHFW2;x%q9%Y6|DsdUwP9SdB8Po8{TKmT6X61hL;3K?d@l&muK}
zO(Bw)_r1BrnU=0`t@%T;zJB+2Yg0EW!atvuS~D6FkTVRj|BPTvW5Y$2*eahCChcUN
zo)|1G5O;KPT5la`+ctT@f7~IM+&($+Y)n^eUcRDseVP!{28|{eL{`;|h6}%c@xM<I
zag{%H;bd#p=|L-K7x9V`*pO`{)B&gY&fDW8<zIut!mbz?7+f0$?Gt{JI<&IA<ANc<
zXDVG5^I`Wx#bUl;Sz#=vQ9ei}R=&dAZgBchX7OFHPx1~Oduh@$lwNvb68YJg`c^<L
z|GC^cT3;3?LCbwwYfc3!S4HjQe(r|!goqRW{slDOcTfXq5BL-R`X{Z#NaFc9lGu@(
z_7c1SYfJa3a;k;po+`_2o7uvy2g`%E*R6EAgsu<-4k%HrP4Heu(KrM_?wMn<4D72v
zM)mJcPcXdHMN>^bJs03`|50wC<}e!8h=Y#drwWJh_SIM`F}mS#9$wx%%?xoC(2&TX
zYy`eqA*>MnS-PWcD+Xd`*Q6hwWz$U>DgeF6d7z<O+FE_PvKCn6HS+vIi9q=|y;rxF
z?%dvXtuoNfqQm=h={`6|FFm2*zb`!rx%8a(68{_k0dfHSU5FDv4!~B~{bp5=P-gWD
zY2&@poXo)OYVTeBm|JHWL-Rp-U;qP#k|e<kPDT9*hDUz#*+WU%ilu(;Ze^S5pp`I~
z!~~n#k1@l*(qjV)uzu70K&k@=WKk4h?`$7b-zW9u<m8$L&YwG{Z(;oQ0gS-$bG1Wr
zc+}<JP=He(S~?y!GiI*KBHq1AK6-N7XqI)tA?#WBv7>Uc9t_K^BWiBmow&vUYy#>*
z^Le*9$m`Dx2SUL5_Y=l*fd4ncA^HXVfH9nLWVX3foK*^yhtL@(=3M>l#exNjP%qE>
zHAb(>)}%e0Z@v@O;mbDG)+r>N&UolAjpo~p?Enb;*`a^0{HPfeBSzon7IhjtermXE
zaRJ{CxjJc7US6Knc+DH!^AcArE+b58yXut&nZ+DyJx~Y~aaqV+TXQZOg;kh{8uL;g
zu@ZShcMzt!L0a<8=r`LB6hw0hR$uDm0s;f{A|6xx`9{*zCc9q4;`fk3!#I<@eSNCj
zQq};&msR`Eh?+wsT)B{E0O@(eR-~vAhx{WyHY@x-K{6B?q2lXJO&>cai=KCHT5q_|
zS#9q_5L2n)kd9&1tbgllLx}MP3|{(x0PVL2J^Ib-&{2|hLa79wiUe%wImqSMvb@tR
zSFoXb*nS3Vi?<&}!ts|;poNfDShH>Qb*g;4ptt<?4OqS`19D8)&b1~9Ys2J?^E^2j
zSw4iodz9}NO#<zYI=Zk@#L-pSxjiHPIy$5$Aw~G7<oTIZ$Ll0Gy9Yu4oV&_Fw$&hY
zL*!NZKZuu9st{y5Eb4)tzm6@Q4vBk(x4_}wOa+JkE!{2C+-$xxuHUj^UPi0RxZy))
z*ru{(z(dW$M9@zL!d|ML8RP#*Asiajgai6XVIdD+QYw@eEBtpQjf5VAN>O8h){Qib
zcd+rrL|&s<g&{=?n7G=y4cn$u?Q6Cey+y;v#02e2ra4FKwxZw3v{=IB@(T-oTpgkR
zb8%Q#eeWys^xjzGNp5oBS~hD4nQYH5uq=Fc6GG9ycj4fNT(9X_(qo6$Ta38zA3<?z
zRom&{)6$p|(uc!x%;(cc`Ypar%gVdHw9t6suP5$keM8<qP%k<HG!E*7X@8gENB+ae
zuUJ4T^F8Z2EPO0E_@)?N@_Cprc9l|Hm&V+npcz%5ARtHsKu;eV8ym@9r>3X#0kBDg
zF?yJ_%Z80w`9Rf`zq2{K@sIx+;MsRC3hS;N?tA%n&3wgVHA!vTWCd51F&=`9fxyET
z^d!nhKF=~eJl?1N1HTe-pQYnS^44<&p6*2iH1K@c*g2_BChQs7x9|A#W|XZY*TwvN
zOqj~iW0t3|F~lImL_h=SiiL~+2^yH9TKdm->~R$c*k1m|AJSJYN~jjBVFFoPHbn2F
z3eJ#|zNrLFO-&s@OfHZQCa!3!0P&A)*lc(?P3CynM!LXbD)D?|w-3<SFO`_J1<g&D
zEn3!|?-Y-^0<`U2rn4k+mHD0=qn@@`-<tiCe;i&Ub5z}1x`eq4Zij$@sNINu48N^X
zFi&XQ{XeJz{&E)F^v_FCzm)x8o&qss^%{0C<P0$6AF_=+{Ht61L(9VipOV$ScfPoF
zC$u~-O-8mQA?eXFbj38mlKp6H{u8n@c~kdteTRQFfIpjI-|#(Y=$nkZ|MQM7qkwFs
z3z_+=JM+3__dvi52u{|qhKAgBNbEs}^sIm1xg}>D$cbvC;knD)PhXn~OZ<zF`J$1;
zi<w7B=vBXu)PH=w<1)vW|IN;-37~kjkDmFJW%}43-g0+8+;Rvw=%GEIX9J7NmU%-j
z_w+iG*@v9;pCwi<J-8d-559`lv5R_wu}(Oh9FzQyk390z08*?Q4k!mn)!}G1(M-g}
zW7AHYM*$-VNlEWaJ>1b|jEss}2Mx&WM*54Sw_MZ<zk7SVf{<<@v-U++H#n%H=WUx#
z?|Kx;zkwG>erJScDy1qKXKsl^L`=36=Y*d1*1aVznEAnSv-1xq2J@>30vjiG@9<Ln
z^pVW-%6Nzju7Ylm@R`@+?##09Q&fA4sveV!S>A4mS+tt{Zav@9y_8Fg7wn9M5#y6{
zun$Su{g3yX|9HSZ;f>lPunup2W8Qbx`l70*R|aY5Y=2MbPYjUEGe^9B!u-2jd@L1T
zz;C2rtFo4QyRo3tIpfRiPsI(5s$5R1MCJVun12&6%#q@*XCbUs&8w_7r(p2(iT|H7
z^yT8d*Uso!d^eDRh*B~&&!E2Xe~PTTZ17IUR6763%f_)WtZ4$ZG8ATCG*bOI`4S)x
zp8X4x>%c@uW+}5yQi9)y4k4s_c3C20mY9n}DTStmw7A|2;-)(tl`gr@(T(PY4=1XN
zaU33#E~KDM|5s5yfr0qqLr3TfPa_abuN#`xRzP7G0`)Pkt*<1DV{Z%<hGAJm^wXz>
zAt52v`U-M#xgJ}qL60730i3m3<52`jXS?|3Rep;O>LLy_p?JR(oXpHj3(!xoE6Ifw
z6$laM6{43U-XXYO;t|(S^P)4>61PS|{YvTcTdH=Ee=Y{&_2H|2T77Xg!TYQOuXV$U
zxlW&xtU>(Qr@wad{W{iY11QW?)Ea$%|6gi79AQ#<3aB~u_fJM&#~n8MKpsMiD{AzC
zV!`W9)lc@(jU}f=suNp)$T5KPK93N~{|yNL{<J`%SLsqq<#9XI@|bX~R!GB|wUvPf
z3TSb>$g66-!{T_kp&L7vT^}asi7f$bH-P<puVKyP&4b8=T%)Sj@VK*KPvyKH4yPtS
z?`O`Q)rrJkldoe!m5sSQLlmKi=RvcweW#T?rQ76g)}D*prdWx}IsNJJ5#5d-`5UjC
zH(%X8{$vm=@b^cAe;oDArVBIJ^(TF74)r7^pHA>kQjO7n{Q9qI@+<OCgoXcc9QaT@
z9msZMDib}22^o8))#1WPvN_f-#hV=|csE}-(Ta0D%P`!(_c&RWS`dYmL|Z5Pg@v^G
zzqs;Jb&LppRcb|%!L)O!>23B`1`gC~BxVY@EfM8pXU{<b)(;?O<du3FQPJU@`z;;a
zcN2ZJT0cX_@(T1GU@@fL0}F~EpH0;4>}=)FFTt<wu3Rs>A3{q9velZf-x;7!+xK^6
zR_b!DSj3&yn6PlO%Va!t@v@$UQtAtyBRpkHEDDFc!eP3530=HxTzY*I3m>7a5OB<_
z?I@p{nH}C!fSSdxl%~!GtUR8>#>>N-bE^1G<Au{sXK!=wD_CvLZ+Q3jhLWV%ZgMxl
z`^dq9H1*^O`y&D5U(3KRR-V{-wX@a{Y^~NZR(ZQXVbk{dU#8Is6Gm&WL1+&8{9RxM
z6c<z`aaLewBPgm}%h%3%OxI9Pjk;n28_NOIit*Z8zfz`5mhx6`+@t$BN4UVt5-0vI
zTfGJ^W>dd*=M$g(T-o`&BEQmW74YPLE#Z`Jvy;H={STr5I51E9qd&5uI&YMleB$?C
zfWMkV4PRbe?|8#ux`l6pV3`MJH~VF9A+R=A;R5i%K1}ktopZe>>wUuOu>0uHcdmjs
z|DDR_2;%i`1%SB7P3x-x6xVo)!et>5Vi;Ga!4&O_O~M*Lid4~coT>J<wq8&r&u7uj
zp4w||zHoq@b^O6ZuNeqW?XMSk!y-KCdy?Xw5mP#vA~X(UzOdDA4qiwHG*m`53JqSV
zz<Lz~(Op?^f5N{sycny&BvzV)z`6dF5q}rz4p^ibiSrSNhnP;87Z9p?`AkVac*5sa
z>>rQo1?mx1XD3DJenWi|hF}6e`??=n?CUj|jur*pshu~)Rn)jSEFGIta#eWt>HL?q
zrPr@khwnlZZ6|;>fv`0)()8!(fB)on1u09bHd-mr5uad^T^RSGq@;x3Y5HOocs+>j
z*QmaR*fqpW!7RiT5ZOCJMm85BeIdw)5Vo__09-+qTy60R?7{#9l7OgrUh@YZ<BzYa
zMr4}}3Wn4CQxEPMjZL32bc}*|&VN1uwGam5r@>4-;-%u7FA*H`o<}J|h0>AfCn-w=
z@aMCl_>~1n4FXW7DIm#2v1lJaaY^ue*D<cb&J}0(U?KnVAfZ!>eNxHoMYI05Zl1!T
zc0fQ0jr%f0M*{Sp!z9M>D054p^1AojQh%85k<O~E+QmOvp3PN~$%ZwsVQnCvb~S`<
zj8rc^n-zpuU{8f?%M$Z^G4(8|9-GP+dPY`&(d28U`PFRCu>qjXIl)-FQbt#~nV|Er
ziRWOp{`iWBv8Jl)olv&F`yCQQLJsuW;*oJkJv;77o@Au2jajxlIDq}Dz46Nfo<lvY
zQ5Du783-Av|5Xw|e*wJq0o#&B0EO4kU8%5g&JD2$)BSO|V8g+1e9N1g3YEF?>t8Bg
zK;*R{hFhOLPA&eQr@-mUP$+}jFuu{|_Ug`A$Gd(qxMnMaB<PJumXMu=$ku;<vw4jJ
z*05z0#7Dgb+|CWuo{p_5xI5K9oO><>n{SZLp!e?N`7x?5f_u1?c2?v<v*wE;;ig;h
zMDy}i)$^7gOw$UV3En;O3{4{I4%)^t@!}S^St{KJriKZ7vXyl<6<!~)O^0}-p0SZ8
zJ=}IErHS+w4&nxlPYSaeRT*boGkVrqaMiH$y-dsOXG5|)&+Gh#jna-V=fw_*pE;e|
zk#y+DtA7nSd@C`d1nMpF%7gp&>U@upi8$wiUI9ZeuVGyK&+B=hfs1ryu3x`ipn8!J
zV^@<}RP+knQ{(4npWqruYy-;XW~b!zK|w+9uZ-v|okN{Yd4b|84}#K8RhZq@N)+N|
zx$BT{gbFtfpyT6lPRe74i?)_;ATN03%bxwrzGjo@?YV2>7jC?Hd9@!kuyqnzomy`Q
zdqimZ+D*d|@rW~Um8#5@N6zuc7fbh95!_ca);BkJ;G*_yZbbq+`vVg<3QI^EUN&)I
z%kI{X-db<(#R1bcZ6%JRAxmvoU8V|4e-44?p*=oYA0NK2Z`u+M7F}9c-Wh3(-L=w;
zBugQ(yk~c$%*KfO=jQaBsFFMk{f7AR$gb7_n)Ulp*Kr|Kd#<=p7=uK>zX}BM^guUy
z@9IVqc(7lsc*E@ZmB(BBw!ucNCkbfok@&r~se6AVD8TV&JAfbk<}8YUdO818I`@<d
z`cG>Cv=@#T#UIzjjPw&ePB(VKOLv1)Od9);^Bl%%yTfT@*I?_?4qss{M%CnF(X}3J
zjXV%~-sid5@W_V&Wz0UD6)h-e5c$j`a>wprf9+6`R&fHdYVo%-^k6MMaCqQG&viv1
zcIyhKU~q6(y|1zarR3W`e+Qpi6`8+|WZ?Yuu;g)&>=ruYEk7IcieT`f=sf<HDR^6(
z2Sj5-xk*>uv)E$sdZWq&wXW|myJO015aPf@gc>4;h>-4)&X1_;_txOsefs9jKM$_J
zhB?lPDMF*$AI>s7ZbNS;fy>{ruE+YXFpfg7Yv0<2R(y0(q(}R58+1;4VR{%=Ehi>5
z4}K2iB7%(N8OXQ^xu=HUq6cFL*odaPLrdH&cKyojtDe#vr{~DbJ1;8#0rl})Q5v;R
zvcqb2eq7fsp^UWG(1fnLSq_cF5mzFCRB0*Pgum#EzJ!N4J|WFu$6&eXs2;ifHFb`u
z<xW@8HSZEZmj3OcgVBNf%cpwdz0`VtCMlJ<2qbYh?EbtRN-3%f`MI{6TV><hr?6MU
zZ7M>s2NpQp6De9I4Z(d(uv{8j;c|f2aN*LaKUB%z7lvHR4-8@$*3&8~<mHgwsn2&S
zaP>P<wmzC)g=wyNsL}i0-p5pH{UVu`k&&?u3r{r9jd{n0L?qHXOekf4Pg*N+p={!2
zk?k}E=ViI;lYJ8BG)pCs`;nVp_O080e{L?K`sz7-VTp&j9uKE)cP4Ukzxb)Xa3s|P
zv~MJHGWElukP8}08iAY4IbYS1IL4PP<B5yf2QQXG_n7bd<Ln^GXF-cFi8O{o+8+zV
zFD1j*(Yw6Pp#3{7ewj&3Y-GoO&A9-Qna4(CCy+g<|LfcE<xu05Cxyk#uG0kKKvJWi
z)kO19LhIsL0BOM<x|lvgcS#3Wn=vslw=SJHI7LoOjOb2-OysqFVRVI1sdLAXOl^1T
z^79er+>i(rue<9rSZb6bS6@UBFi6JJ9r^yuX?5L8*|Ql-F0sqGU3WBU#HDQs^Z(%p
z{C-AtCje}Ffu0<F_*td7kS`C-)Qa%;!^V5rB$C@7Tn6G<KNn(((9ew+#Mpp{^76Hb
zSN+$tfj|%)<+OdhX=P=_YM?kxz-Ay7C}AnDQ&xP^pg#QAy%n&Tk)C6OdfLY=DA-?^
z7D4i__c>w_0MY>P+QH6GBwu$4$(8GpBO}uZ=oI4dg&kxblMAXtDub7V4IT69Gx0pG
zQPW&QpKsZZ-@(D-$Wgu@*)x-*G=7``mhAj{+VWTnl1XV)%`GtzceZgfgM|t%x3x{C
z^P8Gn&wKop`uBH-53xW9@1km|8TJW5m$((ORfp-h9q<tSk_S4x62oSqR<Clg)Ev&<
z7OFCn9|p#MH3`xrx2yOXEA!!82<8-NzxF6-V&9XeP*)gv<6Z`vug)RQBq1Sz>B$eT
z8rW1{2?M;NLj%i_2(Q7ujr78ikNW#VLko!gYQ>eX+Sl!+kAOO)6Skqk%^`tsI1-Q?
z`g0{|Owm%+l7w0!Uvr?=F~OdRxA*x$BP#zL=66Eg64Y;hNo?`T>_-jx<a$w`xn5KV
z`|n=>md&-i27Qv5s^nN9`thf!0@ykne|OqE7$AACiKPE$Z}AvlE^po0*xa{o&;-Uw
z4CR;AY>%pBcTbDZy{>M6wnI;$?&s1SOXuu5q#p<!VOFTXY<*ybInPP4Qp@g!=X*V&
zAddYEnua+7vmiOqIa1*?IusI$j5okWNF7kg0oA4gSaOmJ12h5KofD_vkdj~EhWr;j
zp1S-vapA}$>s0X=pFd7ie#J>)K7qdu2_dH@^4to-qdVc13+$h-Iw5$w2|VvjNCt`*
zAAV1Mc_~4^B3ApVr>X7f%Iuo2_l|O)l2AK%=DCWij=<xJ$L$c~Gytkp;?qEkKWB*b
z6RG_@I8?-b8Ic%rL?scF;=9kGkN88t`un>@7tM;VV%<8`Z<)VncI&kI#vNcSj8<KN
zi|a(1<|m9`LIJklx#sJaWE<|x3WU5n1?`5dDp*0O#r*U|$v;@_j~zfO)gabfwy?19
zBDYVXgM7)U&oAO%Ge7dNFpBB~**0b#ojc=*HKu{at5`>YWGqem!ovD^lDoxW>Bq0c
zgG!hTp7@%kgIeO?iI_THx&$&L1(bjkuo}E*1?QMWesRbD5LkqcGroc7QafzfJf%23
zCik4^`u|!D3t}}r^vJdoNF~OQ;#`#JnfsP1P2kALRkf!j`sX<ZP5HIbJihf5<eq`v
z9HeFe8=Vt2zJ_)c@~~RLElnC(&RMzevgPC}5}ScC`xJy`k(HD4Z3&Jg0x*rZTF=!s
zsP&6LkvVy!b!V&bymu~8?CorAgk*u=U|w)6<KN8Sh!@1NL+}G<3thFaks+b-hy-zz
zG&6X)^X#yevbz^1%FoN49K+|TUiRDGSR}7TYZWs&^sycW`MP!zH@7IBDJojbO-u)c
zPJKDuvhv4ZF!GMXmu#HUhhgpRUy^1YDgDXze0;DGU<AcV^D7-#riR~)Cnd1B5h#Nz
zEL@TJQ$11Qz%))x#XIZym8!@~+f1z4DnE@wtAF75a@G7yZ@PZEVM`kDP9(@c8LxhC
z36^DyPTx9no4fFNB6Oe)tF+MGF=J>g^MjVgD9}Oe+F*}Z6te!Rui*hNq-M?yBHSmB
z_`KhT<2ymDSA0Dz(!4<M^yj6vXONyFyfJQH0*72AXDjDM=nn9pauyaAcoa^4e;>+q
z-u%<k=dgVoLlC0`PsrCC=o>r11>UrZclCtTDtVPd+aH}0(q{Zq7)@ODRfZrmx#VKN
zQ`ZSu)x5u+s$;~_+`t&2SQg1Y3j9?_nS<fp6?mBu*bSf>o;~U2^a<dRP>J3haTE}%
z3ThbXl+%h~{xWf5l8GD$$uDa5x2=m5$cczDfEa1KKHq=k=FLn1#R_5SMWZq7ysPM#
zP?27-i7zyBCi?7bNc&BVVAHGJVv5&x-u~7n@5)!nl@U=<bdv_<fIq^kG`n}4K}JSK
zees=p6r3L)c5LyHr}pRhMN3W8N-lNI^tZ81E#m>FC=jY%zq4s<C|?@DGkFRofjQg@
z%EUm5FF2(ow3aCayy;RPR#$VOG+qN6&6}9}X~Tv6=b^7Jn$Co-Y@5&X+BB<O`uDPY
zC(aCb3&}$5$wtT4=<Gwh(n%&s-MJ=4f5g=5*dc^pP>uZuqCCR^zdPM_VPq20+JwT|
zI}*Z9Z=gH+Nj<%K1Yyi&+Y>=^)C?r-rJ7+JT{|W^`tnTO$+N68Fi#aQ^u~4uq-_vh
zf47S<tV882pFov)7QNX>grG-U_Bx6j8b&%0{ntSDEZOGL;?OJpyUQl0$pzSJLj|V+
z#j<$JruARxFFx}luq%~xd026}jPzZ{RL;)0f-qS1$8RvZ#9AEK_7V$N!<H~r%EQ|7
zL#8g}c%Qv!!CP%z7GAdgRZHiJ;@97+|JQAgprXdqx*NwAOEEqx_aAS0PS$AfH+40h
z20WBmE+NqW!1#d{1ou=OX5^l1&_qMWs|Z0?eY|MYo<z-S%xpa)+QXelhy8rwcmw$X
zAt|?s4~%tA3(SC8r!^35w?U+G9d`cAf%ZT*;?K9wly!pObX&4$S(A0aYrD~Wh-4K3
zO3KPCm!oOe*i;pEKx?WHX&CP}G*o<h`V5F<XRLDGmLc4fz9@_Twk-v)t;N2X1acLp
z+$)W{8S;0;drBg%c+SGx@`W5V@KWJdkb)@!^|ZKE3tTq!iOTuHZWy;Tyh^e`Ghu-~
zsa)?W@kXW!@1~waJAMCgipd=&R=(1tzj?pasD`Dw(Wp56OHBpx9;FR<N~f;bnEjd2
z!p8>(MugweM1SZub}z!UwO{3CB6bvth<L7e?b@>x0jg5UMf=k?*7Q;uSkJ2yU=EXH
zoH_<djUB)P8=*zktblQ}+P45q2fT?~5O1*nbp`~xMnHZbUup^q3&Z?pBp{PnfMX(^
zg-Q_l))~1MpcS14yv(hYP(fr2oAF|N__)~=WW*ZR#&#RD$DC)o<YAd-0Yr;dZG(H(
z52S8eTTbyxoIl!eJ^EIz`W6`lP(0;(5&m8tAQe1-lt<+0MNqiV{FY@QM%H*V?+`va
z6X^Ib4#icw`_gUm`{FJ8N}34@--U&%r=P^;>pa2Jm&hf%K;X7922mlT_%BMipP@;Y
zsO9$0o#a?*vB9!O2K<j-{oNP+KNYEFQ<??2=%x^)5HE=I-Da|0z4rKK>_9oh_o$by
zOh8voPwB0Z%}#`rg2DsVD}@E{zg?{|m=kUKg1wczzu8dkJN{6s#$(H`O3bc5S4$RI
z{<G8%vw>~#7wEXIIXb?CUA)g)TU&Scn8uL}$gqQ33y@$9fwNmqP;G2OTs=!gRajnL
zUY#QEVx|}Yrb(>)9w>1~&hB^4&W<pjZ}rg2Q0F2r-@C74uWg@%aBboNS>het>@`B)
zrY0o-{tfPccy_ZXv>#l4<Hqdg4+8~Rbf&@`T8>X@GXMSp2u1_0%(CJNT-s)_TN?L)
z7GS`#M|7V&Y4A_`6T=^%CX9L>9_+(^Cy9IzUXoVj0E)*{pSqezbgdVQ6m1$Sx$_d|
z^RAvgORREej3LLA)YOnv(=Zz3&}XvZt~pxs-pByb64>A^ubCFks*4Kfm~^!l3jMey
z#gJvjD$2@Q&d$znN4JJS2cRE%U(aDL{8bQZYO@y%5{8j??xmqh>%s~J!4Zf1GBQ6;
zeLHsc#{;MO>kb1)bfo^2fIjjl<Gpgj8)mlL9EOdtz3Wn83P*q+hO2Qvr&9seg(-Ge
zy1OMtA#g1{m)~M$2}j?+FP5Ju$r9!NyvS$hh&SIS756IN^6$=7SzV@U*V@XaD2Mdj
zm-a-jtSGFNIV1msDr%E#IJ)>eRhwTYt`3DB?5iFaor{*i4Samds=l;WWoIvUl)iG{
z4$ZgaC*-VrnMp{kdy9m-&~pxE<qT61DmjRTIK$yhLUW}bF!woP5k9`lgM)*1whan^
z!Sca|@;e}4)ln05T45y+T6b1Jnqqnqv^NMG67jVTS&*-N5>j0q1bF8Q3-XO4AMsDx
zWD?^u2o2=ggRso-`#LSkR2US9$0KC_0Ayrm7`BO-z^aVZF@fzml^a9WEo2WS3Al|+
z_KE`e?|x*xW%lIY%oo3?ifmci(+ihs|K?uCa~3aYK5z@Umfgz!!yC`;mz}Al5Mdv&
z^<sp2S)j@IkD$wUq%Pca_!z5wQkX*HQ{S=fD{OaKObspf2P?B5Ts>ceA0#wTiCc5A
z1|(O`9Y3w^`41aP=v8y{CbIYO8hE<vHv$6#=Ru3HA0&HoVbM~`YkM;@vui1^;hBw%
zZNAhq$h!LT1M)RMKPn~qX+GS+ne$AdJF&jXd+*<Oa-@O|M37b*oVVQgSlmd?RFMk(
zEfsLPz|*Q0^RuhL&xLI|eem3;NCIk_q)%adJOs>608L38llLQ8eto0hv-2^oyoMIc
z8ef<%P+|VLjcEOeJg}(dP0<#L#wTbR8@B1Xp*P;lsvK#DVNh^ETzzazd*ZNXj_-3=
zKBtQSQz^}9K`!=zhT-$|;l~4hcXW0PDxX<<-rF|njC~tXdJzS#p%bJwV$aio`5=id
z#LSrS>3Kn9kR1~9$=EWJ$B^Ay!p}VW2YNj%%p)Do4F%8PNr{ZIF)}K|8M!0?V<px#
zv>#6L&QFKWe$0qcA;fBzO8n9a7_@dxd1CD)ZA<)Ll_Z+43CYh-(Dm49Q0OM*lqh<j
z;9_raq@(t=G6afvu=cf=f}<WRCIexl7v9CuJPPpngz@sF6W@aHWFAGJ2XXhX6#sQv
zzB>*$A1-5WZs2bpL47B9eLVwtdVmjf%Phfr#iZp#=e^^t9s+k!NB)vY2|Ub!YP8Hl
zpczXAZc@l;F!^%RTYM2mF;!XM7i5oz8Ju)klx)9z(7w*zrpz;sfk^PaS?f}@!+!t%
zJwL2rIem^t4(3>0hp)tjCX2dGfABz74xZA6<*lJGS_>P_E<;3)*n3%0(hKYC-|TG+
z2W*q#eY*!`S8tv+Q$Rq#*CoLqAw+u*DHua0H9Yq~J5vM^^057~cL0g-RK}ktdgLb{
z5gG%i6~YTK6=_p!Y6FKb0O1g_L$$+IhesvGJlT@7x3lTq`Gm8E_+e;hFPP{VO3KCg
zj(XYD*5Cc*Xkm5bWB!3L-QD*S^GuUj8N{Rak0eJw<$jx_Ct-_r_=Ua1nO%h!Pb**i
z84QJ43Iyn^)n`3%NHfQmIyA}aBG17jotKr&u3ZRxF8Wyi0WoZ#J+Sh8us>Fi_~lb!
zdPlfp>{rH!EmQ-)If;>T1SlHoGwVaP?!33&{|sTd$>RdOABaN%GJ9hDK%24x%d7-y
zy!f@r*RV$TmHqgbw$f=*ZYhB3*?K-Pf9Hl(kMi>JXNO#0z9m(K2@7?*={*XNO-5En
z4h{U&hN2ar*d95_-{E2af#G)geF&)WA0pt-bsfb}vaCib_txO2&KFcCIu*~yya0Z)
zLu6f!s;kwPVawO2EPtw<?wQ?9d_BeZr;UMz+QGe*3p|&PtsnU*Gj83t$Xe6j6UY0@
z*Ihz<T@T|w1R3u`;z<eyWO&?}8|T|QU}$1}`1@q+d!0i3eDBx!dAM!)&u)*h0^s_>
z;}recn>yfs_NK{V9!_C7WJvjGP0;Z2YR6r5d<1*+7O#gXL-d865x34*EKk2Od5hZH
zTk__5Js)T6m8YI8a(2!M6ql?)YA9d`M}YK(Ak~o5?oZ=`)8|$s!1(-(23>%_+0^(u
z{W#*LAvxJOUZeL~mBrp|pWMz;(QS6z%A&sQw5q#MBh0y^Am8QvJ4%qp1BbJsmp39y
z>Jq0#Eg+x^3hLURBhf<o=sm%5XkDc-o^?~S&c{}PgN9drbl->4UY~HzGyc2Zz)o~`
zE}74Bbx&P&6_`45-vopHC$%e9LC!Lg@gJ72H$Wkuxp0`?I9btPzPI@F*7>xH5lJyj
z4B6g1ODnTVwd!DciJze__>!EEJf)4{sq^$>wcfP|gsApZOdqn1b=EIxf}@#&V|OP@
zJws`+A%8AsR8NYBJ)tH!phcPse+TLGjk=s$2`QI<6Ego5mNk`v-i3@2!V`gPn%IhK
zU^(SrS26CITb^Nwpgc7(=hgTt3#JHv<i+)XkQYf#?ZPS-FGjB%AVap(XX_nsDA33u
z^3>7t9RIJ#Q;V<+$GK>Tr0OCN`z&T}h%<K`_bdy%w(we}?kC|7*nRqv_N*{Ivu@qp
z)drs&U5lc5mU`XV#1HZ;Umwc_Jcun3;omxPl}$w4UU=Vy-4b|8TDh$HUP_k&7D9W3
zHY8Pt4PyXx94thKjD+Q5hmJBL=_)wN;ff{y2Vzw!Rk(WIo>FhmBGZSe?a$J{QrYsn
zqFH0{I0gc2B?-8Lh^m4A;RacLDNOKxoqoE%v<O|S9`JbKSL#wHdL(uhcu!LZX@cZ8
zXE~ew4m2KmqFJ<^mnWn@G&WYPBYJzV;bN}OWv<~N0bO)tq@z{T@D4<;*&wz;O_-xw
zJ;^2~FJG~E0D2R>FeJ4HnUdWi(w@!;lYoEcNXNm!u@n!%Z{wzRgk)$sE;LkH$Ympv
zcWp3XM)I}EssDn>si}`6Kv6^OtIpC2qwO*>vfECwzekvs575xi4i3_7<ap80q|lH#
z`-No8yJvd)8^UO?#REFrvC{(luVo5aqErav(9w^hXW=lX(;A6%Wj&%b(lU^e$|}=D
zVflBsEBW}7$-FBwuQ_NKZHu<$6@5~@3Aobq$bOZfnR&bCC+nh9kHY1!xf5ZKHz+8G
zuq{x_5asLZd*<9ZWdtL5O_?Ex&rP$;!;&EJ#5Gaf*z4x?-f8Z#6v4K0mwk6%-m~nc
zaYIMPNBi@K0HuA!IJNkFC#`V1pxT1YW7fD;C$IMw1Q=3ifBsy(C^f5(Ni|+^*SA9)
z%Lfh1M=FtC6Sl2vgY3Zb#9`<-<U&cL8zWECh_#%!Jfs(OSM9ECsk-2er=8)#IV=ME
z0ewcQ8*y(Z`SXown0m3XQ#+)LKgJGfzhviGQv;_E3*&&Q(0|LoA>r(RaK%DH<GVW5
zEGzD|mI_-#pQufMr{e>10qE()B#hpDw?5sTtZF;_`s_tuTF{<9fBwenW$!v4bUxSu
z#`|Pt3if$jPL)Cr3JViVym?^W6z;1M%dQOSrl9P@gI!}JM$!_^Wh^1$zQMk=y`6Y#
z#Xbef`ED?*87Ohh?x`>V(K-_V^r)Pio!b2saxKeO5qKzSwEzAyM>OC?pU7BLnQEY@
zmOqtlx^yD1PlD_OI(`R}Z~rMgS?Ve37XpBNk?`9LXws6HUBMrCWlaLqxw^AP85;Qb
zAi(7sAwXH4A$*$aQ@_P@4$bA4CIW3c3W0z9N)!6J#~+OvD6{J0=EhZW&rGRCxnYu5
zHFzJrReU9FjACUbH4Y}mV}~Dif)RsyD)m!<BPI<dqw8|3C)um10!|t-KFZwJ(j95)
zTWI;3U3hP4oGVn?&-Fl3_eh?UO2UJA$ONv$62ik7e0-DG6b4mdw~Yn;pywc1R|*+o
zv{Uzekb>-h;^ev&p+9x?n>Q?sj18@JEzJNw04dJJOb1*dGOSe!**R|&2UD`hopj&;
z4PBCty9H+&*WxGb3-<J1e=>Pd!?@#FEYEszP97-qj;!hDM9zk=X5+wDqQOk)N0S^;
z!NGI8nLXp8U(E<2JPuX2Z`4`dml{Z`;=2>-MsmY3`hhE_awyxkST;YgzYa(<xvr;|
zgn7ZCbZ*Z9lgK?L<73t?-z-k@U!TOM4~|jM`pXR7KX2h7a*9`ctFVN3tM^3q$`=zQ
zXbeK~%_=A^2L@1{@=mDc)tma#lZ>r*l85wLB-mG?q_+f6eugY~c8o#0qS9^MD(E?g
znv{AU>?6kq3%*61rbF@`!6D0nIcGt8yc<TISlHRwb<UNXWMpDTy8weOWEZacKUzP)
zgP5ZRPvqjQZz3SEp7wpTN*iWkXQ7w<?k<yhk`hP)<b5phK93_(?)1)H&h?sj2<4xL
zeDqOEQ9sAW;>8lasH4@CmM7h0%Ir_k^|9sn4>a^=*iuc>s9r#NKt}p%{XxW<tj~F+
zu`1WFpOS=F7RnO~y^4G{+gVgDUXolb{#rfM*QVh)bf$!FlH#x5#m9h8e&AObBA(yO
ze>;oXII%}muX`(v_phtyh(r^ZwNNJa{KpRJ4v2ih?@JhCf2W=5aQ}F_(s_8#<Kf3M
z?Fw!A-!429naW(y7hB(=t;;Q7EWXi*n<(Ka0yt>!Q%P^mp826yYyiC6-`(3|7|zpu
zPDUy0NY7B~@cEtDJKTm4m{(t2U*~^%@&J$pjs7BICOkYmUeRH_Q?#_SGC!+ec2eBq
zXE^MtXL0>LQvZbzz_UzfwFjO&!)&4(r-9(z+pzmud+@%Wp9N%_mtWqT%XL;(R_22d
zJu)E&YM`^H9JE7!&=RzN@`kp79gVT_i09VF4nzAl^018Ishwy0GWmtx)wll-PXf<9
z5jlbBrcN5HOdJ%xW4NA~F5wcWiOO-d7Sb1k34NwXVQciM=_c{rd2rg=Gbe?9tUUY9
zNshPV10S^>oGP*$<=;N`$IHWc^Zs`9W&*F&88=uJV^p{Ddc}s0|Mj=xi0T+A<G9dI
z{Q7L+xQHR71b@es!7aP~RI3cTaSglU0ki<ZV+^p;anYwzMu#6HGUp4TK0OuBfHTEO
z$6)tJjYfn@hN|y5sa)XV;*!RCcPZ&J(3dYmB#w6U=qR%{mKJ;4Iy2D>Q(Jj3^qXe@
zJVEs509NWo`>}b#&d{D(?}O-V5R*d2|8kOkyoD}Q9uRNM1c+lyYI(&OcNKaj^Gz=P
z<r|M2=xpdOFQ67})oo|MA$WQ*-k@Ih6>cXtmX8l77JRGg=Rg)Ca~M;QP*z>gjCeo~
z_99g#a52sqptzzUEoVl$ou@z2wP8jMOVYJblB(1UR`V^9Q~DkCR6Z=e@#PqEBAiKt
zQTxFTw-T3HCsm}`&4?FT89Ryp@@g}W!OCasl)0%7jo;T<3QY!g`|3?sd`G=E^Cvil
zJakgf7AbT-izOT|&4ot|2%@vt63q;jFmdR=thb3Qvmei{zV{<>u>4dUzZ#HdQ$grG
zA(28vT}dewoJdd*kLv?7=^)rKcnP4H=T_s)>NL}evC+|TX=!QB)2-)9UL)i?GSGBM
zXiH!M+GV|``N!zyX|ombPP3c2M-K0Q4pp70xR-j0TTPxO(Ch?FAPv5w!**qwG=7ID
zI+pOm0(?o$<`^~-@w>}X^dyW|h?X}ubHGc18J@c;er7EN7pql|T>c5(Sg%L+*i$OL
z$?nzvj%>;%ALnpN7adBo^kaS~ziH9RqL$PdQpou0<D`<H1~N0^V(Nr``LZq|_z*Kr
zuSTVtK1ZEnn%!M*+{T#k*ZPXnME_8v2NyfluYyN3O`b1J5CyY0!fHuEoi=*xiJE{u
z@(#ly?S~(5FCq>Xv94^!z{j<qFm?I*Nuc$C7y&Le3a?LKe_@ap)Cp2yyfkx_#<1Lh
zX8rr<a6z8fz10p#C<=&!xS!d4XomLJ@fEBM!Ko#3Tw0_p^QA&RYMH!~OojPZ6#lOx
zo+E^(Ug^;(f|Yp%o;tq%7Qh7^^F_6sk;kI0VJ(c;rocFBUfgFB3d?ApArfQ90R?A>
zX{`S@rgivBUA&6ALrJ3kNV(|pQqfE4I@5INCUoq-d~%lvx=okzJX^@tFMm4|f#y%K
zR?aHtM&kI;D_m+RR6JkfE^7#DT#6un(bK>=|EbLMd(Ve`3g2;eWJY1{0&lDbq}EDw
z1#O4qpq;{30>x*9L7APc*$vac&J4vTgA3x1LPN9sj^lKJOK3}s7rV!Eno^{v)oiI5
z^e9=lxx>GY@#78gc<=9C{(^hmlb(sG8))^+7cN9V8$up*7xCF*KQ?$Ilvu)a6qV=h
zN@F%-^Nn4FXu12R{t6yPQd#_tkh<9Wt`^%v!TIZro8y?ZM7ah#7Ni(bQYn!AP`~6X
zkZM<%Qrc$VHhqVI#8WTKhY=Oxz9m=ABS|XXV1XI-Gov@VYq#gG$C@SG|LdizVqMht
z^nFeFxG!@_`2<0r%83h@^nZO6rFkUhdS1jmjd_@Jr7FUGc1-5a7&n>jeK+Jg*lgMM
z&sojgEpgW>v0|UYmtt5oXTS^Vzx<t`?V0ifI7rz3%L>UZ$!M*2%KpzuynNVAt{(>+
zA4tEXxQuHqzCiL=KowP?gmDRK8#_j9Lkg&d{yGc_cSHKxJUl$yYPZy!tCeA~Fz~{x
z7xrJM!sKARla`9)?(#8E5@5^HL!^R0(J>36Qsw|V9!z4X0M9YwO3^-1{l?l_CbYT<
z`pXzU3#Ri&Yo7SadX6R#Jq*Qz;1^`UFA#%Ykj8h^bR_^oo(Ttwk|WLn3PpYXB%Jp$
zfU1Pg>garZ$S}S))~42UYg{V$T-%o?lUjuDD|C#4x>O--UrNdewXpjRml6|~0zci?
zVJ(75<7s$pt>o9XCvngrm4n3#nZQxBln`*nLQyZD_XYJ35s#nTEAD(yEby4MDpfds
zB^*o9e@RW)>z(Nkt&3e>&!E|V`*R{zl8ykD&_zBhVd}rKO(|f&s)@{+##~VRb(e1l
z;5m-zP8~G<SgJiBEDGg+BzPl5*r_2gU7d~B!UZF_X1B?5vBFw*u;lKg?)-{8bqjS?
z9quQlvSM<SBX>Ar8b$$ZAF6Q5(ZAp&(eUL&iswecGQ}`^_s1wuZ(`MdNe|OR%>@mR
zWaj}j$4tOo!)~lv8FtjKt<7qeFbdd@U4-qh+$E2a#k_XoQOAx2eA~?4VxOXBx~WxW
zXEan|9%gxVZn%uAErGYOmC@C2{Gp?B_%ECLFAp<>4_`?jsS^h)Q}EEktN@UIlKhq2
zOOq~abY1VTM~^z2Ex1EU<tA=j;yIn{p4fi1bButl%v&K*lsnp^<X#p-W;_nW&Evi^
zGZti$UShy?juuLPbm=56ZZ;g#Q&`YabotB{IBIpqx-X}p!H9$*W0GQ`feNQ@O3is6
zn`LF3s8fY~S)Orryr%0+ToY}k{-W7TJ?(>mqE5v_;qyA~i_ECCp0-}t<yQzsT)qho
zde(SUcEnX%OOjjCXc#bxXH7eNH;oUg90@!#PRG~Tr|P7sMLga8uf6GF61@`2YMp(m
zr#wiyhlbwgBXP6(wWm~tWGA<Ee9<I76>8_;$}1>jTaAKHY9gPd4A9WCfg_*|JrWj;
zr(#o!i)vigs#9U$8Y({Xd^ekk4fgw_#bINi<boJ9cRE!B$Y7vP9bOdgv))`Sx}g_0
z0u$8cZUS?N&t*Zc#D+Nh(Zh$_594m)FTiZ}tL17*dU}6&&P)S}A*80djs5*`uvz8d
zH*;G^%&nEvDE>+ijusxdiY7mNMfq-w%J+&`Xwl)Fp2I)t%1mBxmf4M_qmG}{)4QFo
zh$={j+}BGS_w)_N8>sGng-&pT{3*U=Z(o>B=X_=9z^_0-44_^l)T{Lp2YPzwQ?Yy3
zyXk*?ZBg8PlQUIqs-3}n3a&wE0$4ehYdr2XJ0j^E)XB=zbjL78lK;e&Id_&TEMFCg
z-1JHAd>XB|zDqrUi$QHRHKNm%5l`MXP+=X?hxiECDyOn}k9x`kPp1bReHd#Zu}R<>
z^J$c&FA{Akg4~E52g;WO+P2ptCEm{7yqr)io@<(Q)h}1p;llZkQ+o$bM(^;5oqDg1
zf>tR}J;l#j4olcA&j+|}Co7T`0*IaXb$xL-Zn`z$hFQrbvU1R`T+Y$aQH6+#is}p%
zl{$nGr5DPS#i<xL6+eDRcGpkBL#r?maa&^s*VIP+o%UelW>=;%(opO+I}--wC;`FA
zqxbOF&pvY)0_XfG%<`G1I=aB`;mhFW7tx8Jp&$B;4?(b^H%rXpMbo>yb}XCG${1uG
ztVGLH==(*JmbapOp3mL#^&M>_yt$Md?84dZBlYMeV_k%##O|`Jii*lhYg(vBav*3X
z_KJJ$3hhChm5GkT#G?vNNjV?GtmOyYa5YF$_}PMq;NK2IYW~*6EH~mcM5a3snI84s
zycFdYm!Tr<mADq54!01EJt3-@5b=s(jdl6~|7(}yTDXWa@1X;qH}^umh1!Gz{VI{Z
z>dCKQ9Eg{Y=)d$y2G=z3dzW8m-0TYP&eNsYsB*WA)m4ws{Fsd#v))xrCG7JbBb&9~
zw0FFm!w?dB3%XyHFy8R)hlRDsO|28e=J=}YcHGzNu}avz9<=-7Cq<A6+Fb-H9iR0a
z^xjfo%sFu|Q~fRlD2NHR*mY%M+m6?We-MBg7a0$})SiV4Qb1DBN6CgBlL2J`v=y=}
z-r@c?>;jHr2yxji1)W4#nM~l{iuD#@Z5@mq-DfWl>ZS7qFsB>l7uJRNLYhxw@4BxO
zoKA2Gx@LuMJ3y$h<F=LO`a9ZZ(x6B<ucn5F<GbzeRv4A6uC37%J2n9vsj@fjeULl`
zR8uI~k8yq?P#mBkfF}hqv<}l{DU?=~lK&P0{N)mbmp+CuuDx@O-Kn*THV+REm)cw!
z&k4jk9~l=%^0htriHC<rG+Vm=G{!RJ48@civD14pQS;EwlZZnQl0kdoOpz8?Nj*V<
zDra=WRV+oJzRMj}3moD=%X(x~x$ZAOj*?w?^e_uEj4;$}H;yJ8LDtC0r0&R|nE5*5
zJYL>yj#h|$id}to#Sa$JNh7=NfnY_5BleKNi^}Wgvgyb83mJL>Z=j1JBbuE6W|j$y
zD4!%Qh>X3wg@X)j7T5uZ&CJ6SiPXMkB%&ZvORAH1DFEt@n8?M?pJ~LLASLk#3jHSO
zIHlw(3r{b|Xf9PVO{NDH<yrL?QbJM@T&|YqKmPF2O%u$&3eBUL(;}A5i1!n-=e3yz
zf%bi^s>g=nG@pC&As$KN@Ipz5Ps9s5We1$iaig!B`^sMJi-CGf_%PI4dM`VLe!u8W
zzFyarfzk%`Q%YbDmT^*0XIg;Yz*Sp@t<aYX2E5a`AoD@A`BYyI=4hNpxO+Dxcn+UP
zU4n4}fl8(kN?X?`)mAZ`KVic4d|XBl!yW<$St$C*lfQfS&H|Vk&J%U$Z$GaCu$Xru
zoJE@vncxCP68dvH8)mxjGgRc2-H+8Aqckscy&zcdV$wQ<_|O$<LB%r3EZu{N-iQr;
z@b&`>bc~VELqB-^JGe-d??a<GC$s{qN(E^jq@TB1x?v(h>?-WlHi4_fCKL$KEa&sJ
zVy-RoO_#W|8%e|%08h+LHAgw`@80|JtsfY_YpY*XSq(~=?<@|*z$!;mt;yPMTuKpy
z1GLXe`WQ(VK_*MEWY8@#I+`9sXW|oKHy@~2B2b2!T4W=Fa)6dl3;O}!<mph-Bo2l!
z0Tnh2R+n}|1P@{SaIg?I0HZO@jz2@0ADuXd1nB?#lX}iKbmx=|Zf!o(RL9(Ubx$p&
zKLwwmrZWagga%-{;DCvyK8nv|$_E(WHDt*JY&s64F31=WDS0GZl>L^R?Oi0+<&ku6
zm$z#~I&CWIYy=ZuU>L5nlq?G9k69q^`f;PFQ%dNaqJ;(Xtxr#r3yo_zg@Q)PZ_DO3
zMhFWz@_#$0y}!FP3(!JC_vM2p{MHIUX;Qlthz{@74SBCm&#Gi|pf23T?zweU=*`7#
zuGMG*-hxXXHSh(rpLv!(6ZeJJa`O5m)mZkl_;}K)6ZjMq7h#*vg*&rd0q}Getd7M{
z(D^J$HU7M9`zE-;G?dc#M(N>cJSGJEL=ih(0!v68tzYgjuN)5bgxw+bde%vh22(qz
z>_$pAQl)6fv5$t_vr7!PaSqknvaOxGihja05Djsy+4L{F2z_Covco*Z(@Y?XRz67c
z?p>kd7#Q(`w!-?rrgV@TEL8`6yhj1LCAV74tcb<O%5T#lFe7`|{B2J=MfMjyvRkWP
zj}zc@djJA+{RPg8`ps5IUpQkK2A0XS2u~x%|KES4cv0S@{uDKV;rGp|I|;BdNe&<3
zEX1=em<!=J^s3PQ!qO677ZM0*_C)G?2Dz3Fw|b<!FY&&2l09{|LvQ{?nj}2)Z-mhG
z;j(z=j~y9G%%*$sp)Tc_Gtv32V{hCaQ7R5d-vn1}wgoOblY}cXV;2#4mN6<(5rRHN
z@guaPE!%ei?<^nIP3e$W>_k|onXQdBemzAna0u4ss!|IJC%|a6w6WI~1K{<oEXgk(
z9xD^P*@S=Hl#PB7YeKZ+Lf?042!Ad%p<3j~S5A@!2b&~KVC&;^1H(aWid3@a1iC9d
z*rsP;q9~4VJmH&7Zr|5bm23}1u2%f$;lpX<!B9E-5rH2G{3$(3Xs{lfVlDgh#e5tZ
zo67un#B%X`mW2HFV`4gEP~IMwl<>ONj;f2Dh33KKbUAEzg|Yg?!J=z(y)CVbF7MXa
zF6T<Yj$6SNDp42qPzvF$Rx`JE@9=gh#`&O?{Zy~O6v)2_U`{jhE{aLRpB8#N&Z|?(
zU(oO_VM8=*s{8*F4Pg_+rOTJ2e|b;53cWoFY>Wkl;@LXN#ES3fZ+a>7H@DoSdx|e{
z1`bKxLb4L-y2s}iqG@7xcXuCV+yPIc$Zpsd3t)Do)3?Vr6QCh_A9|aN31LVrehZ$|
z6*GTA%do>N@f?XdSJbsUi3--NT??5a$NbIvxIIkxBqWNU;+76ql-s2b{dfz&sPaog
z6+R#-N>cDET~~Yvf`(Oz%!h@{909ze^?~$T_{Fhkg^vmt0vxHsT@ctV#X3RPpW@=I
z#N9Po`JT^To-?hf><vH99<5ruz2c#tTOr+(=Awl5b**jY^^dx`57^Z_)dV0Kq=ZQ0
za%h%2n4Ua+ngQ1b9|XmJ`Y(>x8e`ej(c(c$sjOrFeWap(GfYi8f#=o?2yI2v$ez$H
znYj&1(k-D%FuMoy>V*E3Jm&2-W`a)_0CR?Ca8vB;zPr0SQE(Nk$P@d}?$-#b1N|hW
zApYb@`J!0vUB?8Msc>B2Xvq8P=Z@IAC3pMog>lZjKFfK7d$ZlI87^C&KMerHOcaC(
z_C5yl_M|i9<kqGi@~E&eFfcSPUJRiMd~l%x-PgOQzB{}khkr2D>`{j7-fqt`oeQeG
z?NBduQTxVm8DE0F!trl}FJ{+gI`6|LWRK0Vc&3$cf>Oj;>E%^m5<jH85)Q9ofeFpr
zYiBSmdxtQ-dJbhPN%*`OgK-lE4N-@IFjvusF8zg153tHE&-+uF2|Ff4(mk^}rWAO3
zQncfRyelRW^KMv&3?Xr0;uK1yU!b}f8<Hd`BS>Q|#c>&6qKhM77uA&jW)2uNihS&y
zs5FeL*!}dOo{0Az0Rz*M=`VUgQfVUzdHS!!E;mxog>z9yUA{0d8iE!~-cw|p0+LWR
z-A#=l=Val%-^v%czWG`Axd8)~LME$?sa}O61EPQTPGzlIL$VX5Vp2Y~Ke+jxyj&%Q
z)foopN_GL!>jHCbv*}B^T75SHaU6hDW)U%|H|gHnUSm;yq}rq_$N2xO%Y1x#F$BOj
zPUk`m&_F%4jO|ZRa1HMahHt41BxuBJ2^@w3rcOQq=W;~o>&S21dyhqkHK0N*uz!3#
z*~Ra>q)O?l{P$nkV%N^5<C-tuZTtMX15hTPeQYSnANbwB&w#o8Eyep1bfp=7eC6@l
zT~S0D2y(#=&IB%0EXFm}77`(FzI_BfytQ|dG+7t%%fVMwKE~^pTPRRrk^EdKwAi6q
z%WK)2e-_$9%6HI!07NY*iRGP(yX{R0-q5S-J1tTSc!z2c9eP7~_~7ldk6e&sww&=L
zB(r`#RmeR_T;Y5U-tdz5_AZ01q+8fVQb-y=Wm?w>>#HtqDN%lFi?|@)ljiW0{n7^$
zS0WTpP&1LB1ubC>fUJTtIQ5FDSDaj-HCytMBKOfTfrM;W(9;QXYRxSzEj>wdLOeW|
z@NeE-n~Bh;3g#<r^6y!C1?&ke*mB!8w@k`;wcMn8Zyr2lHxzf4P^r~7_GJPH-UXW(
zbD@7$zmJ9_eE+XM?M&d)zwyc=Xo4IR26xXc-=}aXSLl*N*U->Vx!B`HSF_)2;_rw-
zXv~N_qT@MO6^>R@UspaUIbjAwYUplZHdw8N3czE!D(fT2zdEguSMMjse(|RLNgN!y
z$IOnC4X58W2T*FmVk!mC?KK(~1vwNtzC#GaEy~Hjq3?<&tNA#kdrGB@$5KXeD0Hyu
zvpv*Io((j^+>Ht}t+W^34i7(pc@b$Si-`Aw&alX^x7KwwAUJ`fJyZ=Ej#@cOBNeju
zR;RQUaE>R*qK*4*I60Fbk8RI7BoX=U6YQu<<o81sTa3uhwM<BYJ^i1!X5dEX$Zh2N
zjVsc^yc`lr+E{ht$2tD>H@j^pRPmc;=aXsYG}(=bn@?{N_L}w)pw}phfvmw$vFUw>
zFd-I}W*cWHsxMc$*dlmr42NMKdzc0^faOEU_<2fdf3%C1LRX6S$=7BG>v@2i*Ka0q
zbVG9zAsNM*ecpvC><lay$C8gHHjxg}kl3p4p?9Z%B3-N<K(OZ6sE+^@K*IIb+V~{?
z!vE_=sVC5UPrKW7CY_F>w2-c`dNQ;%6IB-6LL$S7P!i~n{F33;hq<YnVid3H(p$8>
zOw70NHR6taNOhJs*-m60a-1ogzVUJ9t2FOUpys394f}yUkB%99NDWO<>@SzUS1kp-
zj=FF|;aP_DLjg28ILkSViPj>K>&~fks7)$H&Hk0oYnSe0Iq0Ztb^2e!r|9Zj9p*nn
ztb6LzsZ0nl&3)cLs6v>NGWr#M>~qCHPcx`=W*L3+WLuD2(elUodW<#do;p3Gk2|K7
z!3buu&#$0{1Ox|SfVG5z>4(tKcaTEC%Ow{Z^<Cw#@~c>5oieE;qX_>}G<gAP1=65$
zQVk8FR>9IIu^(2vF@W5mb%;hPp~bMrhl+_^+u<Tow3D9zeyiV%J?wm%z=@?(#Qpr#
z5vn@+Z!c^HdFIC6ROx#Gd%^;?0h<YQM<M(N=v>Ml!vwL>*uElTdxMdMpTNJ)0&^W|
zZTRv5I5xDn3OCh6ymrf}CEn5!KhFI8J`j1Hl$NMF@J^hSq-L*0;0A`!{Drgrg3{rJ
zG;^=Jv-X;#TK%YWvEh*47uSBZaoCd+L%(vo&o})3hsvAnN#~xha>UQ9^8r<U4SFwc
zaPcO)XrQjJ?=D#1U>On>N#+{EOFwbeNqFeg9E8d8bH2dd_5~&ogUB366m^1PqTCM?
z3?N@%q6__)zd^};(iY33j3;tm%uOSz{ja!2&CVVIR?Rv57O)rkqJT`+8Yr3(b_viH
zvS6olnjXZ^N0fK<I|!7?Q9p4@Q%b5HLe{@%e4__`1?WW)XgapL_THB@Z^K*k@e#$Z
z<Bt$6AZ)-wwKE_tXw_NQ8sCwdDd8sm@ZLB2V3GQ+=HT0FGqb8cP(nlu;o>Ds{CSbV
zbN<trT>&K#<|jr}&m50=5CZvnHhiaqwub@dp`1}X1hkyOQ{mg4+)Uo8)mz=zVBBYC
z)k^o*6Bw=We1b4s*o90Jk=_9`QeyIySA}S4{m7t2;5a_47w%1t3;Zg4If0?*{U3Mr
z@drQQeKLsFqdAZ9|IziAVL`6j8a6DUlpu{#($dnM3eqYF3P?8sk|HfgNOuTQ(g;X{
zL8o-52uLFxN(%3Im~(&MyZ2h}ark4+KWk3pe(rmWYn<0Lrh>c&Yw2Dl;k04@m;3bM
z5N>{PxGspPDc7KJrisWbH^Vo-y+e!Kf6&n0{sxI(mV6lckPN+7Cp`E4-hC@V!*Pcu
zs$XR3dPTw7N%l~&M&e=G?OTSg(8iBzMN;`kLDjLZbgB3BSeGKS`!o8cxN`c(`DXsl
zk3VWpMzp<B@BK_9WyMLm4a6mcx`r^EJO@nye)ju3`YiKs?D$@td1!xi=BTt7cnu3`
zGu1C2{eOK|jLRNR#D$zjPy6vOKcRu>^|_hVFqCFi2U;6D&Ijg|S7<RTqdPi^EoN3v
zkTyJAH!$-}1W`5){Vji~_|O}9s;0hqcj4n=Te>@)ivL31Cnf-oTFgZXMh<kO&$v`>
z@R+0v*i1fz(h>b;1B?RMu4rE8v%pt*C(u)IqkJBr+@SojHZxWY0BsAAz&2;<ADmJ-
zE%#ED)3LXdnDvl912v50?>)W(3n8s4>l?`8kHJt!IAMZJ25ZaS6c0%5{x=!r2O+b1
zDO%h!E~&~LUD2uly7ScRe<xq5-UX+npfQwkg7YT}q%7?V#rQHdaXN+I^A&rG>oL8U
z$dsgh_15Yc{YDBoYy5sTR&A&1m54E?dzYJ`4eNj10L8E+I14g#5dkQiQkX6Y1GyU|
zFhM=Jx#q2!_h9yGT-5&ZfcvkPCM$-DX$IK1`y*MmoiaFmKp-4~zYkf_pLynBrO$>^
zCwK|oZ{$mQ6*eUcn!oiqdFDmcB^2Tv?O0bpO(|mKWr2Y{<pah4lwlm#J$yS(<Z?{N
zE!Xyq^!?3n?XP><ADNcH=gl+a=zBeDqt(%Mie~$}>2N8K%TPI}9nDF^PSEUn$2wzr
zfP0D0<IwI6Q!Lzo0M609di83+8HkTOqt@Wq<mBY&t)D-oAr~XFzrUaO-4qHAC%b;#
zRAC^<b--p5;M}oOI}WpWJYm-~*AA6ucJ{>LGDMIM17e7%$v`oN!cJ%fqyoSj_k)QB
z&4Y%~K9r-;NQ71Tk3R<2eD1_ezgPQx@*M(3M8=w)JX(;Pd8$F`51&Nxga7%!l)M_6
zt{K1wMgwyM*qoYw?qlS~$ryRalNit~ENfmYKI%c~>S&TB?0cu<R!G8<XXafvE1nlh
z?kGy;B@}n8hQt{?hrH$c|MPXNnhvs!ywR<!xEXeym;CE*I@)inlICad`K&^=az@HX
zp;ioR2E3hwk?D@#@G|}H+5%@iN=e(FV(w=M`ruuT$t=A5^;hHj@})OWGg2YSKGkC^
zV{~Zrp=fk;6)E7y6dsq`J^d|L@q1Qx(vdZOE-i2-x4d|<;%OIKvz2-=YKRE>LkleZ
z&ZG(6jr#&rJ%v!r@x9#CQB;g36YPkj^|jMKK_EOQq(Vo#A}==%i1H1fqk0PO|Nq{r
zd?HdS3lKp*z~Du*!h*p51Te%Zt#3YS&|C6qeElc@xS;ph*@<nrVFk3+pn_SMHDf`c
z@C;@X!D?f$1Rnmr|L>oM36N`MturjHp@VzspS#Y53e>VdsCP^+I<OMDBY+VRMOi@(
zJ*Oxrqv*}%GYPuzUt9&RFiHB-H1_HS3@7QK#e4u=Abq$_!>fp!l1yHCX@a(Th-wHI
zd7zkhr9F@9_AClq+@hxJ+}zx#+uq)nXy+V0>U_?F&aq|LP0{0hD?uixtn8=2LCRmJ
z=Uu3W2`GW+b0N{sT}+%-OMg#~9KX2i(6zR;5f-16oA)upo7rR)m1%=<`7&1c?a>N<
zo+<)^KZ3T?S@4_;!qww<4(tc2tyy1qE(EEoFY;S|mp^B-@u=1L;-nDxtv6~ycn<q~
z$2>lBOYls2q3UQ0YKU0Tm(PFJcBe^#6mgE*(nv3~CyI$b^5b<wmOh=O+@ZKin@IKp
zF8?dX&j!gvR1+@g$1H(vP(~D9#ShnXhlj~Z+NC<Q4So8TN4Vbq@AgBgLwH8x8eR&&
zWCa@PrWB}Vn<a|Z=`n>8p!H4e8qz9~{?=$8onR~Zo&$#F1WKhlw<&G8s48ZEN~`C3
zTpGdaP>54JhH4#s#ctUYP%l)^lZ|S?ut~qXI<K2hKb$apgGYtH+WSBh_XB*WwtI9#
zP@l-nHK_O;6s;it5EUZHgHRx9J@}P3jM~{b8$GvqRq|Wt@ZbtoVI-y1NZmMf=MfW2
zW`$LMPr5&doo%@O{-yErbA3cagvN_3Y;Y>*x14#zBlT_tEXn()1!u=({=rtp9R3I?
z@BZTt2`2H!eZ0#fTg7*}&Sw@zwnv<Prfq5b)A@fRvWt@>MVkLbWNHHP<3LmJBXPo&
z(IN_fE|eqgHotYcV=!O$op0^Q>YMk^@EjLTq}A(<IQhRyeYZ?TRyVFdD@H~=dK0!J
z2)Owb7vF}oe9ge&esD$a9vr+Rr>c_@c38kf@C|snWW+eF)GRFWX3+;KEFXOU`_dyn
zrtuC#*uPN$n4NU@YkKMO$V&%E7`^Ha=Ja~4)2f;_q+ihl(DFM^qk|LiDR|X*ONo4n
zf>K1=B5po_pj$HyIQpEJa&K>(zijHI0TGC>i_rSiJD&cL6R2kuSNqNZb|FR2)70@1
z>~%D;CLLNC%*NTdqA2%BM~AWp(l!lx@3T(5Q8(g_0yQ7!-HQ15UPo}t{o4S_ICZYw
zE&K{M*AToiL6_x)%CJK)wyknpcINmXp93a^g_`_dtxxriK!f!AZLmW^`epX_>;b*#
z<%u$CWZVlRE0#@ZInJvqC3e0#b9JzI?QUaZ<30lUfiIepRH<ZHygiKo6uC^e=P^-n
zjEtAY6#pLo_CIipVNz)c4xH3%SpqvA;q(^BGz1*Y_nh124sacpE1ek{6K4qA^49Rs
zGj1|Gtj7Ae&$Al$3pPbernPx4)b#0&V1(kCU~MmRy;_#fhLQ$nrc&+5t#a?+Gh0jj
zg3Bkx{*E8@XfS0_cybkQ;vNAz>CK;O=5>$F2ng%WZtHyFGVS^0u_KEq{H-=4NawS_
zF8y8Tgf3xPJoxvss=Vbx{aevnR<|!r^0xJ_xf2<Ww2;g8pg^Wi{uL|I`t<9|lZY{y
z8rKLxhTS%~vw8O)?94r5z3i!-2h>HiTLr&GG9>X!{NY=)exI$Yg}lMP!&b~I7W=8B
zhky2Z?wKd_Zm?`rn7Obk@pmqzi_oH|3xA7nICRNP>422Msjn{|d<yve=MO3RQrH<0
zEUbXZ+?7Ov31}_1bw;`^90tKF+c?(H!cf6&Gs*WTW`Vo9ukU$EN-|517|hi{VvDsa
zIsl>{=irak$OqH0$fJpxaYy?)a}i*yzD~+8;>$lkkKH86zD4?$?Z0R@pMs@+DpKLz
z!Atuv`gdeo16Y-b50w!C(5@_Ced@f41Io?XZjazh|H?q-g}jY2UDLVdkB(}Dj<m_|
z9UnYWD?5HS$bRxFR~qGbr6&WygJ8ofln^Vhj)2&YsPKb<Y9Kv>2iQTH?%Fw*IWd6A
z%eLvBmGiU{gS}r61ZNP0C=2cO??4F4>+-yDP0d1|*0-TSf`jGsm%h6;$8+REyTJ!d
z_XP7${6^`e7GM)NdhpW9<uOE)0%b2n%{Ssz)~r4#Oy&1AGV2ca$8s5Zmt+a~!mlxd
z$UPUj6Js<N?yv)y;0r}QO}uGB*jSoAhEY(8pXRu9$3UWBxWoP~%#Ve;Yc)%an#>j8
zC2a{XiBs9+H#<t5SBZueiq>YjX5vFtm_MLy3^g+_#IxZGIy28r1D|*3VXcjg(Ey&a
zn9&<{#y@H{Upf=@pNW<DdY5A<A2UWf`^`8ymuGCKtWqeyXY`QWGjNhbr4%oe2_lR}
zGnb~~{}{#~%GXgM7bwrRb34XRk@}Zxze3oS6zLcsykBSt&8Dum%?waBhQH$7J`k|Q
zD;F%qyrKW}3__`z&S84mDbRXHlagu^QZ_RH2Ym!@ziHP~GamYvq>hh_!#Hl-=&RJD
zFs?tmp!D5tr8}i{!^`r!7k9oq#k*+i+xKc)G6CDthCI#ARPnDgQeCJe9vtS0ha#a{
z^Cd#7Ahn}M8P!Q5O7~9*T%1UeBt$Tsst*kcB0$H`Y~Tl#__fPQmO>1=kA~H;{MMKZ
zJ5$#IookWSH)+nDa*Vp%&NWEB*dmzudJpw~W(wqfl*3OGL=V!2M(@77toml{iVp_E
z8m2nOhB^L=Z!DOiCSB{6x{R@d;yw_(gs;y=PK48^q`ODn<>%+u54-O|bDLgy1$3gC
zk7hw0FxK4P98>tMD<QTAJZ!@Ub8Elg-4|JOTZbMIcw_}$A1~74St}XiId3cv@jj!!
zib0-ALYP6Tm5dlmvolwwt`I^YCx(*kF1uS9^<OJg&~KqJx*}VFf_lt2%le8PQ&_Wx
zNyxzwxRZ;qI*J}!xKKDcYCO8p9AVzpe$4)Cl*hh9ms_t;F?bT@C0XTeh+U<nh=c`W
zsFXfdV9dC~;BKx%4C-}1xVf#zt{lPNH@5K+=$+PYeH;d0*gpu<y*z8JlMMSV;gK^i
zKsGTsP+t}$-wSY<CbqRT@|wc4<*8;WMOy<Qw~*zT;|*trS&PPag>Z}n)1Q~7zdg4y
zviF6oh#{a5ZHk^fIYRVE$huqqP@tX<wZ79_3s#5G`t_A>X3|3eNbE(8nX*lrarklf
z9SpUgl<#kFDVSs&4cpz!nVdi9!K6)hAZJm}IA=?k)N^&DfUI=BMx4cQsk_C7>uy)e
z9i~&SCEUb6Il%J|00;Fv!7-8&jXI)si`RYmwHwW>GU~uJ!o2T(c5aDTb=z8%gJ*o3
z^e1)aDZ+4pilNrr2oY3JV4NO*0XpJbn1uGNr+$GIr03Db?**lAnZLr=eJLV(Z|`@4
zy3r1!75zkFRb$j^^PVI|OzCmDW0T1%AtG0)MAY$lh`xtfXpg$AZyG)C)ib6i8#@1}
zO)<nh3+yBG*viOTaN_{;x7e#t6;aetJ^DRQ7E-KK$eq{lop-M_>QjFLhyeTmbruvg
z3hbcD^7WC@%W1?DZcbL&P}D+%)?@wzz2cxlYAzhwn_@2gtoP&n554qYkb~a%7Z$l{
zn*sUnq;eruQDX{Qlyk1j5jZnX()_}|oA6w6Bxi`PU+H0^^>BsubLrtmNl}4}uY1wE
zwuO++@=#Or?Y7Py1jyvV`>u39s5s(Ck2Yj0iHX!O@~h-?cK4%QQ&zp2Pdn;|eFxoI
znr~%Gp+u&`9Ncw5BuJ<)1lX)xn~Z-h=Xl&K2mT;_!a2p!ojux2`BXHnm_TdRr+Dq?
zOVrkH$4gkG^AKz;>bcu-tE|a#{bUzPB%C`WVoQ)FLMX{A1?ufXCU1Ay&d`Ub|Ir!%
zI>S`*XX{;s-8{+27fb!ylE0tpDqjd#SuauDYTH3Y2zkO(p!@Wz<xmFeeSRjp^>OYw
z+$4D#1I5|aE)#N|W#W{tOy}N172)OKdrRNs9j}1SOnTd#KgbziylLxU@N#yj%>FT&
zU8$-+-NkvaK0fdZVFKAh{RM9oM2$Db3z|Sv@qKA&G5J~8yH2pOJw1lU(dPNn!aKdm
z!k017oPRx1gbss-wBaGWJs&PN2B9#u7@oVp+rC6zIlIwOhs=iVMBd9kG5N<!_j*9P
zV|+&~?2uLR^Ff(++>ILUDEc({Ag-E|&96=Fs)fSNv#wCxn)*|>{F*Q_ZhsMezb6f=
zisAFv+a2-Tmhs6mm%Rnx1oA<-XVE~k0h#uAqDFa^Lc4~vQb<p7U`5*;Og%_!dKiU^
znndt+UkT>2^|=bK9vxqKWPv5~YZ$#MAZ6Q(D!hd;4lX!Z51X79WlJX^?F-mDx4B6-
znO9o&N)|;UsUj)(EYg{Y+r~1WY9nc8ur$79J}u?EI>^|;Q)NQTuK5su3?!5wBmS3U
zjxp}V+Y2^Y#@w18btp=A;2MHX+>If94C9TPi9fgSJ*iFi8JCDR#S`EIknwmmKG|~c
zBjDQ2)DMSX^pwBpu`SIiN)b%Tq39vmJ<vkYm?iK9Mk*ca)G<2x4%VOaw7)XVoz%e^
z5(2LRg&$q9^QMXI2S*3%6ULTp!k8;b7l4NtAi!R_)DrizN}6BdX5f_)znJG{EGRxG
z_{*TNQmQ%4K3w5O{o#>>x5J6K`g;t%{9CBdp1n5k&5lIFA?i*MeWEIVj=;Kb)Du<m
z0m?vRFJ6i?JN@bD-8;0W%1NHrljEQgL9hS1u&@xz6aqs8FE6jMx@+qcuule=ko4Yy
zBi(BrObR8>fAqVLQLWCtC`ej=F|y(>=z5u#mgZLZI?{Kf;UPY{HT-?KTO(?pd^mCY
zJA&k&HAVw`_fzC)?gm_KA&_z%x=D!nmhkPD<JXNpX(orWI_#NjuO%H>LK__M?QYVg
z2X|9S$}OSl54;&&5aqDg$!LvVZH9ILPXN2j!&ywMXNVnd04T{-D8MT6DTH&iorOSv
z%~;Rzd?#|EZy1i>{+Cr)rKI^+o#)B=>YdVF7$C6#ISx?i&TLba$PGZ4wh-iq=f$x6
zNIv!z=XEo4GXCi3`I~Z5KgYr{gHmijkEfrwD%wzOF>r_Py!K!^hS(Lq$RIaLaBw$3
z`)Tx)Rp^W(Wsg00cB=STo(h2Z2lCq)WlSdOUrEAbKY2au1IK8Tcm}P*)lb)+&X^m`
zzs5z^k)}D82H^hYeFbut<;QFr59TFU*McoG9>ig7w|RtrU`!ZK)YueN`}nce)gkRi
zo656~gy&_+F<-GocEk$*;+fnsuiHF_n~x*c*17ji%IV*}dXhpF1Uv!CO874|wf;#t
zO=@;b<>IF)-ENi5+9t<J=P-`X+n)=@?pZ6BAR@arHP*`S#`n}(hRWdtktfs3Rwi|V
z?ROqSyI+-0Uw50&WqVfr1q>_C<@a4w5ZEzE%+=uozhjHt+9-$g!f2EL$hH*-0KsAT
zbS7bNk8&s;w*4wWH^a^Ub6grVBDF4SLoU<OR_1C-?XhOeNgE!X7TE~SQm3*?!D0?l
z^3#}`>4swY+DV|~pUkr7J*;(mA-(b*N<Y5mQ=imRBB(?KM)tJ4m}UTmP&b`O|7|OS
zcO~6v=CO6U$N4^h62bE%1vnHJbxDyV#)6!$&VM${`0{D!qo*&LX%_4>Dm(6eljg4g
zydVAFn>UIKvikC#Q9P6V;@g3TN=r*y_G`wOFNoFByW!@^^Pz)!kz3fgMMC$SS!92+
z2|YNP9z9c?_QbL?;c>REWO8+NRfUQ&=v@D+b66pX&L@XOZ2JZD&n<Xp0NV@L%Wzr|
z${8|0O!vm$@$rB0e5Q^A^PVP~-@^Pny~h}9SeFyK1716%XT^jzm`w(Vw-Yfbr$a&a
z?XgBPO1!8EZ4e~$4Ad>U*j;`qpa*)5nt=h0y^O`@mic`3e8Qa^elOhZ9QWz&`j=aJ
z!66|r1*}QSYin#WW=nM=itAF2!0dG_X?f3s(vZ+y&6EkdV)o%P3M4hPM<t!#-rpzg
zDc1SS123vFU*GVzJH~)v)03tdw9`08nw_1VRV|f!Q$0I$qzBkJr>B#;tG2Doq=Sya
zU$@FIu^aN(u_-WInse3FWfOmuX@M|+|MuwQq${4cC)T)O(b_dFu`0wE-;Uo$p<ZS%
zeA_+gs`0ZK18kO`1q%8rGEY(eZS6XoI|ZAOx;fClgVf0=XUe>~ptYVoclG1z`(oFu
z>d=m=Wf4;;z=83P1Vxvvy6bT>XpH7e%`U6lXHrSZatV8`Hnnt@6koU0=rX*2s%0O)
zhJ5a*ed01WT3C7=I*PFIPX-t6HgUrKHQ&49Wle7>>1hsX`^LVp&u+u;e`2Zwijo!h
zUn&~c?%>I|)i5DLI6xriN8E&;3E9t~@07n_z_&xtrv7C`>2K1KYrFeLN|TnKUe<`x
z#rNt2<+db@ii?i+UDe@@7CXCfVaf_#>X}QDPJg0-S<;UpS#8p&ov#kIv4Q0Q4_hZl
zG%y>1cs>TPJ$~M^Et#&FKmgC;uF8PB`O$X>-IPImo80#O(SYV*HgZE4kJ36?vob<o
zSl>!tT=w3`yo>waPW?^vVKN71sakCkm784LO89mtD7DkZpo4!4c@KF>d_W#WQNG>3
z<4y18VDf+gpNmk)Q#duCS9~LS1)qw7IQ`lT`!h=HGhe_L4&RGQ);v*!myAH5-rmV=
zI<zN<>N1SlrPjyyVbnJO6etS>S5~luD>e^UW(@*0$}A-GHX}1LI~WtX*%@fp4S65I
z1`xG4PmnK>P-grbA-h?r4<1T)!Tz`c!@@EVHQ(gsQq%7_%P1%)i1=>k=z>r+P?*1_
zZ-w&RZ_7QkUdJdZ5UwdgTb%^O8U-5|h&!F1d6v@w6_^fcZz52LVw~Ykv~gl$V(4!$
z@>P}<#jGb>THkTG^5&=;5ult+FRpkPt%4d0hQU`pKAIBWUFx7Y;(fXWwd8jOcAAl@
z<dpUwkTlsr=kY@jq^k~(?5m{e%<6fO3nvjN6BGHmtZ}RwifV<vRO#^FavwlRGQUSH
zlLiDB&J{OBoGd<!rJ1^hOGp4OsZ_r3dC*GvgAuTG{$3D{NgLQ#!Ye2@Zg%So$S)-g
zrP94N1s1}DUd;c=0#LmkJ}3@2+A|2f=v*eP?nz@rHeV2D-Vk@=^MtifES@Y>%-ki5
zo;ky0yV6o@7cIlup`FUdcLfPMxurK1^?fb)&G$3$d|+aO|0+V<^H{wKtuE&X0C9iG
zx}1li9iGl#wNJYoHa8(gaIa)Z)DGIMvN7qYaOe=%#)`>4MtrK%nfv`5sE{2dsyeta
zIGkA}Ui^xs3CLzJYM*mFb(^yiFI|BC!C{LewV2<^x<mH1m9a4Ag$v4C@8X5gb2ZfY
zLxj9ACp8Xqc@DMrA+P4s17j_wS{&h(og`)n3B6B$;@z(_q$yCwCARK|AjQ+oaDPL@
zx+E7)`5yXIZO%$NKnXYl&x6gHFi}O2XNQV2*iat%5Wg@mFb#Y(=B~I!7jR!UICuGW
zOPcJO<Nikq4eyA9^AM5PYw3iS%aX(?Ui`@GFQ-Qrcs|7}W(&zrUo3dx)g&O{eLhjf
zXfsjPT8M|CX|Us3nl)l8<AvAN_kJQmh83~iyPS2S5D1l^oe-5=!^Gega^20^zB;a%
z(6w8xZ9L~{cf-r9@J)-f!eJd3hVtymfILmQIjXiN-bx}$+eHoee+D`qnM&NoaJ4Ir
zK0Ir7X)nRpp<kC)54#P{|9`(xt33!?p$!iX%+g1tBpFER={rtAxwoWzL7<vXD!3gA
z4J7txk33$akt9D?)!GFv=)*2EI_f=CO3PG1bTB0MTPXgPgy*=0U;u&8)m`JSIA<)V
zOtP;Se5PH6)OuhmOtcsmlLckUulwoD;XVZ4UA2fJr)?h0C`;zrB&MbMC{~LU;Ds{!
z?aF%yDF=aqnmbJFrA|9@vPM2vNn0b8C}lM{$&1y0faCzi9-||T59$Pz2Cch=r7Sxo
z&hsQVb_+f4p6~=?it<@cRs?CH-(@L{!4C11XRK>;gBe?PnMP^(rz>(V!IFw^lQv_{
z-Zs0q$e2`Y0wX^rB-x|vx@c3EETy5jtFxZ~)AV}h*!$D7d{_%HP<WVjS>dg?D4{}j
zLxWW7*cVIP?6J3&S`te=pVu4uR5cI(f>7faA=&4=Ic1q>IwJfE547QT*kqjS+Nt{b
zH?PyxFfb|VUn{Y|wVkdIVBs$L<7sB|3UESg{EL=@OGbT&PGy5anU>L`as}J-Q{bf#
zk{vToD>Wa~%PWxG_~>tS&7gVpux2IxdD;5+l*=ixb!d=b`T&BcDw4}4KEubf?$m|X
zu)fV6cPQd_VlG_-eN+gk@p}Q%?zwM{4ncxt3RvZzzrKAp=q_ssW!kEx4SRb4GQh>}
zXG5u#5E^E?9oprEsyK6^#!+5k3!K+wLv~U-V?&+kn%1U_<ipgm@!GY0_J#B@s|+wk
z2L(UBK9(vtv*TF{h!MF;{#m&}GpHniTPW|ak0IemHHeT=Ee_b`{eL)-8;CJVS6MC&
zhJ<Ir9nf$%@vw%ciap@zX)8rXskhb}&Qj$3KouQ>5Un?6pU_Dwo|}1{ee0u}u=Vy|
z)1qO6j^+*x`kzev6VG;ODt6gFWE|QIM`F5#xQ}?^y=Zb=t0>EFw?hIjySD#UEvH_M
z(8!(*9$%AVF=Q$yUT=a?g*4^c7CwNpb@9>W-8VnpKmG3S-x-==vM$)XHV7T=lWVwt
zeZ84ciX<>(MBn`HtdW2F6DgXKPMU=hO<7tD^q4jrT=Zeae_9cIt(VSAy+Ag82rHqJ
zHFQ8h@!~C|R|iXzPQU1vVG19zD{p)n%O4CJtwr$P)SjliVY!)(9V;G^i_Aw4ZlzbO
zhGn*&LD4NBS2h<W59tZ?)5ASMlDw|;U(?dkm@5!X0+fB3z0fv@EbIrtHZ1q?xg^*h
zMj#3q{zY_JZ>YqyWTaQ`1W}Z<{(O5>ng_Zn2g8Yg*)|iGSpy!K+UiF2Ob)ofG+3Vw
zBJO#%Cm&;fJpAmsXI3CYP8%po%DrT75&_`h^#}YXec{*P&5Bu%@Mm?I`oj=@rwMP}
zC`i!U!H@9sK9qYnbnL0AsV&9466SJh?u7MYS}ydQocGJz00YI3(k}B>2;kWI3~U`i
z{o@50wojI%!v+}VB_&)jsj=5wW46K7MIy=Oa+Up%X89Y0#aJ#GFpPaAM0+YI)cM<A
z3<msC`@Udfmq?C|*3X~T{mpa6_oulGL3eeOyd`a2oBP23V)I)Hm=Z>h1AB4)X!uY}
z(>@ZPJD5w~`n)COe9z4A)@PFlV6WT94jkri*autd(F$EXch(XF3rc7GlJSWT#$c86
z=he51pPOwjQJ26c@;t1E>9n5cr6xtGee=cL`jH{sNm6;Ux`?ektnNpI_uxU@c<A9-
z4?ut2bQ=!Bb=56*o_O?C%SBMP98|DB0vlQX8ow%pidX=o8?VHeyw@f`y811aCB~vI
zJ4)8RJ&fX3ph(Fqt8LxmW3h5;v}KxEMMT?V^jxi@U7+c?7a{END)BPapL;F+_JDYt
z#+;M64$f=-=F728<b9ag`8oeNU>_{eDK%o^`+Q=dvX@c8EJN&fP5Ir@nz>hM2WnZo
z!fI0Ije3Hc;%*FLw@hc`yISf|ypC6I8R_F#5r#071WI6#a;vbHoq*c;E=k9^+lvzh
zNW5!lX<J+0tNXeGi17PyP&gzo_IaL!qWwpU4jLZvb-U@_2*p7!r2pqTh}5iy2?3F<
zl<eS5A0S)_<UFNOQJA<)e_x(-7SeXS4wgEL$LIL0@wW^9Ir)*v$?Rg3^-inN3nnf@
zuCSvFEPy=YQPK3G7p88Y8H@%n^IIs|1+B+v;y6I#0KcHC%lNW<lj?*EwZr|(HSa;$
z%854SGSghM>mEp*IP#H2D^w8+00X@eY2Q1+Vj4OIEh)NYDgk@wvfc0O<Cn4^<Ss0#
zzFo?$@Et~V`Oxq;1$dU;slvnx9I0+^D}IA45i0uycQ9LrR9Q6bJk+4-g__(QV+UZ+
zM<y(>R<9vPX60KHy?TYkfO_~&83`j#T!#MlGJ=801Ih@{gEk(A&?k$zMzgBt(Qx|?
zFTwnCaQkQTBXfDp@tsv@lw!XzUeH3YvW;hk?{1J!5e22nci6}L>Hj~({*pR+`)w!L
z0fMf~X``P0=%aW=Em@_a6VdwWa4)4;ZliXT_}ynX{gzZuLSB6?PmdXXISa}ur_T8v
zf4F+tlNRajor#KT)1hF@Af4+msmcQTs2Y6Uj0NMYzn1XoQC!Ltw@C8&NzoQLt>^`u
z$B_%KAr5{2c(y-IxxZQjJ+|(E!@Y7THQtnJXb-qQcnj$21ePrZ)W%j$yW`SoCzFI&
zd2!}<KzsT0Z~Ot0OY@@CO#SrgVMF4{>yqZuZ*QE=RWtZU20atIZ}2{B4i?V8FMq$;
z)9L2!pRz42k1rVcj+Z?CYJlAnGLQ-<<d3|BBnwVqPTK83UU3-qt?#2&Y!{CwWEApQ
z4l=w${{l|SyJ;S~L0XSIY~%f@3JI|0XVHNOB$RPRLqmgS{EnH8uv^o{Uaf^OKC!X;
zse3jMk!_DoF;H{~`sW;_o)EPe{9pMAet4-j=#MvQq(iE8G;B;#$Vt;vIYK@l*-1Y8
z_oa#t#=*P4)5#|?q-4^(eg{?G^EXKtwfLkzz0{Z+?e@v^cfJ$B)v)bn;d@;lv-9*U
z?q3YR-!>rc5VSc_$?wJAEw_>zx8L}T6A9HY&b$6>N}<o2;A>)LW|l8>EwE91_aV_g
z`K@0my%lBn-@kwTSFqR_43vx>>$LM|`&m)|h<oe#O#A}gyH756#cd18lxjE~iZx0_
ztmnfL;VvTSSy~~bT)flOG9ZOb$}VpdCF?>72@(Xmg<+gVlqbNPTf!I}@gr#!`ukH?
zC{yAN05zw{b&uVCI0{~wK-26pt-aLwyfJXU(>Q-DW!$myUSnabV>$;@WNdKaV)?^n
z(XAR|GhXpTZsRKucGxh?;9A(y5RV%kokD|#&PRv;mj*$>Z-vL_!0@6it<6jG@y7H1
zx^T>r?I#JGx`C2pHYy%P9e4$Z)c;7}5{suZMSCvB&h0w9mrtZ`m2NeFKLayr0o=sT
zTatg(*V5rlJw$qSXl?T+`N%(=gHA{Bz~)#VeCTt>MnRbiJGQq-UzRf=S9<CjxYDh|
zvX{2tT=RR4BU)pF2KvJ0_no&3eyY0sX8UWY$l+EoN0uxlfuk?KV`={115WC-G$Uyg
zv_&uBarp3$@M}%+ajAxsi};xW2(9w`7$Tl2;;`vM^Q+{<<f^zUC&#|!gYTA=;nleq
zBj&1ac2S7QSCrT<fG;qE`GS&n7qW{8Qix}67OQ3k(7cv}Y4HHhz0Aydf+K2k*mk7$
zqrG}mzgY<Htd9=s)Z?wM&|4-vfhHe@EHx)5K6@_wv+tEw2{dyC>|J9y>*oE?L^jGb
zs-mobs)IR4-1g__6-X@pv#N^q?J8?9Y^h*K-=B2aca-Y-oPPViH1c*&P}yJ%t)x!k
zz-)o<z&!WSh3h$fzxcnMLW|Ne)RV-Nl<ZMB<!pdu-4NBJ$+kt!z1=$0?+O-=Kr;pM
zUr*v??4GmvfEYRmuOdT%a2bk1nQGM4X-bY;q0Oq6!WkbkEnyekFyx|ln7I&%ORg;A
zsyQfh{Ct}!ataTSv+Xot$aO=JROsy0K$%rm32xsk31S1v0AUdaKFi#@hpoyg!)qE^
zL*NKMVFn|}Im!Sa;rBcm5`w(bx!;kB;NB$@NLO48nSjw6ose7L;*OQSKJQ2Q^|w8E
zlSY=1*1iw_vJg1eEVavEnPC?iFL8`23UYl3NKF-a{r$t=Mpknc-t1F`H*{Oa@#;x6
z{hE`qE)MkQ{u>^ebS9r5Hag4YQuk-hQG<Ufycz?Tjp|sIh2(Fd-%l2j2DN~ml&9=g
z9H*}KCxMOr_r7=E1;$Ezb+vXQIxl;Zph?AXIfcC*P1u1vgcj<u{Dr{4X6Wwo<)i4#
z!fSK>3LSr6gfAfnHa@)-uyn35=^#XqXz=v6P<f$kkH<~3#aprS>%Y{d$QS;{a`ll8
znIC+d^YI#?Pv5V#T~91t)}nc|T4~9C`?sVxw(!bPlKf`kOQ_13Sy>gIu^3|b@F`r3
z#U5W&=9E+;{uB|GYwy3p9o^g>6fTg!`ODqRCG1xK`ooRLWcAXl+os^dJQdTGQb3DY
zSSx~;Rn+6h;>wjx7*P#Ako;?MFLvuQwWQb7k~b-A=gHlwJLxUt@Nz~>y&W?|Ff|In
zx1*7em@Z5m$HKU5fMRLY`hw%T!OyFB3x+HrHIB>ok(EKF_JA-v$7KFX{Sr>BAW^H-
zf4iF75%1!BYh%Au1NOf<YhtT0Er#=R>A`C#&V>eVc(WkCkBQUgSXorzqszE{cI?eC
zUKz4K+b$fQw^7Zu@O}FGA3}N#zD*{)5KbGD`~za<6l)^|a-g=vTmOMj9s&Vl1=128
zUqOw%His^Yt2rZ}7c<F7a#d<kvk6nen%Vo~VisC`lY1d#x3N{}vpUN&r9e-~E|cAG
zn`vpi=2}}@i|~v2ZE0ZHYAlsjFre3v^?eUZ>X2zkDLXl&fFX;4_#UXx*DoK|-ik{S
zuu%k~mfA%A`np~7JA!P&J`i4~)h0enPLwW!VLZdL<PkC)ligzvQ`{1EpC)RP-~=6`
zHH1tsHRpKd6Qq2=s@#kQr4Pa9yik?g$-5nY4h8~3QEaW4*z1~K8wZl=AU<W7!S3M9
z@t45Qw@h>vVD%gtPc<D$#+w}F_JKkYbD-*wcWSz2W^Db^G4N^QBYsd_P`B2vC*{d7
z7NMso*P~6a#LE_D`DMA}g$D}R{RJ{IMqNgv?zt%~mUH(D2H2yLb|@IBiEKwOZ9P<8
zTlp*KZC?mg#t^^2D8b3reW)|M{%f7_`3Y(3f)%Tzq~sfnC~AMoeGiouibWUQR)N#)
zTElR{6t*XvZ^o!_VNK`-Rr&-Q39TYC5C7GflKVQZc{=cVu-9QzqDt;Wm6%wd-fc3e
z7gw9`;(SlP3BG%hlw#%fR#Q!l(Eid@ulsC^Y#A-m>u&e+!?`5^rTg%>=$W(1H?K>S
zZ_|NtZ!>!Qclo4|y*oKF(%`XPXFZbO2&JQ>(fe0VK`T#h2=>JpPlAm_%Fs;l6^xdd
zDuIJICK2}*O}ZfJrzcs}B#kcrd4>n+ee|7ma<iWn=L`IPNkNiD%@LgUzpIPA&SV}_
zY^2iRVT{b?M+r~uR*j7syzN7FJH~Q*{gKnzv2hRW0?s!X$14G9f7k=v(uGg?h7EyF
zTl^N~3rnzN&G`))E+ugr2Uw)?hSk{5K-Nd9geU7HyPT4c!$P1%s`$NWr`Fy=HA|Q=
zw7kn)h>mV9t)vLcI-ce4i07cN7c~l^iWFA#=TO*s_i22B9=h<1sIUopUwZZJHle4;
z#y$)9aWh8lMNq3i4HRdQI@yLa03iW^jH$OpIhGNdrf3|%eD?c`3Hmu@<vvvDU>!1c
zTp%21Tjz5AH5~M~SdT+{4`lIdShNzfq(wgB(N}7G%{D7eF<fsCaR|n9>atFHKtE+7
zYpu+d-_cNUlhW6F5}oN|U{*@|X8w<`{xi6V)$2YLGkLxz@+6$K`hX-rB6a$-_=*0@
z)9<gq>MIa(%<SpBrh3|y$d(_M9<wv~ZtN3GFroQm5yq91rTp|0U=<vK@fMMv7S4dx
z5OkGww2kGRej!l;HI~lVxr%8OImLKzis#8Gx|IID%&m+{Lc6dA&9b7pdD^Y8Qm;G_
zN3@hv?J2+%_$>!pSbXIppi!!40#5`w<m@OodBQ#h1&BWKHCZqjP~>{;O>vPfNCsm8
zILMXvUd&JPVv9N@ATW8+`0}$CF*%d`Iv-UldCm~JRD)&X0gA-5<uk8U8y=61TKoBX
zPy>+xM%$~bBOUHfe+SQqt_t!>bCIoUFIbY=Gmc>9&C}(+R#d%GJ>$h&zJnDaGz`K)
z_)W=F)(G6wnosZmIsH9;p*llSXD$k|vh?8*gg+4z64Fhvnbni$dlTz?1f;}FCx!6}
zy9m~G>4qS5yW(#_Tm!`a1#cTyFlbF1v($R)b_Xwof-Gwjj;g`(ZaO(8GmJ;X1pp3Y
zrxML{L?;ht2w!kR4aA;5=I?LD%qjdL^*vn1N+MQynwk$Gm%9@FA(!d!NlEY349K$<
z>3)6e0&3S5yQr<%vdV^9P$rG&qwW=u@$dLy<e#|=A;DAGWtYGHbD9@Ku&}W+V9XIn
znkf2`^Y`kI9N&Y^6)2l8=81i&Bq>cW>kCR8Pl6a%#v46N<O6t5Ps*#raITj;0>o2}
z#96bHDeks@>LEW%d={lfOXmkeW7SwNF9Zq{k<nsMQ_3BO0FFhKn|TKhgt;uZBtois
z2A*b3FUONq98Q(46G4k_jrwv+GWR)AxT@MNo4H`hhuUAJe=V*1%JSE_vL2wWr+3`Z
zV~_42FxXIF$8W%U4}OS}FJC6pemo0&r6n%RhW(6XrDe_z)(q0rjV37e;@i(wA&HTY
z!biHX6mq+chIGP#k3;8EB!seXOkNcwR>G>&dTjS*=8~A1dF_)a5dOxnml3IW_SwZ@
zea5*mbA(jC!XldxF_bcD??&v<f5IP?+>d_S1FCbslRZ7X!2OL1K~(ZVj*EaJ%ZsjV
zz)0QWTm+Iz!M7Liahine`<M~s=BOOV#UF?WX-<z!S`Vtf7NxDaq=~uSQB_rS0y6p8
zP)?ZT;}6E}u>SV^?@Ne@$m!_od$1x%L(?!^=;8sH%0}FVL#5mN3vh4)w=Tb9s(aP#
zg|<#_oOqZjDOru+S3&I!rbkCy82+=~m)csSS;pm>d9DLG=Z@}UXe>N*&i)ITRX&Qr
zSNHo%94RHW1*cSgi_LH6KfPaB7$CaHUn>X|&N2PBj_`k?`)_Z7ogG<p8RhZ*8QfbH
zAkD7CHU!jGTlU({5%>;M@8&^U_uS%s<yYyGRqsoZl$)?3X+e2H6%i7mk9s?2F`vTN
zDcbruWPI%Rif+j-eNVE7Inqy*8ueNP%>r2P-!sbVM+pSCJOriUQ(@TNq2w`vt2uT6
z)Cx$03P%=m53pE|mq-s7wv^F{Y7v961LoI;iHym``t_**L$M|0Rz=Bv8E0YO=2opJ
z06~mEZsnn(BA)-pWcr97y72!G=6(n+BW~Tui);qE=r&7dqpB_fEKL??tA_MZbxD93
zs+Wn&P?ljs#JO*xC%8(mR^JtybCW&^!}u5?Qz?`{e$0RmWe8%ltgkdSNm8a1!V~#;
zWMr=V`^237W0mgCJKDM!u&)UV)#=sQQSQRGWX3tWe?ch|N%lX;TBHsSE8fEcc|@8F
zk@P%{MXK;b+wO}~`q;eY$&`&vsFmM2a!Mgq5w6rcGKuSN_Ar?}x1d?_cA-xwS92qB
zuzO>b0V|xbqe2%)-)8ddRa!$f|3WbBHT0>}3&_P!u@5BY9@NDe!|N;?1V}Io1o!d>
zinOQ=_uR6yea+gFXVO@>e=lxI{X5Ni<d9&dVr{w>2WH+5MO0b{1_s6>SQ|D$ypFrx
zIM)SI4zE=R8lznUde{EH-Y@nZ^@7J7Mv{KX!jDc2c9um$NcL>6=Ytre3IzdW={H}~
zx#fL!BeSG3^FGeN#|ynF--knu@%tH3_$4r$gjKI%{})xwNf|105k1PQxafs6|EOvY
zHEXuF?LMew2j5{av6A+t7weRb7&E%M()<D)CQ_MwODO{k(It!QqWtXg;!jCSk;vP1
zx(0$dVYM1fNDLy*sM&p0NhDW@WREj`6}Fqb1W4{(o|)9E;YL>E;*Ax!n;|E$N{V>x
zy%6{4D8?1uks7uL4Ooaih8jUA1^+V{QRj6ox{_a#wu1|K*VR_xiY=qq$@(SSbp}i^
zPB^XFhVCajA4rCSQ+@J8-%}QWkT~YIwr<OWkUCNl6r~$1<jt&K?d@d-^E6qqvdcV!
zoba;b^g|2s9lv+;2iS791WJ?>Cg5u{(#p4fR=+$hd<uoQ!pLokb=UetfeMc!2V_}H
zKFi|uJ9sS2c8GV6uN*E^O9x!t%tFbqRV@LkhpogLjN98k?;?3Rv;8-ysB9kbuD`uO
zKp?$tmmfQPb~bmG!f}?(ckAhHC85JC=JR!Hwm3Mb4I8##P7j=78X=R5omZg@&w<Fn
z4(#wBHLfv2gJo(C$UTbj&k2nk#Y2MSXwvyF1}I5r$B)vydFu{rZkYaiHet%H3Io}o
z%L#F6kk00Bj?wESkv@F;VwwtL#o!8?zmSzBptt(oX=+|((wQL%QF~!Nz;dMF_O^1~
zkv&Ner&g6;a0H1L?uqKyxiauN2YvW3OvwF@_1toEj>3J+M%do91N2HBxSP$)Nc}>B
zgB3}JeOq4&8^oB-d0(S!QP|Wv20KYUTsiR?k9xEHJYj_!?bI!^SLj168D?q{A~}n8
zS5e()(%Y83sh9k@|5v=&KNeYUQG!F)lNZPK-}y%ZPW575?@RNptE%I2k>5W+-*<}u
z`QUL4AI!()Axb=fgi%$70~KDWbcotU(tJmRecZ1RUGTX5+a-Ey>PR<sQo`=)Qj_yy
zY~^+s;o*9Dp9Pp4GRo&Inya2aBeMC1JrC7~|BAMnY|)P_1`Yl0UOW^q>XCG1zziKp
zm6Y8Jy~jZro>E&n1vh1BW{dk)tvLuplD2`!n4R_OTOa?+n&>~l?`<@p{z08ZI#P=7
z5k0-_{(RUQwbggFb?<=*&}MyFvM0u#8GoG(l2AfO*)<KsD%T`K1YtpE0+N1B$q%2c
zsB<O#V=Q1c;Nx&NZHRx2rL0My@HPC$kcr#SNnGyioCK0?HubIn-eXHhvV6@v2u*!a
zsxm~@$!5OP5=d|5&nySxWZ6J$4D?z1P@{9LhOiXVA_8&z0$a-%B8xKif+b;FR)B*;
zjmS2)DZFp}3_S?`S`;=n_x|G7V4*Bkd<~Dk&3luNW;S%o^Jzfhv>)ZvVv+@j4#*Ru
zSsRNGmf;!Q2jQNco<$$uyrxNic>$AP15A>(mXV+;?)4|ynKCR#oS<}nYea6J2kQgg
zXrYQva-y0H74~AUkqSwyJvWN^`8^YRQ_`uhpdjvd6Hw8^`1L-d{t&T5q5ts&TWO%y
zZ-}jKJa67Tdj)_K{uDg`$YzVJ_aTty9(2J<ag1a%CP26W=w(nY0;Ljh>e|zxTrJsT
zZ|@LHp%cdj*&(Uw|1-9}C{0EYtJ`STn?~PDtG}r!Ak4zWesj9^9q8}ovI@kiA(!N=
zK0dAfsNIvPOGV_e!;^TWM?y9#fEh<S_Y@`3qBs2^lHtGhaX_D(X&V;v4Z>X|Ypz-2
zVsF?SW04aRRzvW-BbqUz$j#N2z_w&>dmF!}dhDAjk&xSK&CZxx5p!lQ;A9PDMm8h#
zy<9LL4uKqD_~Hn>Gz>PR0gucmSrMe^J2}`AOj7#`W+oxX@a2xvRk~|rL`TB>(0lU-
z(abm{F|iC>b_tY&ApH9=PpMr#{89>LEVYR;XXhM4iS<*P)p5F~0|CmI!f+V`B;W*p
z5%vKp{}I@7Od!)*e1v)Jfw5b>M(s_f$STZ|OT(Y+=L1vNhzJP_lPKlBEUuI|bWj!F
zP+6Yme0@L1xJqnkhREZl9OttnQ>xE<|7OlD1Y1Ot5fISIa6kUVE{-Mq>H8#e5!EiD
zg-TgnJhA*3`?(S*93P`p%%hM)wVD)o*&Uv6^PxpGvEFD#9K>?=iD@~^Tb^ci^PlJ_
zKO~w@HX_q4K9V7jY%4tQ{ZIlm!AT4a3(>cYw5aL4=y$*T&>!xcYQ?do`zgWiTyPFD
zGAv+$f|+B!UhNl}ob=%YEym^!SPZO&B~bDT2V-sf&mRFc3E*}RB{s3AuP>shxjAS}
zGX}~fIV2uNBi56kWm9yf{*^wa(9UclTDb_cHW{#{WDs1~ex(iuB&2jp4Y-u)O%Cg3
zaYaHuPmX&36qIH>C01xgkDY#3tS8;gjC)%$=RdiLbRtwqoBDfQ)w)k?__hRDWPT)(
zJiZ8hTTSuS+bRXmU*{7uD1$7(misvV9=AE6#ndP6zaT00S`*rQCh+NiB+);RBpM+U
zQ;pH{>*k%--Ji5oOWQ*(Z{R-21;=qnFAsoAa0VEYmaUE9OId|}jL1d+gw7^k>j7nq
z&~;)v5X}Zi6%#l)#DvItPSaFz-pubaFEqGJ`22(wn{myz1NlOsaf4(IqsQU&F4ogE
ziAX#lgsz#n=G>BykN{3p7EItg{9hL|yIotHXih~yRQ&)ir%Ov90gO(vkeGk)uft85
zKuj{Fymmd}Nc+tgP+_AMYKxM7lu+e&;^2h+if-^DiF$pvfYkThnF7><OPR7yR_4#g
zMXwzu^`1w(V!IO>$RJSqMBQjB-At9=>IuO}{3x^bh-now2Y+z98ppFoBR<-tLHHU3
z^nU_dFnJWt58RLIkAAp2_WoF&a@_nYR{TUyczL4lMEJAorfXf_@>&4;;j3ksE8ML=
zKl|D`fH;}K%8!0mDZg7=hj&-Af@y`JRsUvRYA;BxKVvIr{IH0SV|DyN_s_f5I?eqq
zwQUAkjG~^f?9X>aknWUOMXHG6jiRHJ`!3N#qcCeB><2^+{exg(dJb*)-8ue(A`I_C
zZ*3SH1-3seH&bb5Y3IkZ2!={>SXdV_gqLgbYM-p$NURY3#VMP7KWimKp@r@mW;_-;
z={H%MU!*x)DS3CZqmN#^=}#-y!4xK@o$cNV1R#JkS@>-*vz&|!;wVs=k0MmiguoJk
zQh%62slu;lWrBCUz8xv6fpfEY@pWCw2s(y)M~FfI{$lRnp_vkVE}X&}0kBO0=kL=R
z7w*=Umg`JdW9%zjfX$La@-5u@%ul~$V^)>A0J|0B-#6ZC0>=2^I!||{)<`^5Gc2J!
z#|G=W@)yAn|91uu$xt%aSFk_teB?)YT#cJCrUzg40y+%}IweE;=t~5&b}*VaOun{X
z8qM?C0ezf#WxUiFiCtMFmsR>eTOvHi#4%2}9DU{?k{;Yk;j=;Uf^(Ife1zBs5r;u4
zmOZf!d+TE%ElSseCA`p=mIm9sP2`G}LO`H=4Zdjlq@Y%m@wW{+51VG+3XV^>muMQ;
zl7RD!L@7FSsFRkH`qFuESa5JOWr##0G>qF18~d=_$3H=YnMB9>77Tw0LTK`6w}pC&
zI5cCMhWqc%*b0b6xjmLVUZGDCdGSG_V5I&thgX#{u6E(ahJa9f2XKYsM&Oh^d)u(#
zp3i5Xowa#VaO!KJ*^`!Y^&``(yB3W_Slj0f5FA6~K2b&tOI#I!z=-Y6fp?tra*B&h
z`<0JNs5p*i@a^NdGt5^`0Flwd&*)C>Zl<KDm;wLv)LMyc1|@l?I-~M)sgUcp3(Gw=
z(ZjX2_f6OPlTp!=cV5lJKX*%t^d%jW0UKB@3fh%^xF&28wP}m1tE*|28n)bOYig`H
zBQ83200e$t4XnOKi($_hSee=|WS&^DrA{X+b8D~z=h}*W@D8oRn$>P$ve%J=T7ihH
zQx&%lvCia^gL<kAypJBJ5dUKjU`b08Mu3lvOjdIE^(Ed~Qpe~9IxLv)arG#|aFQDl
z9PjOb*%Y?1sep*a7o6}#81k(7pACOwPMql~vq9Y&d3VK`Nwr9u-)Mlx{WVy4fo=02
zPi>uqVZJgbWt~ej5GX3Do+VFURUj>a|LtBwVj9{6s~q}&QHs35vItoaBupl#C?kXq
z*F-@Z8sGkQm0Ij=-T9e2g(6KsCOSL<NLdyB*mswl0pD|nwzhi3Ru(TUac*_!1F`>0
zu0KMyNf-w-2DecJ!gXMWmBkJu%nc$54@1dR!pMf6I*-pLm5|Mh@jK-P(r%O8)*~(|
zgQTw4?Ro<(=Mm~nv*VBaDIvx(D2!~PL@K=D9<4Qd6ROHHEcn^+uik4pZl10?vK<)4
zKZ23}aEamFQo1{o6cHsfAsjxn0o#olR%q_~%z8C0qu1Oqxc_F9+I$!I;;<NO5qELk
zR(yTrtLeEmavdw!Sf!)vd%SM>eMn$Smw#%}Qj1j*#aMg;>GjLeJi*zr3fw_VcV;Dj
zX8h%p@$YXLoAm{zQecn$f`?WNms$7q<6i~Eg&nL))TBJd^Tt-<ZLZOgzN1U`;xW+U
zE;13&(3R`^-aY;e!=$8pKdGS~Om88dC^1fDT1X$^AMnMvW&!bWuPu_qU*mav3ZFUJ
zt-FGXX%TdB<vqQ)@+>N!^zoc$`@RCs82`l?t+<Qz%+KdV;WN`-%X>Q+@s&7_%E^89
zJtj~WQXYw)eEAcuV6_OREUNy&uEai6jY0kdntAb5+9f>!96x~wd`;z(FW&^9ZP~`V
z+}zwp4<F*g5-ke;M$}%2K@|*+`SR0iPbB7G5w{ttc{ybDqoMjJ{}?EoZH4acoM3!m
z6{?-}bg^mzzoW%mOVB+n*RaOdB+tr%xPH+*=aWVbTGDkc5v>2nx44ALieQ~2_p!gA
zqYB{vd(gOoJU&HfWNgjU+umg{sZ8*OXm_|@rSd2ByvaGqUP*>cr#f~rQvCY;fix>$
zB%(>!p<&27;3locOHbq_1<H9qqHIvo-Bp7QF(<XWY0J7}ZCyAkEyem{)MeVPFXq;@
z?Ck6yQusy~5kr24<*_{D)o^i*Z|~s=r~vp%{8{eVCQJ(L8v5GX+w*^mXy3Sb#f07}
zzhOEP3IwZIF^C~f5_hjsO+L$k$^nUu`&*<@e_Qv947-uh>Q*Z5y~BRi^H@&ZNuwXO
zwZBO@^lCKuZX%C3&9MQ^U=ik2U4kN7jw~x(;ONTHr6fw~^~a}N*T)EFE9odK@5#z4
zIA{GUgeQK?=Fc33Gsgw4$}cT#HEKTkfn`D%#aJ7-?C3kIxoL`|<^vj})zS1Nxz*5B
zMsZkaJs@=GpPc%(6PkV>d0p)MytM`KsBRE-eu(iVv@&Uqa)(Q9bWG+{JapL_a!{R)
z_br?AD_p2R(uIevgDKFel!nLK8wZJx_TTQ|!==UGwToQ9N14u)*zNNS7D|ziTBd$z
zr1s_RZ&yXbG#%kQ*`lUlKEB91on$m=I(n%*P6^?U!e$7xJX~P|&OIGarer10Ps>#)
zXOK)pP<Bm;`!e+5Ma}{6le|^wM_}#K#h3LEu1rHZ#g7a72%1iw4!JbDPg|MG-GcQb
zEqI)VwWl!BzFLksR`b!^&`9YgJZ9>!qr+`?AxL!;iC+fY-JJLh1%64KH@PH+H5J<K
z|Ju;-Z~r`sb-LZIactSnKM}Aq86rJgU$n?eTlV4-iLduNco8n{yu(?DPzZ&a&W7E8
z{SUsuU5{yb>1}pYs5<sHPlt%z|CY|7-6mMMi!)s$eYB@p?UUz%F_8Wge-=%?28Equ
ze2=8yZx>aI82a_94vI0z_~8$ASUHCN*UaD5ZGS}?jvN2?0KzNnsmu#V1Q@x9Q=ZnJ
zKMara9{Rv9QWi!NrMaKhO*PYDdai~9{|{8@xo_|{dTXV93(#AaUVswhm^pKXJ(LH+
zl{1Bu1Z`_}0<<FL)^C`??TjSP6oftfq=G4oU!+<38cI?ck;|r?u>@BGWQbJg;tr>{
zEeWe>X1RPUB~vnWRk85Kj8!g|&tSd(wTP`5L*gpc@C-G+1$TUV54?NPzgc!BHbVI`
zzha8O>_-uQ1~#ZiCC5M>iL^&1Y<7!w?Pe^&r^cC6xJ-JM2b`yR+1!p%TifydlI>Q|
z2KTE4&YhR`HbTEt2nY(G$z7yfL?$9<)0A4^!%Gwt^V6B??C(o{l$u${bycN)D=9%X
z>ZT;=jGPf|DXd3Ss7SWV;vN4qloTj_^A+I4O^sqhzLq)b`v=e(=0Iee%&U+Ynshhl
zvL7xJLzkVhHy2)O#Qi341$~k%MLr_B^#JpS{Zx1;%adNp;lg|AG`}8}XWU%;?IRvW
zFV+*M@%^O?cN98l9Z)Xp5Cg*b{UMSs15Q65)l<Zu0gX35T#WDrm(X-ZMnxIM`d@B`
zHJUb=LUlj?o&YBxUuSD7ca^=nAAV+Q_l@QN+)PMICLdujD{tDWg^+Xhni=?(`Cn{J
zr$Bz%1H_8J+J#3qu8NL|DWoW6lZF-^j<NV(Y<N;3=<(<F{*N~!t5jb&DWmDJs}xt^
zI#iIyLgG^-%|F#)<XbMsfzw@2=!urfXnLekfp2;bWsImMPOw!}16hHwW&6(8;Fl@l
z*xeLbzMZ691uow$`pYrjiVin306Cr5qvS*B=Th!|xcxI)qZsyt3yE<iHn#ZziHdI<
zt8|$(o@?1Op-Z?q@9_su&l+C5#nrXBnWd$txNknGJ$jT6$pQNy_%%Pr8C##ofJA0Q
zDO9|Fzc-$fr19RcbtE^;vyATI@xKA+U27YkV^oeBBSaj@kd-IpRUb&$+ubEZr?w_A
zG|XB5OV7fhpdU71d$!w<s_cSijeGHW7xi7;{d&#kX;40&og0IxS*K*qy*`Zog>8c)
ziaR*eax~)NDZ5KON+9?UQHd7ccEP15sVFmZx+Q&!bdC~u_u*K~oti(eM5SJLtr;q<
zC`}V3)>F`9Th>G>(JNnJ^gBAV%>=^}F6n#<8~QGxRKg?G_Fnri{bn!yZ0Pw()qY!Y
zBwD?r$$-9`ZxVFr_A0TFku){m!4r-GB+R#Ur_z_EUgtLn35^795=u*PAdPx_BG)E{
zA7!ZP?EM?4WY5KNB*WW=yr5fw%akYVSxsEBUY}}A<hcyhvnQ<bb*{9(X1iH5hF;3B
zMEL<M8BupQhcjSUzuU=HVc2{nlh0~HAFdQRu-armA~ktjv34bE3T!)jhmU3sw=4<N
zTz$;-X8|)$_Ou^&>iJv_R%!KVZfXj&LA}}x=t7W<R`=y960kTh!0-BpjzE?VRDi?#
z!Uh^|qW<?*pnvA(Q^@9)DLXZ8h6uWU`;mEWpp2OO0|L~Hj54UQRa*#s0MHK(56>ZF
zRMhMYb`qp7VVPs{EEn!&%sh6I(rEPcRw9?nq?R>+2le$_{ct)M7Zjkwh*NaduK9d-
zb?RF|ah;;I_pd@Oh<pMvBg^Ckdt}T1!`4+mWtDDifsgKxZYco;rMp8Aqy+_~r39oE
z>6DTZmF`j`#Q>yHkp?M|77#%|y8nF`?|-i|vzBW{$6-d!ci#Q({p=?g2|6&5%Ua_@
zW9s`KGDkZ>H^AAXQR{J?TG{+05I_3hq@ge(3a(`6fXev8#6*T+L#B3)a$r~#?2?ff
zk4}v|yT0MrU_!NAbo{dbGODZk&Zl{zEQrf%0j=@f*WkT$F@~7FVw<SQH2U=eJ+LzY
zOX0;4+1h>bCtLkuqo*HbevqB7C7191-7T);B=Q=??|h9kEOT2`$G2DZq;R7Zw25Ut
z#UXra?==ut6DZGJ%dLO8_bP8iYvTSoArf@7xHsFia7mu%S_<8pdB#oTp7Jk24ZOW$
z^FuWD<{>K4@C=;}|8`fL(ibF(<H^?-_CkOattW>Q50vAieGicAzceGX3Tsx4oIAwm
z#Yi`$uZjOx&AU|h)`jya7#W7Kx@8l+TFJZH*R#xMA2T0LyNJq23B}S$YsdUJA@caE
zd)A~MeB2Cit>nYPef+C6cdJjYuCEJ3b-lH|F!ABThv(k!0OSf)+Lxgu{Q<IgC?(y6
zyc3XKp+E30v^Cvmb_!~>vK8byeg~JH>YaB_mUV~Y9-C1^(j8o(9sFt}-lo0St}H;c
z3uHiBIka!pY#Q^5ox6r^w@2@M4z~(?|4yc{|NZNWGi6KLJXLn{?jz#-WhS<ZLmaBr
zd)gOpyu{I7zC(m<X;<bT;G+<%K|KKbaYcau-eJo$=S_ub|M>u2DyF{u$RmRm@3OOb
z?@H8rGq<cRBhRUVB>dq+Dxx}d{qfvO<c+!|dT%SIM+}ohhVvvD_IGx0!sd5k-{r9A
z&QsQc6gd#b-_V2?%NbUM+*%C>W?Fx}er6wHS*k2Q#yj*kbz^lu7(h=_qTiu053)^{
zDG9Br5^Qu1CX8>DhZ>AILvkcx7I5}x|1>^uQgWeziD19fW<(}3HJ?VPwv2mH&ZN2t
z+ifx$N?P=2MWuA&Hr>vo-I;@8wye;Mqj$-FD5kc}v>c;c3r0A^U)wOAq##5nM*>eg
zFE>{!2#XN4ViTW`qYjl9b8}^4!Z^vn77PG#08Y9UIi!MsQxMhcdRX$w&--L9Y5PWu
zEq?OMIZ(OdH4ByH*m@`uXDl@K!4IK_2hbOPhdrxs)Wx@R+Y!7Ald4(?u|YLp?jS)M
zDF`+}H(lC?$$tUSjO}R*ibM+`7#G?K9}Q7K=;cGq7sgj6t>y6!?+D%?aewVA(z?me
zr%Jr2f3A?+olEhcCibRd^u{%luX<3{s<BVD4bZQ&{4gOxGV9EmLodl&%wsR{Ht{25
zZIIBJQCqonHWBkqn{t0>cnW1r3CRj+FRgzHm2S=3tiQ~8N`AV|MLpns<6XjL*qz%#
z;!db`llLu3GHW?J^+qJm2dq4z3SbdLiOI|RIn1q_Ei=CxtveMhiuPajevsnnUvLUl
z1xVEju!|=fTsIrBnR=pV1n3LVlr(Uo=tN??Dlr=?aaty><d{`n>1*b^a+=BOu=q&O
z%DL<(b48a4#D#MyX2J#<*|zr%dNnrK!d8?_!_%70>4}Ppin_tabUfuKYL=&tx0~`c
zQo==w^W=e+c@}7a++sX8aDyxW7b6kKH=P8fvUNuE^GkPvA|Yy0g<qVM4$pfDAQ#H=
zfJ@Lpv{q<ABOlV7TTPn*Fx@_467*3VfN+~vs#w1#50<`V-1|pR!5E-|iT+i=1o#?q
zRaJVzualKozV5a+JeuG=T~%i7c2%wbTuseO?ZH7NKACZ<VH}*NLN@a)IpTt0l*f~I
zWxdoS9xYW^#O@Uziy<)7H&*zKMotE~BV?~2=-%GZ)T@0G1l?1$hg*0cTM56JE63iK
zs$(S>W|#K%>}R}v+?0l@(|M0?kg2y*&hrYVmvKOl9gUKrqNT+LQY3Fo;8s_7J_{LY
zKpEDsb}4bkdluh)V$Dj|o?57447re-UQd#F_?1+VVcPKles}v<>~n)RpT#o?5#r{S
z^~nC(?#ydtd7ayBcNjxWMa7B3cilyvA66iHU<VLAjF6dNsg%$e&;rWLkXi@MC&|fn
zdZA9kPvK_dR8(Rf-j#<>j0?<-3yq-@S^y&&;Rx~?voK1BN{c;42@a0XV!fjH5}Zih
zU*fOtuFsA1<+8fWbzB0nMs_B%mqDU=u-6@QyF+@Cm2>9*($Y0ZAv2cL{OO!u^yqdz
zh&u>&y4+UR7$~4d-}%&Fyv!_2M|g|~E`7$JP<fRenLq^Xd<E0$j}+JB9`nvBtgC#^
zJh<z^JuXbU;eZaEYcA{uF3Fp&7{C@N?(!>P7AYx{`9~575KN&M`9z|j3lpr{9D8$K
z*tBxQ0mBhVElC(Y{`$-CZbwpJWt3g#%NbjFBH%kvp6>mSCLvOuB@e3g&MB8juz8Eo
z24M>O`zA)=VO6Ms?Adpbv1AWcJBs}sN)UhgBM_UsT+*S3g?)=PX+kh2S`YD6q&+tV
z_TQ;HtA#>xKkLPA9oNysJgD;_29`QM_m5;kS71)2x;y=pcf^_-Ri?!mmY*~*=O#>L
zEZVPkQa`)@OZc9&KSvn$m#fry2-9?_^`+nOQM?!suvHLb+VU+~#@Co5VpotG-3VH}
z)?#dW<_mnTx&i2e_)TT}^!P0gFohTEd=JG>^W8A!HChurL7m6Aj`%a>kU_aj7I|?g
z-$q*!k|8f{cu2?)us5ur`W%8*s}J~fQPa=sAjvfgyzrj*Of2ECU|%sLM<lv}00Q+E
zaf6+Cc|&Mkw@`0WI0`KU;ij8S8VCbBokxx(MICocU9DjG{R~(<tq(CJ{B#frJdNT`
z>`(#B8ge5dcIK<IgXgWwtCGy+(1gOEn?ZccE8NR3C2+K?v>>4nNaP9(CU}APymQQ_
zVLoNw?t4S%_=$g2VjkQBkm%Yd%DSW3gex|{Z&$YEy;cnboE*_+I*-7M(0Dg<RSq3=
z?ID0*p%zQ!ZY&OZKC0?S%v79l?B9lEO5FR$Urt{;9C9CdD_!VgyZD6`J+lH$xs?%)
z`AKP<%=x(3v7{d=f<dyx2@&fIqjlb|^y#>_UIoV6vGH5z^Dal#KJyy7@A55k_l&*P
zQ}es0o#+SyPcg}wH)|F(EFL%*^W<;w>9_HxOT}qCy%+*}UuG}vNsLeCbc6Rr(`5WQ
zF@M^_k?OFzkW%e5Azhh{3(<HgB93F=TQ5_&>6{H?bPQv=07LL7@tvo1d~`azFuX&>
zw-e`WU2*#QmOWew*5A>fbY+(0;}z_CX@7OPo=PQ1tXYHd?rZ0%K+|yEK_dKn<!i4?
z0;1DLc8qvesjzQY{F(?WF+=vIz6V|(%}HFC8=LWh>Dib_*=!OU=3#99s*M}5ar9$~
zMX%)fW?1Rz?-aD!si(X@+=HlqZ`~Oa-*0;^4aXr-7)E?_hQ2^niT{4hP~dG%CK4l#
zc$3Dvw%`KXhHRr!IxU{qIGS|s4)1ut0wq+**yHw#QaH@)lw_46qU*aJ1}1ZDpJw#>
z$A&*Kd^;(&=h5xHiYD6E77n?iQaAl3MoA%ZRMCs+oO@4HtJU(zRjUA>8!8}qZgse7
z%rZNPYtOnR-+?{Wf*!fC{tnf$*Sf&0c^lQSjQwP56SM_-03`R@jD13~XD`YsJ*ps6
zy%#ShaddNF3KE?o!g)W5Gly{VnFtcUw9*dik>%FrdMICFS3pzZr!u>F55`?g(wi!U
z>x{dwMNtaz-aSSP;DZlweds!XfUhu+DTm>cIO|lcQ3OsD?3|o*-$u$Uu#sRIE=@mI
zXtW^F(p=aZ&0`B9FA$a4K`OuvH-o*1Cjo{1^!7a_j-k5jj|_~#81mhJa(QKCJ5foF
zHS#hm@mD^u)GE*{_7^86lV&<GXWIU*u$+<x5I)c{-gcK9t<eMhc|@Jxu`jnyZuoMD
zwtPQI_-D-}{8ue+ED6Z54D!liWgkNbiJU)i9VRAbFSNhxW@Riik<8#yraDKeK6$LP
zgeJnx&CPQcuDN(~yxONe8w<V9YxAjiPc>%r0(6ysiE_ol+#E&fAB_4Qlm{pc7C?f-
z<|e6faK*beCEotR_8%_*EV%J<0Xew#`Kk|xD^{(YXiMwCIU15O7Dea`iom!$5KWj)
z^T`qDtVMWU*z{&=3JpV?WX|NHF9S~kX&#@|4c*y252lukZ%pdzVOoS}wbV{4^+YZ?
z@GNB%av(gAET1PWDg&q!tT-$6Dr^p5<GavIEJLBuuOj~e4Op+Y(CWX2{ihuzRp|Xa
zTW<%1#SJU@V<tfi#e-IQgx7emV%Ds33*~Ch>2p}yCdlcP4~=OJzIa|i0X}vu3}M+B
z!o3rc6TniV)DVcCosSvvrJ<q8ztrRdNh%MdWUiK6(-$=PsOs8XPWujH!ASInlajh?
z4wHoM$w+h!r_g;y>)ibw;(1c%!DFB6_gYqdAWS)<U?ns!r_ZAeVNE+Gb3meVDi7xg
zrk6C+Xu<xBMMTflo*_k8>1l-`=|M-@Q`aPzzPs+ql|DbS%J%K-q5JfCWL5L@HCwya
zR;fFtl;qV1vP^dC$Js>A4ZkLuCy6FVb8>QuKsM;MKBrNWu?jV~7d$+<uidze^A<cc
zs$9OQ^&aHM)uw-QFEO(L<Amk0H~swW3&|5^@G5-aifR20gkH^1R}T=-??FTDw(tJ5
z#r*yF__(r_GY0w~3IF~4mS5_)91-7?{x#{*-v?R7Fo`6hpTG|YFe&;S_d#T24!EF@
z&@D0>UdaR>Q{?{UzWO3{GbFbE98T)n)!6Zy2d`MMyDkT8-ZqKAddX)&8|_V^mDsSe
zHiHBYSbuqbwUiO@Nf#}%6wW!n2A&%&*r5pQxU|;6|3resv1>=?CN};uQb)zz7BU}J
z?Cb+UCmWVxB;JCtcuS2>MShhWGGzugmKWfPJz3A@5OPYIcKz0V63r`Vw929`y)aU8
zT?FPPkCo+G+|Ix&4jas;!JOdSwb#yBdKLO>4yrnD-E7T+oUfxZHO<zGd85<ac|vZ-
z1p%mlVI)I9`#FPdRw~{e(`Xh6q>D9O3y{a^B;8qh?^-pVTV)e3fW5J~N>XevxYg}?
zYYWZ&O?m^E1Qy<`_cdZLvuJ@?0yp$gbwsR{es;~ICq^x$MXx{+39zW2Gv)P(-GE{J
z3W?j>dz;+l7wiGNn05OG9+@61<F_}_q=i42)ZKe&>BB4ChV?2S;Kk|N8-gs@=_Dl`
z(X-7#s1s<N6z!!W{aYOD<VSG`wBG;I6Y{^R9Y)M))7+2y)cFuzX*7|v6??KVB()ln
z@Et)TaapIu$%JLLK}!2a@&Xu%G5g+}sPp!0RHzUY`cytOHPs{?V5mtMD7+!dn}mIb
zIg=8>Ge2FX-o}t8YLj#>FhM_S+a0{Uxv$CI^IiQ?Vp3eu*G8`4byiGQZF@@h9x;m|
zt&17@_KKJ_1%|pme=)=@>w6KWO$4QkzF07W-X#$pQc~x4?3K9pU<xe}&rf*p^A$kU
z%Duo=aQitXW>z}EB{v(ImQ~hs9{=0|hJ*uRh*R#0U*6z>HwJEW!US`vMbvZ6LXch4
zU?L4zyU4#{?a4U2Up=1sj<fp^wfDXw`l%Iq?-lnq8zC+*;AEEjtAk*^V78^WZl*WR
z53~8Jg4gdKO$9Xq!LjVqr<(VTE|qBS|9`~l!*-jcbqO-{Ym9V`q3{0q8MTOs_6(=Y
zlY4wsLzUle^&&7gTC$0OZQotJ+Y#0U!|D}ipUD+jau-KmIWes%zc!rIScJ_%KbSza
z#z#*BWUB~#J$A|`v3Rw+Qm{h06r|bg0#ML7LYj*${OwCG5s;o^wxPl3vcD4MECE8F
zs@0657mIfgo4gBpI9Oml8F{gW)(8z)<=y9(EW4Q<1{~BU=hKJ@M#&BD2S|N}L0y&7
z36@SsRQvN`T&8=-UhzR)!gX@yS%7Zr)5&LhiGBl`xvbMR(a*4Za^)IF1A@}ql}(Y<
zspEbPj7Z_~hGxj$Ks==LKgMr^>Xd4*qBc};+b@pCu+o}vb7@4k;=E>0Q*y5^Ds3l|
z!CO^eo~0OEwgR&pAR?i&i-(g@+$|h#1qWd!zxUHCDobdi;|=C+yk76IiAzmtIE+(h
zaZ;XrArGOix+pR?M`(M(1H(i53^z>ts<V6uN|1jBZs`!_4SQNAuj6)}(jLSR@BWS&
zw=vL8MKjK^qP-w<(fW}L^4V}Cevf(1{Tx%V#T4dsbU&&e{RIzaM}1*NAE~AIq9gMw
z&_5mlN7-)w_yV?ao6lo)B~X3ttVG`KfkdRe4_ab0<K0jA9XtE?bbpA8BaKHfSz?Cd
z$LjZA-<_%Wz_&89E-fr!VEW}Tpi#O2iG&Xo0w_vGPe&~cq6I#xL&iprbG0kZkdfBB
z6nL%^Mma|cPIuUB^71-R4`a2ZR-%s1R{*Hm%-WLe4#uAsBbo_uAZ?(=he$LMe-Tm+
z{$bC}j?25dd9h^V9FWu0#aY6V;GDW>_9j8v$H^8+bXUwWC>y#z9b@Cv1bdIt#o_%u
zKq3*N(Vs&6QEjYf<Ifp|WRkmX`k7xStrSS)T~feT9?ZOd*G0U>ZU7tIgA~cEs)PNq
z!DB1FQ}>zA0v^5tOaOws{QPUY-Pli|Ix{~-IF^B7Xj=}}8ihJ}c!=C9f{O1veY@Yz
zDhsd-)K{}Rn#cxGKZ&<LB_TDN6Hjclwi0ax#`@b&iE4h$0S}}thCv-1$;yf)>CDT}
z!rnA=v(Wj9(hgI<fuY0TsxC0-9YWa+sE&@Kz*n1s2rHTAF?hbqp>jYf1UoABX@!dW
znX#BHh!zxFQADmvUI%Ae=GoZgEZN|E-1}QIVvRMA$%1z7DRSxa%sf-nwNdzndbhX8
zLn~k%HFBe(hcr@rHER!=d;Dz00~rk_z_vNPKAV@8-49;?-s~RR$BEh-Q`<O6fs}($
z>1+!dVEo&UL?C*MdJ%Sw57gGycFfJR0HdV`c2=EpGeQ2m!oOhJzkZI@15g+_heSeW
zqp?V(^^FtLa??ii!;;nsSiv=DRq=~ZauU;=bk%0a+pb5Ak*w=~QrFE^B`hd^sHv+9
zXk2`IE9-^RyWLQEd3m?64HwJ2a+{;Xv%10CH=(gHS{pQet@a^cKfhY2x4+;DlD3XR
z=lXM?h#c`q-2m-8?H!T!!kVc=Y)Y}2&6jW9SY|E2ZT^Jvy%bJiJb>$ZV&_mS*zR5U
z_rU-4N`_V6QUA7$pBs8wxR2^XokY)j)3v=^R};W-aG==346^L58QpCEzFEJS?NX=p
zg1y8?Wp;av*R3Jj44BT9mnMxa^CvH(d#!v~^$tjv9%5o-d{KYew-V!Qb^pfZ%UVx%
zju-H5l$V#cm^YgT486`L()GIC=^<Y0#CPkp&wlk})HkR&P1GaG{zLo+&HH>nQcJ45
zJFqN|er=!LnT1=B@LvGj3M6vZ!3GdpF8SYOzIkK)(*`WIM`jt6LmdmN)uh8al-H6Z
z9iDldT6m&>vVI<smlo&AP~4o}bl1hZw>B<IFW=^dcSH^W&Ijd8*ZHoNls1=8MBs_N
z(DAa7v2RD~jhNG$x%MydgFpb0YheNDN|Jf<>I(;P<`Vj8^Pq7s3mxIbGZnvKzcRwp
z2JO!|Yj7tchCG-i2k9GE%3{NF0;1!MD4LRka>oY21;rM`aYRC&u$CQc?toSzY_U9#
z|D^T!v>sTy$l=U$;SrYRL$)+>))9r7MKnQMt+J)%Otjl1hA-_NG549OJMHEf0xA-s
zu<0=HQJ$z8C9GqKp=lp==8<Ub9Kjmj6;d~IITYm-|E8O@_%Uo2(UANXLGdqArJubH
z5sNqn+k8@9f<|<=F=_k=O2~nLO%{)&-@(rqzl}i9U#AJ2OLw1m<#f}WRN6ks;2nnO
zzLOHZtHm-sE7J+RPU(bR7tOTHrP45Ryj`-q3v9M?;ev(<8JMPstQLR;jvFayQr<~@
z{rx3plg~Zm385dT=77C)95X=*Vc)UJDjL6s7N~JCr@~frb#%f*x^}dLn4VZ@6S@XW
zY_FpUsAt&DEy{v|4ACyD^;Lt{p_pMep4iJPlSXs(ZaW=P5#gG-ABM3vCd+SU>yL7p
zPT>67_3FhJ!q={(4Jo|0i=N0+u>U~Z@Y<ELW}nPHE!_vlT_n9b3#Yz$qRb2fnF-BL
zFI6!M3JQLCcL4M?#oHLL43P)?qib%=46d^ca6Y)^@bEUv@TH*d5~d!~7r@fL_g6>q
z&>Qev5K_ApPzkk?x>amF6yik~5T441)JzjN7!M|j)@z(l+pL){yV16@ANKwW)4stE
z^==>P`&S<YVfxZ~z0i9vdD*WY7E+@m)BLYZ00UJnpEH4YZf$A|g!1hJfx^$g99<0I
zUqLQrmDs6`Q5JN%t*^OC*Qe==o5!2Tu?_M#SOT^vRm4bQqfgN(c>s}@<aJyQ=~nSe
zuQl^$#o+1rl6lUwG~l^VbeZ-7toR?)(GAlk@{4-3Wl1*QvHeCe1W=&*lI?KMdfAj{
zs!u2WjF9~x4&!3SX9ggN@t(3eznWDGYmuj)%-u(?JJnfzp1|MEASTAb#56K_rPl_6
zaQt-1VgDAR0I3W%4vq-MC&9585Tvo0n#DLo{#SEt8OX}8ds3z?Msng<)8x*a=Oh6d
zD`0{K&zWi+#~_iUyx@j?putHUMqB~Ro$TgXUU~Vvgvqi|pe*k$5QqIud+fVnKkG15
z=WhQPdWaa_Jd3pXCgD%-$77d4_l%R2=7jgma#C+?c5Ia8&|QEAkiXA6*@pdMN9vJ)
z0?@o|f!x#d(6H_Wh0Ekop)HJfJxItdWh=>yJ@sROTbdD3Q4~beG;K4qr~yQ=vzW?&
z%zQU`cy@s}xM*>K81`VeZU<u+YF;)Ov5yX<^snMezj4uWP0Lj1?~%nTnN(RCZe<7@
zgl<B#3G{9iJ+8-)B6&vCE!JhRvFkJhJO>L|8pwTB0#+)@TfSNJP$-i23EuhD<8>ef
zkVfua!vnbi=lB^B0(P~;On}cX)ttT{XYc3d_Z9#qX2aWtk1JmQRlE=(8Uf<)>=`U7
zIDLS|VBPxV`3nLXeK}eC42)v`P|bgN7yT1(@7KV)_~lGSY;$akRpBjs7zXc-YBAtg
z{_Zv`HR9I!c9@q6wQ~?MSy!oXkP!Gt>S^V5J>q$c<^Axygxch62)3Z(8jX@9R?61P
zn<+YiWhOL$_XPRd;jA7<g3SmHI(FA$;;UC8C$=t-u8<;WhA^d?ruKkeFY$~=UaB^r
z>lhokwx{2Xki)CFrO1MPqo6E2NYAu95JpM-ws*Xe*~eyQ@DGKKtsk}l6WYLg*cxLR
zC<I#0WmRFG!Um>%_vhq+yzOw&n*Zt$V%9FsHVR7u^M}k`ctAD71W%l<sl^?9>{2Q4
zGgkt>XD(2Dq^_qS8#9IKFm%#nw!f-zBwO0d`QOzzcH`E|B?eiHVzwR_sR~}0<Y9fh
zaTmx_YPHIh!Zx;@3zq^ldEG{f^?53r%%GFc1?Gs((~C*<_23SbVptnNjNQ&K?z3$&
z<sJ2M*{cGU8#?P_h7w^PPEu#p;90k2>CC7|w37lx%qx@(@SHY$sk^trQyn9Rh*E<R
zTEGExSEIh<sZ;e<yr6WM)GUHmwuuViI9h#EW%COy58nVVnZ5|iN>26;dNqz%Ed(Ho
zQWy0Qu(PxGg38KP-~OP!VRf=hKO`0xB`D9lC>bfvh9^VJc*`bt=*4orXV-jTQCE?=
zp$$v1$;iWByR&Y$!{}{H>?cT(C~JqC=Cla^p|oy`HY6-e7Z^KS;I1g2gt)hC@$jOo
zsX-ftSrY(#Bc1Xs6RNX1!HBBEy_>3*c>INlM2=`V9i`sfCjS=(>QK1w@eT!|`fpY;
zD8g0-^F7g?=n?#v9L)H}`MaJ19qR^Uymab^LKp#bX%Dt_Nvf_O@A4>T4C%kT%d#Hm
z6OPgSxME`+>65wC@E+r`6{Dfy{>JAX;JR9GfR|k`Mm+e?-m!j}TohS%%NIfun>eDa
z#iL`yAITmZ3et-2v$0FvNq=?6USc>iSw;hwxcWPD%}e()OgN$mr@X%y*4-2GS#88d
z3?*uw&1pDG_iO-Tj*Kp6gnWk|q$B;|Qsk=~FLMG2zwYoS)E)M|Dh+PDz(h8yU~O&&
zRz`tl0fxzXsemanNZm_B7Kq>DqU&r9py@<H6pC1^BmEG2Q<It`$O5$Qjg3*l9-eG8
zE`*Oj_#u7h?yk|ddb(S{m<b2v&h<30W}-^WPPU3W+lV^K<GFQ&E`WL=n+6a@AMC&&
z?&crc?2nfHVaIs^oGO(1FP}k3TDQr8NlO?|B$J*~#~B^n_5AFuE9s>w!8$K5zT2Wu
zq7mU}khesFIgp8RW1~oNV%<>mJ7n7yfD7QLqRxYE$6GIGq%QT?LWCFP(CymLk?Pwi
z`mVl^4KG(4Ayx44wBh|IDQ?6vevxC~h(T6{MC{mLKwM<6d`1LmC<;Rmme%Jc>_cZp
zG_m8A>;3A|QAkn}Nj-)qTiGk4ai`g!8z9kV5yBT+75!YnVNP7vG4V8D=wq!PvTwU(
z5cyVCmK1r@x%yZH%>@9F8;&Bt6jYm>?&PdGTwp*rg`P~g_HGpI9$$FgyS7XZXeeM5
zX?_dTW6tm);z(THJigXH>e<rLqPY0w72oZNcZYsK{BO+fpn1^Wf!jx0`$n~snRhU<
znMpjDI9DpM`}G%K>LIGtGE)}^g8?DT818E3hoMt}B&G-dGIjpiCerVL3ypFwFV-5^
ze+@5JwG>{AODQbmGOxkWMXLo<bra_&P5W#}t~?SSIZPyU?p$XfC1s1X3Ng+oC(=AD
zlfTFL7aW>d0&O>hzFnuAaxkO&<G4Z#Jjf?!{-VtAi`Z*kwy;P8e`XNjzE3rLw)YHH
za*Ut~QPX;5;OEq<Acm7`C7GkWuGt?#*N*@6zk^}bv)6p@Q!_Q)$$@}JO29l+dXw*<
zyKM0FONjky2}?%+wnhnm0p*W$Qx4RhQYM#ZhDP5xf4Q%PK#-J*>=0!6GwLUa_e_!g
zGfH3?p1tk^wU&~|&`Q1Rf;lXW0&zmW8yY2s*NK-c1e^;X_CXTI5s`pi6jkVbn+!st
zCBm3rlM0Uy##9+>P7I~RT!M!MD(V92sE`1G3Q_OR;Ef$&Db86|u!s;~z#6#n+YcCm
zN{rw16^*kG+v9i>$Kg5Tc7$?(us*ik+*4A<6BwcbKCmL8prX1}&gdJn^OQc5oOhCm
zkFWIkMW~8mK5+Du{a!glRUqP1P#qaayHdqSTg)xaRpF9z^uQPvwfAAop=TW$-jRdt
zHml8mx;twu*L7F8Q>B4-K4n@aUHWFiMSy=uxKCC#qM<XGf>!VlWhL%$q*Qz_XNN;P
z=>3DB!02hfXF>RKEE*}c+Q{3#8Dl1W{ROgrb1?omUcg&-nf%$+2MTrvGHp+%Z>F;@
zwAeG;s4ow*EYbo~I*@=S>A6&U9UZmY@V{$<IL|qJ-%{S$+tYzGZPe5QV3v&N5eCyA
zHRrXxDYs>fqd_PL6wIc!B^G<WY60i7a7EY=f;gEt<qc|K2kNi2wl(wdNK7IGkQPE~
zYt(5Y_aQ*yS&~q1C?Q?&%VlWkE2u|Z8&{N#D9mAoK~m+ti)Cp@oW<a<y}pk}{}X0Q
zdgX9DLDk8{G|N$6-gMS%^6%E!GcnYH;SF==e+V)Z`j*ve4rGyhes(SPz{hpOIIIt|
zYKI{%1betG5HoW5j&rN`Xnr=X`PlCM1^lzC+YeyMP5?(tL*X3Jai?3HMETj+z|F`7
zB0tTy&l6>$N}#&M{6gVFe?0#B&KdXR(a-$tMMJfI$KWtR^ZBFklZUpGcMv&S*;H=<
z0{RYL!VcWRwOMnO-5<Ds{!hq-grB2q=dtLu`k^hK+cVd4SH`{G@NRm+)}k*s79Bn2
zKgf%4sqmvYjD$+-&TIhg#af%+Xhid*J|(=n<2(+6nX|Vv>9em=*B2+Jyt|lmB!1`W
zn)J7i%S!9Yp*l~q|5{K&km0J?9ZK!j5H;ZJED6@dt2zx~sUxZtfVl^5qi-2>FrJzv
z7!!7~alN@?DtHW2z~f;$*DU#vLh$csB4(;Z(mF$egFg&12n&BB+=z-IKopIW0i6a{
zI_JdSTg)gtC@Lyy!JgP{kY?hVJ-)q#dJ%@12G&iyURn?r1xFfpY^kWbDymKr(cCSy
zYVz`5rM%<p#yFCh4S98NiE)%$2RA=;5S?1SJwk^UX;%IkXPx(2RiUZuRL9*R&sbuj
ziiY4s&ovk=aLw`e4qOhbgQ$G5cHc}frCS`Of#c)B*Y)*gY`l?YSU1%%g)QFz-^(<$
zDU5XF%OzP^e?5<nUfvlOKH?FQZTLAk40PEm|K#*b2TjQr=EpvdVbyVFtLG3HpJvj#
zP7h^rb}IV^nK?18?*|?*^KSxDzYW6#^IQy0-sHu15mh0|RT};>_K-Hn>oq<tbf%Qy
zHm-x)NBe7(PUcS6l5S)_eRds}l<`)i-<p@#*I8V!gqJC;{e`zYT(00oyDK&fhHEy-
zZ&Yunmu2EGEm&L-x*;%1w)4~?4nFH-+(5Sj`q3Ij?aC)9EadD*2Ct=rI-{p{?wnv0
zc5uQvF<7vLI^w-I<cf~>lWgwD7!D+0PL`0mO?oPV41;4!{?ke(Ax8cB6o&DENee#;
z3#$)$+7QpM&^4VQwQ}v|<Zv0?;p-PmQ|@;9tc**Y<sF%3nT#dGWfeq558CwF;}!8>
z`YM?)>HrIr<99gbk{&9+gu2JsTHY>M=my+7_iUm7acyir?&*~s>0J7Sssn~Kz@tHd
z#oB25K*-)_{6{)SE=|>$vDIHBY;VHdY_E5y&R1t?VDp6G{F4aIGK&zXD>R>HmH?_8
z#n}=K-K_AzU2%%U)P(sWZS{QW@`YjTsJw=i4x`n8`H1Zf5N4{qaVfev+0yrV?3~G`
z42++`RMf=(5CQqmlNsicy9_e}sF?zw=2f6Sy&<MnJ}>c`dAFpSQHLfSE&`VAd*G2<
z$@x}}fUX9dJ+JWldj?IJ;t-Jxmu+L6+xu35AvAo`tuM29;Pt)C!_%2;$p7rM7~0lF
z7I#%gSd+G-2d9yC;3_*G!gnaZqdXZUJ*<OGXI&N$sy7VQS<IQtk7B%#DiE9S|F!yv
z=N8ZL=dZc<PHE6n@FME4Wt`QiOe8A`D}eWMn+#&X;#;Yu18L2eW-OFc#jt4x)lm5G
zsSM1jRGL}JsoqG^^0gd+mii5gZ3^H7@ApKytvwlGek64+g}kUwpkgB>qzPsxMP5|f
zv@0+Ad+59+Nm^)jNHUMYzE}o*2BiT$x?|E}wQyPnzVC5?@CA|bqfdvg%!l)7=i16-
zUY~G=ub$}3p1sNR*~Q`7^_v*u+L3664%rqCO_L|)(YuBp_LPsmVVLlakHTJD&hnG7
zRP7u{nN!Q=riH}tTgrEWq{8A7EptMJNW0OPrp-w!faZ($!}^BAdh*j&{$c#`sViWb
zzeFi6AQGmEfp!(~;NmHicm=e7(Z`&2{lWn?#C<mV<z7gKMdc*fyJbqT<YN<p4j{UG
zV3~p`+9tiFXV^;k5{@N|G2p&*37}glU)ZA<d4+XpVtZZjSHeeFYCraVLB_8h$T4#Y
zBGT{_RaA7H%XY@8O5;YdP^FQC%>dSZ523ojALX*=4wy5hLFnoAaXbmVIJ|?pz-1{z
zdZYHUxb;*SGJewK(HIZY(()EC)+bl4)T%VIemP(uXxv1NO}RE~QuJc{OWk*lLhi4b
zus*~t=P#-hl{5|Pc5<MD(gB-eeDA5+{E<~3pmen?e#T)2jTD+s*-jK|SjT1C`caN>
ztSX+{$R6|r#>YT^_wHR_T&0}e;5QiegE1Z%krEk~$qCaDt71wwTt)hOOQo3+(H!{@
zJZMNbN6#~DSP-*14EUcACtv)xwrO-Tkg7FW-9K%MAN|L)<If2Sy0$FNF3dK96Nq$D
z3|Vs=z6p&D+L|Xh6C2iWsPesPX2)cdDNT0%g_OsGk{HP|39kl!wXW;@c&mic4l_rS
z{v~Xk<>7f4^I@7uQrOAK2_KdExI>-{V8gt`#Kh-n5M8PUv|^3Hy6>$ps-do<HSa(;
zAD8(M4J9Q<_;Z@-L)g%eNOYEofKEe+#Ku$($6g+q5=?2{=V6oL=s)+wCdIRDPMVOv
zb+QheWzOpKx5qsDw*n24yq@VAEsh&A%-?#xbg?49KOd)XlL0&dj1Q<`9E>t}VSEV@
zVgN$>_O-~CP&YQC%;ARa;!r6QM5l~g#Z`NA8uau!w!^8qi5s&;(=&D;@dQp6eu{sD
zNDS5q6m$kPypHb8l7v?)eJAPa64G@e>-<rQSB;@`c1AKw!{tGU%g*T47<v%;cS8Kp
zMFw1CN5#w;kb-+w&}I);DT*&ek2v!|P>U4VM=|ArKLP4=MFbyB7sRIiD+<zo9>J=&
z=IP~$Ml%$mWPQ;ueWcmAI%|Ren11JFvTr>-Zl5Tmt0K{3BYtV79TqFgomXQZJ}Hsy
z`T#Uy`0_Z2-Lu077#S@FmCb>V2?vEUM-wjRP~a=Up6Puj_$3qpX<($-+Pe>gDs0zl
z--JV8A)W^z@Yr+SIAq%%LAx095wa}Vl7B9J0Q?A#n79Dq5Xd5}VI})?{O&vh1OsPO
zru1F**`e?0(96m}zUx1KI9@iLnp3a3FrDOZ=KeERvBmM{!48`_<|M0bzYonL$*1>C
zU4W+}Xg|m+?y>Yl?Vi;;I#4;EoTKfZU-ey2_1}LOSBwg%in=A}GH(0UYvuu2C(TRk
zVrRx=8ScbiyA8T;NvX$>cikU102nJ;@Z}oj$DZCWQUZNv`k0me^>p+$#C<&&Bmvh2
z#41F?Rml>{Ugy1+qL7EZwug;|^t{G$ri&5Oqu$)waryJn1bM}IKL0#M1t-U-pbNY0
zuMcvP=HwXN{e6tIk{S{Q3aTWy<v15(WefP&wcB*V#SH~8TG3q)tt74Lw+J5BLi!2P
zlBYDgzZ}M1YnKHGIs{;1zT5vPu?WjZv#S2m5yvW67PNTK=bQaDCQVU}9OvZZ)CIKo
zJ7`R^zr%7I93LzoGgul_iD2oZR(&6)mrR%!xuITG3*J)B&Cl0?nf^N{j+H+X(FsQ(
z9wrp?mqd2xS93?&4FAai9YB1$?b2vAPJFxnQg}NaQFOM3q3)7T!kRK4>M_q~bza*q
za2ErdIiE!#R0}AW=d4uTVv{|ozL?~Z_WIe*hxNHFR%vI0w>CFgzbFSG??t8%I$Wy=
z4P4zLrA`|*);brzjs^oOUkm>E_X)pN!ay&nHJ~J1yPo!ZclZo8zItv!L1c9^BlGYR
zwBH}lq_R9-$Q2nmHPKEyeB4Z2j3AQ>m#hAibTXc-92kXgsd-QO3IF#eN2=iWk!o;9
z>GWsxjwkRg#;O*am(Cs7xIE;ptZ<I%*C(Zt9~38g{ck$EANxGTyu+-EZWKv*dVf(6
zvjC5_@jNg_h!_<vtHeG!Zgy?)NZ@<7lYPgGbIjz8ixP<u&~B}(M<F6i9%MXss=b=S
z5%~vfNf64cKosQ*9{plPgnqOK!W=o47$qh{egxcUu@E6BeYeMb2$q;Y6?a!w#zGR8
z)}~v}GBYz@e|wG;ya+HLl7jy_>0Av67#-Hq$pX63t~|AkBL4?t99U=XM(#flRvL>h
ziBynyoa5N3_Q61g8w>3U=U+3w)>^_xzK&P7zbJ!|Ko%lifmOzI{CLG%H?UbRgd?&6
zgb~P>66IQnf6ybF^J9A~v-LV(O>RK~mVRE_XAF}*VOmUN3td7wsd0a+x*?sR1>I>J
zyia4Uy>dL4<e+a1ro@m&5AEthw)K_a^5!gIHlFG;0fZFee#br^;|HlhVx^3Zd_<zE
zzYkt-<F)5JUDG!D>igFo)4cgx{a@n99E?_|!43}mU1&1V5DS|IHlqbrNS;0=e-X#<
z67!)g$MO)axS}Gb*1&Gq2Alr<iQ)D|NsjA{2jc^3oQpkzY_Y{jYwV@;-D0<wc48I;
z6FshO3|@EY1v8k=Bp8at5xCcZ(VHCnST*1U%?%C?cF-?dZL}|~OFDOhiIkC16V74(
zg!Q3yFK)He=KA`l2!UGF%a?OOTto7zg!ngy@`qEUZ#{63s~~w=t|uF$%1CN_F|a=k
z-kzfZ7pyGgL%&Yy`1s5AfIobkX?LUO{`qlgpeP;QDyD3I;+*Awt}eDXN&$y7*>$6}
z?)Q}^_$#&Cb0It&2Wf#rCcrK(c7u9(Ar?k^_F7But}4o8?yMIhLPK?shKnEjOB>!-
zCgp3)KU|m)4!_~8Jz807>-(E=OM=A50WE;@0V@chkQLZ8;V}&R>T?1bJ5T{^0Jy8N
z`^;_Lh*Q)9i%tkN&U9c@&mDafQ3^I>lTz}$X;i-u&N(1{5$tD(knhKm-T4elNK-+E
zmSq|PC7CQa7S<<mM!mm@ZC05O29vq(?~na2JMH~Sd=~8GZWiTCY4q;&Lv`%#vO)z`
zr8eozT+Tl4+QlDrKkm(@aB{vZGzm`zG0~<ACRf2#qA>BGm2jddSGB2*j?0K>i+K(d
zl5r)P9!aENSSKeZH`14SdJT{`U6@563TV8r9WgxttP?YgK^gaL%1z(h5a@m#Xf-gf
z2uOmdV4Yu)N23hIGz<wFXJ>AVc@KOC(<dRC>5`Y+7YA6ZbBU1<HK6O`y`t~VCPB(E
z{Odm|RgdQOkkm#fbf4P$!RZd1c<l%4IB!h4Hjv+9usG?@|7bI{=n$UZZRH*&lCtQ(
z)_aZBcl9)W4h-l`VTyUzSS!uKh6kVT2S8r(sn!~kogo*k+Hu0jkO~6|4GLOXHCV+_
z_<VQLNc%AJ6g9n4(0l!6TCU2+cBi=$`Z{58adJ>Y`Yy_J0M64jwlXl3rUn0|))(vf
zzF<jGwsy1JTnYREV^_&ZNSJYi*x1=|Q+|MBW7Vy0#%Ya1zvIIphdN%e#%#UXEiHxN
z8c)8u)h7IS6Ekfd*KDeI%(#<&K$Dr@N{L`u;?Y8IGVj$3PL^Ey=a%io&ba*1pF$P5
zjI=^T!~~~*|Av1AZrKdM_>gSRc{FL}_VHFycr#{MVxR8vF^GBJL@rp6e)rU=j1X6T
z>aVD_?@aj;nLKoGWgtr#Q>Ojf?V3W2cV~*{X-w|~fO~!)_#K?b5rNm|-Thr;DK${<
z=Qkp84=i=e=`eNQ(}3|{HUNbEg@?1VmgaG1s%eA>pqU^~4KX@X-0%&tV7NCEEwTU$
zsDh6PoOA~4#<0vuh67&ud<U4$BwtE?f}k7`(gG@$2OSln{40c0C@%vPuG_!1wz{}2
z?lsO>ytn4PK}t=n3XLjTHT&8ua`w}Gq8R>p_Rpc_oZc#4-<pUok#O=nt74Lz6kFn}
zM}VHL`J3_{dSwoj1(+e0zGCS~Ny^{9e{bRf>LWWB<Noe0E;c{BCM%&i?_-VSCcjr+
zzE*ZN##hSidhc9N#4L-nLsIwqN|8GhAu(<6Y{x=RIYo%r_9+f)U3Jl_<vRLVvtqM4
z*{CLLUtNOW!$8$zTjvD}0sd>T9sVb@!`KH6(ZLjW8NPvT9tjm;Kv7QSFDpi92!FiL
zrvi=<_AJRd#^1Zd%l@tVXDpZ&6dA}un68R?_D)~knQfP=o$mt;znwm}>RmTh;z87X
zm8#JAUw1DG2w;9ur&KHxRGGWy2|2_-9;1hJ0e>(XwmAPRb_;R!f?hZ37cjjeth>{1
zQ*8ehOGQmX)&Z9&js^D3E}W%~98ixa>z5lUsHsIvZZvZ>PUh=BGpGW)@gQpBYpkZ<
zX)0&+T6GRpLD2jUvK2|$Wz(a!$f|Gt_vg<{{JfU_Hu&_bG16K17t11lpVt@P6<Odb
zFQC^mywg}0zL8zfJu73|X1BGj5~8COS8Hw}_ZQLD|1=5-v^#gLDiiV^uE%{@<6ZJG
z0~wkixLv*Han&J&PB;hjt2onF&Gjqf+JN}|Aew;$sgts{=(+|iJ@0fML<maU<cu}Q
zt14x^okjY$KXQfFDOB68m)h^r3SW5gV?c3o?4=4a^hJn1lHF#sDlD=uUXFY<<T|i(
zti-UWcV2aOT~(r3``Mi$yD7Wp7SK-Sz&!)KKNEwfvp=GT^FSdz1lZxweS6AphK`t>
z@x-Lve*`*IXe#1Qzo+8eYRD#rN6yS9D45ulE@kiP4$P7qpaqN=cO~FM5lpznWbx0(
ztc8yhK`V|PVv@3lzaQ$mUB$ETz2TxC8{bt-c>EqQeo!QuSG+<*jHs?f3oZ_zBX1ZZ
z&?E5x`&Nd5H2x8~x2550XO@=^BRYc~4kh;?cz6TeZ8b_nMOt#R{>Fc%HmZOhp1L-x
zQRInd>y?uyY3!OBid_gXHxb{I|M?MqqQ6$XN{adSw^<oA?I@MC{C@f@t9abcu`2><
z&C#N0&}(Qn+Z`dU6pIFHob?Ce$~0>Xy*OdlztT&1M8Qy+A(Rj91)~*K_|P+1wLZN;
z0PnQHb3-$V&T2w14(^CZ5piOSroO=XBb^ThYDKu5S)80^=09;I5Q-p`3)a@7TGM_;
zKnYM|^4ZWfXn`J74%mmRpi60-0+pyjArPL7zQ6Wg;7tFwcUlW0;4U3sA0M{kr@=>_
zZ<7kT%v!CUF@Y0B0boZp9ynrNdkGKLg8T_kKff!|a~zT=92@(xDJ&}<e!=Rs=Sxb;
zbJW8_o=Cc?zmz0!WDHivNyy1(^iVK;<pDI(2i}U8#9Y<9sqfh32<r_(3hrpRMHuXB
zQ$%#3O=j=^>&kE;eNw?1x|ODu`0QTJz4?a(dE9wECwq|Jo!?)BCNm$>Ftmdw_1`*)
zNk`O+_XRy$Gsh3a%T$E|t<c3au;n@e;FZSQ*EJ0=4qgdQdSff}9{!AlaO#!6-pazH
zw5NtzqW|2mmn*uJ9Pv?T@95}==zhdQb$GZv5vUa;T>*Mex2+|k!;)2R;8h_GhB>0%
zPLTBzyfr;Dqx5X`*RO2YPT|A|fLKcj8V>pTE0FyFrb&o3OFIM(X$hY>69f%}CXn2W
z+z~%^aO{->UdgA3^$q;{L|%L3{%ijfKNSWvj1f}V{*VJJu24^&D{09f=OT2CmVV9o
z8wvH6cA`Fw^Ey?wtskoVce2|O_um)!*lyfEr9rFrc^WHUmrltFd3=*BD8rs9$Qo*y
z#)ReYY2n@sDR1lkg}asMBHvO3ZRLITHnIS|F4_~Q0bB>Eh`kp8@PACiD7hf$bUHhl
znGC5Pia2FiGm|u&T<bUE6ggEwLQs9d0s3jiJjEwrr~j=$WhWR%(q6RNfB)HE6*ZRC
zdH)9YZ5hfd`#I!ie|l8t3Adz>6zYF`Xux+6Og2rVy?Ex11kd;<l^zZ>N=uuGcT%uy
zmHKjO_+2)v5TME77HvZe>zTt%lkapqxd1rA^XCg28*abW)>xY7o4_T|5iIc7(_<fv
zbVU-&9K?dQ6~a|v%Tom$CzzCLweQ*4af30%C6X6hI=QL_qhKIl0~{Zt?fcsKmuVU~
zLX@~kj?VadDtVFpt$ZTCY}kbPe{xSumRPod^f#=bVtv{QhHGB3VUHc*&Izvv>C&$E
z6)qI2GMYOz-t`%n@BPY(V|nrj2+35{)pNjtT<h{>0+U*=xZxRj7nh3Fb?{gL{ahY8
zCLReL3bMjuLEI_7WGlC#FNd%24#c)4H|mmiL>@klX7J(#Ksghbz49$`_zB_CU>Z;M
z<UoNyX(5XK5e{<OJYv4do-9xI$Kg4J4p|Q%@$XSG{#$kP_y39_W6baEn(+)dj9C|9
z{6UoZ#pIF7=~}--iq?#YGY7d+?^1+3nlbi`;$HU<1SBR>zyeg}_HbF_oHP}W!R4|M
z(;5Pw@R%4%2ilG8gZ)QWZ)CwEr%@j9@d4&*<55ru&E;lz1#Q9Ot%a3hFc?59EjR>c
zFZ6n?K=<-!04RmM+{3%7;IwS&*VH98EZj<tK7RUc6R$<o-@Xs{r5z{ptJvmn$8X=i
zakPGpK|91i>ja<%5vka0#mT>yla)Xny-8I}NO}+bB8`WqxRWE_w1Tyq(yvhyY`i}n
zfd442@xv=wjW<&>FJltWf<r-H0Y-5#1Mc1l>*oLe=Z=?<e$^3|E(rb4YU+G8VE<vC
z?a%{o&gi<~qxkgc`Q;dETEF|-tY5|V3z!#U4<<vc_o8Fe$Pt=65<G^`LcU+dIty?D
ztIzS_HnK4dGGwKcLu~t^r=|Co1opt;JbHSmZtCqD!V{4>Bx}C9oYOhJ?ggAAtHc6m
z(A=PxCG!?>0o4Ss_r%LxNr`w(8?%YP29UqFnZc>Wi#mop@O-GC^Pec>U*FR&lAU{p
zN_(3<x9@TfeHPEI(fp8gA(-A0BPv2F{rCMl!=!QQ6xuFsZoW*8k^T9e@dR8Ck--<f
z+O+-8vcG=QCi$a<t&%nUMSLO9lbe07UEWqdP_j8FnoP16>yOKH7&y6fVbxJ+cmEDZ
zH5i~jjL)!sAfDPQ7J7H@GMbhBb%~;;gFL+XtU-4?-6xv4HJ7zT8?~n>jJ~o{a_s=A
zzy>;Y#GnocPPxF8uzIw;ww48vF0}lnrwHkUd3nZH%>h#r^x1Q3=0Z}Ifs#bmI^ycO
zG<?CXCll>y1Xk4V&QbRfki{V1=(4S?t@1aBtmoq8?Eqn5KLDyomda5Q7)@X*!z?Lp
ztfq#?&cUGr9S;{jKR-`~gz=<FoJ0p}+g~@;|Nb@89o==%B=dII$_L@7xKh2$nCOU)
zf50=V8=T}b*l162jq4IF>l&oye0r6@V0X4)$W?GYjigwI7Q2%k{Xx_JzlrdZVs>XK
z>5!&+(JX<e`1(i<_fX#qqC#$1NSz>O|1Dgyi3op2Ng`79AH_I8)@4z5G{+T>oHR|f
zmnwnl&^Di-<n_%*fL-c-8Vz}{eUCe-W5pL^ql^?7|FGvE0JUEbg!w4}VMP+;O8vYc
zKymE>ZKQo-SO`-R-%Uq~TiqJ#bDg#m?|hV%5WX)M3X&+Ejw0h7Lu{21HFn}NxQ40_
z3mMVF&fhok-u0<%M3mZ-DOUxC3ZuZ22JNl1x@BMp{gWYx?+C|aUcyBZvrqyqvmaQ(
z)i;*EJl@DA-5QuzrK#m{31)2n<5d6Ez4^Q1fW}>nUK=0hb;R#iTK@`a6cde@x$H-a
zNpgv9wwi`#m$JR~@mQ6!o5rtBd*8}-j2`pw|9#qysweMtc{Z6s3V_hXC^<F_+(DZ^
zqYzt8<buyj!~b?2o<e(ri|}cD3dlly;uEMCb<{=$-lVui?qT51wv)LRf|$f*)~oAW
zTZr5fPQ>vQ+HCFW%K$m$1YlrWN4^p^2P3l<dHhhYynL9Qom~i7nxVavrNpo#1nf$<
zeSUda<M?2m_@tRhcY>h_5$U*r-Xuqm(oWQME=oIfSOflpwjkySEDJJxt7i=E3O0b`
z8hu1eV&MQGZjBMTi|8v6iLB8c|9ye}Gdsx2CZEa(I#^z1_)b8SA9XJey>6iJNzlto
zQ9$A;em5lpQ7nvbM>?KZm36{0KqE9nJ!c-n#>J})n9Op2yn|Y20)DKWr8sKEO~a<y
z^?uKoEX|;$e-zqt0uuhW|Mq7?+CDW8>XS{&A4iKMAgbW0+hmpIfZVn#>lT`nW<X((
z&;tJ4c0T1xJkk{meI8x^+FIsCQ$mq*bC=-i1<<b_6k3(;^}w7(l=Cp-Qts{@aW1BV
zWhG=j<wJN1lDTWsmwRcX-OADuhZP5$n?=<2!J)|(_QuM02FJ?6F(SpNL&4ss7l^ck
zITbIAf>|L2{Kt5DWM(F_92cwK(SgBf8L-!f{+{MrCp48ncY2_f#A{1Y+Cn&f`m7wy
z?f-s@|37Q^-$kz+S`4O-`M@>Xqrf@aLn2Q!cMGkE*;MBDBR+JEh=+HR73Ij5+mp!D
z)!UkKMNub>{PcJh!KST#yV#>F%0C}7TEL}G7B=-=%-RpiO!WwhpRczE|L;?TpTi;!
zkytU1R{gas{GZ<*!Ikvh;$t|PXGAosYp&bhrxFTA1Xwue0uJuhE`%h9fTKX{FU$`T
zw_+Ubl4RXR2S_;^CJ$14{%}m+m4+rPz$g%scG{BzKf7NL+nn{}$rA@9pe?Gyq$3ys
z@!OJ!9>8#{A*sw``5LUN9&3~S#a%|nM4_l#*E?aghX5-ehwXH*4J8GrD;Hq#-a7BM
z-NE5F5f2}VT&8@X5=AWsp|<YTUi~<8y$D!&Z=+C~EYhIND})D{i8q1bqyJ`nYXPt_
z|6G6sNe}Jshdh|hwn2r`yKFj9gxYiI#8lHvxQZ(i=IFHc%#3B@{u^}e{D2pK#EAI2
zqh7Rf_D}L<yJ5jv*OTokVu<<CUZ0@|9cd2#_dundm_2Ax(-!{&XfVRDv_4XQ?`BRE
zSg6o}p>|pAaUO%`Gr=p4e0G$ysv7Rzib6bl+Kx_qo^;voV{U;p2117H=`R?>PraNf
z8pm<@iD}Cc?D_Qo%u~B8NSSs^E)+%->TfUI5&$nk{9J+JmbeKnSJ|EuK0|Q9HTxtB
z+QuIZE_fnH8I)iOK)gZ0)~@Egu|QhB&;+(C71KMco<GMD<KyM?S4<E;R#=46A>c7G
zb;Ck82l^43M}!0f6fGP8yx35b4n*5#Xngq`dFM&n3{-_bp8@jAhP$YD*SrIKWnA9h
zJ)JzT;rI}5Aj|ARg5emHa!|jW$Q&At1eT`nK}iGk*#k!D`>F>QQy(Q6X&4<hf3>i(
zbaGmG&h!olAbe|)x0slis2-l`w10Xs6xD8CyglNlUT#JKo1Vz`1G(tCwWsaL3IF`u
z=usT!9%daDL=)!B+JyZu<}b$5Cu--rx>-o|=byPj4LKuK75g9l;TL|M)0uTEpsc>B
zi3nKIuWMzlEyy~q&=0UFaxP|)dlZ^{eE4f`c0m4BuZjTbJeYMNghDVzlrN(ZN_Yen
zNErY`c1jg~0syJ<`05Y(_#<E$s|a1PU_|AdHP8w+#=Tc)tV*=9&<_Fh4?HZ^4*WR>
z5)H(J5)AHyHL?LE=z(bmWft!6;4Myd^fyVcxD-}<=Q>Yx`m7EcPmB_sVS^mgPFmCx
zt|q{#ZNX%3&mrQtt%kKVE(c0;>f{36f4l(x>~P^V{h&PgNofB2!<@B|>9o@q-riEO
zo5(DzELM<$V#I+O*o%Se2M+SQ>My|5C$D#yry2~E0isndlvJ{3EUt`{cT6u}dt*e3
zmt_VWh26A-hAmYskt=Rf7w8U%f;{&9xQAAk6fXUNL>vcir1pJh_h%Hj-C=s4@-s`k
zRC<{I>>@8sUb}yh@Q79I7O&+kOkoQ9{Pim2yxrysc>ce4H~ie?gg9?Jqg3~|-$2&(
zV20A)HM3TA1@Prnev_{Jv3y6T()q%vbu>7FU%8KFQ>e0uuB97O3K+g}WMBZgsb+@s
zx#33@J#uhO6oQ;`0Ft9=FPNit_xI@t6K?AU1qEShqnc4C-|}^N5O?=beyO4j1xm0k
zKzqn_0bP+Q?D!N0>yEMk7?YRWo-ke*sVv}kY?g$5P9^{kR!R_173K@R($Bq%A`z(o
zK%BYpRuA9Tu#S-C{NodX52;KQ_PLa#EEqV2$$y<5In0L|$ab-%1&6Q4!ouP^c<;JR
zHDON81A@oBkadR&xcDJdKMw*k{0vFO*w}(0_f8F50@|M5nZJOVb9}9Oop{{fe(i6_
z$Dr0k5a$mNtc#i7cGoEF6_?e!ZJ%{V!eTw{dV#mm;(t5uh=X{v0(Mx)^xr>&Nf-3T
zPMrNjU_=<KIWT2lUi@i4CxmGBgWLe~j*NELJ^jNXQ{UlJc4p(Ly($-2f)Rx!aG-9#
z@SXh-hiuq#T);?(f|S&ZqM+WaF$69KD_7GtAM+_p^%c$6=|BkWg*J0TAFuvOnz$0U
zVi?`|ZH;ODHKoRs&X6)~l1bqLj;mGbBWv5*W9})M2BRQ3<jwy3k4$a!%E8nEnTKtk
z2KO&W@#xfGDg75SL*`Ge1IvoVRmlB%a8thW*M;L$3C|eA4!IuzVAG}~3>wqORj8PF
ze52=zSnYkr7LNT;eHYVvS4(rw{xqn_R&|k7*=X%q)XNFX$X?FA@W)l|oY?Sk$CCCy
ziQ=r8Yk#dWv(LdbeU+N1Uw`{!YJxptv;rp~@#nv{82;Qi#`P)k5@(Ld5mKPiX&)VL
z&h-8f^Up=bu}$(A?$bm^dLQ^+jn+gRja-_9ksW%`GdX?WPoE8Z5FMb@Mo*vnT82>x
zw=~^IBRF9?B9%Y8^?H?!960!1P<y786=-_hKDzTed;u3Tvp>LP$P-y3{i(qa-`d*R
z3uWs&Ou_xjokuB;CdQC_0votN0xJbvcT*pQdMb6XKDhU{_vhrk@I|lm9mEx*Ptg95
z^%$mA>0tV$43<O`+WPMTG&kRwH2^y^;FeH>ZbEOTa|l+GE`^#Ur{}O`AZ`K75t(aa
zoyK05GjEuFCziLLQS>`LgUnK)onlI0Z9t(!Hu*y%vTcCPZ^AgDIuhBF{PzbyJ_$(~
zd4~lnVmcw8t?KW-!qVq{G_iQ_;m<?C3{a)|KHDH8ckHXjh5y*|jD<C%c^3f-u7(0e
zFDfVLI~f5VCF6VWFbi#MSP)`T1m@0%G1}s*@eRM>DP@fqMycd?2fx&FMC~2ldMv8}
zKrGxoT5yF4YQ-ZkgtmtMegJeo09K|bu^|LP5X=dKY5VSfJoVHZAy`!_Ondft0%6br
z<DW}FK$q7@&=ID!lQ<K>TPo8gag`6UrT=*E{%TXo(tri$a4VDyvTuJQU!2nH$gTpo
zt2&B`sO5?J48(N=`aMmDMyX;v0)kus7JKpeYMBp$gM(x9WJtED-`t8a_Z%?UJ18)0
z#l8-XX&QbrciZC>IPo9;dEU4=oi{Zv)=3Mf&95mP#R!*@sxs9Z{M&%=8O6IC34@+p
z;@>}qQ|Q9vs+ZLY>RA!vE7u`Lo{F0muAKUYnV?*@qa`aO_rS@`CUMuAA{YUyBf%YT
z@kio`5iTMc8d??jSN+=FepSSJzPUOxv=KtF&CSgV!B;P|_k^|q#-(ZI7SXo}e!GZv
zCf(#lT;;N10ocj43pjECM6Xb{es_X)Nga$OdM5Dd;cu!iW#j`s(F6Ue(wK-3*VUiL
z!!^V^T<r%BXKwbXsxk)$YvqQsu~m`CYG2XPBe^2h8euLWMM0uS+N(f%T53~I>mGS;
ze`Uru$?p0(W=dYR0N{}ZqtIDUs#EoqH_Rz@gh$dwM!t5}-l)B}-XYMva(BGj`eVn^
zQGS{$fpQ=&59N!~M*qjxd%$zufA8ZFA!TN-LJ^_tP1z%R?}k0HNwy-hlt@`6duMOm
z3MC_ZQ}){m8JYj{a#wVJe&7G+@wo4Y`|)Y;e4poeo^ze+y3Q-bQy4H<LOb%WOfvVC
z?&7VPAj59$6_X{yfT!F^+Fba0V=5Q~=}L9Ey?wG1D#O{``5a<m#?lRj<LJ4|;4Wbr
z@lue~6E`i;`{&@|{reacWFG+~@Pt{NcyMs=9xuk$y|B7$H{O{ADWQmqJU*+LPI|K(
zWVAl3jXO<}N4|w5S9E-lS~X*iLG{W2l+L}L=XUaTB<hXH0;i6Co1Y2t=fWkp8suMM
zpY;X1!(adMjmKQ4VeT-G!0H8~Kh1TH?a56$p-?JxCf++qP9EGY7hEl#kxw=dYT!ts
zjjz5<i*pR&B9#2FtQn(n1BQwfyX%9jTH1igzAY>yp`~3^GwPO>mJak-pP$|I+TA7V
zDeHL)O$z8z_R8wR?&C{;3kp%t(1_xSl>}x%+y`JAdMs6M{D56()<A0I+_DsY^EvZ@
zwUm||G4Rg*I&8m8%mjDSaM3*nw{8;QA+E*D+wpEEN8{HXN7fUjtNl9b$tca5`X=F>
zC7=Dl*w~FAI&T#KswPP;<~;RX4(&*m!8f#jhq=|6xcBC%vZyZ4?n>810<2gT`!{78
zrOM466}FhE4FV+&jm<KTMmuyx1i3|B$p(KP7#v-nw8?577SE55Dbd2iv)OjbFYLWC
zZ5xzij()Rae10z$y&m(&X`lRa*vPS)RM+|abJJ_D_6=S$-@LXe4!(<{IjNHD!yUE^
z3B9U;;)kx&1SpJ$b-*BhGnZT0YRU~$2A2yO55vP<F7ok)Vc>wP-9TX@&)4cm;TzC>
zG<N4@CIMe@3mR@i*Zt*3K2eeRU;DtuL=uoWJ@Ec$&IQHI42*tVFRo6W=ne7AP`3H|
zWYI9Zi8r^Kt$%z}W^^1|^YyJg*|$bX6-;gAE=58@BEYd(v)U@;0Rq@!0IDe?OiKAM
z^@=QB0pNmq9m=j&>%9vpV76t-vRTrPraXxIrhMQ@yl2nJCpS~68t+@T_Bnsx3*q->
zI)C&W#mC+~U@dkpp9^~W8Pzj?WjLyfr6GIDlI-Z`j$$_Hr3!MP#tS-kByNwPqcN$t
zmq{9%P0h?@8<=JfrqxTOpz%p_orogWHlB9AlPMbU2Jd6e>$?x1QZ(^e7Ut)xKuZK8
z#88DtU=;o;l0l{->X~gA?2%SaRlGRpyr-zF+zSLe!%vxm?vu~7xo<Z;9?$fIwH|`o
z^M#LC6S>SIfL#4m4V>Vd%-;q-C*K~DnbeDHK2zX#c%cry@>Rz8tTbW2(ftwS<M>*+
zL0SdF|LU*s{Ft=_-fA9zVw>Ln$>TjWIjN}@N}q)ZPi+$D3g*Ce!qv<;A~dwH0{4AZ
zN*#S;_-8qe?955k1nngcH3kuZ*P+)WP<Kbj*^XS~F)6>6TH3C5rWjN@CceEOtH@R2
zT!rPEOO_{(9%j{f<S<7l=@yRsUN}}iM*FgK$<UaEFWM3B@f3O&JFm(-!HRx?8HTm|
zV<90))Ns7Jz*geHcir3<Uv|<P;8LYSUp1Q22$~qZ5}R1iTpN6X2%YQ60P_bym^n3z
zGGzP%rOU8KQ!lh^U0K>3lhf0cMOR=dh?u0Xn>IAS&-*McE*ltM<?}Dad};#mXi#cX
zbEk7jetMjE2=CjGXgTZaD@l)Kgo(Xb>T8%+5ZUzACod~G;?5Tk(8wY@!x8uK(}?A9
zt+4~z$TsYKL~^ONq02L_f`E8V1}uKBoFKlXYU3XbWP?^-acNiv-dYN$E7RlCTI_|7
z{7ptA?)G2h*^Fu$vuw4x`wqhLlc;iJMrL9Z!(j*#grU1uJCKJty?YMCo5Mr)8^At8
z5`!9?7vZF}U+qK*Pm+=LPJK_xp`xNmt%+cq@aiDyDmRpP@E{N)M=I&mkh{r1@*~kq
zpPg|!NT^SpboPI?`8a=lK&B=5zL~vJ5%yAOrEEj1nmTA#nbYBA`k=wwLm6H)7;#l(
zH~lOG>#WwdK&>l_qN0LfNgnOJ`71Fi;BCxiSkV=!^WG@$YA#8+A~DrprwNC<DLJ<6
zllvu2Y>Ag`T}0wZcb@rc%Z1L%=<nD~daZ4)CN{k*OSfjc<35`b^NHgE)^=$3XagMy
zNeWzh!KZ=-kVfrH&DKk=Y&rRaD#-JgHE`w(c-F1l*s-->T0@bW-!(v|KRhv&*t%3>
zeAE<E%%@11_8*g%8nM1!Fo?P7L5+Si8pVe(<~_#q?UKJT|COZ9@?*-&KeGDxsTwY}
zz}5}TgcZCqm`svvGBh+4nS~YS)7|u{D$ajkq5o`ToSrvnpT0=Y2l834g6W19LS7^+
zC_Db{RP3`wu9N!3Rw=-IgC)~%Is~CIbiMFU2f!SM?di|4$#9-D+IG}WpFE*w_$v8f
za{Q5)79=<p(*B$A)>0+UPYC|!z7tHKGdg0gkR7YvxQAA0bPM;BBp%?H9!L^_ehcxD
zS~4~`HRk3AwB=dYKhpylzy&Nc_G<!o5>L(#+|#LNGS^A)tGN+yxuu-TFxX+uOVHc@
zNVJ(?3zylakaCRP!fE)GD4*zHT>UHVaQoI8ZY;?BL-?Q1IZ!-<A){W|dgX_UR&I3m
zKvtZWjIV>lSx$P%>dJ;WMVXsHiysR@!mwS&nKL(^)~U)wZ^QU_wvo;!yJQf!2X5Y}
z8<6E$^318X!JuajoIHBKm9)9JS!=Ou98YBTN2lpacCozjDavXv=7*ad@K4zKe3^Yt
zQ$*pl#G}t5e8c|N#|j|o%pvM*`+R7gWINrv@S;cbGy}mr+^7apjbLgRdt||;$~mYm
zd!g|T*yW-4KE4DSOu1eKI9wHFXJ;R4wBA7Tj;WT}5G(Civ2(Q<R8g09Rn=jR9b{`{
zI{W83i<ke2xn*63&X~duV(zaMXFA8BU_TOFC<u6XWatQGj>Ibxxf*Ah3`&(l-#Sw7
zI~Ft5rhjAa$s%}&jr)bJf?GAQ$bC)GGJB>epT$aTUKyRE`p1*p%0<B#_EMnI0Jo$g
zTelENiwf-Jl0X}XT&^nt@uw9iZW_C<oMf;5p@ujOox~B?bMk9AKB6SFz};BgKYFJ%
z@WF$NtgH&)q>x=CVoc_k)6~l)?s?=JJ~-1ijYfJoZP8QDrLXM0z!OZ9QSTf`yMf_v
zh&(muqLC7Ea>0OuiA-aTf{x++<Ucm~wafvgJ3zdsn19<GryJWQIG1fgQNq?({ypHj
z$IquZ(HjfjM%cJfPo7?Sq#%CeFZ_f-t5c>ndTCU>#1hrk@8+uBfpaCbA<5y$uOaxZ
zSeMI(CqhLcb0l;Rj`2di-5i=1%KgcxHr(FIb=utQlctT>_aF3hX3&hs$<Zh|?AUaj
z-FZ%FvlZYQ8Fudcc_~n-JIpRRwTYy$=9c{vYjaLWXwWoKh^k@wD!5Ka!T&V%cA-)S
z!|G6j^<fzJBD)V2V&i$sv~H*MCvp7e9=u}_rw=lTzuQ2=d=O~)!PF;Dp5*q~^F$&T
zhz(9jRT1PtLFn-LbixeH8HUTO^z{GK_wZMhpr1c4<I+2Cs#kNRwj{{IVE+}BIt|6(
zxQJcQi6a4?p7T521c`8a)|&=;mtH0$ZRaXv?D|pvl{TWG!fJSxf{h{ZS0wtYm~^%V
zY)zyM>lIy+8FZyPdChMCUw~0k@cenD>dtTRA8Ec(u9Vjfux?E~tleF>^WyRie+Bz-
zt~g|A&oye|N?2;YJFD#Tr1B#*Qd5{G+->-&!Vx3QYU2rlZ5Z^5KOnlu%c}+2L(R_A
zb+2#ddqWIT!%iRlAdv^nlm7dRLe|>P4(iUEk7`o`T4!&J&Jm}cjteBrmpzCu_DDoX
z#sjlr^Qj?`d)apfGNkg({o}5h1pe9xt2hv!E*E+mjuJ(69-N7Jq3jz^xr%C&%2G3p
zhUp+_3*74Lu9ADBJ(zOk$lq>7<Zkj+-m7rv#cJoTK;v-<-aYzhj@?459O3_Q1^4UU
zZ*LM&UxNCKBI@#XqA_!-<#rO~Jwq)0yo9KPGg(MInC4>jIyHQ*xA<ZNm#zs=ssIb_
zf!EA{9?`1<(aiX6>~ebv$dHlIAGi=2-@W_c+02YZJnZ!J*!&vQ4K%Hg$_V}aak~BV
z{q5R)r;9$|Td`Jm9KS58-!{Nn%Jn7TR>0qt3zP~z_Z*6Anl5uh)>rqsw)YRqO%R0T
zu{GF)Bee{sB0UxZ*AlE%SsQX{%gE1ff1!ImWE59YN@})ZE*FW#s249#8{^9{Xgg&d
zc@1s)H}1L#6@`4^Ds|Cxf6j1S2=`CMJBCh*O|Gm-xiz7NOE2fuN$fo8axwZnJ@%1b
zf$e?)$#)6iA<2K89~uUV-XxBp=aHO7sM7JM=bq7&7gEzxR~stc<GjCBQDi-#h_d&X
zzF<YzcHDcOyAkNXg$Z8kx-sk&a-@6Pqx(ojJuy8!2h$QYSJ#r&#K^8!KyIrT^T&+~
zrlV)DP8jXzPy=|i%rlUmp8gtW0t<JK*II4Gi@~Zc3KB6}md7zMY3CK9E#MMz_P|<Z
zn*sIJO6V+e!A=6YNF(rdtr+^b%Tud;v<rrKwIV$mDbXU`C0^HxT<@ukVCpbjBk7VM
z*r}}vW-YGXr`!3cZ~<6BNV_li9q4grF1<*L5gm9<COTxJS~Oa3;I79t>>$;#gqeM$
z*)Vxs_kLOnspu|9Z&uGDL0e<`<L_VN>uZY%RZDw`iSu7=HvzVZ->o)621;(#)vi7H
z`F8Ds7NHCyS&L(jM(#-{*b<K`qS~3*{NCV=+{#f`08C2V!=obX{Yg^=*bWS1oNl0I
zzopOH=N46ti;J7@$StTgYoUzc6tbh<7PYsblbf*5ekf8B8X6j(Xw3H{8mi%@NuE28
zXt!#}q=~Q?f}4RSfKa(d-5$~&EwhfHcVi2$e6#Y3ja0vWy3KXi%@%8IVWg{2nx7I4
zBlArD91m~p^$Em$`1%WOIgYCD{$aQ`E@8V%)XC1+LuwQv27Z70JF4p)(j@5TQ)FD)
z;%rQFN42F+uZA!gB)*1<>^Yfl^pOCw$F>=`t0IwTU={qKf0)>LsrY{b^WY&)coar@
zB&j0V$cj~ESm37nZHDUQWfrzipSoAVw@p2#LQ>59cw1gOM~q7>Lw4*g(G7TkpA@K#
z32U8&kvq+9JL0+80!A7M3bCbpV0mv1V}iz%eh}>;u1Tmb$w3O`!@QaI=q_)`(_>Y(
zJpTcWGD0cfBzl-%#$*p=+8nsZ^p;O2%#EORT|1hs^|8rEN4*w`dz~km2oY4F)#(-R
zfAxQ~492@p&}IEFHOno1$#!R{W=^e(FMf$TIw0TQHwGdlyk&_<0Q8X(k3_f9SN(IB
zr`r5|DbS_Yr$I6{14Wh?yF?+yvVU=l$o3<P&8SlgEgWy7lCJKmv{V$n1i0WWH~YK8
zBl1@;`R2sBX1h+Sa6-h~redHptb~#D>ftXS5U+DONSU6uiw-ODgRfNd(cIbyaZiS>
zNts2|xH7t2rcF@SP*lTm%hqp@LCw?Rpq@t8?Q*--31=62?|Q&hDldX<kI8_SW<sK8
zpBx!!YHQQ>^sEXAF%WxX0sA6>AlP~vufbfR(s}kZtkAkER3)=ezAZYYN`SW>B5rA!
zu~@ThcRfM4pLGHA7DKPmy&PE3mkH*I7~|h5xQ?VM@ALWhHtfJ(P7eeF>`NP<vB`za
zEYWwgOW%UdqIPRAQ4+7a<-#^Wel3zqe_p4@8o$0>&`8?}7U`NgIu`pBX+db`Obm+n
zIIqGCsa{_8>|i9t(-#Ho%5H5RU2li+Ca+}<a@S1o+B(0W|3@M-_dV^A*9gU&Z#jMS
zw3nkL&7<7m8|H4x?bW@hU5jU}j@CH7$_Lj!jTuAcXn<h&Yb3ExKYS3R@MG<*0h=SE
zi%tcva!^Fni43EK!1zA?YhIOnE{rY@wdLo?yKfs;_GH*%lN-;I!{fOJZ6FF71*Is&
z6@xC*^g%m`w!teT=MJdwjkqT#+*>TSn$xK+Ozr)qnk8mp$>W>&p6dxeIesu#f@Jfz
zSl`G#Q{Zw}fMb=2^~1kXeVv9MY=4~v)Q#iO7Z_Q95uQ{-TV^zbqb$%ZyU2_^UjgPS
zNGuTN_~{}-F|dO+w3ApwtDD2^6Wv}Ti`TQ*Ii+sF?pS-TkXBTKo5tMz#;e$6=a9~^
zM0rBRzyOM0v)NZcv`->G`(dIPV{_XL>42>mT&-l7m@kI1V>o}-78}gOT)OktGmI>*
zNAXs!eaoKhT_)^@e;!bR$LPd~`Wbf$a|D9oxp?E#urMdG9$r06&lqog4ewB&cq?%0
z??P&xWo%a9TjXmsuR>vMV}U1vJ5?8yF@xmi1vT{7SaWZbZKX2tKNWNqEQka+r9abW
zZzR^=A043=a(eFpilzY~pYsx&v}fGjVUqFJUQ*MnLrjLG|G^SsOQ^x`v%KUt6a)G_
zYjElI6P9dlG-GVITtiMuM|bD0&C$NsSG)xsR^nv<XL?~J1T<B(VfVu|*`(32#$OGb
zKmW`)k@HDKHS7rkx1>9o*EfqiJ8-k&g4J74?{_04@ywqzy}M*IAeGeczPd&}+G<V`
zZ+~~mhX?%LU?dR|Lx5gF085JlKv(%1MhESryEsuw<ad0>r*qnsBdCJ4INVMvb@&>!
zWaB9a{|RirW2Bp=(^rRgCe#$9-CZ{@hVR*ioG-xk|8w$?97c%8>VJ+DzB&Krw;@ea
z-6&?&-3T26wag{0c^B`oogmG7&wTRxM(<vl^+}$&Jl#of&n;YAZ(I5#?&$(X_N$)>
zFN;3fgid7kBU54*P>M8{j}J7Mi-mjf?2g2C7G~<_WQdV8s<^PuUs12y{PtuS^8)D-
zF0hq9|7@4>tiGcVdZhSsXEE`Naa<TEZZ!*Y>IEH|6ws41iu!@R+#dV6Al0c4_>3*!
z>K^<3RY#zw&Sv{+7IOPKJ;dp;IxQ^8(bAAa(a#xK56)%Iuu|-k`>NU1x(IYovbA%y
z2`9ruOvts3V>`+%NfsJ&X19Gpg>5<t!q>5G&gw7CPG1UFxcuit{OU-aZw|?Oy{vA0
zg4Rm&GWV&Tv_4Vj*(2Za*)frW+&hA8iF35;grhV>D{LiHkbi_~?5;iUqVv`<Z+(ia
zd@v;`mkjFT5zA%N_-3z2Mgjdla;UL3j)y}-1{8^Mt<BBqU_|EJKw*xME?p%!=e+7%
zkFRy&k&u$I2ngsxUA5XtxG7T#l!$DYs4$;9_fAYuk>ki?jE(MBR>nUqP<*8emUC&{
zHPX<~cms1^D`1G)WpLZ{sesx>8<=#%nRhhL7ZM97R-OCc&=TxKn>F>6{~go8J(r(&
zoS2vg=rVty6)D+M9KlWR>pRx(VRGLKdwE;{FCJX^r^d%}uV-+QYTbH%-js!{xMaij
zrIckLL#)JY$@F5s;J`m2M_`X$J1?C65-Nm`Wv#ObNz1*dI{!UO!-KP2WVSs0pR=Sv
zV{Ii2XFA7jSc!D(YrCIB2I9>Z>RpE%@^8_MpS$WaLzKrwBS+)mY>}P5<yrGNV)YER
z*V}JiX9?mwc>;eW#0|~Oq>z`-pQnS@1PU;9{7eTfoj)Hp8pb?}38Mi&7-kmUlYSkT
zot-UTX|+Qm1|`Gzx|FoEAl;j47<iDJIa4-I2Wgxu_NBtBM|`v%-YGRkW+U}G7e*eV
z6lqc8H;8Ecd`XZpYmm{=9m3Fg(6-bP90}02=}k=hF1Ev@g3X{z1wed10r60owLFee
z%hp<-0{+ad{fX3yZ4_Q~<N33H-b!&397RvNl$sXaC3GXHobp?pE^5s)AyV0Yy;F(U
zNZXnC{mb<uKN5lX*(3~4hlnzp`a5zoVejz-wq;Rct+*`0*$GNZLy}$<4RcXl{8BOx
z%ALqv2<#Q=$d-=VxrE6p`2h}O4a{^49cKFvB-;GMf_)3KRis1&M`34uY>=w6a&dL`
z$3WRNIXOwwoA>%PJroj_Ku}{xR&l<8cipJL0oX4+r}>7G@I^P1s+8Q_%XsaE(utM$
zUsZkXWyNPV$oSCQvQ52t+($|CPGj+TQVfF2GDs`xnb33tV!qWzNSYpt6k|uTZnOR?
zB%E&W>}0wXpRx#*)y6iBvG=z3hmi_CwMK?JwN=$85=N>bWSqzkk1!wCW=J(jKjBmU
z{jYJ0IFvsBkDN$`4APb!MUfu4Oa~Pz(f2ihZ(9qPTl3SyTP$2MLL1b%8I|*H*!Kgw
z)bk=7TA7#=P@DbCFJL3)KLC~VE*(!;6U~3bn?O1E>Nq~m>W52AjEwwG1q|*^KC8i2
zRy_I%*<%~HbZdG*t@(*f6t=U&^0->hK^<7DRw4a}c0mmh3jzBw9aPkllK?)pu4aYc
z`}@FX4P<h4D}psf=OP30%zZZxdOvF$QH2Q=Jv~%a*-g)75izk*F2e}vx(nz4#dohp
zD5}C%%zU{=JB3u+_@tKaCr&MVD&!r!C#mz$Y$!6-mCZe~z^BG>hUfRmjp_RZN8G<X
zYW6bbLGiblsfg>na%?{<i(li+9m#{6Pm!m3>*+Z!slSRObg4R}>4z_UMADVvcGk3h
z>Yg%u;CIai#ZN6j)0Hf7%Ums}Q;<jXsWf3~L-KOnnf%Ueag&q1UK$PN{;;E^p@Oe2
zk`~v%wcqpwk99QErXqHsM^PDZWJoIf-52Uh4e!U1GmBtp2D+x4I8#{krvX+Mri_;9
zWmSM(R={L}l~kbj(e<s%cUNu?-<^(c>|WJ;cN|w6^>T8`i*)CmSQ;uUce#d=)`1Y)
z9f5mS<)D2Rng!ZG-crpr`xwZ_P9%y;O5R6aWEeA@7#*wT|Mu+V3IdQni=th-nZ?7`
zooN|m7CxJrzV(VnI~5j)|LM*4iTlcu#Y=^s-3V?J7$)Z48#tqIlMv%*vinSk#GsS3
zLe}ALJNT*_r*DngL1yus*OV+OzbUfo#_qa5DzL6*AE(QS(M4A1Voz+>*B`CdVMs<t
zg)u~wKez3xACLcC9z?Re0CdxiB^T{~2eDSX^3R<4P?Pl~5fRD5539}xlarG(E|@Yb
zs3L|9#;O&wsV}3TV~2WF0+?(ZgOfY4FJ5HM&)Yn>e;*l0fWeFw%ypZa@03rHf&)yc
z=U^|ZnON5n;Ad+=?2L}z*My11><?O^6-hkc!1Nfp4W|G_;>4m27bgZaI(*p!G^}nQ
zs0yQ)zb}0d;@-_U?IK*U7!<#-d2^`R^GfcbFLm(mR|Ml#-7c=yOeM)ZHj1BBl^xx=
zQbqqQY^Dyfpg?kp=FvMQe#{SrvGbMM!ENO0${RfE654iE>%<myn#7z97r}y8Tl@)c
zH-p%(cX6Kk)Z>-nR73JTXajBZ<WKf3*!_j9G#_>GZDL19kYCu9jqu!J9C`n0!i3Qn
zEbGB9U>NT^)ZWdTgVk?Mfm~$;jFO+ZKn$9N&wzb&eeGm0wWrAO_cC={NQ4z$imZ>I
z5sHg--CNq<YvYoGyZQ(NS@r_mfI8gSwPd5gtgL?%<c7{jMr6Q%rAlJ-U+;tOKF;st
z_GWzghehUcuXFx?n?zycapk`Az9{VXasC^ZD$jgt;~5rP#9p_>Oj)RE<!ads<ekL7
zoo)3nN(e)+DC7VPcipk>j(USv#cY7Nv*1-%yx~{9HZ8WV3*f;u6a0=g>0a363FGYR
zy#`L<iED9-N6_}MMD&IYfABAlN=!5!KV~KW`8&`-=88I%Sno_%t+sZ;%u%OiVUC1I
zU=tY*`kEsskC_~O5E6Ufl--*k(&IxhSZFzJfgw!xFeIb{8bB9Nq<f6$5R|B>=wyFY
z=PYt+yP+UMN2{mYDV=ulqr>rV%)^TXpZ=n8{<YDZK7B)yyoHx;+7&R5U)Q_G4FvGR
zn^tph`$jDYNDhB1e1&a?<|8UI<dZLJY_=2Kb)!FUmxWz`a_H&o)QmopbjBSIf3AY#
z8+${NxZkbzc0>ur)ES5b`rHCJDrV?gz>5<U&iQwkW+{&RRcQ42V4rE|@PPzaOPbFZ
zP<;i-qCY2yi*0n*0-hl#xS`MP+NA9%3Qo!WooTV=Qm^Wp>*oJ34R@)gx!HZ2yFo<v
ztH@9%7Kjp5V*{ei!iy`nr0bL}9hayx^n?5)$W;e+z-9Na*Y=q5UFXn-McEXjNJnR^
z3F$Hr{MGq3xziEoT08}v%I(ZdbQ&ki@btnzKoaTRv=(%))E)S%0i!#(V))G1e=g=h
zBw<vGC&e+4QnU2D{%im^j%GuJm!O&f<qxMhJn_{FYc8%EsjYAD<d`J0QhNi0Y}Fd~
zF|1x}B+JM>`Ie=79^SBt;ZQ0vva}5KG!%fNPD{tE4x7QVzT)wi0O~tK<vAy||ML;7
znl%_v0?V;hkqrY8nB&XK$})50*-ZN5vbST=d8*V{gWSmjfN|vLuF+!|w;fw1MCKU1
zeToVb_A~A044qnd5HXqXQwS_{*)gzwaWM-98}n01s=8cJ``)GM%C!X}FHA^Y|5&8V
z!`h3}dv*V4UhyV-o^SG?#<*Z(1D&{cYWNv@Lenk!%h!+8WhSqXmr+kv{O^$`*aF~S
zd=vQgis8Io)qLq2Jh7aQov#=eUF@eZ6A7qZ)y(3+S!LK#vM|0$5DHRdI);J1>~;27
z>IVS~xF(H?c+~tVFptk!|5&?sg~;Q`N8#&(y#vk+Ao0owo3H|>1!#9$@Ksxznp8mQ
zqH#s{E7%fqFq8)vk`{`Q?R~#a<Dxfx!Igr2SN!Z43~0k!q9pMElIfw_-<0tJ)-4)e
z<SA@4Wz1I5Z5#8cbZ{HJH=xl@v)>e&wPq?EF*#l{37p}O8YUgu7~jLov`+f_wUc9N
zTlk{kQ=J;{KiL|YY`*%*z=VF}YQP^+SAg#49pMhVol(=JFWS20TMpeuB-AVd(Y_1v
z7?h3o=YX68s6P&>BNv){Ig|>Hdfge!t!-a1<uk$5JLc<PkB|+3vR*@6!JqdR{$Sr-
z3?&$_!{GK!)vbon?@(GIi)ubKdscIP)onb!r}bjPw^nAaLp9Bn9zt?DcGHg)X^4yf
z_|VhKWX-Em+u!P~MR((+pg^ZRiBu*YVEIf#Q79p7bFmCh%spMFRP<?vZ6&_0;P=ao
z*Yne4T%JWv#vRQ@oM?rGxzQi4c(+!v8ZPKbYG1lZb2Q`Z=pfU!-RDucM<NNjc5hY!
z#p#D1&Zp>o+q}F(FQXPjM=eLQPI6{@p-Q)ZIZ*Wm3Q%W)JiCQkB(x3FcsV~x>9sCb
z<Li8#Z6SvTIW?8XmO^<o9xziSem|@-T?VByN1jP5(mnv-9DI^TK@Z_E5h-IO+V>A7
zqhmh7OBL+rXMY`oJ<`m88O{ZQ4n}mp^Os?t<@H2Sk^B>;?<{z2INEL13vb`HDs_|#
z%gPgQOU;7GH)}l+?w_BpF7<boximXmF7g7E_y2gUD3RBy$n1yY(Lp%;5$VhX+0zdJ
zlU`eWA?&pvi~2qsA~NvQSR?Ym8Vd$x7k)Q|;mdSoJ%119hnl<J+y5QkzS|NV!*>Fb
ziO=(t@ojP{s;rCZsWYiB?F9ekEu9@B_CwxXs5}E<f!OOeZy3$dOJJc?;*XTfq>UEi
zSLY)plFis2omxm&7vjB=b$3AO?94%hlNC&D1d5x}w8H!|OjKYgLgJ2~$3PUQ(2w`n
zC*B;}Hs+aJbBVqe*K<(`e?R5vqyx?&7-k~ohXbX}Hp!G>dKUk;yKX62hN1u4;vHO<
z#KI1*BcB>z4IG&KGEqQ_&HTK!UDOoYbtBJpL1aG0$<if_f#I67UfKo0Lj5WwrC{|a
zhL6WXeNKKCH5t9j5E<Y#-4LIEuxY?7MH?E_;S433qkpu$gbCmTAP@Be<8_p@?S7`?
z(vHn;K1*AX(*7?hh~74Ln0P*XXOu5{i3G^+)FDGwh$DZmTnaICJ6v)KV=kC>Ko?W`
z=+(D7=ueIHv-b7hHakYVWVgu5YO^){i%ildV^Y74e)?tH4Y{D!|08YmKr!6#$^~8W
zFU%f;fSGuVx<K*UMq)qJhV_@X2Ie=h03BRZ8`g8z5osi*8Si3Zd-wB!@ALG0RAynD
z8G03dpUpKE=O9TulFK{J#UNvQA$<2OLRLy+NIJ?4+%y`i^E+MvRJk>9%(}?O*W3RJ
z^eZ!QPOV?A^9@hz7!@qto4t$Q(AM1ZC`%5^S{d2Sj!_*n5qk7NNc{($$yzwOD&hBA
zelYpgPV`&}rh8P!XDH_UMv@`%W7m{1F57~*>_hI!o?O&WZe)rB&LMtcLg$CmmfOC;
zx}n$c>gRaH-2181mx84Tj(%AAz)9u9%OO5X^mj59k3Ba1(3I%e{t$E@iMw&*vZ%+&
zwgN#cd6F5S;uf+Z@o^nGFk*3mII*H8UVNK)dK(8Kq*mbK9APl^C&rz9^ZmWuwfVxf
zOepd8#K@(O3>thNpu;#na(M$(ty!$$lt3Z^NK_8~u}s<C&1j#O#CtR{B$xK4;U_VU
zmsDNe!4tz8OpWY!I8Djjly~8NE4>MmAF@*$!0CeciU~5W{-V_+>FX_BuNYFI*ZI!z
zUUOPh)V{=5cFBe5Pg8G#<#&1zm)piM@Vwz{Z(+tFMaR4UP-oDr55Rjp%-2UN2`Kr<
z(c>5@_67SYvQ!e;h9QqU&N8M>q`8%$TNl}Rb@pQcOh(3JQ3PrC`?h%KBu_VEy3^&y
z4K4UGI@-Vd#KFP~C>k&?-hl7-6xnXd)5B*EyYK2)W>#(oc&2IDY9n*{MDHy*Ak%gC
z6mBa)S?CowLx<jda9<jOsgdoR2+{(VrW-dO_>q}sWaZ_N5i97fGqdX>K4AhYHU`+S
zwv$5FFY$0jQ|abgc{uSE;Btrmsa?m8{|OJGyL)Aef(3Vv|F^ZwCP|+iSOHi&XMy0$
z(bff;K+g+x0yl?e+!h7iKo`LOL~z1lp~t_1hv>eHi%3Q!Lxx1vUVC})3!dOp*U;|a
z#crV8YYqLviWIqa!n6G_LCb;#U|LKG?ys3z>@OXOKQxZ-!)~LqKRT7&tm6O*suA5B
z3t*WuGc}zBj;<<<MP0yDGXn-8nmudX5w)!?J-}yD#NF>>i+6UEk2}Izqh)e*hjIid
z{7-D}aC&11lV0|Yu`By};?W>Z!->>g>*|qhv$RQx+Qq7#3U6_$=5Zo|!)J#NeXf+u
zVsqk@g8#-`zsAO?SrfTS{(lt*YtVQ&SX5&W{s#c!m-(^7^SE&7`#zkERri!<B;6O0
z&Mw|7fb0VDsC7MgNa&$nDmGl2lUdr7P|IqQ+xCMmURZ+rH=NV3aXYA@-!X9J%%6kt
zrNICOWm;N-$EOrlzyVL5KJ6k(E(&msY2tFy9Eg_7Vc<RlPGiVB9QP6!^3E4rX1cSq
z`|b317$XDNQvH<9s+b=n82(5JsO-ROuS2lj{wzB7&IM%e%I?l;n|;^SF<I3N<R&%H
zV9nV(598xxFo|5L3I#e{Pc>o0Ap-E7>Q4xDjQ(F4#^@l!#GIx7^Mm{wg3}L2yiU(U
z56QvhRLsmO)6BQ1ZJFUVh~L#SkVudNB-8j{ngQ-{(Uk7h&)dG;-xeF;_1V_i(<=mi
zGv{7GpTb#TVFS=`YpJO{3FFwHn7ed%rFw8rQF6-y0%Lb~3>l(AVH(60Sr6d0^MGZV
zB`9Vlw_#s6EV5x4%`IDj^&*+Dd~{L@<juV@bY2=Ywn5MCO|kd3S+B?TJZro&o1JiR
z$JUzi@@cGUWMFVNstMU+B?#hx0B=V$@j89<h9wQaHmbV~REi5-++Nvv397@W?-Grh
zLA`woi~`C638Nd%|4Bu@zQu>pwf=uZ*B!`w<L!Kul7C}l{`$U7lB-i<`z8f7j*QNm
zf~?xHbL$!D6X|!~p{Gwma5-yLM0TCKx}YzO)!sBObY=u=UWgCMvZbN+!ZAW@92`|J
zZ{&8KQ85G{@gy!TZo)e*wcjs8T_ZXyp6P&%jPcdjr%!jo*3E8kC|oh}f?Y0x<6&H%
zO8379iCKW>=HwuWHu$Lm>k4;aq<T=-hH>fB$B*zRhO=m`&Tb7mMO)A6tr7|Bc<Gq0
z&A`khQHk^2aXqA?5|*F1>^LC{f)UmvrigdvOj7D{N=GsQXxF3QOuFe+;9R=5-a*;i
z2z(vZar}(Ok3FIrFF>sRx_`br=m+Ih*18_ie80nRcZq;<c>%`Dqr;X>;3o0mdA+}S
z6tf3_I~}T@BC{=gT9%&P=5>omhJt`)za1cr_h78AGPk_!>eU@4f;#zI_bZn$E05}}
zpY~ooyDvcae;y||?*csjHBXq>(VyXh&mU6QB1Ec566V8C>3gQQvvU#R0%TJN;36>1
zn;jOYe3HRHYed(MahQdfSw&KE5?>Kko99CnxVFLzS?)N(ESenrFFZ?o-q8YTs+2o5
z391g<=<itD!4uKkEOU(HyXQpVNPki(*~0g#S9&G(-2yp3VsHn0R+uvn+knl0Isl?6
zMFr}g<o;iQwsa6^ZuFAUhxi_R<*yBaHaET39!cc&(w5PssrymZs)l>n`z@u-8}B!z
z38A6-x2bZzDM9Nw@j+n1t1k4pkzI%+p3#XE0gNif%P^W0S$<C(u^)P^s*AqmFnelT
zXa^&nl_kN3h6WZ6j-@iI@rLkDb!8DvqZ**Kz>e{lJK6@nKy5>#9}`cNQNXBV`5#0q
z%ZWY+I&hbzkzp>92<c;f;~3HGhM7Np{22dq(O!8>)*)l@c|uCDKJJiEokB6H`zdW@
zd7(!t_qtjr_aQyYgv@VyDzE*C<~_FA*23|MEZfx+|K&<;9u!PoOxJGzzAOjlb=DmM
z>c_$mRU)50jO|EGr}&p}Xqg!^0))6a_%94awx2TsdTo~ew$6h<K~eEWHFLBPER>@8
zmLH?z<@o2vu)B2Y++VCVbJQww1_|Oi{t0uH!aArT*MI~Pqdk;=BLKvk+BLS><lEsM
zrp<s^Bmwij)~0iEnkmuqHBjhdVut^vxETE`$<uCVG0F{d*;bruAg&k6y2m*x!hQOA
zU;yOg-ab@Z(+%LYQHpxHw<}*dhV8Cz0{j3<UYqBoFHdK#yH3z{JiJK3!hd_=dU4;(
z<ra#Uxlq|$ooq4r{nYu1lj@W%_>SLcR#XV=%QIIW=6dr#{dx^dwZbJvGYS4`R|&9{
z&{_!>xR!uPxi)SEZE3cF*OKhwW4ELx(fC3K<(sds#pBRyy*bY;BA`WhLzt$zr1QhJ
zxb27REM<DEx$NNk;)WP@l?%L3Ugod|Ii5Pq3smU8@M|NZRVXt$!}&&Z;K;H;GY%{U
zb`Je~6F)n^s#g|5SYyF(2*0id2(ZXXLapp?oi2ctL?w=4yI(?@R~Jxj0Ni5*-b=Lg
z?eKDyoeU$tat&5iII`W&>OnW=XzrG1%gbDINT6*`F=P;Nu_k*E`1r<<x<bSs^2hvH
z`(G{4kK8V&ouB;sxr<hVRerMtAj|&qeIG-wIhZDmHn$)DJGeG6AX`nx#YEG(i28kj
z1{SD2lZ#?E+ad<CwN=+7)`D8ELz`kGqE5-mN#A8AQe^H1wdZsdc{KnMUekdI=Tt6y
zhbCn9i30n@!3m~6f2S`QrjhqgQv}tTWV>+Tf)!{~tw4BiS64|XqIn(mEP-8C`pPbe
zHkg{FO-xLzygq_Ue=;nLW_8*X?W|ayc-ShhY-MxU*y+a;gKyxlK0y9>>vv6eu-y&E
z!nj)7fi<zP<{D7&?(Wj=YrdL3v!Cr8Vp%DHhFz^IWf(_li`=>jTP%<crMgk&ye{5j
z;_DvPc1yp>@!@y=M^28w0x7@9En36=mWO@ELZVAd0jvSK-=5w!WpvrJ`#`%e<kYtp
zUvd?HE+}ZGOUdB?4SF`9bDF;1rk(YktKZQYb%p}n6rCx~-XGzgr>CdCrT0DrsNLmE
z00wG(k2nxcvKcvlkph5cFC{;Jv~?Bti_os`Z!dt8C(jTQtJmxYGbOZ_IgEcQWzF>j
zx*RaA3qOU<M>pM=&o=#}S6f@<Gh%etN|MY)A!N<9uqfyl`aUsCgG&UN9HF-F001r4
z1cTuFTXCpdOmD~BTsBB5d!up7>{4fD3zqPm>}WP&&Z1<cg}yh@GOC9+txgGjqqVIE
zKWjyCNv+zWV+VOAb&q%6BK#}O&jV41Drmj^71`0l^%b|rv5%PXS*><dui?t5f1OSJ
zM6gW2>+7pEXU9w^FN_cdZQyBHluBg8hF=Kfx*k(xK=o9>$XS(Vbzr5>1A9p`UDYH0
z#Q2zBT|K^7Hn{#)6xnahrp<M08*9QI6b&^swS=yB=|9>f@M*Mb=f|K3Kv;LCFOiW_
zk;yy%6<g!H$4&1Ji0{IGPDWN%#m+9*oP`?Nnns-Wjg72KOw(nKCs7yjt(R_`PK<lC
z^Ut>m9+k#qO6dodpgTL?(4BV+m-?_t9#imh8N{=AAEl^J(f)go{vUp9p9E4wEWDE7
z_^So-8zUN{{!%-=9TSfQ#OUGH_jvlQyL-;^`?Sx>tMPFnJsA01FSzhmI}v~23A@af
zL|4?h?d!!9EQz-sPvsFPF69R``$Zw4%A15o2BStq{GOZn;HK<3B=L3p+czLe@ejL?
zhg(6pdAA5bsbP32{?oNz-rKuIO<lbh*2%o3M_8FaCh|E*`mF}3*{8R!&o!u*x~i70
z7o-*|*JEQ;B6VxGK%@jr(nCUXc5gc8*FGfou?OD$gAB03=RaZng}K(aK!%t!yNG#U
z>4soQ@Rul6LlRO{+3Nd4D$tnrVN5yw|HKr-K}-?8X8Jp(zGK@*kn=lLLy26Y^CJ09
z;f6&^&v8Gl%fvD_hd6j@>VN@@1Qt;dzEbAgwzz{_22ipxQpc~y{o{M*g(M!D*oRY>
zeKEk-BN-CXF-k&-KY-sn!*_(Q3AoEM@5H{%1cwhx*R-OW&VhOYD$))_8gOC_G<B!}
z>pT~bKQ_bzrK(v`apH1+WYf8DsOsWw-oz>szuwyf&=`u?53XOjrMM`?>8pg;1Zc+9
z=P<C1Q|fg#-(9HQ`Pd`B5v%L&utZAf@=?w#Fv<2XYjfacJ9(yCz3+@QvMXMh+DaVK
za5El$wGtn7m_uMZQFpS-IWc@215Ie{)Za{DekEV0_v5nYug7{tf__LzN06FMTCI`7
z{S(TFSGECG`}ZLBWh+9u*~J^)#(B9*#}V=4bobqg@sM>!r^~aNP-{Omu_Q$!#vo-k
zU><jiK17r2sQnOXUyQML`O4~Q)@cHmpa9t?6Uf<uKxF}-A_;2NjH05hNqz=2Sjv*s
z4XY(%9F6V5Vha?ZQRB<Og|a0)I*`!*WEX&V{hsij`p?%1(*ustwPzpKI~^1iMcr*!
z7%rBYxqqu5&<s_y{cBe4A{ZTTc#pEl$i+DH#RNs?Z?4}IHZ%hu!fG(vum6<=-Jdob
zQkBgCMV<H7YoWh`Nd*a}&Ca;W{fCgCdUXZ#mKAE`y;jHgY6BMASsGxz9hQ5&OYlnm
zQby03*H|Q(_MC#uKabJU7*VaPPtOp2E5m8|rin%*;Qs7s1Hom@u{@vMEJQP>eyg4T
z_cIAvBwbzI3h?VLcFCH?*9Q}gB4DUmv8o|VD2je5L*<a-?&q_oKsu=cMS~d?q$-;m
z12Rc?pan}m{(>q9(K~zvy+dA7U--aC77_5d`rT5oGz0CIX21La-%N7qDsr&6zs)Kt
z_$N$nUT!JNdcyW!$Sww%4h<*-j~rIg$XD;q0Ul%j*{Pp-74M9ts#ydnTeD`*)-KDr
zED4J1#bSfS7GnqW@~cd|_*HN4T<+5Gs~KcAGqJ0F{YsQ(SetM?ej`zCN@YPIqen<?
zLdBXBPrS|?y>!Si6zrzv3O6zS^alP?Cg9KR0gP%M^Hyr0ne|xqfk!KBGl(Z5B0@P=
z>0*V)Mw+4|-XzlYYZ@w9-PYQ$B=bumOK;}kWo?(8FnPdvg&;lU7-))dv_es0xg%Nk
zrt{iWLknLpm4@_m4R)NNB~=R@Q9GQf)>-*2ufAOyM+tY<_c#e-Pa-W}s_vLqQ<Jms
zQOzL1v4fi6SkWoigmD;Y1mA_ACfK$6RMqqK_=68y&@+cp??>l0tE(1#ptD*Xkj=K8
zxqn|azMt(~Y~Q(Z&9LHk*F&^oHvH~HPow-EWEprkD*-c7<W0ae32MGKx>fXW#qut9
zV53K4lh@_}QMTx>0s0##Hc`mC53B!>y}_k&9vh1}KUb$&d3d@PZe?e>H~zR}y8q;k
zURj!q0Fgiuz#Eoz+|)nNUCJBGFe!tnvsEsfKrcQ6N(Q!fR=T?>k4k19;PdsaFt=qH
z+v{<Km(zb-4YWA?DeKPo73<^A^rUfJx7_ZNZWB8H<tKO4t>*qe9~>Nq-xwnPuODkS
zkNWE_`Bo!UX0PXFCLh2SMSaU3JV7YfX1(G(Bb$*|@lD%-cpQby&`|NwNfEX;H9D7`
z8e(ay*RpOa&W;)zB`bRmcnv*<GO*np*nG?%sUzw$B+!DXQN9i5us3DL0;%lUp;=i<
zp1J?0PxjYNDl5}bh`#g%(HFLfZYxsp+O$J1LOAqH^6*kA3C=R6ym>S84=zVn3*V^C
z;Hn#u0LIeX^GK{fal_mK-7Ahn0y>BlgVjZqeyI&T-m7a?81f2_PzmMzpL9y|Ug%BD
zS1eVWS#ud;zVYX!AjnUw+q~Pq<=~PqOOrLdS@jud_^ub8NNuG5Q?4Mzd7t{h;}2S<
zBcTMRSVuZWeI<Xols#F<K;t#SeZAnad1KOu#+mW1Nrvv(8rAy@uOzBq;Vy-8>8aDp
ziIOLZz37-P3S>6ZZ5m;nWc9HnKe1u>@)~E*3u9iM9{7;MB{jtBcp+otw`jBz{tDgv
z@;VSb@&Hovp|LRwTsm8Qfa0ME?t|3>P7F#{ulj+UAvZt&a$bq8)~!<IwvG_vl@=?b
zwlGv?YlJl^U}7c_;<kAOv7Mpg`sX9-moG$}9dm+=_goC~-s&TZKHzh;aBzz0Jivxl
zg2!5suh@!G^HtQ5p|vq?_OOR|sc5{V9AB%@dDZb;`Tr3VBpl$N%4{AF{yIL45*Q@-
ziPR=U<ffNth-GiFWz=s#-Tf0QRK%8(IpR?`UMGd2j{@p3h_dqA3G^RTmMk4_u>Lmk
zlQokD{;wAn|Eygsp}SWckpB*W!cB-~(9!KqLV~@{ocRl(7jMV)glZD81`kE5n+n(=
zlF;`~_WR@L=&7{{Joc-$Wmi1(qP6TeL37nLSdrNo45$3Ht25;oVfq3OZ&n}(1+v}*
z2Z*y_fI$cDCHzdY+`+WU=}X$Ps`!){^{<$UEs|zS-P9$^s_)=+->{(mQx9$C{0XGp
zylMBD=Km{@v=0J_Rl4mzfkXm<WIx?A$iNE0k=Q$B351%OmXYMD<2|D9n+C@VGOK5r
z+DD%V@)HUDm}&2eP8`fy*shD}?VSx+^2v$(4h$~mTDWrV<cXTugjpdc;mtu0yoq7M
z3<-)#>7?QBk(d6-s{OrfN`8Bja4)VWLa?3cXW0ZE2tavg=kiEoC&u!)WNHaFuP2rC
z_GLRlI@0WTe>I5|dBZTHgEWnIZL&zKYCqO%S@~>Yrtx+@XMJkrb5`c?9db7`S?D=l
zu{NuiSE{cYa9l~Q%TW8gc1bAb(L{_*k4$P$y}oFY{&xnlMKVx}#-^r>JdHBPr-A%7
zc)wpx-is<+dT6{hsvRF5h8-C>l%%Y8FqqWXRWWw?aNf>#1S_nZZl@jJj%kIz_*UG7
zJpoNZMko6#Zf*gRc#HC?_-LcJ5HK4~;`qPLzD^87bbrSw-l<i<1!be7qm%NH2FyWG
zT?xZ@lp;LptbBBrbM)0^fLJmZ_^W&tw}otDDMgRPS{(eq_a&T)J@F^p?o)H{a^+9V
zHaEI%<i3-mY|rX1_+WF0zUf@L>}^zX-Uq)Yi;1nQvQm$UH{bpK{lDljl?TJg;kX(6
z-z5j6o`jRP&CF0xq^W{(_vWozE&yC#6ciljG!vjj%R333nPzK}&F9Ti2J0L@_u&R*
z#oxojN|CU|vSs9*=`JXM5sbgm!kILXBoFX@E#4}=OV^|mljcxZXic&`5!K)^(-0AY
zji1r&pb;%QQI~ZLZJ+RLWWfGs5|Q5tdi3PP<YWd|&VnGeP+RaOquWeZ##8Wf$$%P&
z-zPOKt(oqr(9o3ZME<Aul6ale!PJ7FP+-jyqP&{K={I0-D{J1p@Dn`v(U%tJ$Nb>I
z7p}{yf3*!I3M8J%#Zq3@^2p(P4K%}DB^1UJS@lH@f3%5XBz`>lcoq{}u|}3zsnRFn
zo}^di%I9WKZvO_LkoWy8HUeE<#-5=#oG^$J;CKS*M-Y^}hSHueR7ZUvGj_gs>o}S*
zE}YvO1$NbDR_VkXn1z_%41&2P`W@d3Va%6`5<CQ-(EOOC%<B~5Q%#<+Cw{Fld>i5`
zwYJ!3->9k()sWfOnJkfg0Vq{#^%7a>R~<}|>!J976J!1&T(^6zxYfU}+xAHtuu6wb
zdQVqJ!OmxXryRsfdCvUY+!~<ehS|(zB4P6VunD}=lzrPm&Oo;K^rI6SU6eSz1*#+=
zAd()9w9xM8J41N|h0z&BU7_I>Uo-~IR6$l1NjOC95D?1p`zHDJV(tm@=>s=t)uXbJ
zS0^u_Xl!xOD&ArV_i^L;%OgF?zjSFnycP<!oIhPj0?ru{l3AoJ*V2jxA1q``lLhQU
z5CjBQ#QJ&iaicM01_!fU!v+CZ%v=qYKkTM9XUD9exbhM??nmQ;D3&?7WQ+y&Y`2Op
z=VQ&Ke0gwRc2*9Vk0x<daH|^BV{;T6xNt4j;FBk^K?nKB#oOZ9mE55H4pmR10#rG_
zD?Us)Np4V~Z6k@k#!2FL&aCsn01{XB+c$Woa<t%+I{6HZ{38u?!Oc&F;yh3poO}J2
z6qD}!V^)n4=q-AwqY0^9yr)~N5?l$fJ<#0p3Up7y1<v{q9OO%dT;Ss$@3ZW0Ili3Q
z*(M{_%dPq7E%FkTaE>N3@4eAs%F4sdb}+D@nk^eM`@or`{d$oy&PIy)=3(9glF+{l
zH@fyOQonv5_BB74P|o^c(b`ko1)fAPm^K4iR3266CoToguSS>2a4U?I2kyF}XWbce
zI=S|%%Ns87ucM5UOTI$1?q=HywB2(?h*se}Xp|%!H@xcHO}d{`Gz+672IRtjG_@;*
zWVY3(4f(-<Pe{J&Y8@^7T=|W{?YO_c4`II?{nE@!84~?%V3^g-dK{VZb%UBYuV^w{
zI;`y(RU!wUb?jH!iN`>BD|83Po$`C(Emka0SFtGxV+;{MQlOH(pmAkr$p<6_#s$6i
zd$R7VtTzu@kIlAZ{e4s?cU<gBy!D<=ck{AZf8Mlj^RKMTa=97s$yNOjb%E@5^}xyJ
zDnH37ieFXwj(C#K{vKz|W1sn(jE)w)xx+{vRbdC6zT-(P8}>SZ*DXxU;+3TU?Rdrh
z)K?4Ae7QQgIYSKwRWyU|!SDkpl@g(wvV0Ntc_Ecc#|uY?(@Sdx7LAg-yHMX`m6m5(
z7@g7s(>=+6p>R>>YTY_=RJ&)sT>X7iXZ>skOK4(-Zh#oPnD#4Z*0^uJf2c=i-h&1W
zu0pgcVZL<Tu(eIkkEso2M#s%ze;y1Q2DtX4*QNH}-8`N6<lMOQvu6r_sPV6JPAOda
z9Y|*_FtzS{>`ga+dx>J-*C@Ec+Wzf-*=>6VNZj5lncwQ1vwgqWXYWn>z<v?&ECREo
zuWGF)&`B_%SWyjd5rU*<G2M?7QM1tRBWb+Lkx>tVSWS5~tkMfLyEtD^ojgY7uAv+J
z8MSaMy`n40-fJD^R6_B6ppiN!$oV;E`Goa(D68i_bJ_VozY!qTO6m2hY48&#I-q!c
zq;Z|^Q(Yi{ohg~X2+-DV{1nfS*xL0X2TKvx($np7ii^Z=#|bzhEul#8lymrYZY&Uz
z{*X9lK}h6{oru|Hjho9f8N@H7t=6AAOLgP(I5;06C759U3LkSpX-EuCZW}6D@0tG#
zSMGq!1GsK7j}w07E%D1Z27tjL-eD0`D_}$L?D1oJ0|89zD4fSluN3_gUC98dTjV-p
zusdk1@<vLxywxXmZX=`Yem34>75V^oXh*CiC|gPMp6d1N-B4Pr!hgP;?+?>@kDch&
zkO=x%3C&eS`k>6)%rbS}NYsqRwKM$wT<c^fz5`RIe20rjG>Q$;M(4sNySLX9_p8=s
zPvs2jndpg`jeNOIqu2QSm$k{F!L8K?K$P)(rKE^q1s6=>v^eSf-o`hY7kfP?aOukf
zXa04HaA3QAga(&|Zsu8+l8J{m^DMpc1e130OUK{e1cK*cs~r?p9kV8Vzn%jVCqIM<
zTmgDQP`~PosMjQ^E4`EFfgd6LM37+#f?#TOUNvXIK68fDdU-_zN^^l$#-CCj7+uoq
z)F-XrI>Btl9W)nFyo(~Ho$(ufm`nSwL8+-_I#jWExkh|t<3>D*1ZR*u4qyfsaUWp!
zsrjx->MP;K|9%IlaBYSwG#?4C01Z0?ybCfB2NduvOWkmXdj_y6((@Jf8fAe5t@Bm|
zmC5_4Q#pg6-Epf<4}$#~L*)%qp`uTl-jSi+OD3Iw+}U;P_Ls$=1mldnn1uoas+7jF
zO1tW%w`WvG`KN;N)y7<mtuzqJ>p#dvm);;RA+6Oja?O6bPe*I;vkwu*&*g_{_h0=g
z+k9Ivk+pu~67ez-)Cl1OJcF!XZaiP?zv?TA=lUFTYb*l#nnh;S#$I$p&ud!(X<8eB
zF(hQ<y@t=%K1>8TZ*?b8oE(i{v28lR_yx!Fp+ZZ?eY|&=YzMv4pCpTL$P1p+^PwBj
z4X3<+Z$!tp7!uvFmnwn8?eOVf&&-<I|NapD>t?<rgR{num1vOR<vkMH(}DlE$bnlz
z0sz!!%FgQ_b8h;FKU=0ZiS-rg?YlIm_&^*!Vc&Nw=&ndfCt~(b6SFA5N()Uq(^E2<
z$*1sh(lU}eH|w)m&VKq+P2pqw%1^=;?{+2HmY$^-1UwxU^L_EVxIkJv0`3ENg`2oe
z{LgR1@V^G<a`j3J+7;PW!s9n71FQ&SLgYboR}DBz<Iln|6*-mg+T4I}Yx3L?V0OsS
zbPP1I6kSsbdNG2<Ya@oJtkSJ5N+YaCIrWsuKT^L2Ac`;yICeZbGIb-!joa917y1?K
z(7T()7^aqvcr>TnY7l4#2}N_W_vTkD#BfMnPEHj>JkO$|FED(SNMf%~yKS*sg5DX(
zHwYLnfbXe4J1(R17~Wd#-YAiqsIw+UqdK{XI{hy2jObFwKn|AkcF`r0XVk?-9Bkhh
zOm`eeX-Ld04gCMG!Ti|m+lD5$?zV__23dMenY|Mc=>q-Ee}{Q{Ac~|4+GCP?ba+ez
z>3VQZnuGa`M##%=Ejn|30V6~~ehr+#yq-PF0vO@Tp8vN29)dT;RJ_<o`wPS>#`JrR
z*`P(wdglIhpLh1dqY@4MMk*l$(372NmkPfnJEbxO)#61?PBkF?fOIp;YD+BZb9f^5
z=Cv^pE%w4peKNHD{h#UDj@sA@>s*Hmmg#*z0g^GM^Zb7GE-1f~xt*pyuPlIFn#<x)
ze{9z~*?B~G2u5x#cSlr=K0eY8RCZ-_?6bQ`A0$K*Os({A`PR9L2;XW<s|vfXSxdHX
zNk{UiJjmXl*LX-^t1VIn?T#+FqGtJedUe_cNj%(aUt{eJeHm=&p#6PM35w9O_@=G9
zx-Ka+_;}|@hSS8^{Fh|4f6yQ8%XQp>8CbNqkx>F71%#0#D@a-OTeeJvKt4E7<5dZ8
z_A?M`Zf?4F)}*h@<<(k&#cD`!u){!5pw|9m&L{Ekd(GwL<*lyZQ5WYG5gwkCovGct
z)AEsg2bO{rS@*Jo5vwtnMLX8?P;py~X~u+vgL;$SSOF$1k;{3h`}?f4`jF0??5aea
z937Fp!aTME8ia&|-7d{JUS>tJT4Po$G@_n5;3TF6-Uc8-o1(tM(>+(zY|m}v(=gY%
zi5^#KgzJJ*d=QZR{1J?AEMb<&IT8cbdAeBzWF#b(kZ};#yw&QKO_9unYxF?_A*!R=
z=seXN-*q{STTVGxv#O|hXi!*h09X)#fCVByrHXCondf9=0<DJImaKaRIHI<jiJnDM
z3$%Xa@Y$Dp>CGmbI~3+$((?Mp#6$KF>zSXUK4w4KP`m+bkrE((Lu^6fy|-rN!T3;)
zDSkqjqhui7{pig%squ}(upTONVE04Zs&?m$_gOfcrsJB*#JWL%u`+=t?T^J!9uRq|
zb(9pQ=!0Qjj4=!#x8}r4;^nTsCk1W#*wjn5)ORVw|61TQIb1BfIUX$*pU%3KU~;si
zM&tN%$r>rtl=bYD@r$icK4fRFH=g?oVB$w|_v6{CfNK(6uDLCG;~$Lb?ZwJgMm*DG
zzM3KjU<%Fx2#fuk!y0keK+x>4#D;tM%TiI`zLhzzu+u=cjhO7AfGVI4W~CvaK0YD&
z%ad!OURzC@NdubNhUdrYEqK@(Vv@t%8)9O@C+`z0wSx#P-dK`%AOSR&7nBnWYDUv<
zF*=n^JW+sgB|3#NSEk*ShGnR<=SKJUO1*cMWn%Uy8!70*<_1gi*lRX#n%6=uH#T&d
z+(vKt_`Db4kVE;BP_y|BmjXv2!Z3^xGqBqi?RcD4!&9@l3Qw>iXid|A6}yTsix>bV
z2$z8GmxL>64eA+TOuEC)o!6~^Bb4*cbDn7up@N$qPEOBp3JK|FY3I2vb3O?p2amE5
zDxfK6!po*DT=dE1PO&AUFa?N`d2I(KJgccz&1|UXm~(707Twy@?-ZF=VUgF$KwzkS
zXtN-<HHP|%66PMO5^2whjn(@>xF2r|3suz`YC$g&MR#+@8~QQLhI_V}KE$ahDVE>f
z2TwM-TYu1BRehQl>3IVf6dvGX823@})7G6TGMAb5tDtpr+N0H2KWD@s7q)y)2o{@D
zL^Qbe-((CR=eBs<5J5A?eTg-1(cF+&z!UtITGKu+43&5GV~y{Jf=?3)JTEfXskaHz
z3&k~;uuM;Dc_y;3jx(XVl3g|`ZRVYKLkLTJ%}&LvOPdoIu5<w$TiR;>A7O7DRn?;P
zfy$Pxl!ziAf(la7h*AQ&5owU_mJ+0-y9`v2hD}LHgLF5dl7bRShk^)7NH=dTk8(Ws
z8}Ge8&KP%$<2}gUYt8u8RLP7#HumjPsXmMImvZj17iY@5DfwzG%$*16_o6zIrRc&0
zPEY&29pmw>dHv?i44~E_Kd+3!(kl?+-`4;4+>qqw6Y2wjzKg-$J54dTL;XDVUuBWx
zv8mF)@0DO?W*u{Kep8XJplRn{<ZeG-X5?6FD9_&a=<HEBLcxE6S$JQ1o)M?%|NW9Q
z?zI{3l%0Z?WH&u+(e{bB(DP)(6*5?<&|H2fvGQe(($JNXzrtR8F{B!`0{Sn~O*Of5
zm)#jGevBLcuZVZ&7lOU39Ii*_1DYA6PwINE_G#67c}Dufm4V?+G2<_U^LuqFYhrp|
z+WSCv+{w8~B}Ea)376ka>(6@ufTGfa-DM3hi+B!8HZn~<*I&zLi3-I{f&hNiIVZ-^
zjduOLl;PoFmhURVa9YH<7EZl_Ffdy~XU5HBcTGJ5w3xR<9}(S?Qc{{pVm*Z{6G%kf
z<%t^)Kc7+OvmXAo)4lJ#x14z9`x&ypNgBk&2d~ch8GRXXPr8Apn)cQuHZ@V_T*h(0
zIl5kvO7nKlh=MMrl`dFFX^vvBcEr>qU+xRxu1)>!TH%PjwYu8aMZxu9`{z>Hf%M17
z!dX>5J`Qzg*sz@ibFTitymWiuB>P*6-Dj`45;)w~=OaT^UQq>*Eh~P#OaBsL5&{E_
zMf%vq{o;TtZP^qVEpiiI3WpHhTVo!}a+->1*I&GNvD(I(|4rU|OF3<um0WF2D|5co
zh<Y31hQ2}Db(Xqd7?F{h%1`Qj8nme{h-lL$pgZ9#TFtWU6JNU>2wcmiwYDjR1l=ln
zGQvJLwe6Zb35p`m(ey7K5qO>B7<iOINFb&7<n0j|qUpYIUJJGmQUOT9kVTR`mpz&`
zr~>|FbzuYWQ*MyF0X_;RGwx-Ox>nJgmU{op{nV%ifJ|>|Pt*N$gvsdn`ny1PGbb(U
zd(g@6Kqs#R5}XOPwOh)!^ihb}Z4d&<3^=bmB0Y@^UeiL>V1rleGM<tAxnJyjeq2%I
z?%`(rFw71A_i+w4Ntl!GPw9~n(ma(WT_t5zO`}L-fiymI-c&-j!lVms`NtG6SMh75
zBxpiNENCW%E?2eBoAM1vIhmR{?=K5gT!!@TCBg9Mn1SV6RawU%QC%y2{UsB6*<{!0
zyfT%z$)6SkpdL|t0u7ZD1J&qlZEB9qI|_#t^)3G5bP(fx4Zg-|V{-Xg7^qH?8Il+|
zb1*od0(w_uWO1Pxa#K^`*revyQoAGaq^=Omm{R*ww^mcK%ML_dZBWzF6q0Y~9bZ;K
z>`Y8>P$UwLT)PW%Ksp&zHwL)B(;6aCOO^#d#b8_r_m7SXBBiQwl5n!!C|y9fRIoq3
zyQjwlOoUnnsMR;3wK5>EGXWvXps@$fIZbvb#^hw<Up9Vz(KLBiP%johj?5@ew?3z2
zya~J{FDDlXQrmXOCl%spyc@eUVhi^(gSVeu?CGhD88G&c->iq*F`6J>^8o7g+|V4m
zV(qgZusJQ44Ta|UC4f|<KPQ0h#c@^^jYnH&;Y8xS^LAzlZ*qTXfU}PrPYf;k(~(&*
z<-rcvjX@6N$lma5>uh9L*yo;!*=c(?+SCnyL)m5bs}$UV&hP*+viO8<ZQARD?wkpn
z*G}@LR5<l`w$Hv%h!fyn@E!5*Y+Lg7yGkwOMf+ImoR=KCx{&zjw|hn!NA2<6V2Ed{
zBzB+Y`0<l8(C&EU&0P93@&WuKJr2J(IoT;~w3AI=b|iVr$G8qSyi+=jH>>HltM8N$
z;KH&G+W#nsxqa7@=aYJ(g_h=r%2S^)++4T(=NMuys@nuN8oW`sc8iH=^)gyjTbwZ)
zGuGY|aVD;E7G#l^Myx|}_sqgfM2wNg;#b{%rC_k_dum45yqH<IK0}ImW{CECIaO@;
zX_kL^LHNV%1jdf}*lFCTvsceN!bsvd%=SuwhT@yOK1^uL7Ps%J)mM9Pa2=u)mYb{v
zeUkC=*B>L7m23#KD(%uBx3AgAMyM}ilq;3d&8#%Dfil~wGk`$yJs7|>tDRoj7qz=V
zx@^glW|k1&@+la<u}9CjGz?k)?438TtMb?^8it&TB>)s$=U(IsB3gtPp1W%412D<5
zLY4{^UKrz8&f@2@LWolD5u53=%GjD!IAmm;sn3vkh-P*SVhLX>D-cb+L=jhTibu@M
z40l*^*JbRm+-U&m)o|SRo;A4dowrNm)}G7O%by3DWTPlxsh<)Eg~b2RE-;+RMw#4{
znBJNDWco}3W(oRBmp&NnUzUMAzElW6_3mWRqlzL-D{QmQi=a6z<vimq*x6G#J~ynm
zFJCbgr7aYs4EBN|GRbD`YDtL(U7DQw;tFvG3M0V?T;d%^y(lZ`2$MMZGh1)l6}T@~
zbDBII@Dp&TSL`LsdUX%K6hjPgVo_k0xjEaLMH<v3mH5{Vg=t_+npefLs7iyID%Ca9
zrw6bEr6deaq9*{>lX$ZR$LAM8X1M85Bm4zaXlYQ}*WUSfUt8WfGT{89#&FK1-di|J
zhq2ypN|_5a!C}znK8I_Ew|+T%Pg}{y-zJKhT;c^I$1sD;{`-A#G6|dkof#(^u^A8d
zP0Mrd3=itj2Ejk|Xmbrc5Pwl@@?^}V3f~CU1U{xW|M9cnJ@EBWJx~1zM_BhJ%@1_%
zgXIfiPals?hax5@uyUhcCT2XTnSZWWA2(EbmkFYa3Nn-M+OeDztB*DTyR9<oD}yU)
zoS-LItLXZ)wYQq@E%lIqti<Oef}FG|>3cwOMZ8QU%zvajPx?Kxl+vCCo)gxJ!{~ug
zywY~UFS8KI7@xX-{F-f7-ZMi;XX8y4YPzNEmp@mQ-2=~Tn)AYFLKPzzsxnIMe7FGz
zl;|yP)W;Uox~*Eg>;lnYyKUgr^D5&+WF<{KDJZH;8m4Tb^|U)5Qo0L`Bp^BYiAs)|
zTyHeXI!O20z(=lA<P8jr)*?Rk=FlTbqr5>Sc<71hkDqh<Mhl51>eM!4ki{Gh_Kh}3
zN0@}r>uwZLCf|lztb@(Xn6|e)l`~e{b>%xOBd&`2vJmi;<7AyV_GC67-6F6Tt2#nK
zc1R3;<7XKW6|XD|igAYPNLT_44HT=z*-%(Sr{cIxo>m;yt%P1V1+ww(rL0!R@-eUK
zvV`#cXn}L{b1L-`eG4z4w%wiHh4K5vvlnm|cxd*RB}>v%f{z6tL2rhgKz}pn3{RWz
zK;azBbWo9`VX&J5JSNC!jurP_fy|<L>Y-1M#b=eAxbUgvYA)7WUuACRV`}TXZ+1Ux
zhIZ^@ci{&uE`O!MOSEI^>%@<gXpXL~=U3@<#9<e%R0LoC+cNkd+{cYKd0uJ=%Wcxu
zGSTF3Pu;5<nLJ;blhnL6PMk1bIrv>Q5lky#u$=HSp&fS3wQ2mX<4=;jzm#N$Yj@$F
zN4cdn(~b>w1pnLa=LnH^NgA9e3gw>EPeXCW`}!0$lU$4gYktHr{V#%sH7Y|cktwkJ
zDa<5(sG;Mjk_%D;?!_s3LGG~;VRdJ-XM_WH{yg-*5THjWHBZ~qk3>bVC-4Mu$l@`z
zRyqsZ5N_=X#-o*UrZGDSj7wMfZpjtxfdNqH)}Xo#PAu$6_D)v1a-Y=`5~6ZvmBXi}
z-^ERHlUo$+&-7$mt|8-rlwZOlwmTJflU#Wwv}?sw(Z?1`*dX)>aYV6K++W^<3_0`g
z7tecseAg@$JpH`aco(_C_I)TGE~T@w&@=!tcE`-tBYPa)p~d?~L#|?*PP4tO0#e%b
z^HtwV4$<%p+loH&g!5UaR!y_OZ`gW?12&v))HQxH-n$+3TeGY$7djowE&8PI)|c0-
z_PS>JkxX_a>K`Gd{^pTcW_>dz+dzKW^cG4Wv;26@C5Q7FKp3<NuMg8f-ih;!_JcXd
zSeYA2gELATGi%S_fIz@@+*=hcBC_$uo)j<Qz6Y3TF|?7Cq-03+>3zANEA~S1+bO6E
z7xh=GJL@tR8in8#<zUwf;deb5v-Je0;Xgou9bt*Se<ltRPq8a-Os)+!ANs3<%T%vC
z1)Oii^CzR-21@T@Wd<zn(-2ciN^(gy3@zxoJ9=O{I2z6P@#7_{=vhCLn#8adG(}xk
zwrtFJm9_ry$KBQx@1yn?l{>t>^EHYy@?H`|Vu<M^f+9}PiB~uUvHu5GmcemMtXI$1
z3#!YRmbz@(VQVRvpGou6d*`SBnou@GI-VpWo2n6PcO+|RJlET}H$%O0wRl=hliV6}
z1SN3VM>-TQ`r<#&Cf85;Bo<r`vhj(yd?nxZ^8Nf}csl~JX3}|6G?Imu^y=M|Dlpc{
z-_s5>12h`wBe&#3j#Y;pv+=sg5lRVcRORQI%2&3QEC;<Eh+tgY+^umg+&eRn3}8&V
z2$ZiR6t3cy$b0LMoHzxf=DP|yNhKk<$x^;uP{T(8DXRm-`kWL;BKYaL0pr~SpM`FB
z-&+{NF5~Jc%*6Dm=W1NigtHlt-=b{t&dzmn3M(du=d9OT`19vaO)pJrh>zt4&|z9M
z4Q?zVDU`uBs%^O=YR*dGVUX}X)%^+X#kE5}VrTg#X4;-6gH44Hh7_-B!&umE>g}4(
zI#)2=ZKUX>pm>gj|HPA6%e+yY$bl5b@9VIs_fFm;BTB{iqBsWLomqyG&<+8>I<pCI
zAi;DZEYqmzF}Lb!Lg%|5N9+_DSzvwImu&Py3$IziVdxapi3(1@_I!m_dzI%~?EnBF
zGSYjg#WMgGc{A<b52bje!<6^6XdC`|j`)k8i!M`9nFjgSy24Z4=g05j-VOraRb%KY
zYd<&=0WXQNsoNA+Nuri-htr)?kc8S*a2HHFDRr11=B=~Cij6u*b7}>X%4O<XB>qJ7
z^rtU2@TB-woH~0Fm7%O;vibGq+}PLJ)-daN^|c$X-s?ZVs7N1N^I=z<nM%CtjVNtM
z6S!a*A$)J~lQ~A<KV@eUH9HXM_Q+Vjch#@Tda(pr+~A2*d6ol3T3mO?*aVmYs`B&v
zuKpOQR)-r;vpjcPZ_T#M7s}U`jlQS2?#-1{r^51_v-;f^eq5548x7g(gZPsnZyE}z
zd@*<=Ewm)FTi<uas=ja=dcEc2{>CXacH&q*=LMt0ys4k<s*nfEcTgq=f0xNPxtYmY
zIqBAbv)M696p?=u!k9%uSVRmL+q=Qc(ISnhrLBddshTG31=stTJY8a2a;|E^2Chh&
zQ7iYYOq5#4MFB~R-%kHhET4Fd+q|2;xy<v8wkH6Kuu9gqAlkRC*cmfv>8BAKx>}uy
z*B8R%(bNjz{l!`!BTqwHE7-``>Gj4138ASLKU(Jq!i{C=H0j8jaxG4J=R8*w6-nzo
zHB}VQWREA7rrJ3M`chZk1W;6`0;Hl0gq0iiO6eugnqxVynYN6J`o$RK`AOA)b)DlS
zl2(xhD-Vac8*ku6<-P?NC~wjDBhmnqH?NKN$Y+oyDk2cSomZEe?PpoRIp(?keAs=V
zmZNo6u9>Am(llVJ_A1B)uxtp%B?PWX4s(G4E_)#rK_m)l6I9#uAZ@u@oFSjcF;eZ^
zRiJ<N1?p?&A&Su`5LQ0kO0fl#rwlpSjtF}AXNj@Xd}?~SG&l!tqAl;~<9#0IcvbcR
zr9*dyU3N2>-uCN`$tL?=1bb$a6noNsWJykPZ*9`I05{4$Dq-u0C5BgD7>+icYcT|2
z12+H>N&Mw^=>L4Vu$<sU({UNSC31zd-?Nc0iQYm^hc%ZiKdl;jIUhWPg(Qf_Ro-W^
z+r!H+#7Fn43^dlSm7wi1C25W}c`%>G$r-)N6moc)5k-EACgeW)=o^{vQOy#Gy1m_K
zZ0%cy>x0>TLQd*$IRhxw=@<@_LXAvZwr+lFQ)gdjp9_^Semj0di6e0I>g^Kdse?ZQ
z^vmq8<~uvTavTTuCj8}T;ht-+Try+j=)LB%Rl&t=IM966G8IJPGIg_6m=j|MAS-Kb
z4$~PE`6L6sAz@azIZJ8?98%647kV65A8kO9s+?5(Qf@WGi!JHkov&WFwh3V{x<zdI
zb;Dn%7GSMC9S?pw7Qj-1WzP4#1p|93oP?n4u4NDc$41ee4;P?EK?K9xkZ$*I{dlHR
zcEcfm;<I6J-m!ABuiEfyv^QkxEcGhOS6)$kt`F#T_(=xoiq!2YLujbh^3&;~7~)7)
zb@C}HDyldW=iaWY*u}3kLhu8*tp=}5(_480nn>xV-w`jMd=9z}k?V6(Bnv>X(I6fR
z7dl!YW94&v#0$sE%`f1@cB~C}B`#YP6VSi#ORwMrPUFL9&DVtax{*3hcg04_cx(sC
zz(o*#&K7`27YNDOY&oR0nI4NCIeAi+Wc<trT;ps!DPMbv{FXL$mQ2x<+>r$OEcI^D
zqC~YHDwd7I(6I;@MEBWK8mU_*;Mt9mzrNf1G=s$I59Nt;Pv1iXxqy#?JahhzFF{E)
zK(dNmuZAg50u=$1mT?$jfn~@LDmqy(TkC`pFy^uQAwhEp_F23`ZoE0Nwei;O?aPb<
z?!Bw*`~z&A(@nj`jh_;#6=&60e=J`zKHzd*ffZ@*)XZ|HsdANvO}Nt1ZGVgOUa`M#
zE76dg`SsA8saiqd;Bu2n%F9!bRgA;fA7f%akWp8Y#yGmgy*V(2A+&OdYx#bSlE0V{
zhM}A#h&QS<{z<OVsO-tI0O$$5e1Z^%iE&vn<7qf(JHQ|VGH@|Dmo-OMfOd@mK;Rgt
zmE&ITo$p|5uRgWaB~B9$;5b{Y`?`~PCzP;UG`+1<$L%k0^*kKP!zTvp8ALcO3Hf+*
zwIDNFf|ep!)+Y^5kMYE-haBbuP$W74C+CECIm%c~=v^^_&Usg$JT}$ko<xdK*7|U7
z+*qwXgvE7|JQhKVZNg2J;o>%A^8>jo$INWJgS>sK!H|LtJKR*q$`o-nS5k+(d8GQb
zsWN7EUvzdKwkMr?x1s>ykOu#oZnu!t#mgqsHALsH=^!0Eeo|xIi6fJShb8*a_6?Gw
z&L_&=orI@^&3qQ$#5W+zyvX9axn|*J>W(4dtnGx7)<x$}Cp}p$2)`~?%?So3?=gNb
zI5w$(;4P;Op*gePt9y_4_k2|lgn{Z-adZmM{oKrkH=qtXHSMlyNYP^W%1yv_$ItNo
z>O{kqQlZMXqheQxLi|oZ``TQ`X5{+Dmdnu`m46{}28Bb8*US0fkh9>Ooa+vs{Z?G=
zhR+=L?!6zfzSS3=g7(}mQh$G;$M(<9#bh)z%-|Opsuy*>STR$_DKjOqrkQ#=-RI5q
zm!K0oD*XgVAL}Q^J1arcfh#0;sw>do)dPHP2$Jk5((p<we^TRy5^!O80W%6*s|-ut
zPgt5g_Be^)EI_E4_-&o-kl{%KFGz{KQu_C7b`X2tYh8w()(%UQZg=ma>w|MyL@%BT
zp>q^`&jMZZoXE3-O$5ykK-JAhY%o`2k;`YHUXUTwnqR*++poqd>Q78Y_WD9bK#Y;s
z=90<WoK_s4JMm-pPrA-h5S}*$0Ib7ix2dV;l@wXk#EfN@;zZ`!&5`XDaGyP%tzYgg
zGLg}&^Qg2BE87F0rFE`+Y(8x7HK}|{AmwG9?qNQa#b6MlcQClZZN3!0kd3>AZ7vn-
z61zfK4aYhI@_ng=&;g43!Akr2OC>H>DTJn!&2jH7GKyAFLKBKjmOVQ`AL7QK`_cM3
zJM4mkeuD8~#T0-{^lnfvhYYRV$R1Rxl=XU%jptI?c)CRdouX{hV+D?;hTS{=VYf7t
zlVEB%cUa|9rlRP4<@C9^I%k;1#LzTOUg@c*xor2W<SK_^v>pE)0~-HeOJC|dx)9OK
zgljA9D3n_5r&}~f3FE1+;NUppAZ16xdmH<VX{t{^P}b&7t;Iu=SH-1PHa>;#XpH|w
zz~1AQ&B=#eTX5N1Dy)ApZ>1!y_e0^`*c+t-$`)yn4=R3P8vKGIpk!t81t_Kk9LS#|
zKD(|)X<LWFk>-IlHkMcRk7%S;uY~dyON}Q~mBKZWEQncnq}$rss+6ffi*@ks%^>jj
zQP^w@)?*HnZKI|kV-xQZL6Tb#;~WY|{>O)30j<XZ$W)zbQktC5)BO6iq=iMcjN0nu
zBtO1>EP$m?-4_q9ioHz`b{{M@-tPHvDL6P73omXf%-S_&l}c&SOK?e}wYcjVtJu?1
zKV!S<U}#=Vh;v&W8nWI;Ox0KJyl<$>@+<@7BaUb~?9LEB96!(Gixl8gR8%a2n0uf%
zgY1ipJRx!=wtCxHv{(B*0R_j8mg!(J)<A#v>nfO7xHeWzzkQ}<0zwSBQGxvDg<#&H
zyzW(FA7(TH*jWEence5t_{@4W71}IuFk--5EYR6|G6ALD-Cas?h;7J0NzGb-p>;a%
z0n-EDT8@b8(t6(W;#o?(SMCeo8qpkev`WcS)cBi$I)r0`0)<K_aLjH*8uN3fBlp0?
z+!nxD7p)nUG`jwJV>Lx@!?O1V&CwUe%oM)$=zqY|_+&f{g@#?jT6CObxDcM~u3f%I
z;ZCN%@LNX5EoWEPB1bVJ)?|l2wSwj`?h#p;kZ0B5$7~9prmg;j(WJ~pujzXV>dznf
zB|l=v$+^Z)#BSmSp86R4R1gP4?(@_XR)3A(R~v18n#;f^@C=W1%8M75Sy?Wwxxx_4
z7t#bfO`6B%634D;cxfNS2;8-v2%8VyQt_Rg%Yx<Hj(O-j6_n;E2(vZ0*mi!7_(Awq
z%0$rFJdt$@HVqmmmL2b#b5F9UWYR#}uxnj|A1GbzG)HX?<=}ay)O2S%WCYR9#<$Jf
zUW?9IgGW^bhKPFtUOO)7?>H5be8A=`0oQ2~rWVlN*x&u+3yXq3LH|UJ3*ltaQf3k#
zz;Vu<9DUiU;fQr^Xu3wUR1W<^0f#1@f^4Nx81r~}FFyCEo6zu(bsvV<dBhRRrddS)
z?LI}Nd+}Lns0v4#`9HLcQ%K6xv&5~_w=cbWvgH1XSxV^3nFiJk_tO>UPIXW)<PYdc
zs=oB_ZQyy86nmdN<UT=N8Hk3N)p9k0#PW11FTkCizC4*e$8Nw(#75)|L}*T-?WQ}~
zSwak94@(0FpH@<{!Vf0TgGnSF4RboZywJ}F^9IDTn$kKt@kpNcw#xVUuD1B^`00)>
z4hwXgR?%mP=%A3LFCp_y`aP21_L<)_Y7+Co5H|{RWAjtnmlv#~bp1PH27PtP@E^cp
z!~hGDg+;B(fwEIW(ct^uDc8N{@6xZ#ji#qfzVXlbKbz4VgrYIL!`wp!YK~t{vO#+5
z)vp^cB>)wISX-uo1m|10o(y1@a$}h?!mCv4dsDwdgof$2l#KG0?HAWn6<=zV8iQ<t
za<)UtZ_E2Vcofqhb%gTiPl@Wo6RCC7E=!YA*u9FQ`nCfxt>Vyir(k2{5J0?;XRkWg
zrjl{$=|DE{*&&N5WEg>eh8WldsZ4__Yo9d-oec*TtqsBeSpXPVGY(E<8TV<+K;1De
z+bEW;g?V*ne)m093-Xx{F5FtT>+OmoIFyW{xcCXACsr5Yx1DdOC`6O7s!0M^Hyp4y
z<V4flR7>hhJh=#$NVN&yrtF>nR)-`BhzC*6By{ki9ue9;yTgBpj3z_`v%Bxop2)1F
z;6L`6bSl3G%RNael8EuG3FV#`ph^>e_@3Y^36y#jj5wJGJ5vC`Qo7&u&R8B9pj`9g
zPrWI`!%g$4;!B!8oJo@bo4u#Pny>i53Q<P@AB4m?*U1xe1EKlNE^Qfp201>&ojB)u
zmm>_c_iT2q3V3`@F?_e@qHvg5Ka2wJ%n{i}L(bnQ`L;6YAy$Hl>m}v4sm#O8d|y1F
zNZXF1ni_u7Q9t~7^V^RM^3L-NJ==ZZpC8`ee<nGgJ}&vENMMW`N^l1%1P5#hhGdw<
zMr2vF<KxU987brZwS3V6P;}IvR#Fa&R602t<m;!kK&>H>X&15?_4R996fe-x>2~ib
zCbB5hp522eSX<_5BxGeMROh6l<rR?QundfH&6xjs7F5cubrne$o?`(AmQQf~N;yrR
z_;Govor!`&$F{ZFZc19wdMypuy&}i&YTM~+n=pa~#EwVyy!mm-8IHAhZcoB^5Cs?L
zWMD|!wr+NUOnfA!H|)B%mqctinRmqRk}>CYSOBs&_`1uiwAXKYjIK2s3-0gv=l{t9
z&}Sul?NU+B{@#~*zJA|}T|*RNB)E5KB>JuQI5mrIZts2{wJtlsrQ~z>CY>WQ70wv0
z6ZM8*X4BILH$|B&H$KLWzl*#sj<(Na|D><Z<svxoamj#SGHLJHb#U#ze*3m+_0{^j
z)^MhF2q2z>yxY3-H*ewIP@#dL9mX|`D_m}*zZ$z>=KqbW=MTVFfUi?&Csqf%;wFey
zOd!&S($CC{A`5mmN<YC<tZHo4Z({6*_n#p{hB*uwjj;1FY|K|O-<M)AWd8kBwD*MG
z($<Aap>{sGV<_W-LvVbh@UrZYAGL*cBCapo&kj+%v7=PDDt2Hr_(;)TbzLmMyLpvx
zNW66GcdejQ6`DHuBLMzj9{)m79vAlrgS^$qH~!YFn!_xiCBVMr0V!LkL6|dU!qRzn
z;91B3nip9YOJ_A8L!KVla4om(-yjX6=kReM8aNJ-M);;Bl+FYW(r}n-TG2ByKFuf>
zeJ<px6ko5=pDstW3B0K8Qz{r-c?+8uE&&)8Tv^VTeTV^eln8uM#umt{5i<QgP@It|
zrB}xxz^uGbnZ;M{etrelj^F7=H2LHdbiDaD#+8tlsW*2lfA(4HJdR<J5WhcV2lesw
zbS=QQkY)rJY6y%=-L<ZRUE2`uAXY09gvKextq{K=6YEB@C}*$`YG34GpE5TlHL#Dp
z8^Spzjkln~65Z5qW+C(PY4V8<+Gqfw4V+jqb)BN%MujVE*i$gezU!QzzK@Y4?9JEL
zN0xTDo<%g{)h%%F4&tY#r9C0A-37yC3*dCnEWJWsNCa;klR#4i2L$x&@gUuyLyGHa
zJ<pFsKfK9t8^t!!2AzLUjr*0;IHk1X;fH*wJ#H5{q_5ZSDPP;3DwwhI`1Y}A>4#^e
zTt1fIfURT2BZ`iqp>k%T8a*{cSULGc<XEcZy$dO4g5=PDVS5iUe+`ot`9_eowR2Kq
z0(Qj+NE)fUgB)xsvK-(BbSkt*YH|(5bSwTMq;!?rSS0FJIfTN^p~}?FyG_W~tc=Y5
zFn-w?egnP;UVvr>_MBs@4__k?hcT;GNoM?2$2NwSz;#igP76T*FgKj5!+OAy>E2On
zVFbsv5SbB>;GC7f!^*4?kg?pc*>;p#F{BeTnHA<+{pxH}mVhB<DQF9!71epF(1_!;
zw`n#|?VK-D<tbX*hlDX`B?V%R5^@;{{n#z@)?x^`Z_!06FGLHpZ3*AIp93HmL{^W?
z!rLMZL^ydT-YKLL;}$X5e%nB6ni761=pgSm?3%D2V5KJtZlkh-CDMVtvd}I$X~B~V
zZPyReDqPQ%FRyglo(DQFv({C8y^-1n6O9Kl15wcLF)_n78@&>3eefQ66J{n2z*p>K
zo7wP*rJ)bP!*7o0uQe;>sIgaqAy5oGuO@P?s7n27@YUMLpBw#-cJx0}%0~olB?_ea
zMts?x7H~>E1j-8I)onV3<ud%qajVyouqC2=w;qx-aQN%=!3dkSFzGVjG@d)-=|0s;
zl`8d8qvw0tjX#eaO$bVRJ)>oL!IC!(f1Q_|f%BTqyNIgH|0LPpP;SURPAsY%r-z7b
z6ArWPz!kZu^WHlj%+9T<C4mgJa&n!V4O&9WTx2HS+nm<XP8rA#GXZ<eBabKV;O2Rw
z`O;)-iXSUKCz{~Y`%#j#Yo5A7=sU1k_G)LuQ^+^6^iridEsU~*>#<YQF+WcDYaQ<*
zT&U}=yE&7v4{xs$s3z8p6U$Z+-dsh0Rz)v?9CpX{_oOY!0E;idlga<|4rmq-Mc$xx
zJCua<=M-x`#1OPcv3eP*smYV4Pwl9U9+J^P&-Yd%n6#aEg+?e+2q>xMQ-)9!S~xP@
z8gdJT5(hEGod+x?7skDPjWRgeisYUYy&~_YB^e`J@eU4Tzr-m2fV?C5R^+R<YR0xR
zjVD`@;V{XdwtaCEayps4MZ+m*>NM^vsU*;n3{qvibP}exWrWfH{Q@?4;H-lTBO$BG
z89aNMij9-K#h1x0YI}k#<#Kc5Sr)b2OmvTRPx`aX=cRoc(d;2jiicquEJE9ji}~SD
zYOx$6{pY<GjgI$mxLkpYH%X%NW$I;Vv%G%TG^2Gy1fhD3$tC()gVBpIqi|Z0?TnS=
z=BcNDmRJqNr2SspffA;H#dNWok2|QTnM|I(Iju)Q_8+YL+b@ja+tZ%QQ01L*<io2X
zTR|Vf0q=QNm>G~bDU9c;IcMAN0zu^<PcFY|!aE}*inCAe5-XeMl-2q@TgaO;VesW_
zK~)b>WU26^Q1T<kPT|6@8F#9?G{5Ioi3(t%ozLw(+m}EN&4AD%?X(Gg?3qM~o@+4(
zkjde?Y8@^7)ooB6gt+Ydk9JS5u6KYJ9gz&J(T^n7MIlOxA7;`vUh9wC&m843G=pO&
z^O~g?@+VK81g*eObrVTS=qFAmd-1kuq(K^Hw4q*XfNDWPM8v7}^tKw;<x~*NQFcqg
z-6YrvwzMObNVxRIjh7g-G5$9-t+%lS3xTQnJawH0xguzLm#J1QrE;ysBldVv7-Fa(
z5m2>t*3FBMsm1&IQmx5|12swduUoT#TrkSI%a{ND>S?IaHE6p<<K+3m!TW;!hlMh?
z^9L$N6}p^N4CX+*>Ym+Ja0R$IahXQE=-<3Cm&ryJE{-Q^VDqq;{;^438H3d_<)v=_
zp^ofMMfrE%30kykXC+lqt<q@^PE3+~MrrR0(RJ0w{j|~)cc^ykQCfP{)E9U$miwR!
zKoOt`{6)?MKJv-C0L2R5CARk^UhjetkJLV#fm&fnNf-g9R_faBiZWIR-pxV!HHaUq
zPR%*?@0YDp<sjAjdKLhYO5-QcsYb0Q!@l_+lb=(2qct)E<sLSDvU>@#`GdMp**0Nn
zOO*`0U~##}1ga+B{2EvxxStpWiVEpci@f#2!H}ccm+`jQRnewl7tRata85g{JFZ5t
zs3sC%e!~A<8g$^)njPotCEi@iz32}**NRCY2<LGMs-mCmUo4dcD%kY8vfFYspQb#}
z`V8HWi7fb3!NwFh16Y!Dj)nj6bu1)8vxl^Ux*P#dm8zYr*dI%&d24SC9)X9X)++~X
zr!c4Ns{scb#z$4=WL|*;>tpWaAM`jl*9nYQ?K_-XcOJ_r;+>|ky~?2KCBtt5AfpoR
z+%-D$n~t`ML7iu5UsR9K9J2s2yt3Vwd;sqeh8V&)QucU*pM2QAK|uCeM7OyBEQ&U$
zSuIQ0ZIrT<7{-WcDE<uHaIu<&=Dyo4WYc`fG+P-<;5YaE_VM`wxLl-YlI2}@#Q7VX
zTSgAX)~QWlJ%a11)PkPddG39|`t36tuPNv;ev_APyV0Qg52rSMP05nZ00-w>ou&cF
zPXB}8`XSEcedm^7hAlOf>L`=cPlr`K+GRNodc$M%mb_euoO{@@1(65r;vYL2hcTC_
zV+sC(I`9tv!WIs@RY3r@A<MYAQu*aKixW+VxUN$wXccA!bDN!e^gG2Vqwn)Zd*{u6
zs_2`BcM3-u%7czSgqL}^&T3ypKQ<0@2&XuH48>XXQH7|z)M5aj>4Z&qyKw1{UgB0s
zQMemx_!-vqG~m>VAVwg+@dzmT5b|pY7~{(GEN}}a1o$6gNp*bi;Do{;%GXqfq%xUy
zC9nf}++HBhklAo@Ho$*Y=BWjUVS*e$7Oh>r-yC?U-frsYu+xy0R!9J3lHRcyFa5MI
zqx@VDxvK4YH;RBeg)v}&g08IuJNKv1xC^hz3;R+JmRYfa?DW~L0=5eZqaKhPY3RcW
zIOls^ldln;x|p;}oxRHvA_4WNnCOrlV4{jRB*XX3i6HWyQ8s#9DSxn_+ogi9^C^@h
zDj1qw0cUj{j$;M#C<6Yu^CilV2_J7m3RaKnB1%LkauP`aZ!s@<$%v_+Pl~)G!gOv`
zTllJ>>+oGuE*Ptn$96AnrTE$ec9P?Y()c6GGPF*|HQ7nnH~aDk9nQDl;612pfD?4%
zKd;@0E`I+fejM+fFUTreflDsg9yl>e*e9T&V(SmOw!0$jSa$7n-8gzCvz>DlRc7gH
zpwHw!S~?>bfg#?iX?f<;_gnj?Ca8SpIEoh_%?BUNlP*~1v_8b&?u;HSJh88?to4!E
z)U8(KASXSVp<uP^^3qQA&vcbfRp!jD6NaD47h;qy{g<w7`2|X}i1g6HV>x~v=|Lm+
zVy~EAFnSyKfw2&&J;-pSS#l?omw8M$e^8CIRyE|1d_cw)cbvm~7LofP=oi!AiL!%E
z>mK$+xpwy+sMp}?_hqXabw>VHV#4;LCFbF9bn-Z&m3b1pzkMz?w&Xe#rs>=uNBKFY
zA3J7mvTs>*^gh@)e3ZC?Pt#o{oggS_Fz+oJ3yU-o)-KnPz-tvt{;hGBhUbHxTr{kW
zsoNY@+Gq7a=rt!3Fi!~~7@TdDz=<irAR+)J5?=D`Jb=vy8MZ;*Duzbc>UEr62B^hi
zuKZID%EMt{R2MnFd2bJuPT-I{xC61v?Vu9;F;LnCOK~f#2Q!kFQ)JyIKxEM2mA>+|
z{FBjsFKG+HWOP^Bn{6i#Y}FttNGP&Pkl$9hAxM#q!bh~Ipu16@u63iz-M*dn)7B<V
z^{@cYwr%X?!xs=+f=v;ej1>VyCYts+F76Hk0HXEu%tvwV;>Xv!h^bi#BiMOc>|DlF
z_msn3p7)a<CxhA;UL;hAoarOVGayzw4!iI9l)ZDgdbL&aCsc7zn^Y2fR7YuO7^EZK
z&$HeX7f<p1Gl|fQAN4IMxv~lHOAh*!`ueNq*dd#G-CkkuVsex~G=rQl-f6nu5*5=L
zW-?!De9#0)cti<pKzDYv{wywa&G}95+Enn92y@TxQZctLrGJz;5GVU+qIT2Ywtp3V
zd$dET(a4NA=*t32@sTLEziZgNUp4HWQx(Mf8L}ZYjE|V7Ro~N)t=O#0qPu$gfyeZc
zP#<;x6e32#FT3Cj3?l!OuckH!V6SiVOg`g@9e~4A{BYv0c@7qi)pZw}$;Uwayl2l;
zk8-sxH@}s}2Xo*NQW!N&$|m2%sjGw>Sx>^9Z_YHJU^y*nDpT*ZI|&nP2U`Fyvp!mM
zv8*GGPag78!VJql809g#^}*vRS_{KuAL{n6fhiH<xj?1}lbBgcMrScV>TPG8hbQ{Y
zZpN&=5`G>RvnxCE&L+QkFN-PU{#8?Oc`R*w2_8D$wPps9Yb7+?v(%3Iy$kW+$>|c-
zj4LPy@e9$6+L`0)Z#8*fH_9M5(vu`|SV~sb!KSPR(gc<OhA@+4!ol%IgvQ6k+As}+
zh<>fafS4t-$wJ%Ke;M1IJ7REO3oFQt&BQA6WRcxfIJU>BrfZaW_gD_ctc@Mdgw#Sx
z-jg;gCnrlZ8O$o*C=_tUh0sGi`X~QlxA<W(szRneIw|yrt6n=5Mnm5s^YU{e>u>3E
z=f(;&eNJ^?v)>^)oIYMa@efV=U^ffjhYZcVtrS8wBopUBs!>iS^~j^(%~~4g*wQwu
z`s&i=)Z8eaRKmrfAEI!Q?jL~?%>&%BAA)q8vU}Tde&Ws$G}OJcBdYo=@6pnKrX<;4
zQ<Ch)3?ncs$vzoUBbP2+O7S}pFYK;{l>{`rgfEJmS?TrndkM4cmfe+~{NkNt<M|c!
z+-4XmU4Eu0XYciaKCJBlKB>^TZ&@Ralc^}i%tJ)*RfMss7f~2o_ukuC0*b#4R?>+N
zEdrO|J|y9*0^`<`)5eLPRNuovY+OYVej#@8AXk99GPjDWGwmgx%X9{uFd1xLcd#~H
zyPU>XUwdvl133HHcz~gUcNeKW%itz&9mwbIc^7|2r1Jo_**q$OS@&fv{zp&$NSJi2
z)QBgO2D5Yf_$$1fnSeen0!EziZsqZdLtZC<PUQ?d<%fBXm#Qa)T`gNU$q_6c5s9Bc
zKXqem&*@k!aW&7CHu8>%spdTFS3Yw4A)|M4IY;d@pdOdNA#lD?*x|Bujb^Wf945hF
zNEJl5RSRSfK7WWdAxKS4m9(_XZMQ#1=Axb!Qz&$saME_T0YhwW<x4GnqNOn~IW>rj
z?DzLPgD+M7l6`!StAFrinxh-;*F9SWQ%`S97p+&TyA2GvpPTx2l*k_*>@jL5zVY$s
z@h%7$x!7N@{8?a192Cy|=CO|jKkwhm4hiEgDxPu2ffNV#l|sW%>O2%{DHzL4AU*tA
zRwiFzPV4GCY3|7JHMKFoiY5|NKdr&aJ_&Pj1RQLV`*d%u1CZqi5bv}DC~7Qkx4H6h
z-oRc9#O27W5rIuegK?+`Y-m}EFCKOIXPIpy*GY)ZUhSQdd8kR2G&(2%4JZpV9O;_R
zsO=rE)h~An#(Z3;Q<gJU$01Zn6lNEX$eR7>WbL*c(+%W0CosH7p%5G!Zg&BuKNV0H
zt;I-0nX@|vGtML^=}xV#soB)K_g4OFp$o1!mF4FJ416<#oE5@0KuHRx)n(8y5gMj4
zTg4D(*Rt3K#SVX~D>3g8NA7U18<(AEDFh*{03$#Rerh!AbJ@8gu3C=<^hiJ%vQd-C
z%zXh%<i6roxyktbsC|JQ%~7&b){sM@6>>xPIxY*I9ths18gQ2D+p{yCPl`3Pp@2wO
zcC)gaa_Zqdcn?qGT71f52c?1I6-W5@DG3wNg(NMFL?*MC2L9g=*FB?Kt$4=TfMyP)
z7w`Da%&42Zch@%lGH={7bT#+xvC5ZT_<v&?4S2rn+np_h;a{Fzyeb-F6&o2feLg!D
zX=;B}e+*4E^36w&?AcPo<Za;S+rj9VDCm3(m?w_&o-3jBOzvKf)M0a>gn3nHt>5UH
zqpBRrrZ+Pbn&#D|oYn>466`F^oW<lSVnEw0`~0eEnax;&$t8s68U~yUk>yuHp=E$Y
zLoXPgfr-c53$RWtgGa!hxy=<1CciC*?P>6|+1GGQTBs%`d7g-RW?xz=xhWyh4!{`;
z%mw5;l*6Civ|Q2$(WBEf&+WPLlu^4WD@7b?pjy+>xeC)NrcexKcYi$@*Abl$(2q~I
zb`A6)w3H1tAOxBNR`LVNn}1qaP|tmihZ2}S8NKymtF<AB=5o8%Zg0mGI|4$lPhM|z
zPg+sU%C&bkvcmMt1dw=`|2el7;T)qOf-e(yI<B(<)f0wDC4(j*z|*3U9Z<PR$>p-8
z#QhzKZO^YewXq$e58_7bY1!H3awMLGE+eX0#YRC5lErd(qV~wsgmB>|`bZEx<<4Hz
zxJAi%tsK2fu9EMoc6UcUdtH22aUqwwMMQTwkXk^=d0N${9tNM8Ye7&*#$y|Pg9VMH
zxF7%hHg_14{LbWzzg;td1Ov?@Z_B3N%}mo7eQf!edbexcPAg&i=dzx~MTP+JYNlcc
zr}g?xsco41FdxD3Cg8tryJl7yVRFp3hdFC$`1Lb8mZ_qwV{KGtnU(c3S*K*a@|5ba
z9ArcVlKiPHH$QNgWOD}p22v_gi~k*P5LSNjCW4^ZcQXYc2~N3eA(}oY_v*ga`^Z5N
z_qZ7#2gKwqs@d?>{(MVS`ebj-qs}Y)$p9edWIU7=fKw@fWK93~sz4%(Hvnr0VGW%*
z0z#anLobM>y8QgClyVdQ<(x@d#Lvna1xMIR;y`~X)?2~Cw7Ne5*o(z}?}l2I(i^iE
zx!l=Tz4mvV1ppZ5<v3AB6=*?(bY%GNh64a$rg}4ddsq1DT3fGc6$k?q^&hEB!^OWa
zATlg?+cIF8KS}yx;apaId+#>Vb4Fh6WqBp2Bs~HuRtAR8(y^m!o@??v=GL$SuF1_Q
zqg6LRZm!eKJ+uKc_7vQ7aB58%WlJCBf=o3V9ksc(IKF(|YjwES6{c{3sohTTGs!%3
zn#JYm)}}(^4G&$oQ0ADBz3>rO4_QI=%YwX2)31Q}zqzaIHQIT<W7hOMs*V7LlAKeU
zBzSVZ<=2PU$MUvVuLxqr@_Uyz8g(sN>e+VuSaOcoJq`@)&C_p>WW_$Tdpm+C=h$?s
zu6!X@g40W8zR2N|bos=2doS(-looG;BPl^1<~l7wjIfyiSV@Bt^Eb5IDL~;*3NEL}
z01rHkTmNj>29?SWkI>Q#Hknf2&^G9qH>Ttl7U|B99qMW`S!$h}{w|#`T*gFma1R3$
zLO_p<AGd<|(pG=hCkts^m4gsFq>D|!_<PiU;xcOG6m%xJ2J9j9gz4w3dVW_W+G98i
zzk)y!NgzWL6oOQd`$ss;W2|zuN~H;xCgLT~o&DuDiS{QU$rvCZ9lA#whk&!;J|S7|
z>09t6lPApK<1qGyNjH6uGdN`^`zx!4&#8B>RC#8-&-SH8T%8F2<T>LflYlHIHbrvs
z^3GCbhjDOM@R3s<Z-dY0FVNT>(3p4|^o!v5KsefLbYh-*egdLZy>@mwSoXO@erR}W
z*51W7Bg)@zq5H4|q=QZB{1*K#7pB3&!I3n2rs3okiqPvS{dv#ov6Z0Fp45nyIfZvC
zZdnIcU>31H1dmHWCsqG+Su4nGySbzBo>F>r9p5`w_K>rGkr*^S4cCzh-*N~+uaj9Y
zKBzHqa;CBPmWMvN7_?4*{n%>EKCD*AYqZ9u>BRxohO3UVlVSNtgfpU|yaM+@X-d!)
z;zV;l)12{)84FzHi&o-dLo}lb`mybR1ba8nc4vpK`Uz<8gn*c7hX!%%%r6^3<na=)
zzu$jVtp?R`_;(Qq!Ne2#hv)-AViE<{)sUtgES2<=3!pRHB7I1#q4xWU8c=@p=wUI?
zI<JgY`uLCU8sy}Tgc`f~Z#(BQI`eRt{Jm07svytY%=LHioJA#8dRkjT8ax?tUtRv&
z?(QR;j_I9?_`i5OJd&h_a^Kgjd`xz<1{9g9IZ>8FW$!?A1HvCf(|Hej#fA#hzfp*A
z2c$eYMKx=KcexE34*lOxg=;FHc4_MAix?C9I8T;n(J}v=){!y_RZa84+cf@w)|%U-
zznk4ZKOY2MW~JmCXpnkTY;%;Iz+f^kH*Wa#=ILrrD;gi<^OF=*VKNl&(*KmrHhi~Z
z>>|pn7mXo)grjWRPY6~(`d^#}$yx^EA+=K=vcqN7df*lp)}<FUJvw}mDg4<sU01lF
zYyw`)j)!iR#?{1(k<^2<kSMa#@NV+{kmjxN3_Ve4{59l=(pBEKi&dF_T#3KB$}4%K
za5ve|e*>4~gMdRg1Rhl8@gDXUC{Ru~t80DWNABk<?fUlZP7DA5$aHX$l5(b~%k>K&
z!V!Qxv;&v-JX~z|c?I~E03Ege_3qrz_PP7|{W*O%e;+=aRpAdDMjvDMMnr7p3Z=IQ
zkE=(1l#|g;6SxgtO@IihRu;ze%Rhi!Vd{g#`94qTqW~J_U|_C7M$^0}mIuES-h&75
zq};fDdgkV7AZYaj;j!6BP5u56<O}y5o8LZ;vKnv}-@5e-yW&76y{`AwF>T4Bk$Oqo
zeShik`ip^sn;X%b1?J90D~VAC(r>Kduu<2JtMR09{Kt;zyGU9vd%tOs3f~RR&`w<V
z^;(_z{aP8oU^)2u+IHEp88Fvrz|SZZo60GEhfNV=MN{LA_m>c!6ByR##eyGUC}E}(
z545!`hQc>N+rV?%CGc|X%?8~+qLukp%OTvYlKcx++7X2h9`>OGUcY|b0?=)$M&a!$
zPmuC*fL65k_b1Z;Tb41rc>iCYjC<K(0R-|h6pD8^i(`cL+MJ>=Voh>r$7kDGEs)jU
z$(Wg|qRV?gG^^C~chzYL`Yq42b0Y3@lr_G|5rc4XLC-!oQpfzgG9JRwM}zI<{in5X
z5QLN90V;NCgZDN+7%ez(f1qZ{%6WvATIi5%z+EZm8u&iFkV4ClQ%`2Ef`63G$A<x3
zv{gBshKZu)J@XI~gdW{@OY~T452ePEW}{a;lAaEaaW6Jf=aRQfz+({!LWu9@)?iIU
zG;#A^_~9|uK@*U(T|7!m9RkM2ig|;76RT+GajypLh~7%m;mzeF6u%O7`{B}QwQFcP
zl6gd)m@tPSp178@f@FO!mN&GYw~AENmfx=KXKQ!gOU~|2*PLNn4m}7s#!GE@J0itx
zFi5T(`@YQ8d}i4|Re3h$<ll7%PyQFk()A(f6LcbC6z;`EtB8A`w$j1lC)BC)(7fvO
z<09P3eP~~lw+g{QGVo}dfQJWBbLcjQK^WZ)QRJODE#c<tqvtE3uglteOSjuyLu1N2
zPuIkCF{%h>t{9$-9k2=^;NA#hm+*Rg4J^D>!CwyHlGwrvZ<{0=&DHo<?WK8u{t6CG
z(5*&-)J+QbN|VW!P+HIP)gba}G6&9$4GaX28A3$y_Fum*#r%G$P@bj=*I`*$492u=
zvMvpS&St~h{(R@;lcyohiS<(15mSV;@qxdakb$nRZx5+)Q=V}X)oIF58B=0&nlvLL
zqWplrk#r1=2lBvhc}P4+fIx^$LIm-%gE7C_q4-Z9DzP1FU>@!|w3SHe3-?vbNxuNE
z#mdage2q-C<c=aD_D%(tO3k}Irc2zkBvnYipYQL!3>@)~YjAq1zO;47y8Ah9g1NTT
zxw@C}%1<kEfCp6oDGu=8FRXn)tSt@FTF*T}RcG`s_8P?K5?~<6*cEyxg+<_xe->*R
zcwa*xZcdPX)O+946;giB4kee}pz&9rQrsf7{zv2LZ!1VT!->Mv&_ev(uJ`O}bfl^+
zECkylABSdf_C|VKQv45A#|hU&nOc3!wLMy6zX<)VJMZoZoI*zEgM}X2FES9tl*V`>
z9gK@mr@~eQuDD-7B*{<)Btf8_2J4hkzFw`teAL)bQ23Y>GZzsWZu2ogwhI=BCd90F
zQ!ZB794IZN!rNk5Dp_G)^kI9W&#5PD3JKw$BZ)$3?s8PP^$jnacsND#-W*q!d}E_}
zmwVsaUp9}2Kf66rCiQ0Moh*e5`4Q!w`Gp>>9(K-O!cK-WanqbFP@?e%2&NshtvN#?
zy2saN`()PdQN!<%hq8E;5?j%7Fyeu?)EzXWp73D?hcLLfygnc1uvTQayX|01$E?K_
z{C~~RP#dOj$@iU9Mt42(p-h$P-E*D!@cnOrhi~p5WoU6*)R$eP!F_}V$HL3lY@+y|
z50fS&=GQ#%(RA(KHuCGUK8+}0FXcr%^@&7R$!TyN4c0HPBRyuL-B_at=bjaCsFtK3
zhIxBO6~CN*@|3H_OP{qsn#Lde>WI%6vBuT$j)dle3$e&`*|eFL6^|xKAG2Ev#5`~1
zo?D&SOZ{NFKR8@88U!t4ALv5vdkZrk16lMhug?8#$wK5Zq6WsGw_~p8!nF@8Q7bj|
zMZ@0qn*Z7K<Ng#No9f-z(cdMgK1%Aq6#naX_F+P5&`Pgb_rDwCum2TD1~`|7zO+Px
z1!_C^IW|LY%%IeivO8f|u`<=N@U%qI5wg@|zb$o^{8Me%p^&S)x|ymV^WhuK<{O*U
zjpEejKmv?Ln_7Yop{!BIuWckjy5Igi8501L{`pyWo;cn|?JlZhhUkD$4got5nhLHg
zzEXO8Ja|JitW%FSR}R;XD=^CM#QWcbN_-ML;XY_|r*6%Kk+ETV2~y73!K&fVN@JzO
zxKuaSUBx&#tBaApzc@_)zp9?Gmd85WcO?%qlTjJD=&Jnp1p=8TE!{z{|L4UrsDqXK
zYn(D2%J)?Cc9TfV>aH(_w{gJ{dofIc@S}nr+Wcb7r$W8#J0&;!6W$zMz74)quv*%|
z3CG#=AH$W!Y#IfaKhHQ|7-5WZu@#I(2eIQ1LK>jmPr*geFkp~nK&Z#S!Ep<QLIz-R
zLXlG9mW}Gj!q^Y!W~=9OR68$xBvxO44FDz!5MdqG_q?vpCr{)sYz~vvWAAb<7ArnW
zlk^?Ov%F(-IS~&x+(&hxl(dC5y7R&y$Q%qA3bx;CBJP{Yc&ENzFKp1Ws2QLK_G<oH
za;h61=}K|iZjrAWi<T|iRn&4T&$o~q_*W_M>(i+~)NaY{g8vEZhi{fk4|usR&$4iE
zSn-9H-$IPRN=;<!q_Q@_<g*C7H_uT#{3zbnk`D8diEVgqW!3N|-~V<vSl5L!$s?|o
zDF8*Pf~TNCRWL5Xkqa4Mvoh}D=N`O_v=@;}ZIp2SOQvfHvWuW%AxxHp+pWU_L~!nR
z)aA=D30G1^*$}8{SIZ99E+s+hqJ-A9(@fry%;igs67J|d|LS*1Cb>5A4YC+<>l?yU
zn<rm(twoQ9lJmW7YhkkHs?}pTurI}<e@n}shZAWAm(=3?E+B71mY87={p$s?1qlAW
z=(<zUzc);OJ#Gdy@I<|(Qa+4QJHhX<<hL_cv8rZ%>ygt<v9Do8Z*HgRJh_WQ#nS%~
zX$L9jW#g9f#Q5+LaoLOt&2*;)SkHs9ycHb6j^~?k@My+a7C{Rj15cI?>A-=C`i88*
z#^MCGX-71hJK5Q@?4U4-gt$uc{(|H+NcFIK@^op=<P~&-NZ4Jj&LV0&6W~IE<}mp{
zBfY1??BY$z{@y-|?@ir-U#!8_x6t0XrHks;8NFb4bXUns`r~72ox(8(8O?Cd0Zh-v
z<16gxy^i_;-QCrm|7YI8U=ZSx_Qn7EYj#p&v4?i^+t99W_Nbyw;SJZ+8?7};qvWZ{
zHEWb~EC%=8HhtYrpxSwSF;F-l9fta<gXGfW2MS3#VPV~D)f`3;;8cI41o#V6-T;U*
zJ@_uWf6c(ZD|{m(7dW6Pg<NdQV$YJ2zCb${9CcDiEVbwhYiVgAnsS7!Cq&sH;RcFr
zL}H43EG6EqrK$1tw>2)4KeJ|gxHQe1>p>t+GyL14FyUp|?Tt%PP}A^G#=QH>OF6r5
zuhGgw+2>kn`f`GX6$+EGR+T9KwCg0?KMI0o1@GmKb7z<UdZYRunGxItLF#fzd9eK7
z&lNB<%ZTu@D>jw{uLRRo&esar4RYq<d(o22>ccz=_l0dmWfKC!&1KD1NYqY#!w|#y
z_pqN|n!iVFgWck$+~eC2KNpYh0e;bLn2O`cUVZp8Mqo(rEc**Y<}1DsL(+VNJ{Xm3
z^s)f0fNZf=Fh(VzGqyp3ISCCs98{bVXP-mx4%`S4gcq!Pj{&eB!XroQP8$(Uh6H1X
zeb1TA9@qV?$oiIGzXC^tZb%x8eX#HTKKCxnjbWr$FOn@xB7HTd`TUdgY~?>6kY#@)
z{oxFoOX$ya<Y{Tn+u8r?5cKf@@y2oDxw$U~o#Y?pszC_RC>EzP;-Z*scUBl&CnHO}
zto+>Wi2wttRChB_<zkF_TVx9>ndQDp!=wHit~*-l73AbC7-Fz47lEWCOQ*8PdJn=H
z+W{VQ`1Ow2$|V83R|IK`AeF0%YB>mPC3%H{n1M)8+Oz=3WDKz%L0ur7YH||>wHqa$
zb-dx!clHfmtXk^;p}E`fWr6(*D1i-2F>&0Y-{fse=bc}-07(w{rg`r$kW$R^)_!Fy
zecBdn%_ph0Pj|W2@IV_t%lh8i!q0Lr+|p&g7prA@{8@OI*o$NTGcO;)G4suxFVE(b
z1QE@DW|{~7P<bXUb*`NB6>AS_S;=2BA-qs#V9jhp*qem9(tnsIbB??4(n-n~R&5iz
z5F3;)(WEVG3&yOco50Zo=4_U*1nc>LCJ#~u3REK<mr2{H8R$@pSE8d1aKR&Zzia}*
z=jJYIjx@I=Uato-@MCCHcqC{j5Tm4>(gXcq9sqWxjbbGmJ#Yz@zb7f4M&0`D|H=@#
zjU3S_sA_oqu-bU$e6yU5X}LSA^z=}BoE|6H@1KL&z|S3vp}#*mbX&T4sPrOx?V$Sq
zZsIt&@xNBh%4fO%!f?Mnl0Rl9KTHx@FTO_e<eIB@z)WQCvfbW_$WZ;^!Rbn6Z|`iY
z(qINI<U=TM`FY!XsO;r@P!8YVt!OThfBN|++@}<r9LOFCLODGd<70p7Kb|O}V`S;?
zgDdDvg3j|1P=1k+fzmF>j{!&g`dvMF2gljo6OKCIB+VGXk*fCYXAy9|^sb<)-X-nH
zsTLWqX8SPn|M*>9481K-@Le|2+g{?4bE~bDTHAMQ>^04)f5POHPsgXmZ2Xt2WrFFw
zjYW;+sQuf;Ac=udX_$B75uxM%nRotw+qW-<F#c*d%X9Jk6GH(qX|k4q5uBmwZP=3C
zVo$Y#-Mc#Z*{KOKD}x;%k8m4lCN@H&$XX_<zUPI*W;T=X9>SAZ$XHZDzBIq{HjFN$
z6Pdw_W?h5~x)xCUcf5=kJZP7M1c-YTF4KOw6GJeGgt$VS0E4jb$cbSQw2Q(%-1r*j
zsb~Tz02`u3hw*42px|}J2Vd$sSVUVu%_bAb^9)b!cdfaTV6#F1wMJvSL4P_egFU+G
zO|OHpyJoJ!IYjn(3QUc?Zdrd$n7xuP*zKgJTp1--zB-&IW<Jwbm(A>`i2tWr|9XqY
zA$RGh|NNXW)t_(qcfoOig7b_o{Rl3K-S+FSiB#Dxt1_E-%7x2y<`H9ueb3P@tR7_v
z5p`k;?LxE*kn~V1`U46u8+vKFixYWz`uqFYmc#yq4IF(D-V~k}rYBFlL8}EB1d}ml
z2|u%jSSp$zEgZxZ>~@ReB8ccaf>U1)P7l(8$)gx8SRy7^(hu&rfMa;?L;yTgt5-*n
zvB5-o^=Q}+WM(GA_0n<HdD}H+;->uB;#6p%{HG!@*Rs*`?de7ew7-Ah4c(-%UYCKI
z>A<meM@}vlZfwVylSqfEUby|=-=0ehslswILjMsbLD8u|#1O9<kDv^h8y0ZB6T8!(
z1SUJ5YsXEh0An-41<v79JRhDy8ekFUd&u^W;lFo#jqqRV4b5v@u#csnjrhIyY7lS0
z?c29+<obWA>aqUDBH+zDr{$OUsm`PH$z<@8CHt&XpiqHkIcfxplIMdBWLoo+9e*z7
zKKXqtIA(NKLR&SYdA_RiFUSgrh=VVE;Q#eYmqLC&;nK(}|I>ku0QA1{5!-zN?+A|9
zaSi7P)CmN4BQCO?xG{2Zmc5*B)aAuLXj2}3CWgg*?MEo^fI5E_2@H)T?mW0uDh$A-
zk`Ml<CCH4k>}0xX@&EAk-tk<w@B4T}WH*p(vS$>@=v5?Jq^yi?dxXr$dTD6N9?3}d
z4v|gMCOf+#d!%HA?|HqD?)Uxq{r-7BydQUu^1Q}*o#$~L$8lCb&i(=#{J)*BhY@yG
z2_yrx2$@;iaOrbR%Slp3q_ttn!!JjzmCLQqdP?8&{PTK;1FPq4{kL+9yoQK2G7Tt*
zIs;Ge=pH+na`u|kzm9X@Rg_yaxcZ;?(>d5Y&n&%wj+jRW?A}LAhx;_{-TP`jZ&}-^
ziYtMwo7u|G@lEgua`P<}ayCcqA5|R-)g|N$knuS=pB35Y*uwv*4tXTV>Wf(-W{`0|
z?NL^&*hGJ5x<17GqhGO^$Dqf4eeTD9!_XxJowF*zr)RHzbWw2?Ewa&-_mDo%Q3?iA
z|N4p{T3qF-Jj(qqVm@Sk5%AQNysCMKQ6QPSQC0HB+Lgth-tN_DsGHL1he1-NnSEJ8
z!s-1It>u^<Ajeck@F(1^fzcR%?UD+5Dx!Zt%ddpr8lu<{BO%oW<<K800fwVDa^Gcb
zDQTK?Sw&Sm^V>U<Xr+NMg*SC6#ssest+X46=Q9?ZYv)iGW67mW(cSD9Mcb-Ik@c0R
zxyx7YJ*+YF=>PMuKjUbzSAvLb^E+a10-IB9ChIlLJWFRwyjGALZf*-g--%D6uO&OF
za2Sx~?LCK|Zu!E;vI^gI#=|ne*RuHg?qRZn?}d3DiZC|adC?M#_gW#Q_J6nrfBf<R
z7*-ME3C=)W`*r^GY^efMVm*{(R>s(A_qvel13wQH3pCxQy*lCgond{o8_RI{&J%J1
zKh)Gi-{+jNpBxQc5FyzT<XXOG+>_$GCCI&L^K84pLTrXzkPI;eQi96(KU;)am9z&+
ze|;Ksdm3<DqI|J@gbVL{G^Xkh3b5Cbv<UnZeFt}UG75<9mKd&&P!C{`2+pcihsYNK
zpr0;na5yU!0Y2ZQ%s^0(5}a1Hf)?6V5g+TD(<L`iQV2d*ddB{+i2^k@NoLN2r2EI=
zN&hE+u@K4K$ACiQQuzmY7XMwb2enYuwxm2XALi`-x&Qhn_p!{%`#+1HUTba-zSBj!
zAaRQB&k3=>R$TbP;XMmdJ=xCZfA+6{b>sr~_OF0@L*mb&aI)|dTeiJ>m=naQwGJ>v
zJq2V=RtF}Yh+XA{behsXyX_p*(Hc-kA9#^Qg7*NO;)LBMHsLuT;lDrbUP$f*GzH_P
zVLiVLk7DNP2?|(38~h0iOTa1rwXZD%przn}Qm--mPDy7Ls7~(;Q4o^SG+c5wXJ|K{
zQz~|r=+8~BVCvHPXI}2`AG8}Sgu^M(aR790J*@ip_Cf1f<0HqJf1ady@q*TSU{YKC
zKCQWCV%pr6o)$w)oY@<6<mTYuKT7H|_Hboo&cFL#%?x-8p*6_27XLF93G1+5=cRQR
z_{26O)x)s3FlZ*fN_>Hs0>fH>Z9gS(l)|qTLhnexvnVVBq~=`YhmW;9Ld=~SXw7dH
z!VaoK(e@QV@L!FTjsK)#z^z&W4kaHb;X@!kFP}Za#`bz&-!r%y^9`cS3eE*C-=4o@
zD$vxtDx@!Fy=?Az;a(GXR6O_Of!Kl`6l9D8(EJ(vO0FYycOMr;4?L&mqYV`==?YTl
zjn3Bnb5@NASJduBER|b$HpWhSY7|y#pbUNg*-WSBLcKAcF@K+G;=Q=5y!Z0SKZJ-S
z#UY=K)wn)jItwQC`)sVugCp4g=Q`+O84iov@gwp*`8CiE^B7f>A=2;S3xuZnkfNU9
zm#<$rG6vsW|AU!)=6CqLbWNjr6s~lRSNa05c84R8aS1SS{23y?=D>;%2`Sc{W)a5M
zHRv)Z_c+c%4h9XF?lgpL2&)a<v|kTZ_mQ!^OlYX&y``WlpW@h_i38mPL@?Y_S^4hx
z$M~#O{qUezO0A=*&iA(`bZsuePaJTx9_v7`eLnz{hgYKiMUn-vjw+n`j7P-Nr$G<o
zO&MuFWAn(I&~XWPnrAzb+>l2B8J&pS?3H2Yh;m`!v?zeBON^$@&c88t7DE_*0grKN
zqg{GI)Z%CP&}@v`KoCqOT-A)>ai3xQS$etRpFX)+`w!wFQpM~l<D_9MQ~_Jp-$9Cg
zTog`yK~-3R^SC6(X)u^LS{`Ro($Au5^Wn)~Rmj}lpE3J|zTa5r&#YGyLRWw4*)`1s
z1!q@^(L;alH>QdTZTPXo{2d65APt(5e$GRdVlZ6@ntN0fhgIdPn(==#&55QmKM;k5
z2VbBOPz?k_5pE!$Z6B0*DeDRH3+uXODvz!^O0J>!Q~-Tc4oBY30~{!T1ukPPU}$E*
zJeda$Zaqjz^7t(f!yEVt{6C_?J`Z3)8fyjRdZaPwFx6QWqWt$E9;mU4i|*c|Y|`*=
z2veQ#UI~_3{3fp`;q$J2fTn?-qiv-#4E!Pg>QPW##-DUOQaiL*6xlkqz3cze7{vn@
zJ%orE^{UHkFAWO=QfN^;S7F`<=|k7u&O=0cSPTo{k~EdVTEnCmd*B-|hu{VmH}h}g
z-*eq+gE^4x9pw!LK)Q?w0ZKgraPjx!lFNT?A%W+?6E}WkBA-p1<<jY~rx75Ex=&)6
z;@7GQwHsi}(`AzJlZ)N>5|>%waDVWv+gEdMGv{LEdS3Seg0t^!!u4)fy*pm7I)2i0
zmSy`Q6i4pte%1zO&n+;JMS;NZDT5F&`40{BAndub1w3Gm^!#tDoT>#!RPk8B{29^3
z(hjgb)daiB232kV`fNH<J*Y(q;Ip}xGbf=qzIk{R3_6rS(35BPP(J`H?`yydbTYgS
zV$l*W6+Jy4DjXp8t$Gmu_3h3iz<9X~Py{F7#I6iry?h#{D*|;$z_E<XLyR?7YHyhb
zOWt30y+mi<v(IDe#J5OxjqQW0I7GGG%VFp>&nn0-UoN4wuMm3>y~PVB=Ec60RI3kM
zY}_lhP}OLb_+jm$U!DK^Kh-}h9p*YQ@=Lfr)~39@%~j%{eON*rUaFd6_+AO|L;?ao
z1KewV`Z;u*y;r9b)@RyFQ)@@TvpNWf@8$l%&}CiUcL)ss0Z_oF-2S_E;oT$zH#s51
zurq_VKi~;+(%8FqFEVI3_k{Y=4AWd`Vv^DEOrF8=&WE&I^vNI@yQ{OnM2}@SAqkH*
zmWt-Qj}!$6eZSv$<n#5V<kz_mHg%2S)K17Rdh_3TeR=3z?f2l^=A}wi$8+C;b!ERo
zS8Gg|Hf44D>4xWt#rQ0`)ddOEJ5E~o92gdVOzj0B{Lfr)sh1u}NXu8DPn}`w{3Hgj
z2ap02p*${;l!76t2qW4o6V_>JB81|mV44Sj3KToFr|({zWeU(t?yVFET_VeL>GB;s
zBHcf5X2VC&E2~(pr{Zf9bj{wws-sVQNN2Bv0wF)UgUE8QvyW5Ak(hE@>*)Cpa&-k2
z3yCT(4*g|TnVVhJ!<b@Lv#PIZHMCniB8X{G`{>riz2R`pd)o_O=X(7_+Q(-D-hqDx
zUK5)kWy}0nt2jX>5=Fdxf3uT?iWovS!I}P|laI)T&x1hmDw{hDda-}B%+&=R<!8&w
ze*uM~6~_)3dV+y871?q9D{&FNJOl=*f#!`*KH+oxX5@JWe7`*VK^YzpMt^o$)q`x;
zGmMQHzw==3gtQpGgw7AoXBW}kH6ldCL-nkEP=@xJo^AN;i+mAAoD0Q@^KxfRqaw(*
zc93cXUr23$fA6X$*{r3Veq!q<Akbg`Z27-FF&Gv}91}LC!h!71bCr&stB#-~BVrwO
zc&;WN?G-pvWs9CGfted#@_>OJ{pVJyc*}%=%<}L7gxha+!8V9fhj_s4B^bc8D}S1u
zEXH{l@a8ZBu3>;4_{>_QO-7flY;cb6X)4!W@o<8AVPxzSbP9)zwyIp84^g;oT|G&|
z+UNL&m380o>Cz&Q3w{4=o!q5-d+QsG1+d3^)$3a*3!k*664?o|OV8x|^S?|wVpCjR
zn%_nn0Jua8m4cE|E0nJTW4dM6zqgEqHW@wz-Y=xvY@pNYsm%1&p5$2vb&ew>6Y%;z
zJjko1Ald|{H<m3gOdJ^c6wEUG6OZI4RC7F55(nly0IN_0vv@?e<oslP>oF0btp4uf
z0TF@*!H|Mj+NbhfOb<N>I)QT+P4d%<He$izz()Pvehwr-^v)4(870Ir{FX^P2r4N<
zg1i-<pBM87`@ST_9D$ziHFz+Eo5WH$=&5U1VWiNn*+C)T0JZhM)we4?3q8o}&E&X}
z5U>|sFxSHZNpI^sR}o=>8KRqKdaoV)QV$=qWP!7}_?dfF*U)>}4Cnhyw}+C6ANIz{
zmdGP13@mPfioR470r`O?P&VE_YDHFO>%n}@#b|)4si%akS6Qne1v(2XDpxfzGcU`o
z!}B>S99tz1cEe&;X6QM08&SZiMEqysL57JBSQts!XuvI=_%35m05`Y8EjDAkMD5ca
z8t2vJG%&MEH`o+rIedSfCpZ&yUiS5=*}mx#fm$_JH&@q1@wmrDlCxzCs}xRAQ85<i
z!<<OUMBP)SWE|g(#ylWvX8U1cqlE4IC}aO%&(1xDpRf%ikqhsyn17H3y9HtuJnC1*
zKo3Ih1p;?(njM@V(g92JL}Y;kNR_=JMM*P^YpS_V&1VYvR-V(S8caFK{S{w!*wWNA
z@s~*fDj-1?ijy-P_omMM4AQ@FVDP|Nj?ijgMHpxoguS7`nFI~21SYb&C&x2PXd)&j
zJ}<-jWQBKFYtKSUGjL6l3H8rU@-Dpwvc2g*nY*4bNmz0?P{~k^3qyc)=`R%8ep)ZB
z{^k=B67tep9f;+f*#>;p{duXzR04Vhf#5c+0M<tWwd3A<G>r_%-^Ts8KGP?G)Qt@>
zlG<Ose9`ndwzN1I*Iv<9=I-PU!xR!B%w(6$gcLw_R0o!|rZ9OT&u7ECbWKnC7O3Qo
zPj`O==Wren7d`PJPF!Ql$1)%}CMe^BK<I3_5uh~`4vvFLpjc{kzOozAM>I<Mne@g(
zCOnR+YgJ{d*po&VMzseFkt;tEN6uJx435Ba_t78#!V2(-kZPi0b!Jr(YrpfzSdsi3
z`zTvFaJ4qx&j*ZeCf09_<eSz8hd0uBtHny`LUCu+*p;Yw0J$`{^bZ<A3O`F1??oVW
zKzQ;vgvIM(IiaXbhZzQ6!t`6s8eEGVIBHwGEgn!rq}gkS{B!%M50$y)Im>N0^fy~B
z&rAZ@l5Gy_)gq52yYZfchpbmW{+V-&D`f+9@$v^wG)!IR2IvjTm21_;iD)*%(3iAw
zt$;|~BRkC7f)N%#%z7qYx5aZ;z?%#)-O&gi{c0L0L4+X1DIyIB3NnJ$0xCm#C<jF9
zCMT*L!+<dO*a{EX>oIDaKG~|THRq2g;k??}wvEa7x@9mpEzhq1q|K8IAy6ElN=X5j
z_?NF1Dw8Hw^ted5yo>ws3Mi^qnajaW-lS-%^P)olkP>UadGnu$ndvaN_H-hqcU9tK
z4D1FgVy-Me_ot_~pp`_AN@4YOMn$?WCuGp7ZJ$AP7X~$DFX0)DC_b&!c)}5dd=`4(
zy(SW#I7c=<ZZSt}iO)dPP|Ci<3yLzX8(;2mIa8K_iGPCM><dVTz7aJ1iZ}iyf12k{
z)H#9Y)^tzH_&Rla^?V%3{P|=Rx|I0aAj+Sw+D&PBx>JDt{OOL1LoWIJ>OXZS+p>Qj
zTfyU^jT2KCA-)yqtD6JhAnRLKxjPRa2bntIth=JtE5PGZS@4c3wAIZbLa}@%WUR(P
zw&(fwRCeLVRlOmub+OO2JqMoE%<+qyDoj3D!#Mq;&=O9@UqN9`t}0WULYnS?***-5
z13~+n&p;H*0hag4H)<LFzdH0u0DF$UGJrt{{vNA=vSK9la&toy*<dKuI{W9(*|RQ5
ziHS9L_puic4fMNr_rKT;y?GHxb~Gp}{cqovuq(_i^~m*v7_G-bOilK>Kgqb-FAv1q
z-~{sY*Y08<!T=Lu#^B9`wP@iheyK&mNks&jMgm{wMy0ZWzfB`#$PZ!sh4Y83W@M5C
zsnN(K!}2I#)FiE?I>IklazEX~f8}&?21np8Qf)6)p};qgZ1D0fHA4U8>78}3Oi~6Z
zP7L@Oz~_g;J)I&VX2D%jUr9=V@#*rDl)W%t?9K<D4;~j=Du5<f1G8vHUfpv6!rTQ+
zOp!FmV(q?h(r@Pjz8Iv40(NH!D;RpPL-093-5>>`u>?PP5G1Mq$I3z*T7~7NFl@!u
zYW?ze7Z(4x4^m^zpFcwRCuH1dgHX~HgU5I%mu|v;p#7mflQ}@be(qY`1s4*<xWW<7
zK9Y{DMwV|Bf!I5PP`tS5<B7^yOcnYih1@0z`p0$>3Uo=Yvl(~N`^P@Kmk0R834G{I
zhJd}!tNza*SW#$y+$Va{wb<ZUoCp(4e?tyWkD0D71D7BPgjQ<okOb^scpncgA(B_y
zqH#?d=yJ94e}n6}kj%cY`$YKz57H5UTzHTqeW9u%mIJAN86`QP(WjQL!?JLZ{Pp0E
zR6m;&>%K=-FZ+`NjW7hbE|Q5J73qF*$PU}=Y_a3W{spATdV6th{|Mc_eV?_&HaCqg
z@C~e55)k*1K0M{9+hnII(AmC(f)5E~-ur6CW-}$-$h`F*B$;@!`TAagTS)f2vp-M>
zUjob}<UJN=vPy;8Qq&>MC>awx2j_e#$G0T;dG_>QjnP*^@l8gkP+TM(Ygh&ih7u@b
zT*BuKT#Z1jw@|w~4X;0c$!=HsL-2NrTsRCF_8HF)rc1D*Pw*%9*iSW-=kd%0f8I4)
zSbpwsS8@HEtJOB5{+oU7S7J=wXzC67`5%e?L7#O^y=_%_0wsGwthv$M-RaLB!fn+Z
zvRHW_TIAqQWW`L7pYUMb)0c1YSS$PTryQ@v#A^$I#Uu(Z;NGQJBgPA`hr^)Mt3!7F
zz&JwfX=FBwJc9%`;FT`$C3n4-#s`F&a76Opw3$S%dQwY6H<7D@2IWMCfBD5z?YqA~
zh`z@AeErlvFe*axxjcEwRo?23n<D{qKzkINsQwP5e@hb`cwR0w9W~OMge2e~_4zY|
zHr*f(|CO8sPX8%W$ng+Bq?1Yd`|;Cq(wohrKfqW<WZ@+@Ejuwd1-|VCYxAufqc_T+
z_Y(>g{rfYsVd^h8X!$NRj9>qV%<PA5QYN0cO`<0WcAI%H6$(ubaxuF1@S%$62?dsE
z+U%7H-oQ7|a2|mv)q$|HlAjK|t-f~)bZZW*Fm9RjU@#T&QM`cN-(no>i5zPnzjz9n
zB*?T<ZnW^8c(&3T4Gkv|t~v0^r~?K=bn)xZ9mb3n3k-CCJR+n2`T0d2#IFJm--K3a
z1LRjY!DI?pjC59=&4?Exv_p<kdSvBj-9k6jy!^d93(;sX8yamzycb)U&8awgw0GpU
znU{Un_k&y~MO2c@3{;h9nr5snhHjJiLr87!mvZlx>9tW1(m&44W)LTE^}iRG7k7aJ
z9?Zjm_r@aQ1N(<vy=`#8kZERWQKZ#8a$<v4;b?mtcvROw;7kIlf?BU#-z!!`??LKX
za4UE@4GhmNe3@?Z3NUq`+4TiFAceIFOt0gPSp|a%vqP{yF9DxDmFV@x+N_U$aAw>@
z^%}gmwIz0|@8B#H_IA7(wl15%^Q=$s+w~J@o0E?@8aq-v;w>TQwN`lDC~X9{FEI2H
zfKuFb9LBu_gQJ<UbHU_LI<{RqztIF@;!AhuK5>{5kka@LX8uflW8>#nS^aDZF<`1y
zT35HwHmdM(Pv`gH5OwPf^VSu8`8CN%P!0mAT>-4Ju0GJKBx?2koSqYaWq{C-LTDp#
zeJ<YRVYm<sHc^<nw+o+cZ=%cPm!WO+x9T+m)L}aqaHoL}PQ&$u4(*q_u6mp+gS=sS
zR;_#lkrFqkQ*~Y>AV^K*<iM9dq&$|u7gb=SUclnSn|gCa9XilHHfdE89^oJ)wZgtz
z2>otm=t4eI>(;fLx0f~Wrn-9?1D33L`L|0J%zB!73O5@=OgbeoAMrD`gzxzNCohOO
zNX$UwQ_YZf)}OvtIBN@&^?inKuinFRBf#*xdvwq~UDI{nCz>oplgtlS%XS~bWm<p!
zXY63j*oTDml543>hd>O1-E|O+$tr;1A8yS=WQSOoy_cTNIGXcp)0Pd~+IzjUeEpDD
zq+tIwGw9s4mb%$%P}*c*09F~!t)B-@cs*on)2-l8II=oh83D%ly-AYUeP!+!46lDC
z)cEn`)o-BRBaOTy=YtHQdVwfVPbN$MjuX%qQx5L8K<0w*^toDvQ!jKrhw7<bmQvbO
zF>m!LOngWs3V@tntQ*?(5)gji9EdhPCkdYeJW?5SGzKN>@Czj1yPgn4s&?n8?vF<$
z+9pvb0X0N#7|a%Fe(BXIs4=R9yZzSmU_ES#<ElZVQX?U_=D@}ZZS&TBLkve58N^tr
z*ia@W`Eh_^c!(DhFZtXDjlUq&HHEUaVJ|rDoNKSN_)BCxNSwNB9QYaS_>$dw+|G{&
zJSeeWgyg`ctBbK{*$K9Cy*=eBK^gnz^E}t|TN4=xQRU)q{9Xw4ESb9ddCezN{`C09
zKkjdOlTL4CQo{1a7y8c9&3!VIJ#5nvf7jWou;4!l^bk2ppntyaPK9!`K>$hGt)MM8
zqXy%9l#mxx|474mD93gqK~PF|D)Zfi*}*{`1NAeY2nvO1Qtfa03Lkh>$rH*rSZLs}
zX`R0Z&ON15yxea<&~6cteCe{ZPIfDp$*`PS`xzp4E`@OlnDeK?01Qo^a(%#?Z{5L*
z2H;G!Ue$(?T7uv;90V!vCr+DcW-t=@C&FBJZ)n96D<+iNT@8Tr+WLy~vTlkRqX5;t
z>G#OSKK}Gg+YqUhZ2ZWF^x{p~0cZjidP_@>mGfcqj8d-^Pe2Ox28iW}U;~!mmwiGI
z_UzylMu@^N%KIogYau(;Z`_`J7yy_E-kxdImmcRbnsnj);>>39ZPlk;-bZ{E&!r#X
z;4rp@`u-b>hV6$FrmF>qgp9y6JH1?I7ojW|hNoHLY?Tr}?mZbV`-X?dE%Ez}Ne0dG
zq9+^=YSf=I|1`Ya4nbROD`=;ru_4w4u{5?if3bG<hTtKYIm4`arM8vZ<n|GJwi<j@
zi}>Z7)1l?(a+b=T)!gM5@xh=+e406Rwd}xu?x^tOBiWcY02vVz#8)BK6gX3s+{k9q
zf|;rxTp~nev7kt)r?)S=(nY-x=m{NX7}zr?XfRs`u^T4><wZ<zP`j60@X*@}YZs2D
zOu9y2oVy2&%hF>nzf4rmqj^g(^o7HIpiFJ~>H(;qlL#N>s$LjG7Q<*xrAc0!2XI17
z0eNx-#=Uw3{E<&dx|i^ia?GR7EqpzD{4ua>^Qn@TCH{F!y!_$R`jw(`*nZ?lS$a(U
z5|qOnU{zd@afb1f_u7Isk4|p9K}l>F=95r+<mH8n-~zQ_2)dNBfDyfDYuWo{NbU^e
zWv|dXOs`;inx%&D1VyBqUpZaPXLxFzR_J?Uf-;8ggNWUUi_aahoZD4UX3G)s!R3TK
z*lhT+NMncFF%;^{!hy0n)-3)P2^<ydNxJX&e%RWv$0^}R9D=Q)edRb^=*(DZuwHh(
zt))7ZNQi48ozv{W+e1BqkG6xWJ>gcCyl|C$BcuG{O}5sBPpx+f&a!WxAY2mWTSU;{
z@wG+&>vJY(a^+xvq1EToWq?HAMUr&Dp5@&IWkE!#K^m4$67UUFeAv8zo^u{t+R}oC
zOqH!VT>B>sSLkGNE>dnVHUn=X)M4x%(X_u;=Dk*O>-OzWNhu4)#U{RlrDrzuHGbrL
z`vDRX*JyQ^id5d;A^7o~5=@~ptqbLG)kmlk^ebN~=VuZq8STEx6gCTJKN<@3^1MwP
zI50&33+PjldmgfZP1DU!zCevr?Se*gqBCo<pfTna+A5va2^%1=ngO<|_J-r`%>#Qu
z-Z+#D6_4GSAaX1V&Cm(sf!c*jEdFTf56G=KIOPZJIcEEc*`Yo|9n&b<d$Fv>|6rxu
z=WBpN$^qjqIcGR<NZ0133E#vxI>KI6OnP=w1I+(0{QkaN0G3dyoO$6bE$l65M1L_?
zve~N-a1>qGm^<dkqsNk=;+68mwE6HA)BJ+%cUR`%*m0|rDQS;){sO;?tf>!L<DNse
zg67_YKW7AYTB@SqKwKMZJgNz{V3FFG^3&_LX$>7pZ=#V<VDD5CGiXZ()8@F*sN|`V
z<Hl9HmO*4a-71nl52mU0AcdQ{pI$|AfJ?1Fx!&?DVJ9~n@C$|D^`r>6<5X*#M{Xa)
z|8zDmSL-t_eFaZX6gk6*mcrCR`&6&D9s9Yx$ltpEhJqmx&@9&}O6i}0nY#J!1sg-c
zAP%R2U=+yzZDz;Qy%yx#K+kYSWTp3gp&x9DR-C~?hbw>K;NCCB8P#qCR4sggAFpXi
zzpaPzty%z;p+6?GFceV|+FHEX_#08$4`2smu-)A{C`E4QI_d#2Z1V&Yq{8CorLJm2
z0np<E2!fvQ5~R@gAZJM+rKF^+`M4|<C~H0S>tdYLp+GU`!&uC|O<0Jaj!Etmig3FL
zD)evYae6s*`-V?NUr*HYrdv`vyDGPBAHc#w2t!u5Ux+RqZe1K@;$bs*5+Rl-aQLrl
z3tLjggHqK226h)#U*BTFo+5i86dmrw*tW!De-Cb~#ofA{w4TR}i9pGK^j#VEdDHQM
z^3o6NFcmTlnM}DZAVM-Sd+A9sD@b5!YJmK`uf(NZFe6VJ%WX@mO&MMUj`w{qn*to+
zKr@=3rAaz<9tK{O9@FLnh_;QXuc#kzrBif(y<ZJpxDo&JIMwJiAaSFrCgbJGJS<-w
zxb&Mxd>gMrPKjWg;0US2u?z%2JCC=XMV1f~sws2=xaaHDgpJ-;$Pb<#<UY@m)tHE?
z?{N}`fJi|~`hC#CN0o-TJ-D2Lp=#=nqvgu*ccC<@OCL?33*?Q@ewpe|*0K9G|4zJ#
z-HQ+tL@$pzYH|na6OqO`$8QZhkfhVPJpP`R5{vvPlB2q8>Fu~uYTCSf7IlkWzsakq
zQ)imfUDE>UPw<sFcOODk&cNL_xX?N9jj76Yzm{ii;b-t<A78(IXgCce?a>6Dp3R`T
z3y+zk?D(~mDM?q{y_$eGERUrye7Ko<<i7w=;jU9ni-p_^u#JXT;4-7h!XRj`8QI;#
z>+lAKpm0G@2m>^>yd#>|2kuG%hOjV3^E9Q_+-~&<c(&`{(VI?3&qBe~-cG^Bm@sG+
z*w_o^HG*glS9jS&56@RS;!AYzv*PpcQzQ5y1^w#_T7mm}-_e5K+Zu?~LbZ*&F19~D
z)~4?`T*GQ#z7S5vBz2<y>z#}tVr&6ridhTz(jx2Y-6!W%^-S0yqwo}hr)Tz&6&_wb
zsRM)XESWoy0ZCd?@E67nVD20%SV7S&yw`{HuSKXOOJ5v$lC1lMZm$tgjt`v2(rdMh
z!fV-^RGO~|>c4TYPDUf{_EVe940>|)s|8uCEDHw8%3Y_?J$fHQF*7ATe3~VH^KtTT
zXyP70j}r^^E03>*&h(|SY8-vS&bNZnd;$Yz-Qz-qV&Am5T5aR3Wn8(}2G8D$Dek&K
zoqVCq>G8oJ!A9kjkejMIrv>e<J0W_V%Kky@c~1@pXr@fFP9w1%B*|w954-|PTr}Fu
z%|oG2+iT9H9GqkWZxO7rd;=;}%K-kcRWGo8Lc!>y_u^OHB|n2Z$1b(g-v=}K=|VwN
zJQi4)M9_>;R!4uwu3?^@W8W>2$~XtfGAb69VepBl=iN-Ww0bhsEN?pYkF7c&qpTs}
ziKX#Zh=+lMjvq`rQDc=1Akl`vyJW-!<-0!M{ZM-$4pTf%7CBkGg~ri5sBw!!!ldB+
zVWDDT$|fA-hw(9QXfAD~Sr9~i5??NLRb}#Cx}veTUs@OWyrD}14(aV7)}DIcDJnCP
z<VZla6~YdQtl<zZg77M4t^;SwRVBxNG0I9Bo-is<nKUqPG<_gkU~qVw;WkU0TFLy2
zvGhw!U27VL2-^exyICB|PEnE~HYa6`5B!bBz_1Wqdxkqp1;U<?9cp4BI@Ao5|EfMW
z3QWL>D-7x=h~4vuA8?HVdq$YITPs$ZH4AxN1XS6*BgmW+foQH)*mbHhy4@0R!Vfdn
zROcOFl5ms*X9ywKj0i}!bc17Lz;lBOB2-9Z<e`+S)wz@z#^ewR+*ZWtX=pf4tuH-C
z&U99NM-Wytz87LoEJZIBF*z>wSg7xoI4F2oCxH2sAYiN|d!j-^k03KSzw~il1#>v*
z1c*fpftc{{&Y5B&K~_8^ZnHC21bk>`Va5<{%0)uMBK@3x(S=0C%E679%!tY~T9%Cy
z6L>;hN-h9D@NL#WxiK6SI1o~Y(EGBF{Ch8#ni9MA2%?!rE?J`-Q!S^0ZDo(){_fma
zKeDIA&B`y6j2gF(cig?KonHkV?|Zo8{npeOakgp-d_D4;!6?F5+Zzd(zy=;ASOvD9
zG%MDA>6a!<;LU;;WjDMK!GVD+aMT^@Tp>@Z1)YI`hp-euC758A9!+|QOheHV(d!AG
zFfflI(%?x(7`)Vqpr4)=gB4bne|YE$R7v>PnATtnq@UZ@_v9M()UZQ|Q@x`rCAlkC
zMnU8&a5gipI*$^%LHbBhHu;L*sJ&8m!CYjag%Q<)L)nl^%_s#?Vuz?vE>*m647vb|
zTBGqEv<r)c9#k<n$SJ<59uU^`2I}O4e-4`O#UUt!BY>Er*VRGmLeZP5!JejHlzoJY
zOBtH2!;q07Oz%^fd;aC(;B{p@8xRFn{+@qOZ`|TaPFLp0$8$VT)`xkIqk>n^=f~`O
zhmGe7S})(-y9HC)o-=)g7MKb9n06+jqqk|2=Sx^w%#*O-QF@w%znZ}Bi8r5z@b%l`
z_CJq^EU<O@b?XJ6idbfQ;bG$0{)eK5&fqgv^w2+h*31nZf_B6$9RDfo4xlM=b6<!&
zLr8KSZ?|m(3@cTn;ny*a4*l%5@VpR?Ved#i_}PFcxmr*)i=bY1PHneSkg*n!C7zxG
z323g0HssxCVj41>o_JJUs0@kgOL5BH<h0r90?N`Ayo|#pF(+M2p-G<X`8;xh8P-2i
z9fs8ZXxuA(rpgkog#3fj!B{SEcX;z=1f281Q?|1d*$JSr<r4z(-UL(%)Ruj(UcjKU
z{iNGP$1Xe}Vy!H_i(s>THA|6@b7Qe`LjZ;W#~p9{1%N>O%oTpU0tGaQ19ION{kCfZ
z&<I=x>z>R>^CckLi#FhUVX8#&@(mIF+~S9l6-s$i39|=P!e_p^)!fVPB4@O?u0uoD
z`#fDsFskjZH$a3d_sSq4-HEmZ{Vox0lb$5KQ<`!o#f4&A_Z04+!{{eG7PRhom~*r!
zUiRaoVeioA><;6D)CEdQ<^-nn@21FHF%`29fPu*tDEnR(QGQBJS?s@<yO+8fP5)4P
z=QqxJJ*8UW$RX&KQ;mM)$1?DvcB+Wg$AsHv9V=-ARY~6MR;2`OGM;NIF4K=pCIn0Z
zl#A^8QC$*Hi<8m_LY#N5J;R32itB8)O)IkaLpAzfMPGLoF*J~bYTzHB=V3SYg4v}W
z8s{jrSuLRFwu}WxYpxnyuLbInc?YU#GLu)Mo<7YabQ<d|%0Yr5o>}M>8Vu+=)LEW?
z-I!=<)#Fp%toPtPaR{?~=i+7_#I^~&{!&9biFGCncv`_L%{3Jo4WrVbw0eKWJ;78j
zZX%fPMq*J$@;2#gF0V@efNM>~*L#N99mORm84`wMs2(Q51|8fv6KD^8QC%qPOFWJy
z4i6RjuI$r+n)>Oy4`7LihnC_jy%!gvty<rv)xcODq_0;S&)zR+Se9BI1K3!GYjm5(
z0S#ek0()^|T0VI~N8-2c$Mji<j(Y*#^3g9+Cs}i?+zH-v=qzV)D6WzfdMzQ)2YKBC
zQrmx`<G#mN9Z)oGHLDoV(gT8mZ8?C#jl|-Sz&8l=h0s<)G|Jl@H>1Q2V<<>cf8;}f
zdV6e)PHa2i40+*wxN&^539~@Retz*aya9snINUZMy2xTl%^(fbSiycHK5;t^<$Rwd
z$a9=cBl(%1H8&a26g<GhZxwa;E{_uiQDDL1XNaxyJdgN&#rVO-)pWb0GLhepl<4GT
zMaoQ$Z5!HN!ji8F)>K8UW=>Z8Ze0mtYb|!SG7>x8|L<3T71eW=HA7a2$S<rj1ZI4i
z&=xIr|B-A^4-0u<!6g4-KrK`_`+@v6ab*M0&eLGeWuG;QctjIw3JST&X+jf?X)tmq
z42sq6q*6T%7c|JxH5$-OWC`O4TDI<_j4<OH!~<5oVtWHlQD{Y@)DYPa*N&`AzPk1`
z5r!?e(yuN|a)FaWuJQ#w?1{%wdxssyTRAkas)$hwO|EpUv;;!7+Oa|313HKITKh;l
zj|KL3LZihRkIDAH&1=k;uvbu8C8;ufLq+8wQdH)<_9)=`R_uWvUn~K7Emph?dhBX$
zRPN<J`2=#Y_$U(F@Q-~b11=LTqu~$W{_*MB0Whc<x>E%<#RAB;gI=O`r^h!GfPu_m
z1jelvn8Q?%JettzCqv&NhKEhjWk%jt*>3@0NP?N|&~k)X9ycq8t!VLP_nJC0zBlTE
zak4|@5}=gCMXUGx65M>^vBy0hGRE@o2uX#1fBb@M_7FL0N21N<oa?;Bm^XVpq4{_M
z+iTO)wyL!ee08H&KW>+zo8fB61_x^{yh_8Fq(oHd+m~}7BS34zwOlasTn>h-4-KZ%
zgeGGx&u%iDaKQ9dfY}S`6iv-T_wW*kOHANhPQR4JJ29OBx8zz+@_tQC&EbBxL27D;
zs%zt5?yV9f^6sdvKIMD=Ap?qI$3Lb*slJAZK39%qfSG>B)IVTIfnyg2h6^WEtAS>a
z1aPIEaInwsugs_jJUQ}0Rj1OsTxzar_dly&xn9m@U?>8Vb7$Vb{NP~lygEstjfM@B
zlZ!!fOs!V|g)M&kr|f!1mgHERanRO`C=65rE4O(7YsZ<oq2-7MC70A|z#o-^E6D}z
zqy02Adx*C0pRaEOf&62pt8R8};KQ;Eg|5iOQj*Il*-;t3ZQ*91GMRr*Z?AJ{ZN+`f
zJntBD?tq0?&0%WJL7&Tm77G`^Xi4D?-tnumW{1attjqq=3i;L$8{<*yW{%#9as1hE
zT=v|vY%sa;$k7rA>HYhaq3O<DgjDRM(AS5CHQznou(hV?S_!qwQUCiAZV)bAtJoOz
zWrFsF<zD3N*4}$Q$Eb4V*c?=~M(sBzkVYB;b3Mq0O#${tXd#+GJw2bnoSep-`bkjS
zejpSYZuI~aT@86T3&f6KF!xjh`u)o)q_9m`$pflLmN)fS^)(I38yWu7q+pNuN>QJ(
z6Sk7P6Lh#(vBxl(DAg+)AWO9d>%sm?pUMj>Q0%<)5e+AYsD>cyI`BC(Tf{shQAk*i
zDO>&)?Fa-PW!LE+pDg0S&>%jPWK7C<PId}d0RjP6u-d8t4e5gp_=%(tbf|R*PMj-%
zwh?BG1l5au7Mu|9oC|2$`Qe<F5M2Vy;1l=-&Nv=~-QPz|d^RR7Z|6<Eu08~HEiOiZ
z9r2|!k|!@cmmQJgi)e?!78#&3k2kO%9N(-B7Dr)@>e7X{!hLfAn*@$($sfuBu>2SW
zVup<$Pn?-6W$pTujyjhuUl=S~vw~-f>s;cq>DEJA$0t#!#pyN0J;~TeZ%!fGNAM65
zzBtZdG%-l<7r21_K4D@+>Jsi2*@G^&02Y#j8bHX~dNPwTnB&rj+S>Hcgf+MnZaq@q
zDWE~Qm{??5AFc}C%OCvv8S9a8zMk>k56<--xiSRybaB%uK~R>LVvphW#h4V($D)yr
zij1N#rV|cTJO+er=k4D)UWT4@2B9Dt7>KGka1&sMWpHZX{QUWI0)Cj(`qEdmUO#u{
zOLzwDH{GC1sP>f!sQl1Ntmna5k7A&@squ!LL2MdIQ_jFQ5ZRDHXP0y;B<#v-D`hbk
z#Tf`8hvUBdDwjdpLraT=`mc7*EQ#k_HIWF+F|<i?pC1kavd8_B5AkkM5>Ckmb6Mw@
zy|gan7-kh6jvEH^XLTJAHONBo@#qnlr_Jm+hUoJI6;%^iqr+d%Hl6zReWAuWv{`_;
zQkP?aTu9fm|KP$l{em8!{+D4OHvd;~P}>jwW^sNtNUML9v00-yv6JYi;JW&ppP$jA
z2&zsXXjG9&cx4HaHPMTe9==}Ck<(06*q1WTyL6MJN*E|I#a^q0ZLV(nLGFn7u*f4X
zC0aU?+0TvP6^LE&t(X?}ID)y?Z8VcQ;BC)MLGYk+RnoTa)Liv6kAQfm1f5a?8p>Z@
zI=bRkt#W{@n24{})F^E8z+J~P(IHlXy!jW5`bJ5F7mFyC0a;%nq5sV3^7%SG<iFKF
z!VQ0rZ9K;R%XUo%Z&8#l$y6UWV1vx)!OH~DG}b`7DTUCdkhA2wS>LN`2sN;2f_8tM
z_@Ji=9Q+6yal88b7xbQ;w{1Uup6@@m(oF5u$;vCwLcF#3r(fkCJp#*C)rEgx!(E}m
z$nAYs<B5)q`H*)bVedJfZqjtnvVJA>Q0r#B_ThlnfP<D?BvkM`%Z_K4%LESpd*L7M
zM{Str@6{#I!dDWs-|PSzp>Vw2d&Wp!#WR%|zH+#7$DXDKN~^M;t;&e#XCPY!Kniku
zuv_aE=w|jSx-AMk{G@42Bzje}sDi;-Lhp{{S|;)W>?uH<@g!jMh+O+3*PBRInPE(b
zV5(oU{8<gpaw#ccl7Y<cIaw}0AO+Sf4*JY)@>Z%O*jC}3p@1V3FCHk1`ls-Smk>F_
z&>h5Vh9#7B(<MQJvmpV*HPK-YWn=ulZyi3Mz|y|*SV#w+hf<z-V>+P|n*RCn5%`)~
zb-nIbyOo<OY~$f!1Z@C5dU3k~bB)D=FjK)8ZAC^JmO*a`nk&Dc%Tf<a!DgQ1X)3Bs
zMEF)!(+4V4T-cNLDJcaNan6Lwreu4RPB>436~b|AyHQ?8pJJ);Zzk+pF=^?N@tXls
z5r<BhL9zMDo;Z6Qxs%8F{>x)H#DQuT`A2cCXcK)%GmR<@;hATj5ar-_XXXY2Hb;Wa
zF$Doj<-W&KPyYyzghWhhs04)Hz*F2Ra0x~Opx)?DsM-?}`bbUFnTB!<N(&J<tXCAv
zTr9mj95Vl-`ILS=R3X~!k7wFtDd_}&vGo%9HEUtW>P5ETX0whN(ZCh!uJ7&C`OPNK
zeACL*(P*dS7O02O4=gZfqS&lsJRgShp<x;}P~3hp0|qL}^=+PfjX<(&<fwJ(_uS3P
z#g6Z;<+~UVI-op;=17v<y`Q^(06!t<Qw1j%m(9i{hC#MYAV#1O3>}Yz&_O&4m+k>U
z9PzDe1dlb9gC+S^Yi^t=kmTeN$|P_^C93I?Iv~)+@0{v6tnW#6jZwadil~wl=Wh>x
zC}WvD##oW}@-gdn1vWOe)~VYiGTHbEy@Fsk8v)XoZqDWD)s9Kn@;!EQx~t&HN+&?g
zZcM$o+d3(C(eJbi--AFcwo<VF{Y;>Zc*w%RA1CZ%8G=O0)m;Yare7S{@&XTZ$1*`F
zyMEwaR8Cc6u*SYCc@D;^wb5<m%r8oH1wQS>JoHl~Fs0r<8@w`xazo?XDo`|<K?czC
zxCzlV$_2(C_X&r-s3LR&IXF4rKWBp1NFBWx=`lde)cCLgJq)Xjen6YyfsIn>@R{y8
zE&(%<QvKg`^m5LYObQw-q_XUSz8neDv-jytuie~`7@nJtx`7oQ?_eRL3g-tv{cSsZ
zS9esSJWact5bWR-5zY<?rJ@aS=TE4gt<2i>^V-)()pk){)S0&FEwYb5bo3uSknCTb
zxyU$C2~-*@`>%Ooq16$X9wDjm43CVw74_v}$#ZlqFF^F>r{(MPmL*2#&TU@`fFDlg
zk~O8n)3UOvxArJ|@v>%&Q5fj+;pL;edG4GOye_^T@pA~6$R7QRgLd5r(=4HZ#U~st
z9*u3oXUrI)C_LS;v5@1V2wiu%{@z^viGkN(n%PFaZd_~gHYI1b#9LGH2>Vl+7H`j6
z?#2dg?dh#4!P+WxgPRGx`>z?^X&#()KYP;VRgSG7UB`&wUdsA}{3D;X@Ybpe*o7N{
zHSwQTrLx{<&BG(c3y+v1pHS4lHWQ0MvGff_BrykfT(<uQlS}LnaEM>D_5@W;*{%=R
zhTi3p#904ugx)~D-`bx9x2P7fw4_pavN4!MD2>thI_{mn-SbGx);x%ro{LmO)a_-^
zbx!cJhZIkh)WbGSA%-Z!=^63@3m#r<emhx|G*U$_Gf4!x4|(Nx#5!BgM<>r0gv9w`
z%n6dcZ|H<)OT<JaFojulrAC&0OiVoIa{kcPqx_Tf=b=j3UvUcUM&~W=rQQ|RJ&~I{
z+(?(SvnpHf_Yd3&ULG0S_C(lB{HQ#k2G}$BRD_M-CnSw6eCC(EVcZQmx#0TV*<TEq
zRSgV&b&hxrNSPoNK)6u4hd@H?f0FR)HusCblXGrs=r4F8NE%+1CA=O9If_=A=2<9J
z%k|vM`AVU^U2u4~6hk4Ace}ooa>A_eMkqkp;s8mYgvCyUNf)=0tuzWHF}T($5!v<l
zwimR_<+{4o+6|MD4JWt5w1V+kEUAd)FB#?zJEQ3z<PVfm%#)@L=x#&fn<<MQj{AG_
zR>YN^EEoL*^mU1UFsigMXREj%H}iZw5vJ5nK0Jriw?dV_6G|Ilh%67FQ2o(XF4||n
zc0&uzzMBVJ=U@1P@ly;?SCIYosy#HydPac*R@RDzR#KYY%XBmjP|Ja6A+7Ub$wyMq
zG0~(*6O?TcW9QZ{g995dZyfP01p5NBO^TGiESSoEh~?uvBQtGaK=SSAI;FELk&?;g
z@r;y<|FFcEH|RU0`Z4GVG`r9)n*|58k^hyQLBdgr>mv7SXzmbsQONT3m4tmRc`nC9
z;HU5V)z^Krs~<xFli-6uh&RSm1H6}E7-ralBkr^0p)uhGrD6xSuLk(1Qz73917jlU
zq0caA$Z%W{6cT8ANhhFj^;IRt#w?lL$3QuFbkEfW6^)|0;RYJkLC^DDS!JeY5Lamv
z!W9vZ+axGDQ0Wv&p1#v?&q-ln%{oj|!c3(oeLqv|yTwNz5vl;Y#L7Rn3Y&cWmy9ey
zti6sFJj7=IArvmjB6|1QmDABXzbpI%u9|R{CvNQedJc07K6eq{4gXJLVd4tAbUiAe
z<?iiwT={7w{jLVMt7-V4O$-LX@}SPnX^B)A0zfYT|0HbkyA?WS=GkkM2lnqjVpQSD
z)D@yZ=wqNtJ}lhrk7dY90fm$ri~qxK*l8S38PT8TOfT`^XtBmzbD!p)&yQ@x6ex@S
zleEBoHqq%vtNgLED368>x|NAp{4P0Mbx$tt@hRX}(!A{qEt7ceIY2|c^UI*0)^T~D
zk9&*Y)>YX2&s24=y?5rPL|qBi`49ke!%q@kNeTw;H~y4e(s$uhYl2mL#%AOih6*ki
zq7XjG&fy6m{k=RI=Za<2iuZ@WHa@Wc+G51KEU7c)glt*_W;+u!K-z=8OSsGXj6Gad
ztS^m%S_OM~%1khji4MsGTw&x@T;r{HoNK;RWx2%~K0y4tV$_c;w!Aqow!pn0wd_+t
z=zqy$LO7Hqf@pdRp23}32s!+4Rk@pGOwJ8e;>sv4(}OjvQc{<Zcnbz_z_+0)2<dT9
zb`Sfmm1AfCd?T6(vWGi@J{>GbTJDmJ<(k#gOpXGM+{OS6kQ<4PBXw4Gi=B7s?<ux=
z)X3Xk9Mw|KjwD_l%`H(IhYpqR;#PDTh&HjtF>4j5Pq>XIJ2T}QWM3}5ntCq4_Vc{e
z?1ZCsMdqDC<*nO#%kR&2QXc>R-Oht3t@nA-@(0sIr-6kk9c|}CgQy+BIuKgsif(_Z
zFgsZD48{VXG_Jot;K2jY;YWETlJX#f!5Bl14>ON@ca8BF{PFd}HyVNjG6E_D1&HQJ
z$Q}T24P`*@x`<%XA5Uo&1uu9;EXDhcg5{YS)Vh-kVvKxcp@(ZeyF&}6CI)`6!IwCZ
zmmlD(z4bB~#+YvoML3U7zi1HJ9d(2JL+W`c|A&(r7uncfli5daam<J>VFwp42LcO?
zPNYoiF-@3wbhyZWI2ONg9E<g1&;B^>;a)2iEx?4yI-)H4h<^inWXPU0->aTG@D0=}
zXb4Mt|1|U>Ytg*3%~uJ@O(b9f^Dm51s=hGvL^iWb7z+ZOHKgmvRBenrDhrcrRVPb?
zDpQ9j#O?a17R2NCp|E^DnVi6Ls~I>qU3LM8ZY)%J_)eD?LYg<I?U$C~yvFRtKek4m
zY9(i8*n$sXnAOZz*Bb1STwD<YBONE{KQ>AOPc{CdNQ2q{Wsa2wmVXW-?w11wVADR?
zYn^nW_`l#>Pf0fvgd4K>-<T;rcOzcv`LYv3GthMtEk7nWLhklK6}l9dJ@|<K3{a7I
z;Dg4Ua5u2ufovG(Tm~>=QM&j=Y1*8=V6j>IdEL1A;k%(ZcRIgswHN3No<nmkz$g+5
zC0f_RNLA}+#3C^g@H3`R-<4^-ey*Cr`=y_hrxD%``_oU=xPJz;MO;mRS6+G*q&nN`
z)ddz>{@b5st_U@lS9)*%+DgVY(+=$UgTRVBXc2Ri-B<|X(!d)Pn{h(NLqNg!In%H2
z<^~NYt8RK*7dKFhMe4)@-_h~sm_{u~a_-$w?Ap$(utWR7*L?9?2!>)HKarE56`~+g
zf+Zd|!BJR<uk#1abqS1%n3rBwQkj1xC{2yh$+TIm5H`YBIvF1xm2Ixq2@+@KB3X*J
zTqikicuDWv%rZmqZ6mg=z+;ZPnWB0hxwenf=F#j%NAnk*xhJtHf7LY>6Ly~w1-`(W
z9)VIaGKPqlUupxa_E2CmOrXN&i~(*E0Nlb`CwK4UOJvk@U}pmrS1mv~iJ+WC!wu2I
zRlBk4tu|9g*7j{pb=>T`C2-E9zC0Jkx7&I_`<}>%;Mjnsmy9Q|%B?={nTnmb@{gN>
zrXt~}$oH((BfM<<PSoeYEeu6yD%!1wR#fiU^z*YxCT*Jkr3*r22SNs!*v~(fU7LsW
zI69<P-_r?!Du99LscJ)T@Ff_tnfnpQ_AYKtpnL@HW#IMV0x$S!%c8zLFv`hYNZtb%
zeRIv1<S>NM^kB&yDrlewGy@LaT?WX~9(BWQyM=9{k>U`fPP$zS-bgToT7P;+T?EVA
zW)0JT%RXT38N6|3qW)no#exBgf0+dPNd4mEQsS%@v>P_I3WZ#M+`F-K+4!?Mm+3(I
zR~h>cFR&I9{T#DWK95aACW&7O#M`}UjFIe`{%uOeh(+-4Ius;kyy@O!+(%n6v|}=M
z{tS23e=Z9ru)o2z1ENL$+YT_MK@Zij2b61@L1y!u!$vDY;dK7V)%W*(ZJ=<<HfRWZ
z0whGp{H-X($EQv77RIyue-q!@REC2B@|l?J`#+MIL!5I!CQ*dxSOzx$hYA6uiK5Ey
z)y3i1J&xDGEp2K&n;!c3Kc{;^sOpn-Byo?U<>(s>#j++Cm!mdFa)hfaWV>C+#)wUl
z<qdiUe)9wdI|4Y>&E894C{{+las5Tvv2=|V5cCM2KsFT&KgKdNs2ZxZ^V)6pU|mWN
znASW!AzRY=>tjOe3o9Pko`h|=-Bv*5$Mv*mm{<PQKqtNxjD(@bdLL5bb-_8gTpGy=
zV!zwaBCiMfy^Gr{_)BR5tF0JFqceF2F)X2kh@KB+qh^8S$MfdKQ`^zZSDg^?L?LNu
z&V7XPXs`!CLS|CC6qyXIG&aE7>B2A4C9B2o!tA1=Z(cx$H-jEbr;CrbgwJwvnXqks
za|wopDDI=?QT$6=9c-8TOSm4>kTUjxww+?v3rV@Nc(3vsx4i(kMICFQEpcHmjq@F;
znRuRV56OOT03fK9TX8r<;55f7cyaY&eDAAiN8$DK=GFLhACJ?v0<|IRt^es38FS)}
zN60>}f3peC08;Yye3aGudzc<rqSnIii!|C;24uv=YcLD8wujuILBBx5ehQyeDBk;n
z1o*23y6xINT~u>PRj^!W3~H1$n@wk?o`XL$!E|>^ftE@a$N@1EHelz7EU-F%Ld{u=
z81o3D2hCSa9G?;mW!iKzYwQ0Z%=@;l+kGsrO#Et5V&Bn1^(0bITiOqDeGF>i!h8Sg
zRwuqHjN!tHnXHcsRWtnls+Lg!6ZdeL`wxTNV#kYYw6q2+HS0KU|4%V(i39B|$3&I>
z6+f#kLuT{DMrMp0e;uIOAxf9U9%r)QKp`WUdygMSK$=~wUA^JhU+i=axnK}cGPPd}
zXdu2-;(JnN=2QMTATAx8IJlG9&81aA;xh~(BLaq@T~A_kY0@h9a1qG8xuP!!jZKso
zE>C3S40mQ#8d3dzO)69a&TlheV$xZ+Fch&QnoM{bTPnEUmG&)wAA=5-fF>s(LiZGd
zsAt2Y$%Daez2Tx|28YRP(V6)5o9@#MKZ!$Ws`?c)V`6rt3=b<OCe4r7Xd&>H!`)GC
z<v-wU11CZY&7BfXZJ%>oY{YOQz=0&c;m|$>Bj2x>6`A$Ei9<Edz}R2J<ELf6!3t#L
zBPiiPW!bF~%mW$$M1OzQRJHWym6avX+EoGsMtCyyO`!<wYzDI7Qsx?P7VBa1##H{w
zf|ln@tE$y0&-_@nl6jgMMh+vg?5qqjm|}s6+icyO9+V2k8HH5I;w|`lZAk$AkB<ut
zq)bcCYsHlQq)n8oAps|4(-fHtySXA|le@W1MgGJLkB)2&4fp4Y|3}QQ#l;LUu9Y3g
z9ar)Ap_n0D|8O7nCv+bZXTLUQ6=E6Mr5L5IpRWrS$S76D$?7WCF1bnIz~e2krT^dW
z=x<J8fK`!a_De4p!-EG7utPt+^O$-C|0WOct0J`_A6Sp(XlU-k_ukTvFI|m@zk)W}
znQ*=2$q|G1b`eeyc69ftAdd?f%;roYu@DEE9qDDX-!py78KV#bBBZ80l(%C*QMz(|
zY^DaP!^4Mb_eGgAiT%4wtL8^LUTxddy`92r0yg=;wYBVwdS*m?hzru86E@`npCH0d
zq@|cb%ibd+5SfL8ClCjCPT(-}I(#q&3)}#c`DZFs-G6qylJo?)90A`GQG6*FPOzX6
zq0q&$i9?tloPe<jdyuD{4I?EbZM)c{9o1}D#v|mN`$@Y)xmrqNemTBJCL``XoPf8u
zH=}9*v}Vu?p~aS;+Xx=G;HvCc$0*w?q`@6c)4;0l{0jPZI>u=Kg-06)%Q+)&R$BP}
z66MQv)%GhpB`%o%&LzyID>W71snBc@uyWeMr*L2Ir-*9wo3)W39^Xz1&`(%4sO{n#
z4z9^nxj{_yw3xJNgd+fk0g^C6HEM>Jk=e7=0FRK0SObg55wS2#o{CZW&7Z%$fmIl}
z63ckLdMOHqq(JG(oD3I&DH)wZ6ueb8SL~q=VYSEY`qF2A<OPn{5<2N~s|am8s6;-T
zdly6D9cOWpst=ARNFM!%24)X2!ZH|JV|~`aI?%xN?VOBQy2knS?>ISBb=3jz$QWvj
z()CagCdzLhSZiT#ahY;m{yKb7`m9Kp_)ZWoPnWv=udT-#LsR0>8{=FvvVRAGcxVzH
zvJ{o~**8}KAo^^yM~1vIi$7G}wIDxruH^Xzj0ep;Q$=K7T~s0`4YPlIJtThs&L;a$
zCfWY&l@`vgf(Lg5b&COU&Whf`;fp)WsQW%6+Zp8P4Bdw#02?DM5-8gufX<^VKl<A4
z_?LGofL5?^USu8O!afY(bDKG|dd86y91{8WgA^HdZRxD>+?3ZmAToD~`ZHW~t@J8$
zKTo$B8>1<`x!>w-%Q+%%elowcp2)R*D9ZT_dpol=jzU~e%;`9H_TS&l0_2<bS6K8~
zE&laDu(Y5A`zKxE2Jv-4xFSxnMjgg`hlN4YGv1XGH)D>5Rl2V)k8ag|x!_Og{{0=T
zYz%g%pOj2>9&bWR_zo~<?s>?g_cD1bqDb=l=xzG>R)766gRGYoXWXBgeXP>!bg4jP
z{#JcE5NBhJtDsw01Fx?Yri%N<6+2uw$J3lIFK-&8Se~x<iS1~twmkW1J}^a!kI{Rx
zM_!odO{!H51kR0jK2OUgt_aZ4HbXW+YGR`}!w^Jg!MVeiK&PMTEeeQ--VT%PSo)#s
zBAf*QF*)Y&$vZp3ZxWnQGkO^D2}#BTH^4)p7WQoO)A`TCp_p!Lgq-ME!(Xw1fQ;ru
z2-NOoP{hCDgDGWa@FfmOWgMfm-a6*zdk;WH56Aorc8x2$27DB1r-vo){K~e(W{Ql&
zSnm%Bdr*GoHPLw!5b(#>id@s)`_1@E?K7JvRXqlK=hypA$R6uC&dX_kjK}IffLx?5
zy6la%@`L|#{HdW3)UgP^#yW(K`C`_?rZgC=VFs<ho|M>)Pe6%4oh$Wy^K8DD^eQOs
zlPsEJ|AOLJj8nl~j%+0PD1j$*p>gn9t@8yQ_z*<wpY4ndR~%&M)NlCun>>9a;&hHf
z{#%n8mMlnLPM>_?eTW9q7hkmyJ5p!5QUi*ZvI*Dq2J?oUjf%?hbNu%SZNa55ExSai
z;DlDGVsWP_*Bsi>o#N7<3hYOtu8L<XR{Jt<+rq|L;!C<t1wva^2^1{I8yL2V3(P#Q
zVq`0j>7PCV;f_j`Q}#4V2!uP4y%w0G)HTmw^dsWN^=FmC>}3(K@yC@@+@v6)GeYGQ
zkgzGu_Ph_+j+~M5U@k$35(G5M#&tubU-rUAn$2W&;ZwC&tT~<q?=0T%J{8iVNn8()
z1*)zm*Vbb7KfM4RfUsC<*v~({4tFu`J*n&tv;JVL!I#f$vD$Lvr@7Ffj@GQCxNA(|
zJ(BYM7=d?h7>n%a`=l2grL{s*msCzlU1%*0`yX_fDjs*F<DWXCB%(#Y0XQl!j35`u
zUx2*yCC?LBHi+dKENY<^(W@v2a<_wstpq;EgV{fXK;4(h{2WI`SqBy6N2B?I2fpqE
zAh)v+WVj(>@}6wGJuSq<@eYvDjUteD1xO!68kB(#vZOx*-1G>*d49)KiSI;`N)iy?
zYCtAz)<xt(0k5O3z)%dyL_Gdrngs{GH1d9(N{*r?z}MU=ngUA?#Gr@8@50c;qu$o5
zQxp<mhu<~SaIi@|csQ$jk@w)J|8@BiqBHt%DQ@YKWXkMsfVjtZ(4*IE%2mR7KE2uh
z$#Gs4nq+yZ#+d)3GN2;HEfTTo&wa2+b6P_{VQEb`OZo{qs&_7C>AW}->oJi=hMTs9
ze{RTVV<kfB``ZJqjF&##g3s+qdVy+~Lp-1f3F;}h?JI#eTTX^rIa-)IkT#Fc3_(#o
z@<@CD2jbt#!vNBkIw^*{SO0Z<l!jh*8HPZSZr07^VyF+M!5Uab+nbWZ!0>vLXI=}%
z6MES&EI+owMME<W2QKmOWSRR=E9Y+WMNP59>jTlt6a%x*o*zZB++1pd1lXokZ1+k)
z#&oMMlwQObGUV4P>~_Sz3;9mpX?o<uo-(Rycs(=H`Fe8VD76N-3H_A|fK@RHiAT!F
zkdc*w_7J$=gu<L!seY;ON0sX<8B!pp3cn}={A^-Ww-SWZP*D+|0dV4`gC^a6{FAU0
zV+5qWQvjKTLl#7YQ6Fc+|F&-oDFK*I4M;I2aP9niU>E5Yb<)K;`AB=Q9Y{uF{3M4P
ztePU`Vo&-Q7_MEnNq#?}_N?@WMcDFg>k^#4BT#X(l9c7-3jGK7H&q;1y0%6~x7fit
zcsx+PaUo+k1hDU<7h%Ka4?>Q>-Bfba1hS-~O$EUfn1~m>BQ3QV;}-L;J5kn%*tt#n
z#&7qe;=}ju;0VMNsCXo!Fpx*}b)vu=B=8E57eY4piRVI^W>O+Dj00*1nlxVM1t3c6
zt7|4HWdKxThEQ_mFd7icPSpS<)}9|zB4Px6K&c&yy%0YP+|3XmpmLorQ9iX#_%;Si
z`q7XPrBLoG_t#tY{`n=a8ze{_kLDP|tFuU=sG}2k{FN-Ux)Bk0KO^IGSc!Djuqmh`
zyF8&(QGg%ZA#-qk!u=RivUOmbQyx9SEz<h_rRyK@bjuf>er|NLzN7?QoOS>oTAqT{
zobYHFfpQM<H!6Nny2!6fSGE1fV(2XtMvb-($$vToRNS~j!tMPN4alg;#2^MYe3$Dx
zr;$1i(4K^WZI0DAZ_OJByJ#vfvSxy6;6**NEgy;-@BxZ2-4+FA(BaG&)3bd&U8M%k
zgDV3Nh<p`+Ja?>Cu7#Z7H=?jUUXroYlO8^1(Zs;qt`bH}kW;lq-l8196A{^7F_tRA
zrdVk{lzs!`XB{`O3Oqq+rVen1O9bqPsq5<_k}u5$5nbnO7vwJrP6CXnOtXLAT>2Jg
zE~8Dw47XMT`(JsB>XtknWaVlyT)Y%!*KvR;{Oo(zN^RbVx{*uXB7{ZAT=j-`qwq;T
ziO4H7XL2aPjVCDE#I5Gp1#K^xC3LbZjRk6{4~+ju1V?-Xce0Mit^j66Y$`wW%%3S3
zZc7g};})zWD4=x-6%t^NLrdLI4aQ@%fXlteoWF-@K`Gu?awr}v)j@Q$-W?MmCXm_>
zM{8eP8i^51xw=myVC@!bmQK-1N1(zTAAseJ|5m;@VDoz5Bgghi?ZOrifX$^nIRs(J
z)X7;N&|vF*GDrrkum;jxI7WZGS3xm+z5sJ6YW49GdCCXBsEB#C?w!K)9Wol&c4B_)
zbBk=n$L+=x;vt%3)7Mck1ML?nXY14w33^Z~NQ`oHdz=;aH;&f$@z|oViGWT_s3|a@
z(5Ec2Y~5M$KbRIvAi_R{1IfnuASMOUVNv3ZygcqR0Ub{%oUGACr*iFAGzvkkOzNlB
z3VwrigsW913zZ6TOicb{%1=!27;?A&$Jl#^bKUQM;1LlLl@*yGvXw0}Gn-_qP{`<G
zlU-5D%1mTt%O*RN8M3p>C|gNH_&s0h?%eM0@49}!KhC+%b?$SYtNZ<Vy<X4f^Rb?<
z`J*4<KmwgTO>@q(pREAwszPFqO-=B7CY(zgUB#AKNLZ>eB2)3dSlfx<I(V<l;U*Wt
zMvleh%4}VJuUVuOPlIk_Nqc!!>}w`BqnGE@<>w_@#inrGc986ahm0u{qEt{5WV*iB
zu=14Dsl|1)vIZSJIo2dXw%Uq!>a6*O_r!gdM~9A(xh|FrpyWAY>mtzAcS8~sYRH!e
zi>(J#cYdv^trJo{dtyp_$Unf|;eV%!%7k(?f(;EdC$9w6DI`b(TJrbjgi;B$6Xv>O
zaZ(77-aW29@D_wm%xr8eR*-AJJST6SQ~@VP`^JUhCKdXCt^{wn5Fb5g@6XM*cX8u9
zLVABt@ET}-jwd^NIRca_8--MJa`aztAc2Jm{Kx8DkAtxOSl<y97#b6E27GsXmffcz
zL8G)ynY;Ql6tYNw3)>Jn1~__Y-1d*{#t+urSEmG?mzd<33dHu*-tU}CzlK9XA9@fe
z)llJyC8sQ!O3WE?73i7fH&|IqS60|+3c67$H0>w(kB7<K=Jd;&%`DydRsAOE#AJ9*
zg50y8#!|rTb944fx3hL;X$F_-eWHx*$v0DL=`yP-_wOI**_SGZQw-+QqhDrIR-r#k
z#AY9~gHX6d_m};E@z_L@E1iWCl=i_Qha3v$6;NUyz~YB-@aBsTiQM=b@<7ykk5rFl
zW!Ijs9zC?iKynNFsMv~64p8%h_ushwuX60LCi*!$ed&@y!klbHNFphz#R6l8l?s0Y
z$aLYg_;L_C0Wk;$QN4izKNxJps9Do~qX{b&b>K#s74}M}5q5Yt@06~4^$K~ITE*De
z^~iVe#sa0d1?b_qfjI~`b2m$j%*uj;1<eByP_$W8(p*^F+Ld?*U3Ac){%}LyFpJbg
zwY@L=&%gPM!)P_xwanv)hMbk8I#6_`g4o@e6D|6HXiK|&pF#KNKnYl)qqvN$TI!J(
z*t8t3*8L|v^w=-Zd8uG)(GE!PSP}+-v>l&Llr;X`tUA!Aq{2B!61nsC?`pEUcTwI=
zDulxqLcY|zD{N3ltLCDBbJ6ShwLQJ@d|>La#_*l~$Di&Mwk&v^zjIlF48k`9_rmCW
ztM|4r(bMB-n-$pt(f8PZ?fwrz@4Pdy{gqFV^x|gaISXR=)ncyGR=p3BJUo^7R*47C
zu3gBf{MePE)heR68|@+8!xF}qjJ+#Tvu#8Fcj?hM4PI)55M3e$U@xx@`a4-%5p{{D
z?COb<>eA(~sS`3%!0or{TE!k{1tJraR|=B2RQT`1Ma$L?-eYBEZh_I@;=Aal>6J?r
zZ7DT)4)g&MyT7#JLTH3BH@BQzUCV9Hrp)N9twGL0II9{hO1zN0_;a}?mDCf7%TScm
zmLTYL!-#gd3gd;=>?UCB!a(!BI1bs1KW_<Eu0djeEpOOiEEp=GF*mh;DpPkChL=l1
zY$BCl6J=x5D@+}!@ie%hSjYs4=SVX|JJ$AvlCnbj5zRl31wJhnz_KW!>v`1lpLQKw
z0XuFSVE2=UG4)K?$>yu%C=s>Ok-i+4Po!3R-Fv%$r1>@7T=5<<70Tw@XVBTw6hv&6
zmBV^(i96ar&M}Me>Qz0l%=?8{XJBwS@T~Y;1&sn=h$Q$kp>JiwpIKr#)qj9^NZ~u1
z`Y-0~%S=nZ*y~XmXkp&SRv7$m$;jTDAlVGp<L%XE6`X@|^o_WA#?|cbAe~s7ks+I^
zD?|(F6YhLdq&jaEfznOFY3!dvZ-H7#jY;hwg5niu4sMQf?xqAc<?5Fd=IBe%!AAAj
z(aJKX(Y|QFV)l$-5M1;G-g{B;oF|Yrgw|1rZ7mdi8?t_>2s-_2t8Nu7DQY<T?2v*V
z1vFwaijn)##0*0u(&aDK0-d6qt{Q*3+6R}ts}YhMIl-RLJ0$MB5Yz86teR1Wm&Mg|
zQ7^L)ZKfQ3vAwmf3_--QxChidwO->NBBD&J)@$Lqe)24&D|CZYP2)MBWkYe%+M=!R
z`jUs>pT+k;FOV`C$PHK=u-SJ|3#4!1yw~4eDa#hKx`ZHEy&?+`|7;F$K8_hco8moD
zq}v$@vY3X#jGyr6JpEyp4_ZV`n7P_mp5JlL=AL~QG*?bg!vo%mSV6ml(U5a}wL!p@
zDnpl+2H17-&ZrO<M5OJ-x<l<VhhV?te$F6*7&IW+(B*)didlHHvmKP;uvY<IkBeI_
z8Joy+X9-ygprYj-{V|CAS>;SH9*HVVNx>-+BF5SI1-zn;s07ikFdl3uIO=xhw&0z(
zeWd%??>Ydl6oe}a*p8e5J^sKFUS@EH^{+WvB$3@7;#Qu-!Jrr}Xs&<Dm#|%)Cx3>&
z2X?Sr`t_ONr`IS|L(|S5RW}g?>}&%pgtxotQM4vW2rq&vya-EJy(Jc(-KYfYY8glt
z>9|>5N*HzjAO0L)d#^#$R@;OZY!eq$S*N+HR{WqzKl2XQGm3s}u;6D%ypv`bS^-<?
zD<XvnRSF#+95Q^y+>5xiuCSc5c!I<dTtK=59$=|e2H~a0ttUOd^1O-#03<~mNRAXm
zcl)g(De>kDAo1q(xvNsRHvdlsOWyo;GPWj|<eTzh2$B=A&kl02`@99eu+5Y1UptU<
zU?oL<iyx_vqd4@g#hGm{%`x1GybO^78q=3^u3UXLzhMsCBnHSz5f+01Pw|)RF@f$y
zz2G5?$M(J=dUchVkx}}>#>BzrI0P%sCf!e?n`k{Q5f*>#5M93FpEXGIekq%vHECxw
zOR!Ks3YB!!Zv*IGbo$jV?Olu7jZcIK7H(TCeUP`A<tLSq$Y#lDC1EpKQDQ`LII;{B
zw#O(aqA&Ayqrz4OeTHM)!$<fyxuE3)r0&pv9$-@@bx3~FSZCvG`x9UurBmxgE9#&k
z;#K(WhUb~vx&pC%Wq*1H65c(*LB7i#7>W~7mpjb%t-moI^JU6=Ji$!+vz7_pX(kb!
zu#?lPlzpccV<cg)V2Q!uc48m_>rc}0_Sedyl7YG@V}1j_U!<p<%!L;N+UkAJGwc5N
zK5q4SHMr|fck-upnqv02L}BVtkTr52M%#})Vz%LfQuSxh29Fh74~T$Qq@l{9Es?g>
z7U2FDm66Y$aa_H6)p^<u(((YA%0sp?uF?2NBsD;WE{Is1wb%u%Cd#j(y6`>0m>H?c
z#UN@&RZrXg9?DfrtBgyeHPd~~e2an~2c!O)8tu@5TmRT*V4)`al^8_SMX+hT<ub~p
z<b4TKSU4W7t0XvXKlOUKAEv@3Q+#<#fB4RO3f<SFm-7tSfS2ptQO@QLo<*q*P?VLx
z22h+xGdB)N0lG9ruMBk^!~pk##d`s5i&3(F<4v=0?`dp3nEVFh9tt--^+XTpO&vkm
zml5{}oGtO_hmF%b^fx<Z4^qInR|gx#reSoB0GL||M15#?Jnduau0$z)5^H7Vxs#O@
z{kHd~Z$dLI=m|qCnr35RB5ly_37{wc6DIRjwW$jlEMHSHA*Y=R$Yqkr8@(jp7zhMk
zYr1Ea@{XYf(tGUyBjCTgP(6xCJcLmv2BPT47>WB@?guyH`|pPBn(O}Q0RHbh+q^f=
zHoy1|Q6D1|XoC@uR*j3;D0n9WFZOqqsT5V;@SR=*PrjlP+{T802iod>M|?o7mnX8(
z8&xbS0kct(cmUq<(iZ3O+MhGjmLCe!jGit&9Q#?oY5mg}){}iGvwIHsTVHspCpdNt
zA&$i&`o)clxCI-KOt|nc29`!Z^IZIO$f^C!Ljontxaax@8(u5paY${W5_c~)8jd0(
zo08dfnQ5oe*uHHDpdQ*fgR<>vz63#n37|~Tz&ea>-|9|%fXr*%FI1>Cq*%XmKt?iV
zVUU*%PVdt<hHK{`1_<pBJapt_GcXYGb8x{8UieiA>Bu74XDEB<`*A&Xdr{Co<CMYg
zJCR@_2YJl@bd&I(aI$;|5^hmXR6#4GFA12NwZ4RMl;dvMWUy3IpE#D&ZYvSUy4raN
zLCw+$!yOVAC%ilRyWRbD+H^`Lvs9+)<t0#%uOS8bwYc_m`u|Nx)T7xiYuLf~xTf!^
zJyqkg`95p&V?tRJG*t@Mt|gacwdK?L37aX%-am)a8&V+>@_|cU)kK*oU_HF(?|oKS
z)=^UV*B=^6M(7^j5sKmGfM6EV4bb|aEcoR0t95AK2~2=CR(!j<Ym&d92b%`)t5+nO
zHvBnc4jMJ}$&7;mu+-?O{$9(l{k;bg&|at<Uz;eMM129`KYO!v)qpldNC(CAd4Nnc
zq=;!KrpqC3;V>bg!$@kb1Q@Ieue~9d7g*b+3c0YEQEWGEyi~nztg)DR;0WxoM1Vl+
z#B2X)C^-GUqPQU@&$sdITm8!=YJFvgkFzv`-m+v3dWtq^xEH=<&`o4+F$F6_p9y)J
zI@+Ox!*7UuX>M($)EKmDoAmi_0R7fe*x1yh3`wJlh19$+p>^pVQ=i^g&I391kU=_7
z{E2I9z;%||jdKx_bA-YxW4X=%;t(f1{(S1hhKDaVKd}q>$@C;+$7guUu-84D3)sB+
zg8lPS@{Umr*t(!~$dcqFXKMO3H{J-rkEfpG&hK`PdHM)zEP2rEwsGC^O>-H<O;fPL
z_QQ@+)_NG20kQlSHBujZx`w3@k>yC5%E69DiiLYIBstJ1>pNq$|A|b(N|*6hg1jG8
z)GYWEYgWacnGG3SSbmn%(*EnMhQTRZ;RP^v>}`|<pOTYF8?`jhWL3ahz`~PY$}?!)
zkR$E}4J`jjncja0ga}<RYL5w=^T2Yo_ME^?{c{nY_khm2NyhkSpBrCWFNC*Ou3n3p
ze(kzF|K;MWxU;{lPBhm^ZvKGY-dtGE3#)SPvmBqQd%xEegG^CNV)U$Xzge}t+%j=V
zZCUE~n;+e-JsVq0y0eW#>;3Di&mnG-sl@Rsug;>0iQMV6QQ`~Gf?BHOY(caewod9W
z2H(M6x$zJ&PR+Qhl3d5|M^lvv3MVT~&5HA%m#<Hlvmw0tFGyLhLBEPi7`qkoQ?9dp
zJ{G;mcgU_laroirX1A`Us2Bz&SGUTz%+g&r;>AI4Z&9b&CIV+txyTr>%xzds4>50%
zkWx07YX{Z*4sf$zZn;o79Tf|0&O?!bMC=o{SjiZ3Z53S3IW3PFp)?+i*N5k-q){%|
zczL`^mI~s;VXxK)8%-$TKOu8uG0HqaqxQDot-&ta0)l+xCmiL|<+mWjoPhxa7M9m7
z3F%@HvsWd90L|Hm*UC`m%6$TY{CKpU48?jy1yOPiXWZ6ND{qbEPe<i2>Wx?JTpzjo
zNBq*qIusUJyAl~+3ogyox@@j5bI&~s%2vdU&geQsW^qk4lLybKD&d0XkaIx>4_W}B
z0NDE|rtMKmsRc1c$n9Gy+JU9F<DJ$@-}sub1&}5jO`qjJ-`-RX9BD0|PTwvXLH)m@
z>s}eVTPX<&<&yC))r;u_QINA;f#@-a;!fF)dfB0oF+Og?>XcymsqsN+klzpEW%__P
zem|Nq=4OVQlge+xAk$=g5FZoxXfnhYwJcZJLfyqx4qY1K5GCHZ?z(HdtfM%mZGT52
z;p5{Z=}{bfOxNT76~BY0rV_c5b<|olDe#ZYm~Pl>CIw*6F%Djs6lgY)=XNU(SdM%1
zMpjYjn@zarh05i7`FnX?e;;TPDQpYz+muo=lDedUSA4;mtN@Qd_0legaLvFFVZ40f
z#i?J1F_ggBSN#~Wb{@@CPxAZGm<U?0%4t{$>b>LUw)#$Rkvw}mXMN$r<O+)ynMk7T
ziXGII2=-|``#EUVAmfv^AUg$kFg>pOMZ@&hSw{S5dXGQXe%?Fi%13oR4>3xQXoS1-
z6ww>z_@(Q$Z7eCjd&wIOA-^;U7|^<<X<%mpN6><?HLMy%ra}XYSPcpnuOA<7=wEqY
z9=kXX+2F->$A``5w(lRIE+&-a=hj^f9<rjIN=(3a1|>l=?14xycaf$k49$m2fPK2o
zrSX82CK31+>OM>2AEj0K>yFvGJ@r>i!3oI+z4)lR5ZH!5gdSo$?V-gZMLc%1fD?NL
z1gCblaD-!4UH6v09Kfa}ppkI6m_S)8we#vuLxVVWk>1j?_k5nFN7y0;zCJO;)*?*9
zG4P=eFb$EMm>XbfPBJL7P3D?mYG<&YlvVU4+whuXqY}20Hzjn^)K6z&PBU=m<YITx
z#0mN<BtbXZHxT5s4;lhb;)n!E-X-9G#6?KETXtF=cc14(I|T-|VlDAZ{Asx0_JRaG
zQaO6zpq9>=`xQTjM;%d1P^$rjYRt9=vI-RnuDw}^`YDnkZ3IS%DR7iKU(6`|+#1<t
z8@qT(2v-;xv%$2?38CA?i^)Cfa>4Tu4`N<__pq}Kie}5BxG<z+!JndQKy!~N!;1R6
z)oB<i!(%8vf7b51_!L4AN-U>$t97=J*bOsa<YKOOL(d)p)sasd1+5e+c<6jQ3R^Ss
z;WxeAds#WO0U<%~>>vY=o<Ld6u}`LH6GyP|ztGpY$evPXJLXGre8Pdg@zA50z4rbu
zHATu88WqKY@;4w9MP>_nT@j$`Dl?{cRN33zAeUb=S@y7Z5JB^fM~GoBZ8BwOJ<SP>
zLxHQCi@ge-^%dw$ec{`ppB2aIE*l<tNa=54OV|iAalae0aNUknVr|}F%~sOxRk=T)
za_=oZ$&~4%BGQsCWK)SlsxkhgObVyQ2r>xEwjK@Bd!H()Egu=ih<&(lfkkPCb)4nZ
zWRXXq@bKS1@9$z`2E276gFoo4K61k(PK>VDz$qQU^PqYkoZoIQ_J1iw4B1VUFlsmD
zoo%%r7U=9$DX|^Zhut!^x^JPSN`;Vm=eA$5Krw4Z3D}HP=|Musr{Y0-0t}^yV+KmP
zw>05@$%8Sx0kNvkz!Y2iiQKQ*-SJ@4&IldFRNwe+6_jFs7cH!bZ~!dH&{_H3Q=tk4
z2M}w10~y1s5#8KdpjgZ>B^$PcEy>7aJ$$i$J$NX~j|S~SfLx?-Xi}Oha;J|l5R)G^
z2CWh|Rt?h}v+^tayTJ#ZA57xET#T0Z{Vh!-urHL%s|MgcjU?hZsIG$*PT7XFUZZE9
zg6drfQuC{Z<U1d)o8btHh27q+Htyl1k|O4Zr!OcPL)wMYzDZ3DO;h6DSG>x=vs&c&
z?YZ4i>al2N7qaLJxpo3o#_Z?heCYA6KYN^Uv6CLJ&#Rv15#`0X6}gG6$LVHMRj0-l
zKCPC<KaGg6eW(1hN<wu1c?vsY+J8JCOo;qDm9N5ab+slo&D1~?Z$FvW)YuqU@m>VH
zp0vx?-yZQif>9TAa)FmCPkC+z<MiZJdJm+|o<)7&=nUkwhJg=Yh;#J@D(1<y>Sggr
z>6i&&TZ@d9>mQFzB}$VbSOcYE&p*$mVrtScYIGg8DKocJ1%WX$B*WVRt-&$>T|ZIR
z;MqMUGvX`uz1(ZBQZL-#GnsHh`8`lMQ;Zg(P_!6$R<D&Uq(vKtLyTdcZLIlQJ%i)N
zkE0aUsCPLp?+RUEbA|IJLa=g5L*DSA9K9?K^gm&~$+7$Tp99#$xUdD8F;2jy_1IEk
z)QyB*w=h^NEVBe~MSHfNNbS~&M7PT?xzQoXP=DLv6OVj!>T+reNT*CsF^>JNi-mQM
zsb~3a&RKjYT=s)SrfXjbF!GVu_hCiUCLy#swFMgIzd#w=l#21!NKq7D!s+hmN$VUu
z@@=3%c}ZZ+!}HFih}6>qQFk|P*0v=d`1b*lqz{LI;S*2zTb~+6k{0Y>TT_?muh2KO
z!rt@GR!>5&e6|j40IK2Q=m}d<9Vp1Kf7QIQdD_Eias#H#g;FF&gY9nhwz28_fNnhj
zHAw3{!rcT4twoKihf1|lafA_uB6gT4zl0ZKsHLTq29cdZxQ|M0@~cooL<BZ_ZOVYk
zUAg>)$@y$!o7?2U6SL)~nVBCD*yhKJ9l=ln1__C3!%JD77MYz~@=sDqMTS&4nHpcm
z614$E-i5t%VYlz61jOLr+hJY6tM|Awi&SlUu=C!(kvzbLryvw{@(_?);=xHk!NoCG
z6AU0hY`6@x5`1qQTE9!3zx2W7<UPr9cuoWb{AlW@@l)vb{%<hf7IfUM{W>J<{k0<L
z<f~7~W<zx+HP3nHB&})2?qK5xf3drzeMAmNIHQJS#<Vw3GvuuAvOq%kz|5k3iDF&9
zc>igQ3IR2OD{d$D&+f7hee~35<Q__F0}~Pk**UENbY5Sgk^#m41`w1IfD9wH)<2Hi
zIfD5D3&`4^OGr$Nu6{rC6{<oD$RA2>tA9K)H=d@PaIgH>)Wn^^N3;Wgw<^P9G6k%#
zDzy%TX}f)Tz?usD3)(@MKK!#%I;2#HKz@bAROVjFt?`=CJnMu*!k1vLB4K_0s2h0U
zk=Ew-+hk=FJNvkzycjx<*eE`v-p4UUeVkyLERM7a^l|FcyhDFQ{lC9Bi7j^J(QPJA
z*1>=mc!eS6O7Q%cjd;9;fAUR$oY9y4PVJzyCu+p^tFs&4jk?NTcGB44XQLrsfr(BF
z?Ws))zLUW>J`o(o@O+(c{ctYDk4C!)-(dw~KPZ-nJ!dKA`PMeHHrUOXp-dvOsR`OM
zjE!Po$d7c&AOG7Rik?YEVC-0lnJ_$G&S8_XquK@uYEK}DMzpHH5LXa|UmAO^=>W2Q
zn2;F?{@lPzZlDw#>W=I+W&koii>d?z8cG_?i0kn5BoLG$R&`^hoZLhUiva(dhUv9~
zI;ucU{J|Wwb73eouWWZ0w}Lacp0u8!ISD2_7(yTB;}_j_>nZ^Cq8}GDPKigcj!4pf
z!M;O`<B5OsZ#((>$V%QNTxi9|xN0`h2Nc4KJ;zXg37<X@w9DStnVO|)dDG4I+8Fx7
zRkfJfJ1c7p4c6Wh9m(A<ZKxXN)1<diciY+jCHyfFM%|OC{q+~p7Z-x+T&UGrtXG|{
zSU7ol<|rmR+9ae}pE~hZXZ8EF$wmkAezxE?nEyUw!<xHePBQDTqr?o-x59yrL|z{R
z%0uc!{4%AVHJr>0kYzz><4hF4?2V`0EDeW`CTa_1G{XZvZyzs6co<`!NA~EpPOdQE
z_UOVx9QtlE+T<h1b6o|kl~{zycmgA8qI;;2Gl_P}9#GX*u*+ZcWU)vv{<Fx9kJ|##
zv5@#bPI|Ca@hbk^RH%v!!ne+y++P(H_LOTP%cZd34DmkB%d#+#rzwOnqV`2&S>`rl
zGs6k_#{%Hd`aL$})3MI{ZdxTL<n$}V1h>^b6o0PzSWPuHec4jGI`QM0X3_<8(I-zn
z(tJz|pY?P*10~a`4^CO>-_~d=r686xRaEwWFYV}jGWPzLmIipiPdqfavnZ(4lyv3%
z@3R3J{>r>V)&Or@jF(w)%7OqF@C^-pTml1U1SJ0iZdP-zMli_}bDE~Wn3udW4FMS(
z-dYx&37o8|&+>=w(c_5{S5l=k_2qqgMCBv~acCM4qQMGk!gp0eA|@bs(gW$)RC{bs
z1xJ@1_WLOHqnMjLFCLTks{^XGn5<<m33{p)c+O={@xws_bs=_C?dOhGm+#}a9i`^N
zK-s-O{fDq^@qfPbX;@A|d{b%)IztDl%efskBjCfp>~62OLDUx~us95_i&rNkzV$Xf
zt|%!nY<b4#Bza^M?N}!vnSbSQgZ%YoZah8S4W@d`-iZz+7xLVBD#mN;|7@GJugfcg
z+93T%4%ZvPw^aJvk+V!tBroK>K69Q@ia6`}U16N-#Y|-#eblNg89zmduv~wiYBL2R
z?LGU;-vV$q<p9O!fXR$*@V4<(`am~`*5X}?6L^>BfwGWcA<Zx+VPiX33R1yk5*`(0
zp=$_@n|X661ERbD%*!7YX?s*bX$!bjl-}a`D^kSJ1fC#M_}KCt62x1<OMbI#mMKJx
z5zt>WTYVMesnZM&o3zU8c|Q-ZZ*ffx;1Pwe7eGF)88Q`^LHO=m&J{|yZ5dQMsabIC
zB5JgaT!?AK&cGB|Y%p`v0!D<r_*9*|9f`TDTy~N)&`vs=XU2j!wi)i^X|@OBZ;B@N
zrZGl%RD5_e@v#{=VH#67O859b)0na3Hu-Zv%cHs6*6XCmF6}@ruyt*<ynNgimY*oG
zaj4gE!C8T*Y2}%4wu4#oXo<Mb0q717{w}1HT=`xDdQ>9UmERr@dX!sn>A|ya9sGT#
z<sAjCpM6p41DyzoeRm=mWe#+SSL<o~1Qtkj3X~2l0HU)XIx`w_y@o!sD>-6nD4fJ@
ziLU3)EcMpEDm465a)<42*JyP2BHP5TvlGyFfY07+W#aWNdBoQ|AlOi%QKh+WjK*5V
z=B;!{EI<Iy6udKKM$e`_^ZjU0TF<+NQmIoSj-kK@);WjutiL9}K+_)YJZ5o^U>P~f
zQcY^T!k=2vX+MY8Y>rJ=PNk;!jgF8X>!N^isJ~Tq(W*aSb&>GSDS+i#Z1&jRv|)UB
z<<$GLTmo|R90=%YyPBr;?*_w1@(TVPIQD2f``H(b;?p+1PNu=^=oOjrxdYyfOyv*w
zl(2<le9*SF{A1I^u)$-j^gqd05t6N0{(S4olU%m|@G*HANaR|cV;(g?LT2xr6!mg+
zcBmv95mf`DA3BfvnjR9)usA+5(KJzzEsvco_$*vw)y1_qx0c$;_URd;r(b({y-$7E
z=e}h$*ein`+F!)sF7I6c6omjrqM@(a8R}AKRnpRh&PC^MrZb>52m?*ju+2~89YPMx
zviA3v5}^W9Yun#V?|}=>3jZn_oq-bY1T7e{jV+2r94Wvzi<c!JtF;g$>XqhERNe;<
za!gYS=S7K#yBVQHH;^0G_JUO{A;J%r&_XZoGq&B(r{(e(w&suU#q{l4;n^x$fo-c#
zU^zwvN1GnHz#C^LmTK9`8+%Ny{*bTx1r;;K`_aw+gnB%{MuIoEx(`yu+pE{q;m@d+
zz%f_m%&#g+Y#8cgXV;x;IepxrCE`Z*@1yNgF78d2Gd^?}Gx23lPlRvnTU_kuc6Eka
zuWa?%$L%{wT5hk(E;B$ut*<)_r(3AaH2TB>oNjM8-JU1z+HKFZnJpJkn~fJJSGe%q
zxE0avT~K<uQGVW-;!ozzKM&AHj=DELKZ;d+2L*gPJ387BDV6HJmP4VFNq<0IaAd1z
z7g82=oR>x{bsG^o2VkFb8_m0oV^b?&YKP#owZ(_onrRPpDJC<2WH}G3LNp;}W94cK
z>}*~9#IX$R?HOQx!cj(eE<a=Q>*m`f^Uv`30l|)jmR!F~dv`P+pm8>>7mu7`HSEM}
zlmVJJW8h>6J&Xb19HIBzL1o<yLa=|nI}>09-rol)d{;;SD-e4dX1xEteq`bC-nj$z
zEafQM$TJ$!XelOS!VO6{y}T<T_<b@&q7~LS6yCC{l*En6rgtb(83w>Ep`W+;WB()i
zfEGZ0Yz8gWV*d1m(i4)fu&;Jutl9wNJSp3i5Rio=;d`xmNP;_`6g0y`vk#OsG`%l{
zz>d165V~l9RTcF0pQ81(4^OZqmjEe*B7!UxJPdUTZ*<ar0M77CEx=tjDIR1HDp&#b
z_5S^c>r88~6aL+}x)0u((+_S_PksMseczc=KwcnLB(dT&d9YxyF+1Tic>R2fh@2HI
z@G<I#pxA6F;CF87+6ghI%q;VyI>g7ZC`DT+-tWE0y#FZ>A_O?%s!&`Hl4sWq(BnCI
z8IQK-bqoxvmW?{c{Q^;3Y0xh(6M*G77zn_^_5`dgl^k2J4=faq?v6SPjMvIWA;E$X
z9-<Nh3sHxOIxLjVvdFl2trDp^;bB^c5{=UktbmbAv~8|z%w73D5mmVPD|6opjUC8l
zDkbE|&-rr>zPI%Z0D<waV(hiECWS=SB+shjj|DxABu3ysy5HtNCIpfll&QtF7zP))
zsSO{ql_GAHxtEb~op&`gFF<2=T(ckb1^<8GehB%$0{!8Qfj1ChihL@}0s{TEugqtF
zXDNh`VmDYJP?BiXh8Z3}n;#)-=xXJeFUyBvb0SIC0i0hzN7~)G3ZSFv<?JAAw6_Ln
zf1R0XE4mV){oymMJK#3qi}#DR7oMCJFg0_}F&=UXmjLS{{Bxuk-CBX7pagJqrxVD=
zF89)<0)lXvkkMWzwBk%M#3<-Lhh=!DAzbiw$*X~R_<XG39_!ul$<j5Sx6<6Mfj?DA
zdPp&VLm={;vvMe#*T!#vX_$4cv9{u;2CjUD#P_*OR^;J^GCUYr01X_X3@vVyecHXD
z-TlP<S_`NMm@g#6WuKxdAo8_8F<7<!j3{9<JvEJbxU&1=B>%CVEX7Md&g>OVK)^^y
z<g`v)i=xN72O~Ti@q>IzS$Fq}?<b3lVdqMY$8*eEIW*z4kjbBUx-_W$&#SU`>aIcu
zdpzLkQ8hf6gp+Zh9ap~zEe#pjrBFAk@tSH_k2*T(0E&^9(U6U%=OL{D!GtrPitFyf
zW{ApZ=)uXIh(|t*2g6GdRG}(qR21rlp_mGN`ZUXEAkphaaXead@iH${-BUvfyYz`W
z$=!-EDe)M_)$h>aDFWDCbocs|QaEZ-eddkjSpnu5bq@hbwo;3(SnUwy9Ch5t3Jwun
zDwny{B#SGQK_Afm<fO!OobY!kBa@^&y%w?A4^j3or+C419_=#b#>$_DN)qgo7hF7<
z#Bug_5ul$~SguMtGh9m;WsZm^*TXBqB*v49D3s-Y&A+cbsrvdvminPK@KEz>U`-}m
znxpLku^Gx5jbt7)h?mzqdO>&n(^K=I(p^8X)ee@?r&<hb(U;~-jLN>TLCRF)tgus%
zO1_u%(P8=k;P0EShlpEz1;_x5`Wfk>8=^qsc3Nh<jyCaP1HJQF#Y42!7J_FojH*7a
zbx?^mzIkKmJ!0%OO{SrSwQ@s~!ad`7_si-ZPhp+fSbDN{hQE^WN#HL#lMaJ*(Z%(4
zfK@0#_ZNt==ocW{Sb+s!ZBQqR-CTBI0H8sozw+G`$QHPIdgZ-ffLZ_Nts~1CMI!gi
z?YA%p1KNSkYoazuPl-n>)Sr)`{WXWY+?TTJYnxB_*x&2|3ZdY**A_6l6|(&oI4vig
z_;g25G6&7#jwCCr%~ImR8K4Iqva5`ZyA03)*aot|vtm-IH||_rL1G5evp$JG!OcOS
z;!8JSUr#8+8jm__EpEd)p5l32zL%~*D3}BB;rw|d1k5~0P};USxdiQl_}puvERJCS
zXQc~auP@4U&4}C83B6l!NZKd|?LQt?DOqQqlD@}ime6~iZ0;iw`y=VdOl&;ux;lh)
z1l%?o8{@On4)rIUgxqUmRLjz=^#kRBY)b<<WH^XJ<}psZlh90{Tu02eDaCorn%t{w
z`^W812##d{TQKuRv*wN)lm+>yY4=iQLw_286<MHi9$VW@Yk!hGd2Y3|9mO;*0kio<
zHE<38ZY_*HA0^&?|C8?XXQ;!!=nMXlHX!qJRdgAM+1p>o&4R%t*-Og=xWQ3d(DOu-
zGRX~c63(=va<?OLNLafahWY@YFNk@a<bBXIPY8z$pjv&qmCieNG6~swFP_{{g(6DK
z-K##0xEb^?hh(c$*aN$xxiy)BefU>=FD_Z^Zn>M*i_Mg*?%>`Bza%fg#L&z2*4|Mg
zpz<3mJ5fV%Tn+CmI-aq?l+ycFN_=q&+6b5O_>_Cl4LIM&v=sl71>nm2AjL)xv_c;l
z5-(Uk86Re;wcQ<FAwJ?>oOAA_LF0;hW|st(AI+~-{+==57u5ys3dhXiAAVPwQEOfO
zHYa+!iF<y`_gR5Pc(|bf=FP?b+x~<>@I9cwrfPKXMqzM)4pX;6`!isxYdH=Fuo&pV
zCnqNt+vAX(SXm#}>r~Ek_zDA3hh~&}vlRR~vf_>mtH1~Ha$ZT2^u@+NLDUfJ$c}3_
zQc1uO28ED3T3Q$I26lcM86?2lRh<=Jc_yH}2R`m;US18<;J8IP8H?l8Yt;6R%dQ!u
zFHX%Ja)S=_*1X2@8m%r_Mq=;-I3=&teF<9^p{aq2N&cZ$!kWHXB&v6fnw^MgHPw$!
zzsxq?tw6}5{6T`vR~JJ_m_9SNQ5N63>)l(zvcw{fb%y<ltkLmLH@}tU&h6Tvuw~Ff
z-e9u@-V#*kyMLN1*R#-E*}WO0^<6<k%&)f*w0~vad(Zku`b64uc^%iUusCrsv9)aB
z9n5mjXMd)Tn`rO?CLch#1EI(?6`sj>2~<nu^Hx&)z~AuD>yVE}%tH4n*7diKzNVbl
zbJ6i`?tMcuraCuE0s2MfpEyKzZQ8@1A-)93)7-u!1yl`<8io2-T#>_-MPrO{q^RwP
zP=Ff$*DMkNxdXnT$A!FkbvA>?=~+7}YVtX|P6f3jsIaWPWD%rjqxzD}rX~+3#If;w
z`Az;q!e_ASX^1Y6)aO+6&8xL$M8x0!p5@RZJKNp9MK?-hNfWuhk7U^62pAL;pCFPh
z1osz|FtSXCcF;#cMJEfHcPdV*xG?cfaNMz22|_VXcZ#h(YH-RU&&u~{7p(UH=-*H~
zu?|psBeIvdQ0SibUy!E;*X;baPeI%w`qPfF1Z?U9tVGlTmx$>u+#2m@rpMFCOqc`{
z1~frOYE<WhG@61^EMc@q(8Ucs9ngac0|ttoh&RLbciSKHJD}09@bXcI78;S{dR4ED
zk9dopdfGj&Aw+6L4lvO@PV*xh4BxK7XRH4%rQNGf(ng68y{13tmqy3_>&iXtExt^W
z0BA!3nI#P)la}yoKp4yY2KUbM+^r7fUB$SA#(8OV57UlK#ox%%vv31U^VA=*#*4@q
zMlmE2K&!XtIB!?Y16vED#9Xkhh+Y;KwktS=<8#)$3o1yiLqq@VOG=2lVN4~2*x(Jn
z*OC~~SpoTBGbkz-ZTb5lV~LU=Bcj;%iD28~b6e?KUq<^H>u>W-f0#Q)aXigMCxVV1
zepqZAOTTOFC#Mo`;yduX#yaMwqR_TM-S4sB{=xfK;ZklV6*V(kO9`{CVeyia2G)M)
zoo$aGXWKz8CPpoK5-DOOTS%4ePLV&VSLvLq779YFaNsHxORPG$djYhLu7VbfHj)l)
z?UPLc{T>q5R(guTt07voIhQ?D+b)<nOcur4Q)Xv^d~Gvt`hUVis;fQdzliJF9~nEA
zk$xV`m&zvvZ8_j&B(j1Z?`gv15V$}z23TGBGkp$_ec^RpxRPEeTp5)Ne8e|U77LH4
zy)3=!5$a#B&A<f<l3G6v6`Cu9F5E3@ngewiRiqQ5$%6W~f+SrDp!-OeRC>9;`%tG)
zKRSN-^8t<?$eUwF&IR-S`2moJSRN%PTlj|qHy6lwX|N;0-Dxpm{bE2n)6uVLyLffg
z-@OlrKH2bpZgC0^e5%i#&!gcIw1AEMS*h$PG3Yq@xTek+RTP6x>7f$0CdJc?C(F=;
zIg~HcdH-U7_Xbnm@xC6KIc30ZX>tn<qR#4aS@vZ`qTCdM6K4{Q)C*t?M_M`%^>DWx
z$6s*nh-rxzEz2Mz5K}S1IDhqJ2i~^=EnKn4t50!rdffnNiDE-hkAchmg+uP`E&GCd
zyKN<S)SxJ_YGp7Nbs;q{Ay34`QLXdAP5H*y{`!&p<5$DQ?%Rq~k%c{>a?Zbd5GqZI
z|J?pCur@<&_@pKQwSf;8Mn9&q5lj%s|2qg0I~vgX*M2v(@zk`m7SLc8tNTNaVdbdG
zJ@Sn7!{UH7bc2AXC5$Nm_T=I>gfhq=*DZ0Gj<@nj;L+`>zTQXei}r4>@CG>MEQb;6
zXcbrSw(iL+>^*-^D_H>U!hpE!0}tV#D8S!)DzA(~`{*_G#mfUEKo7ks9Ng5;P-oEi
zY+?+mnZ+Tj5@~VISxAE}v*m|bgbMP{oyLAVX4kbr3Agjt$Xr(dkuvz~>6!Amd}pQM
z=n(HJq?)vW*5{dk1~kBUzUU$9y*+8GiA8H=$e{%h#_f>OG+jdHUi@bKw#Kf>FVHS?
zjE3I%`OR|RdB(i*hzk6+1wVE<hv~zCziNYKcG_OoWY`@R9_1zgl-Za!0^E_+(Q^d`
z%|vkhCei>Rug1CVfK#syBn=OZcEx_)>8`KjjL%K)&X!967WcZ*x23k~wzczY$C6I*
z0Nu+^gli*NcLS?v=dE0*f2R!lu|Sw&zxrnL_(Gtsu1CFJb`XQMX~Q#N2QyG(_6PB4
zm%;8GPC#wGW&{98841aCtswY_x1}Q^OA~LI1dKGQyj&<?7<Q2wR(`n#Z!PhF0lB+_
z)T3;AJY~0F<5f`lh*!(ts%8Y~^xoq_8*hQiHX1fGYVQJVrF>-}EKg&UotR!PYYFjA
z;H@g#9*zu)S_?X9%uVLb4ao4ig~e#Y0*tPYA*+H`e8yT=UmvEZN@HRgn4%{CPEi!O
z9!C3zsJ&|kJX|+rmo5%T;3X`l)m^l$faJj10e@J^y>TR|qa+)T2N_62$Mfn5Aw2Lb
z3<oF3^wrap$k1}C&b17OmAR2!XeXy&)E#mnfCCi)@A3oplK5V?+kLc7G9JJ}TwHrU
zJPW$LLeMFEc;%3u`r?JvQ5W^kBjAAGTH5jY(4W@PW(NtYy=z2Sm&_CMOw@hJ*>mcg
zH3gfqYmBPhbqHPTKby~{52+bZs6+Jp@F(nV#xHN{*o@_0^`FTOj)j%)N`PILgp!h-
zOZF4&Z4j@D)!ujuHl*TtA#JLJyL1`f4vvz0H(fl%O2KX1>!;dV5dTx?ws|tPenh&+
zkG_M75t-R7&r7qqPOPc@%$&dfkb-+Di@hIpD71&mx2UbDFHdUt-hv~=zzX%Q`3TjK
zJvQ&}D8?vl3KEBJjmOolY@IxHDt9zx_FY3$uIP;_jrC_xytex<!c}|1U9(XyLf0<n
zjwh_m6;FTsc^-K?m!=>#nrV@kYh|*FP!`$}1?XSv-ot{Y^NUC0(phKw9{dD?)Dkog
zEkw}MuWd*Ztl@R(Z1`0)n}OXd*^hZdWv=l(wBseLH!}!DpQ&VBSSu!VdBQCPJ6h0Q
zm!UUorPBZeD0w756#4lw;abwpq%W9$7<DI}w+*B%ZydoawCl)_7RryGUKZ0*X^mRh
z`6Mhutz@n*6h@#-(A!&Lt-HAFnAn;^(}^q1&@8ofp9inA>0Ef(FXwGVH*xcH1{|ZO
z%u1hMEIr?}P`IA4KRpni39BeUh=}Q0#eay3M{sdvEHZAM!LEP_H<6kXq457Ms2dFV
zu%7AFTE9U_sG~CP+|^Ix-n-T0w`}!hX02iUmgCaMhl-z&)xZk71M@rZR^L7M=xp<0
zi*vG;j*L5QCM^Mt+QAp5IDG$f6f_AtfG9ie!1_AQv~dmo-Y0B#6j?RZNFO|W*u5jn
zF1T8VkOQDbpSu-63vDE`Ev5aDWb6_ODTnz-baeVP9$f@_hGyN{2<-f6<br9~pnHoN
zMV$+dA;E&!M#%T)#fo|33NEjV5_QSBNzQ8FU5<WMfYXXIBoI0+4mLilBgXX2452BL
zMtVum(z#!Z@prjSZE296mZto&V1BypYg>hjIH0T?3VEEulN$zUq5}J1R>?*{5*x2J
z9>F=a|1@4gs;A>b*ea6r&4_SyJCL!@)(Ie8W*w9`vtJuuo#)#k*iuF*vS`hg-kpO7
zK33><vQsoHKxs);e3G;J`c9GCvM`m<wA$!X_34=rU}trl&OY}em4s)rA;cHv4JD|a
z5X%A6^G_(DG<mBQNwreMu!JTIUJ{`kKCqi&pz|y{Oi%3c19KI6q3MRB5{?xEe`H9i
z1C{#kVUe81mgrNIs}HXUGAPMs=eAWH#<<SJ-k1VuK<QE&HG&t9`<h#aw5F>4a4)o!
zEVI0l>r>TRA}!gLnqNl%$kLTg(<pK!coQG+1enrt&g5+kudz<AK~H|JP`OBUcBHqG
zM^NUZ+<zcUN__j<UX(<PMt`sLUZ?Z_t@$&Y0HOjlY(D{-TXa83N5_qD9q5PjfnT&J
z>v)eywyDx#+b8hg>FZ{z<LTU|V(|m{IzcfO)-F1f!z!~>PrQxzKaVAWyk%ztBw245
z5|x;Tsy)gYoO}&kAutocq40EyaFhZ@iXvhI=Ed}2r-Bu-0X?%7@%V8|L$Iadsn`*W
zE6g`$@bccjYf^5-Jr7OHIb7qoGNYPd*P)fS5`8lkmOc69h~T{e1h4DN2?B$A`T(Uc
zw=<jw=D0o_6BWf({WdT9Ak4YOw>t*U=!L?Z`}Y5B&i$X->3^6M{3rZUxYhpR*u5c=
zD~LVXq3qhv4I`=@Nu0pP5tL#O<^T&*!4dWZNP>s%XUZ#F8l07sm7hStQ2<FWEL^H{
zj)dJ%hJd(RmT_F+kNffLiImoEZ|-JR*;Le5HJ!fS%IVu4cCS)8d6-7vDU}gsV34YO
z-JwDH-1A2)q!8A>XP+;GDO?$>U-3IwuoXKLVD#pSYFazkH9-%&uHzL)ZS9C#poN1m
z_ltdXIw2TrY|+*u8rvOs+nWKWqRA`b_<kBJB`x9l29d0Y+MWoiOFKI--Ex?EdL{F0
z&BD^^h4=7VG<xYw_ow1KZ!?wut51;OZI+@UoiY6$_Q}Ar!NNAo*(EYme|l}(?!8@q
ze1stR!uQ-T<E6edy<`BfrlyT=?N=+%Ph<kI>%soUc|QWbLAY{?_g7O4E#ri$PbZd(
zbk2S<6n>F9@PU=^1i-jhc}od``%%&a(q(I`cP@?BnjpZyoX*K^3H(8hwcY~CZ&b-|
z0enN5b3Ih11n`{OIKo2jZ}4LK!X{HBciiBuLC<IwJReLR>q}09$WGr3mtpjUkI=}6
zhI{ObHzBo{xU2>~BFa7^9&jsKK>{-<%pxOcmvi;MfrUrN@P^(Ou4bw2=kwqOi-2+W
zit|S@G-E27c7_Fg`#wy~yyjk^wzXBE6?TW0&r|(Ki}{3BXW3#c{aJ5eu7h-J^VUU7
z*Wi_}zY!yI9?+8mSBYmNi0LSQ6d6{y%dkhQhUUF{l4|j)4o?KSYUP)_@U}i^tMH$C
z^f(+Brnq!p?`X&66Z_uQ1K2E(CHFw)3GaKFKLFV8mw}!UIRHo-F0qo@!|dy$OnaGe
zf5D28H3j`0nnM9E+mVoU0a36jojUEKrDIbhL{4+E^!eBp+m9d9<4L>~hLI=F?}T;_
zTH*|g6T$)Z+j_70xe^4DhQYMrbN2qQ1*|1QBLzLHAHiLE)^qdw?Q&(v5jCt?*t9t3
z0@S|vS_G_f|8Vsu2O{>W$gX@{<&Dk18B$jHv)XgJ^{2V)`a`eiNkuH>eu%4T`=Ka(
zr<lXNB#*UHc0m9~_-<!c04CJ`ByK%TyC;Xf9}P!tFH_gYI49QW$GD=LI$t)2O7z0T
z`p;<$h@AOr`Q7^*pvQ!={^<Mdq}}}ok_UYy^bZShkU(Hsi`@h@?!RFW9|x>8^;Muu
ze!_=3Ek0U?H+wltTM;Q5-@kv~ahA{Z-P_U6W-G-KPT3{;u_Js-Em;+jYOnF>>976i
zmA(+khBjEgtY!h|(dENK_&5gI_3A(}8QOkcY=u-AW!qQN<mr`}5-->pTE_o@J488W
z0PO-J;b+I{O8Nkh$r^l{Ir2*@Bwip8mWz@6z)s<}jo7w(d!$HCt=b?mDqxu_$)JRS
z+a$Vx(52;QT<4YfjW5$8n)EKj^I~=jxcHd-b^Pg9ATm+^0Z`#Oeml*SsikEi_Nngf
z?zDlJHO$1_`2**at9Pnq{v5d*&)mFDtfosW?3a?ofEJRK_}$#a%KdY;J^6&_HK7nL
zPZ`+Rc>zz%;o8}KhY~y^_&Ek;stAPF1OKdHwR_*>eJ4l`90P$jlAtziQ5Jl>KYGKw
zxtOy;N2mG{E!CD8*>7PAee>&;>!uLX)eS4A1J=NutgjvDTdPIyM}sohVJAi;|BapM
z%znK}{o<QRX=K2bcYa_o=G+AfaVq>})pb1}yE7b_ywi~u52o{YatGjQkzTuJ4WMwa
zq;Wv3gX;PXT=ZV%)=g&%rj9p0zCXePp{BP!$g+s(=&(n2*i;J9FT0%Dl$UEC<!b5}
z^cFtZR}|ix8^cZSboV*~I;t7>W#FhDfurhnNrVuSUtHSy?r?XBTX^r8d4r8XL8_0^
zc4l{psCb=hqE3gLS2>r=KDWQ7GwS;Utunc9%!=(hk*2KBG>H;hkOV5gCuYP9&Ys)%
z*Gb5+VR(qJtNAGhZdeAC7(xT19ooeRjB?LrI_(4m#tXBV)KsBVr*7CcrS&6@9#4yy
z@Ue8=Su|QYYq`Ij(ovH(n^XsE=NW{3hRf5sHD3P!qq_naK+wJsWaAH6Z>#zUlu|0_
z(qk#8FVs5CUq(5}J?@|@Q)+-r|1J$vAmR068v1(Aow9DxcTB1hHruQ_87_L`N6uMK
z!@^%pDSV4PwWrtQ`XzpfFL5(3?xv@PFbC|~qK`-!0LqV(r0@}43@gw_9E_F>05QPq
zd5da>kJ6y`JU#tpD4d^bsM`!=+3=|R?>sspD}hXWg>u2Z`+NYK3$FYZ>;N5n*+3u=
zNcp-P_YxNV{>_2(zUZ#a+tl1i`nvLAt+J!lp{MMNfb{v+nJ`|@>@Rx?9EfK?n!nc}
zDg3B@hwZE10d44=1kRgyK-hqij_xHcFA*h=q%t$6G`?sWxbYJucKPXm87YFqE?&T&
z9xwe|4*c=sacefW?^dmS<cf!nbtj&{&0M=v$#!>=VLsu9GMCXSu1Si;uw{W~g(g?|
zU34G&g^&Ea9LHQ$LI!u^GE?z*08mXQ55xCz5-X*Jx>DobFe#_=O1R9?I^lF5zt^MN
zf}TZ7sZ|B0>6P!W3yn)MwVpKkL|kO}&o|0LlDICXY`y*Wo<1=E8m2Y%v`aYB%E&Bq
zl3Rc8e?dZpJtv~FW_Ek+?pGi=AGi;d>#;i7Y270rLl>_*|Ef%pz!T@+P&$NKI*E3+
zc2RBO3ionE%a6JH!SSQRImb1H`&;?$MWgRagi%mYi9}Te)JCVp#pjRr7@g;bcq6ni
z$_>z`G)P&>6DToAFQwzXc6YSUw0&W)KnC`fE*D$*(LC$Uhe#_T63R0B=sREM4<EED
zeEI3?R(R_5>*IDWMm?sIAH|dz5kIfGW@7*16~b0MZsuN6jx9mPhcNu@0efInn51L;
z-GY^2a^m20q4)waRW9+SoM?YmE(7QHi|2I>_vG8p3f@=rNx_T1ufFbr%d+ufE$7sJ
z#j}r(2KF8Kb4pL$NqAnu`I|XKt-t37CyB*I)U~d$+eOuQ%npxuQEP_BLs)9b2@^TL
z^>Isz-`D)b3HYre8l+Q5R+N4#(kJ?OhU{nnJ6CQ&{uBT@T9H~LNX+Y9f_NqQ%nIaa
zDMQ0Jyp0TF%T=qv!;9ZX=wa}|z<O_x*xuGZ0^(WAE0Fbd2;Wepnn6u%<h1vxN;65D
z-OFK%Ea=oLl<+M*@84NA-d$e`G6`;j!KSrj-ZEC#JXUv&!y_f3(adGng{iHBG#`9G
zD^yd6I!=*^@fQx(0R1jU?n`%N+u0!w#jNO(-S%CJQAM}wuT|SDDr%hJXvRO?pruc>
zp8QYQ#+TWf#$~MjgObG{t@8^y#*=|QOHj7Is9wVT{q<c>+rwsMw}pZxgg4OI*v&ky
zaGYg?8y5kfUF@sa;y+QBJ?LkCnv}?ZzX1Wcdzgls!o}|(D0v5g*Eyg-R3X$`VeU$T
z0TYp;Y#PaT+s44{rvN+Vmoe1hx-%_L<lzJ|qpqgS3Lj(N(f|R|JdzXYZ=ZB+sSeWX
z=Dj*9hi~TajnyoA*568@<}M8_naA|d5x1>L@o#H%tl~psh3`{$`ckDe{mi3usR7Wx
z#UuU8rX1SUs|@W78+hq3pGMK%^d+Y9CFtXqzJu!YXQ<OWlXjtFT6yJX6cu6QzMis{
z_wd_)p1&OvRHCWq|MeoGw{a48Blu>O@Oeg40jM7B0*uAKtw5T%n?DxM0wZ=bZixaS
zMk)noeer1V<1W1a6TJTY*v9oHJ-3}vPIofylGe{*f?L|}e@`={ztKoK`E+CW<FX(q
zY^EdrEYO4~0?&Zl<NFX48NF)v64{Hw-u0W{T9}6Z8lyi3U1S<4*i{F%iWpf~<Pb6p
z<<~Zdq4_l@DK+R_v9QPUB-mc2Q?6<|4lkj{LT){e<3r)7S*1|h77&~*fv<+A>uxXG
zZFrU%QCVu#-i+eYqn8eq*`UGwNHdiv2jlewZMH7~CjYxMu81~!Bt2oR_$|HEdF!%A
zD}kyi$DDHg>;LoZ7WQ7hn#lJ3)E_=21C(B-h|Ggn4umVhrb}Owoc4}g7(?PxEt+rZ
z3yYA%5sWa3Fu6cLp5ve02a>RPX41~><7lq$nbxaRFrt`UWx&RQl>v6QNU5emBO*?3
ztuL>8_^c6Q@@*!N#0aU3!FHqo`>t<Rw$5$=XWhK=u^~QB*Q62hhpEzGLNeh#Z`EsG
zF_uAwBTUJxA`U%^>O}k-53`>kO(QnsD{FtGP^xX)bL+>Ghn%PNpZ3@o=H7IY*KnIj
z;>#0wFzEEVu?BkOBpc}GT&fO<fJiN2>S0hg(uDF03^iM2qIJ0cE3bjBd)@SgFV^3p
zx%U_G{m?k=qZ)}Ha|x!qCr9w#?PbpgD(s=tH4rPBzvuxM6X0mJZg{8i7nx4|UX)cw
zhgI!QV|(_wuH1U&g4*#96c9y#>>j0oM3^%W7X%1z_q)(nCN^fUIPyH$x5FVm%+e$g
z=0Q`vg_ZLnBKiP%u|lzND)$1E6<TMCfcf5fL}0<I_$*J*3Rj9)PEqS9WV0MEcNSOK
ztSwS>ru)I6-elyKkXJj)6866Ota9$7-J50cd1)_NBA7kQ-*7kCP<U+DXwKreZOvL4
zFH;;A1@PuV3e90DVyj|&2mB7+;kkRy?-2JKRt?(CEiqU<J==Tw>o{)J{smwpJg^gi
zmH5c4*nemi{6QpO4fdx3O9WKZKpJO}J^QUGqc7Y><uJyJQ=r;nv5LH)0zP{7E-745
zo<GO3t`@u2X4l~UyOGUhzXP?|p4P^0Op{|gV7KW~L#5%KO=tr+=+nTHLM(p{wt{dO
zf8S*pF~t$SB7)qKC=>`-{=vfLkJ$vCl&~11!3hw>SH-tJbbk~nv7hVB>;@_?eg3h&
zRH*-N1#kjBjj)qm!n1r0E@R#w0TorR3fsV-lPbAHqtAadzKVls#x}Y(*!WxO2d=UQ
zK@Y;yQfw()b_+fghVmFG!o4KA1<s=kQZPSkObTfG(||BxPeVmTZ6kK_{Qeo0lxt$C
zto-};@AwqSTl?y(<P<(237EG}6BsAZS-Zi<@f%5~qx3}zVW0^gWwzs#oH(N{iKuC~
zmON6{H~%d3njwK2N;4^WmSjc#pAU88xhD(EWEc?(_GT$ukqv%a{${BXQBKaumK8{_
z2U0)YIR!Po5F=~94HgI#F9%;~<sY%ltb593C@K->*5Khv$F>%T@0UIyHhMvQSf`q~
z?9r^j#r|7j%&38D(>+MR!u$E2zGkwiVdB%10@HA&1Z3F4PQZf&9Le>NQFZVY^@}Nr
zoL4+Y9yIi?yN52?IduKZ{*}76_tzg$01NtcrZ>>%16ba8h6B53(WM1@-v;CaPJ_oj
zxY)Ms;88JpI9So3WtB4=pXK@8RK&t?*vwHchMfc%z2k%R0?k)JM2up}g~T=c|GqP$
z58+i(0JLNp=wM<&9jRkrZnZDx>LS$n_FU<J<2fA|jbTpubThIv2kkVQY9<Ys2bCv>
ziRW5mgOeQB>E_o8HgiD2tKq)zs?d$f5uTf!!5YrH_260I*2){7$Qx`oU84=sG}ip0
z^o(~+5`jiQz^soON_K!H702_xooROPh3nl0iv`5TU%!h2HANUu#jifFmuKt5_=^<e
zdu};5oJ5((KViu)GdLOV_gUV5D9zi<z`&)C?5a>R-$^P^@B6O4fe1-y2&Bdrd9)AN
zkCB9KgM||3DMV)=l<?aDbx>=A>avj#?Q@2R(ttq6X3+a0&JNm*V`1CrJ-<)u6&$b8
zA7xb=W0%`*)tSs|H=ljEo<*~QVFS(5&fe<}ib`;#AJPyE7mh<xnOWf3i~6{tVXvFs
z({q-LFj_fP#D_@{Pd0@_gBMo#7QzKD-cuD4JpcpjS&)n0F1LekJ_XzrT5S>lw^iYv
z+d77CSR2ODsXBzr_LJ{yUyPPspA7%D;hA;`R>PlPfIyq0H&?Kwh+m&n3<KU(4-#D{
zst(DjzYC+BA$cf8Jo=PXF`eZBsiRr9&c925B;X|-9m7}jV-pf=-P%xM<-fXL$ryz5
zwd2JsF--VElS@lqaQOHCLWDpTAe?(T;?2%LL?zRbxBccnB~KIlDqMVYy0%QEEN>t;
zR84LzE7joV3HT*GeZSxi!oR#G0?=eZc%Bl3&<XkHLT&}x&qMAVVuB)uP*x3xQAO|c
z1-V1QIeJB?nV8BKN1-okb2=I?gI_?sXaUInRBm_F5xNV8a*rrKI#!dO#mDeS$BF=8
zPvJXxBd_`Rq+qi|wB1HwW#*4p)tVRFa`8;?`Hm7u5yxJXc1ZxoZ4*~_J&iQy-dWNN
zw|lZNc=P&EBCG$r_xS{1nuzoh-2Yt~(RcZD6zc#9&*H@yqCZ$u7^7rp!|r3p1Ti<+
z(d)e+1XG}L-`va9@bS?B{^BgI9SOc3jOP8crQkYKU;~}O!g%fYtW*j%$;op!CmcOz
z$k-c`T$ac9U|malc40d;)fA7*lHn{A#VCOcT!5$}pdk}6Oefz<6OCkN&dkgtE`#nr
z4E*m3FsRYw30`A>5_v`oV&XU0S5~9=`R<f@J+Q7RZ5!Sum}z04tI=*69<BHCKp18q
zcmTaPnAqAGdi`i<Yq=Fg1XR!MBqqx3J0=IP_Izvl)Xj(#`FFd}0Au(~%JXClruxu{
zUJ^Zb4rpZH0HnbqmrwTvt#$^G7Co|FxDLO+q8Wb1meb_qWEv<HuTFn$>*o&_Q?A`=
zuzeZx^}twyn2%r%@kx>EA_&7E@0TJ*JLy4VDk4q^cn|}yVlHuJ*#sM~c4Z2797RIZ
zrx0hS^F3B6VhG;2_iMd&H!^_MT>;*<8<my1%MjAY1<?vM4W2-nv%o9nSl&^H0_DEJ
zTdHqSRMiiAHPI3PHpohha+2eKqJnRZP#kX@3$IeOoz=yw-Vb@+)C9@*M~-~`pz@x)
z=$tezb7oztqzd>5B844haI%D2ze?8`)kRZWEXbt52GleR=y{I<`J1*UfgU=1`+a!$
zow&v(xd0Y{-2L=f{a#Daa5via-**qVUuLizg`VU(8RWx-rlWeLq~G7ql|wRA=5iLq
zBe~7zmdP67%`MgH6fV6xhJG4f`{U~C{SdsX4nB|LXd`fcHSD-)hV<OCb7TGRQ3CEn
zuX%MXFGlkD(YQm!`i4hvil@Y<a@STxP3@@CR)a|MRd}0{ZDSnIL(YNfOy_fMgr*4&
z13#3pID*SKNIn5A0Av@T$mii)=#%+C?4$9TaO@Jevm%I3i3AoM{DKxoK*&!y$2VZ7
z!a<?jFf9<tt`iPXM25X8p*5-jN9iCoOXuTBw3KHo+p97T#xK@JcSPv?%abjgANNUj
z+fH0V&nl{IUYF`)=_+_1XuBqG{h-giJ920BU7_80&BS?nY3tyK>lY&ap5sSdOyRcs
z`P~_y6CTK{;=kYfSby7LdJF=Pg|Dlqk=nBW%+600_51HLeFQFrj`+SK6<}>j$e6eP
zz516?OL7!1YYA%<{0}ldAiPk7oEotEh#1nbgBue+GAubBUx;E)Ykz9WerZG)p`r*U
z{n>1mqrq?pIuNAcKnCQT;RK=E(umXLw`dRT?bV*gsEw5qV#;B;m073lZm)xm3xryw
z!K=^>s_V*9v%FP+`t?bqA;Zk*$1*sYLq(UX_*a+oW?-wy)t}HN6mgUpCDwCkdqU}d
zhVmtyBYW2Ih-pGWFRT>Bp~r}B3J?Vi6^l>QE`ZSYQbC8P2SA;B|KL8cII8iL?4ECJ
zLRTqX#%1ammrOtCPn%AvRB|m*E4RA@{of;q8F@{m3_ro-QdYl)_*uo#hd&K%>P7zZ
zJ{&~;_b#72Ed<5x)s~3@ANPre1ap@Mo<cV!4=0ipZqRpEZvjLnz+S86z$8A+;^SD`
z(EoIGMuq<#>sDmp1szbDTaQ8he%OD`y0laI#z`?}&yfsrVW@dQ{|{mR0Z(<`|AFH<
zoQQ}DDV40UMM9_}%FNy}N@Qg35l$+Vva&@KIkvL0LK;?)>|K!@tIQC7ulM1)zW3Gd
z_kaBF$K$@Q`&yjy`Mlq+`5Zp`%CdQ&LPoN+0EE7g{Xw;AIz!;<NP177o1I0K^754{
zVai*BeTNNWF&P<|W2%>c-7o*xhsYz4R=Z~aP)SNau|>+LMrT{$rH+6NO7dia<pEfJ
zFbiG`2LsouJuNTTfD41vexM8bw4S*PaVdoQJW`0L+*n_Ia)bNg1tP^F_@w+;9M-l3
z9T7dmj)C^jQ#Y*yIZ1*0d^Q(PW+FT2R2?IRO@TD$2tX=&s}aQ?eq|E3K<1589+Su5
z?>iYEmZfvJ+YJv$mHlVmVzp&=b1t{dgQXAFe#S>wdY@M=N!|W+2=#!&u=aWM{C*T$
za7@bvr@r>*8$pwDxnfPD5rQ8g(;7c~tPcCF!C&#;SVQG?WO$mr)I3RI{t3-BB<W51
z`#&IkY_X$=UT7Kv9hpZ{9O?#N%ffs5B76@t(Kg^c{cH)V+Imm=@Sa*I4xGGp3@{42
z_w4!BLwBhtx;#u-#`-0=3mG0F9ymXc#s|Lv$3&)3-Ep<&#|HBW$H9xB?Kgg{_*!p~
z5-A$PG~Z(=Cu7%sPb2d{{Xf6QY`mA(rpN>rhQ1Q@KMHY_>;Lg2n}Jj_jP_dslR>OH
zFL5BhKMHVVvV-k_FK!fVk$$@7s#5^)Om{L<SYoxvGK3O~Oqc*OlfDm{%nu&Vo0fRJ
zz93Q_+`QO6tgR)IL*?RnYY!X9mt2>@z)LuyW|eQLiPG$_h-QeiW23;^ua@tI6Ow1>
z$&BG)pALfi;Pft%EqM5)uaRButaqBLHs+x8!ey!d+y%C95>m7Nv=-C9S0OC0JgyMp
zr^h)Y;s6)(a_M~*ls;KeJfeH5c$ogDI`l;wf4WI_T9A{V#Q&);?6Ro8Yu+Ao`4HHR
zS3i$c@?b^=Y?Gd2<ZkbbfVpAsg4sZR7RCX=#)y~3N<Jg*El>|iLULvx@CqWv#L36u
z;VHLH5rQ8iS^}kwpF>uP{-+74le!Lbgcm@sR04BuE&%~y^#<Jl)%G3V%mUVQ+m$XW
zJcLJD97Y|~4N|R72iA<9(UpMK4%#YI3XTETGa4%SyVNzUzWasE1CeEdz1aUe&IdpQ
zbXa-(!luQ0PjO@6GkJvp&ZZ&4(NCH^(tSrTB|snd+a3=O^gb6v8Rw-EJ;oocTpX>b
zvmeT;2BN3_8ua*q($QVH!xeQ7TOlS8R|AiUN`eQ=9vFBR&C0&krrJ>L<Al&I+p2>Y
zu+ZG?fl1w?nu6|3q%|o&5dH@kyCD@DF|I~bd?=IB(`zh!1Ppz9ZyQpMK>AH=!9kR}
zl&gFez?+hzbAeds{N@frr+1(D#@b9kI+UiJ=Rn5?wgH{X=G8~)k92e3)~J390mHHh
zL_^l{-vsYw2PRJ$g5V;BnqSN-AUze4shxp#<V9!Bu2V=hKa$k8d${Y{cemK?Zi9Gf
z?-yY1ST}xhT5qzqSViSdAE4SQE?hUw?=AC~GRlMkx=#cWCn&A~?Og;fI+m~W7&v#6
zXEM3%;#a?`9@dd-+}kqlUq{t*1C4a8A!dul0IL1{K$nptjA~AaU*K%5)yFOVxAv_B
zf2q52Lgh8j{&}i_d5EFksPYB<BC=N;3D<DebQoDgU!?k>vsLImfP{H+U*&>w3MA!T
z87tGNH}BA0r`kcFd;$hq7wnv2hz%eM9KhZjrfMNV9s^oZhWmP`)O71KOR->qw<`cI
z5$F%$0gV$tl^0o6jeN8Z4Q9w2@G~-vFC2LG7JuDWA^OiO@f#z#@v-zHfq_W~+8Ywb
zMk%(RcWn%8?ut~FpC9027KN%f5^`KUz<{m|C`XT}hX?T@u)r)#)SCbkk~_LGdmhU{
zNVx|+VDH-T`E>r`7T_USB9I*tYk~XZC&1n056ltdWRcrKysfMbpc-k-%TNysC)TW(
zT($ghUx0<|td(8-#NtjHGqr?Ho=KmlR7NUjSaS0JlBHuVZ)++*gEDn`S<l>C?mN8r
zU+hFP3FKl`^khr!(TgJ`&s{^^Um4aY(%!Sy%oIL_XZjKpAj;0&#|l6!UV{Tps6H;n
zX6}K8!Kcp_?IM-n{pb!XF?5y&0QQB!y^0^9N9aQ_kZq$7Sl5IP1bJAcy@XZ7L0^Cm
zI*v~<3JMBM0B+`jt%6~RHqD{y^P0;o0qYhfr7ok&HiMOA?x*7%{p*2e6U_!Z8!~r}
zVKC~2$4-F9eh<N^0dCO<EX!fdl%N$kghC(hDa_c*Z_ql8>xPrto@=ZRq`<4CmB2GW
z80bcU(*Db({XMEa?@V5y#RsA-0+;^)e9ZfW7Yg}zxdqC7r*1z-mu13RbQy3v0N9Rh
zPJH=DpVA+MUy&kvTb3>#lJb4?=FQ9gqjSn)R!BwzxENNZTzOgvFRnEsJW!1}K1!F;
zaypiBXI@7&vG*iD_w_!n#igd|gtHcoH(1YFJd(D7fRlU->wf@H`Hr?%?oylrV@8vI
zE&z%f^fsh0Zu1QnKng>zRd6PC@EK^QY8hjaGlQ=af<1(9CBrR96}Kt^#+x4+=*)Ob
z-UZ<sfVB&9D4ij-P-mef4AwX(=~`qgV`4f`!^+HjHFWzYShwQX0x<}+Pfoc$^XeF5
z%5}g1^QZ3uF_m|r2klIN=}K5tI(PvIX<tqDephbW9CsY)z)koG4}nsV(^a?@jo%o9
zdd`9Q0ZhSH4j8||N*aWoU`+mrXv0u)jgl%2s6uTvzT`N}^Fb%t50pToZFtMW3i-U<
z<^Y?Ug-Tr!LSxa<Q-q(CS2JZp8w*UH^%i^2SqYqwU5k(c9l}NRwFqr8242?Ji;q_<
ze%K4&u)jW7Sq<FcKfts@uzwr-Q_zGE0<!G0XnB9Up3R54N(;p3QQ#W+crfnh@-$GA
zv(;rJ<*wIi9<qHDhn`L$HDG6&3gV4N!;%cE330_AI+{izo~!E}x0~DP$Sg~D`{9Hh
zw7KP<%K0U<Zm=5dTvN(3QFparXymgSN<c9!N;#c;ZfGI(nrz>H@&z78!ZD_-C`fL;
zR|>gdR6_&LqgVGjDAgjRj>(wgd-g-LyZKJy6YPVi5N?$Gn<5X@%@;5TNtN~~0+C%(
zo`yRXthv+S@ags67KrL|qmRlK-Do#@Ye>(2Ko{wZ><B9FsA4I3Xp~=4s0wu0C?Mh+
zTTPjYZY&OE>kA>L)~S*moy`dsGA!TpPnGKqrP%s=mw|X#IhnRFhJ{q%Tn=KD;NAQt
z!e@seyte^CWO?;$7DLCfiNe{7MSiorQ~t4g1hLT60ja4s3=7Gd<}uwZMRoX)5%}4H
zc%u}#Le5Mb-r72uZc?8PACz<$;daC$WtvlVLLm8s{wfs~(&O1nQPPeU-$Nujl2I29
zkZUhfDq}R=J%R)==vj^>+zkvx-RT{bPn%EEiw&USl#18BYNa6GQ4l7K?JGd0f>5F}
zn|jH@D%9S71y)Aq2CiC2f!MY#lrPszPS|zwBgzec{Xf8@9BqXmmr^L)`Vx1^sBm28
ze0Uv9vD#8>@c<5^-mKtp;pFePN>zrV;mOH@A$o^n4EYvSjHqM?*JWMoc^(OIe|^`a
z84N&%wA~3!Iz$dxx_4m*;eQejXma4UrX>GY3*W3jLME$JpU<Gjf;WftB4DW2h@ADZ
zXN-}g(i|G|`&pa~p%*rSy6oxOOj4}fM&=7oYf>g|&U6^)=cT@vtCqM5U;x8ru7(Gx
z3>=V=uGn9Moe5InjRIwH!&vAs8Xf#W@T@K0?97BupEz=pfY@fpEauQ}@M?v~hwd|c
z0M+YojfK$}n0j_UQS~$V=X1)Wg$%r>keXX6U8A|XJR4h#H%j=a<GUCJ_>=*JCHJh#
z=VTNcrHuOn0f2>AR0$M==vf3+i$0W6>4uRy{L%Eu2h;TbH?#%9s|2-8f<-LPmJZDT
zq>$F6?pA6TuSF#`!hr{bufk5!=#(4sehozsZV%}I0nCLh_|BpAKsV6KyFCl043ug7
z)aYY2b2<SyN5BhdP;^=$RX`-i2dJyw7ibI!Z2X*we}o|s)ng^6aw4xHC!~4r$j`NR
z+9EE4bas81wP%-#F|%4-)Of*|SvaldH<#w-qKcKI1`x+akLlZWv3331+M6MLa?M9y
zED=~DDS>mpX+8@d46N@z{b~#ohMyn6Zxq$gt$$XTKPJj?&u4SrntK~p{k-VE3{>F=
zbRiP%+<6AIVl4*bexeb`3pyIK?A+evkW55&dLV?+E_hO#2}YEr;%YJ~-=O_4k|bhI
zE?jv6`P$1Qv{VW@Z8?N(g6^@E$ghzS_;S_voBcuCRfq-1LURNeYDjrHOnD)btn_1H
zOTXMM&Vvfll!khAW1@Ke;+0I0i}vad#yMKc8zWLZ5p%))4?M{J>~B0`V!BAy2H}E2
z-~U;z+7)rM{z$6=?I;vQ@!0;g58n+<tUzJ$HqIQW=-uZ?-#qcd*^s4PH9miQ4vL8o
z3NT!T>Hs7+qd;V<A<qsNf6NO@qR{5HOY*7^f_i^|+~tQCRhDB=Oet3Tukja>tk1;k
zVo+7^stKxs3m~a=zJsRzryvxVuJ<#g{%m1Ez64Buk|V_$axAP1*hOQe9v-k6-`=^u
zJ8Qp=0Gp8Kw<f$6sLd9<*ArIZs6QgH60v*4X`}Tp;4y#rPp|s8wP#1E-*!886htal
zjVM<CyVYgjLp&!2DK<A&JQX7-Au6e<#lY0i7PKC3gL#%R5`dBsP`E{Qvo6iSX9jj=
z9@^zZ^}D;duE0K`m-ahSIDu9S;Z$-aI!+QBWZ~kbc>FS0oXiU#mjY#G7-T9y8^w*J
zE(s#`88G)LTb-^Lg%x`^FuIgf?_aGYWL@lVSjbxU=yz!nn1xsn!X+Vliw6{gaY#`L
z+Twz2{aZT;A;G%WfofG+5%{shetIzb6v`h!hAYc4tuoU2q3~S^dw{SGWsGXJ(8-f0
z=eX&{)($dv#K3+_FVA$X+6n1q+pU*GXFBUlHA?fS7tbVl4#oos*B^#Oz0S`Oy%wfZ
zCPTXQ3a|t!-pHSy?9qmet;K+Pm^j`xgwM6ZfpODAL&`IcY3JqfkW*GL#Sntw<tM=`
zo%e3vl}rekq0!Ktpm{_wjW{S83*P}NiF!cur-ATs%v(V-5SIK6jng?3Dj$$#&!r`9
z2q9AssHDqt!lEI(;Dn(*-G_o$4J!g?TCiG^W<j_S1fVaoI7~?1GeVC<uEDH`nlL8}
zYECggtL_#PGJAF3t|1tV94b&Z;6A{YU&#(j^%w4gv*&L`=o2(*r~lO5Vc@v(`{$RK
z3pZAj!{s!VGg<M0TMC01k;=U!0to#$py~bvRNj<6_Z;(F#If7{$WR#04FVVLSi;kt
zs4t>JeUZtG(bE?LoH9mHkqC?lG)>S4_@6*|GnXXdCLjS@Z#+K+Ve>!|_(&60zvKXu
zfRPT(Q75-Ol~W$6KL<b2ezENVI)PxA^z5CSi)C##;wK#k`C-B4TyBcJ1f89n+T+tx
zZU_1pQOSyYnGw3b?=9Vsl<g7>pNc~Ibqa2Pu(x=2MIx}H1z^IuSDl=Re{>D`>H*Qd
zB4t<8sjK<O@d_j-P+3)YA*bo!LQA5vA+V_lL#fx3bn0(WzpIzde6GX1?_=EYIzd*B
zIgc3cg{aIfQ{M}*DS^1sxADsxe_A%Izi^Xp-$~Cn`_)!U_86+i!34`tysyGDAE+;e
z{(pWcOo%8rf(Q%`3>#C6Y&#l1*;ZJB%<5}4()=Q6su{5}p-wNPiO`sEtXj}R%3D*<
zDp6IW3PWvqZCc6)vTVoTFi_k|Qr;{#3g9q+i0*?hD7t=Ay@yVoUIfD{Z+MIQDxIkQ
zK@I#5Safv0lAw0!2J1o4+fWBjFN*=Ss(4VemtO3)@MyXrR_Ig5!y*<C?9ef!TNTLZ
z`6adD{|YETK+~yjRy%+8cDw&L3`)kqtTNZ%wnR1;uD<A+P`aoZ<-LDzS_~XZPOpPm
zT=lY)I=k*c>L5>}?@M!fwa@$cT&`B8&e=P~kNy(&{Z4`@y7GM54_kW`MA9%4t_9h%
z0lj@#e};dw08$;w2VXirMP?UMC02tqa5{KF)uPL+#icgR{tS$)xPes~2_m;|J)SSF
z!;}}5^&pO@9OFLCgp3B!ZqXaIl_m~E!T>S}gv;q;GGpR7zy&?r9^KLk9sz+1P-ErZ
zVv#6kYZ+K4-lzhJRb85s8n-`kE3C~~T|{_N<U3dRbwE&G{Ev`PQN2S2EGQcediEay
zh4<hA7f(^*1|J9=gjIZ~G|E>nW{nR3sTrf+x2cDJx`C8nIMRNBkC%Z>C0H2BC^bD)
zI=BZ($$Hd2%bq$|+@Tv@2OLfzUa#|O0itvB?+!TD!zA&s_whNnM}anIX+(l;Q6Q1a
zluz2nH}6*hBqZQ(z*!j1XIlmO^f4+z_*Hj@w*$+8fc!!@%XF__Uwg~;FmBI%9n-OV
zv`;BXy%*vf!B?hPIM*~l>wNlTdoGV>$FXA#EL~qjAqD!Q(h1)F+Pg=avt)Lh`Er<o
zWp74+67j)&0deZ4zixWthiRb^nrt<xANPHBe})<F!JJ?AnrZ#*(}|(4d;xe+e`)3E
ztK!?H<HRT5phc$4`4ZvMJalGWL8<DHX#gebL5c;wvDhP1O$hJ)w1fY@(-iujXAyFw
z>IWtf<W;C>kXHhe;Kd)#yI}FhM%-@W`Pk$T^S+z32M6d)ky>4eePUlHfh`2WY>$1$
zYg#0oi4xC%OJ6)F6inToZDvmIahQWSM?F%<tEiu@dg_r~LdalR7J=ua;=eL}6ft){
z4|ME#m>}*bIOI3?;IqB7$tS?n#Eu!Vq6i`BG#rTT@aOMWdeqjqPg^zbs~G)fgeXwh
zOrYPnFay0JU5n(giXT5}*8H>3X<!CC2E;A<@$VZiJP(^|?X@9)Z)RwY$2QnpYM_dY
z;Fr~TBCn~$?nQO%j!SUhBMPPS%{SQZ-Jm!l93GYSWVcQvCs{;uBqxVtfbn&4SJ6Tl
z=Zvf6zUyn!o}W$NKMUE<y66@>YN`R45DpX(W$0wiVUwt=wB8Sc=v5lb4ot;AXJ?n<
z=!iq|LEC9TeNi;CaRML_t$1}U#<K_^VqOqfz45W!`}2AWJ1zG!FA7zm8t<BU6L~Bt
z<XjWPu~8;^I#DDjumw}G71F>Y=T>0(;r!%XgkLN+Ny+{zHUd9gjhc3JErP3FEaxa2
zy$l%WSa5{P&4=z-B%uq;FX2OC^P3J7dNhU|2sM7mIGDYzQ?a|P+(CAR<?@veXV!o1
zf)Soex6uK<pk6TdmStcbj@$L1e;UvFwQYc1#%OOe<pCyY+B>;heb_%8nHF@ZwU@H)
zMMASxvgbF@z0go*_oR6Zq*p~4eDF`CinH|>9w+wuLI0=%PNYo7E`gAl5z_*@QR><0
zrFSH!gP=N@3?RnU1F0hohO)@`9VtE?tAM(bq=p-BB02zU^)wg>b)1T(x320dvO#8v
z;e=rZC<kI;3futeXZNN@$1QGyMVpdmk}|8f!z1a1cI|Q!Dg^2IAxt@J4aG8Gs&2CY
z4qob6_EDMC9;q0`9~bZ?EymU02dv^-I|CZwM!$&J7hK0ay7%dVlD9l|dhrp6=cmDf
z%P2sq6Rw&fh>|@b4h652x5pi~0C&F)Kdy0D`5@;#B$=a{a6)M{A!@}|r=WHx>uSfU
zSe76@F8iC}!6C?^dv(L|tJjStEUJ8n>0;NqHFW5AG3HzX{H+>COC=JHff<z#w^Zx`
zF^jqW5(uxlj3s$C0wFo6@01nQ*WvJCfUPwP)`&gn@x8Nh#JqCb7}>W6JzE~bufQ>Q
zSS8pUPX%6-^jxzL&rB$QJj07B6D0HN%fzHy0qSUi$d?9MfSWJMj`aO0SdXiem{ON(
z)Vp;uCsN_I^z7rS7j948yuonCdFxc&KOA?0q$|kgg_j9#X(I3dDDrtot>d)_f$8jM
zv-OTwQRi&!E{d_VJ3JELO4JUtI9=pl=GfZevQREledD3f)ykb0$}YM2e!7<WrGW8&
zYZmyFd4Yz9S&4cv^57Q}9k_oa|CQ0<$rIqA<_(@1K?BVvmKV0)tk6@V9R!QF^27(!
z!zmaF5QX_9mW|d|PzP`kxx~6M!RTogmb=Im(&S;%9AJs|y6@7%dO9V6s`nFGAI}S}
z&8yj^)U3aNm_M;ySwoh(oyIL$8^-2E=DU)|ZF`LSbXtRuqy&v$O6dS_I^}5#U}-j^
zovT-S7^;X!UK6h0c4oqBPa2KEd4s_G%p?OwCQTq~c=2K7$80wGtb=*TNUo@6-{87O
z_BV5&mPUq$vq2<q88~ra?tI7F$7R%aQbKhtXX12*5x>;teIUn$8F5&38nxbr4ypov
z7pu;hM-tAj1x&C#-Jp#J&N3D0QcUwiB1`=j2zo|N5eKC_Bu=`n?|xIOR13_116)nt
z5Q{=b5bf*H?ocy$kfCrkCgvH%NKI%#1~^+z_T)!gW5lzvXMT7EjVeC|u5AQL!M-O`
zZZuGy%Dk$}yUhnUTjT?r?8`Ch9;&7Z4j0*iE3&^wj3Iwh_M`^2%X&3AOWj~iT^P(s
zOBxVUoiqQGRXsA0@6i-(Cw{i4U!*Tesgd{Fx?zvVf>_p*ownl2F)kl?YmOf_YUu_a
z#H87>E9URpA_Y`dTFtvR5v>0=?EyT;VOF19Vc2xkoWua9JqCgUb|4wxuWF`VUv8z8
ziZ*TcA%A_)y$Q4{n6|OC#=eCK88@J`oUi4rW!ZyIApiL1)mO2|4bFQsTED+U09JNr
zM=`&C@^{t}s!0%D4DNkXuJrrY&`u)3PUK6g5IN@w{a7YOhY-zi{(d;3=*?xugIshy
zbi(JbSwF>0062w&<cRy`9s%Oi*ruHjarH$Dj;?kwgjYHj&w3k{SV(Irdkk#Iiag4=
zz{Z|X{`=H3P^nDEI@dpvEOemB*2!cC#7DbH#;ZpQP-lhsENeNgz+^HV0BM<epbI0*
ztKS*GovE+ig`?vQtVgoCPzqhp6_jPwiU^0;%gx-nI<u{dcTat-Zjd*{BoE|}<Td4N
zQn=3HfE+S0A8lx-3@+!3I~UFZP><?jg$K!zg6Q1}05{8}=>a3$Jhb}O75Xg+I!?2o
zzG>Z(2l6<Xs~7oqyDt>)!btvZyL1rK2`d+wli=L??RRPnso~IMdg(S=UO6`^xa=|Y
z{8$&KNJUsnxZ3wB1LL;z3N8_;6XT})+8x?LQ;lx#aB2=dZwD#SnP+T0Ss<j5+>?=-
z_U7M-<qt3#YEy^qJtQNe;1?2h{jQ3HwOXe%%s#V7IGMNh#aohBk$uT>0O`qxAdI2e
zu2O<mMdJ@7iwim%{V%eIv9^I_)#I?ge}WPB@{a~lb2+U7^IDxm-)v7PK1LDb14NJ0
zd*6v20XrrTkNldD_5nmEKF@-DM@9N$fIPsMmD|>-a$>%oBka!pn}_S&kgpdMmk1{z
zP(e*E>^LZ0dZKtZ_d2}vr^7|tU49HzS0gvD(62>9u#{;GBqQ%{V}EY0V^SF}ZrzH5
z*uf4?l%|Q|$p3Ai!ZT~Z_5h=>JeK*_NfYpuhT#ob4?7s@{G?=bl6P#&&gKXY@>wnf
zd~vTzYH}Z3bnTm#aIbG{>$;=&gNt}4@HyrY<^6?px2@IzysK*;(@(5z{Sj1En6My0
z!v(vKuH@Whn%fTB4?iWB?AI0x`%|I9B6=eF6q~v6W`qy2-zB+!lO`gFzaC%`{z-d1
z9R9ddKO*MX)wYx@EzSTYmG%yTvIt?|m|>$>_O|za)-o0uIJ#xmcTKH%mTA`*hJ*rx
zx!>yjPd|#LIpFkuTEeY848^4U8o)xN!{e6~*SId50m}(l$Ez{4^V@G`m&+i3Q0Uhw
zFMdGId7%B(Kg-twDnMoF%SP5<eFz%(Ljp}m;tj+c9c*l6^9KRk^1|xF{s!=5$@=57
z1NeBckn+oNwXSEbw7UZ+D+1Q{{bEXqApKhfY;rE$^3t0~Y4a8MMwWIEX;e^GVVq#p
zYWG-qSBY8sNn2$*wqS&>4?85YuH7B5VYV;>-4CbZV5P(Od7*u}%rdSEjQsA`#z0m3
zYN1XI10$i%H33pTAPC-yN_MzFpCpf!aZW-CTL^4v7wKx6pT`>>bn4Y10KsUI^GhGS
zA>nne0LnByhT;96cxBGXFbJy!&ILM`L~SBNacmf?og?95c6ym66(#LZGt*!aVGM_q
zaM(T^QQu2}?2&xK#Hn>J#lrK(8DOoN<BlLq4e+tg(KJh6enX|Cs7N(&x>>7xX(F_-
zNz`MXUXt&0J>Q1^?~`?#LfpqmciPEq6BIoxH(U|5K_9RhV&XF1NPD<!vW!?JSlTe4
z%dx7J(8CG2z2s2p>GL#jo4K#W;^;i}>bc0JVq$XQKL>BG{rkO>(V#N1hL(*3vXkC0
zjLwa=1p>&Fvi~-837X_D!OWiV=Siz3f;ihn+SVaxKd->LXqZT>YR%nVkd2?jtD&jG
zC=4k)nofJqUIDCOB6?TZrTgorCHI}~xGlH?S%>aCc&eT2T#@;q<l3X~YmOqujPWat
zSL~%AM_dN1>;6W~cZjnHuc$12qF5-(-+6CBCyoxyM|1#dq4{X86i$Z-R2)<3-dCU0
zik|N{*CO3&p`3K4eR&v0Ob)r~Ftr1$dqw)V!2DF7K4R&V*ur#2#AQUW^u+rcrR7=X
z-0-{dI1=Y#+33`$tssBhz+B`oD)>wg!CYb*mZrDTxVPDRu_=;Ot@_I{F_cUqVJD1P
ziEW3n$d!8f5*HpSQYyr2P#_}$<ihLv846-$&{$4t@<`D9_TCiSsgug1O5UUd8qPdt
zs6VoO*ZBm}xfwPaZtOT9;=kx%SU?ooh__F8S~qYjO`+-XNqZCX50^E3&DBO;grTc@
z-DUwtH~!5bBAI7w)q{4o9J%>lR9bur3`mECf7~trT`vPIw(i(SGoaP7oUjh(>LU`0
zO3A*|QpjsqJ~$*})L!G-H4fo(7x9*uZY*FyUosu3yG`rryW`_?5!QuHKDp~3W?-`t
zI$&MxL&mLR9t?~1_X3Zrv5cH>wc7Eom!|rTp1Zf#x$k1W1AaTB6W@<&8XMXXC&1Qb
zf+HBh-sF2}3Vm9d88QQPugN7l!E!7`gtd-In{D2T2G+qT2}3)Q!mJ~oWZW)!9>q`7
z)6wVYDIG=9!0TF<;^p}wWpzz;<i##+4mIi~ijMvcu3$-(oJX`<nzx2Py$zD@fVnu=
znau+A_6{$wx+F{NknxsMJ5TBIaw)V@<oB1pNkxT3df^9~kH7XrFcl+d>mmHG=1E6@
z4W1Q<l$V~_o+}cB)nIt6PjuLi(;rGiw-+boDiFdxVOZx<cZeUJFT7acFIx2AX{9dy
z;xCc;2obGkiY3RoZVvnx82FsaypuBcc?V4q6TX}2M}JwNAIL-cGobGL%%7h@&>nqs
zG!?ztdVv0ra!nv&=Rqe^H>l$n)LN%GgTS|t_q02fp%HZfSmvqEFrDE+Ew`)Thx-Sm
z^uQ~5IOe<K7pP+ajKB>I+&yDGcUI_6S_K4Hr(6Yup{UU@OcZyFfnExES(=1+BnrXY
z6Y+az0o11gD}?^eTduWt%F{Y7Vg;9Bm)oDfVhygNLw?#cH@}qsLXJMD4cUx_AEOV6
z!XEb_T0^A0Kz7Ck3W0X=q;j?97}-bi^9@>po9_U>LtZpIXk{i47!gg-bMXfpJ}GSP
zLYlD~D6z$0B>gHD36(27bamw!n#Gl~*REnD_ib0LrAmWY>!Aa9Rtf3NLEcYxr<H!)
z_=P&M_>DP+yKe*OBnvIW4LNLmKHYtAWqa}a^JhLcZa}TF(ZU`lnZ4bGV0<KY=MMv;
z7glPARb3nYqN>twM~Q8x+~;G!1LbFT0%VBy+CrCQcfqA{@=o31+u)GU2w)tp)29d7
zb7xI2K<Sc7;X=^P2>k=BfkDr`U-vIOSyU0Kin5(>YF|ZaFE5=gmA-&cfe`33M*^(p
z)Mtc$Kx(mq*(?hfU---YZ5TsEWo6{O_tG*+^lB*Jv)*mJCsue*-iS>NGyevd@tK6E
zi#>)U;=MASVF|J7_L|fVPq;7+5c?qRM8icD?mp6?sSHC2gLCi#vgh|M$k#}~3wYO~
zC9?GbsO0%Tib-7@7N#~JxXRYS2^1%k*%_FrJlQB2ul#{p5{au$v4rfeHjlZBqq7>{
zLB!*JfbHx1^K8Mbt5E`^S8!Q=cxH|_rA{_Jc6aNSucH&k+lU({Ts!aVj}$e27iU{&
zXTj1womK0{PX#%qvq_I;B>eO=(|AW#T))|w<3-yG|LPoZ38WbHf%Ykw8x=gN0W?&R
z>;wwyNKL$b+aH8v1*%ROY%gQ%!b+(#+<%ro?MK3TFu`UGxwNbsum=P<xPa!SsB0=g
zybcRNB~{hyLc)@gJpwgB)I_I0znx1@{lJ<zH4R%Hfve*JnAy@|=q*(j&iY97j+H-j
z!n$Z!9+GsCNAy=DUNO;i{mkJUAgafxW9x)3Q(uK=erO6u2VRR|1PlG73hAQ@BIB?g
zkbY>0AlLjyprhw1zpKn{PJpgW&YNdTU7{NEiE1R}#<&{LMWN>yDfPcXT7lzgtjQK~
zoS?YXs`PS30TR;Kg?88H@Y0)Lp7N;TFv!zEJ^8+jby5||WlN9?@3kTtdBF<ia{FAg
z`2@Td^SAY3fy)4c4FliU?^1dM2J~J7m9XotoOECP{uSPA;%p%JMiCbbG3+7H(j}ot
z;eI|-Q11hAt6krB!?Y;(OwQi=R9v1HOK<UgX!@9pEPDRx=h*l$R@a^O#k|v`f}d`F
z7GFeX>UjdS>MyP|?0)}bu@CfrHJ9|{l5D&31VA3;^MEw#0t{>lb+U9Ca{U(4&ny%_
zg@xADa_W90U^w<7VR<-cR~qR2I@sX(QN05c#vd^YAk56rq1VNOLYB}YJS*$ybWb6W
zzb6Eip|3*Kb?AsrHSuj7^m{62&z{Y4|DsH#u=Q%7&g{4$27&Wv$MI*|&!DYXd$l$&
z-EcLwD2eC3CkSD_Fhdge*H$Lk2H@EC;jJlL_DQ;=?L2&1I?gr{W;3Yws-r-HfyhnJ
zcp0FxoylBf+h^3EtV96;WIDe0x1e<qng|FgxLP$NJQZ2o(4vf_(@)y3-@+GkYJe&4
z>oK6sx^2ChQ%UExOx#~ld^CN@ZT81agXV|RdgP4I&)>Z7sV96&_<5&h1h3u7+OMUs
zz4!nC^0}GN>v7H5U7u?DwqGPgjz_8f(dnYwFaEwk-5|U)EPRbN4KaCZu3&5qR!&=o
zuDpicvU<j#-5*4Q3=tcCpzU=reW;I+X{7<kReje>2ZC4QV3qGc+jRAYKYY(n<6bz?
zBQRO<RGBMv`ub)EC>E+`T|;X9Ftn~n4}N+*54)eCzE<-Uwcl~*?lgt=I`ox%h?rbl
z1#{*`Sa0^X-8WEFK@sXUjt;f{fWb33<U9FXr;@l&+l-b^D@`%Wof8EgD<&wNo2%po
z=<nR$Hioa4C}iI3gDQ8(p9+^7D0fG#o(oM8G1mbq01`lI2N~l<YuIf0doqGN6|aOB
zsAeJs)m}$H=bpnd?7HFuV0w8Ob!Szy6=PEEd(Qe{=xz2!WS7<Si5%(+wj&Sm2UQOI
zBhY}VFZ|O@bq72>_M=%Px0#?6XJPE&NLS`5mew!HS>7W<CXF|bU0M48&G6dkG;A#-
z7|OBH`hjc1EiP;CF1&v-?*IAy0DD*|FSq@~(S$q)oy`jHpNYsFU6g0;cdVxj!2Jmz
z-5;td5V=K)_~`xs=sO<9V#Y=<nO<woAPhR9@0Rj*fc_;cg}=vig8ZpBSd9*}A+9<=
ztK8uQ6(ESF=F{88C7|Wl2(N0Ot^F*7Gb2I-T%FZ2=Wn-3<TK!BmH}vr!*KNiWF1)T
zR;06((JP~O6oaE{?C2AH0Oz3*+A=Pfz?`4eJ_*;HUXLt~p$<7U8E8RY!&UJ>ne+;}
z71<*3tnEW<s44*<8D_UMVCnwAz7U8Kr@^tw1=kneP>CTPx}0lNjDYUnb2-Ieml?NL
z1D^wI3|6f^QOm|O80k@XXjB8-?lY}HI{+W!Vb$-7>`o&=4qP(!siYiRur;yJk>WKt
zulQbrDY^QF7BvcQxSceQs@*PF2}$vrrBk>|B<}97Rnw066K~m;A~o5bmeTum3R{pu
z?cC50DliD-szO~z6{mtGn(M-@Y{8haua*W~FI<eZI$QU@|GVqC4Mj(7>X8TO8t6Jv
zY+k3=h=&e@oX~SqrO^Do02z~Nvn8j#V>Q?+pCPaqAfZwWZdjYiA(?4o|Eg`Y$gV6S
z*{4y!4AzOGK+JyPPoAX1lEvSpg!M20uDf02$^5QYkjC!0&{?+8j5xupy$Le09M8wG
zlQ6?()YWf6^AjYb(sD&EHYiiY=);G-^<TiKv{z*Y$~m>sz6|xlAY|&7<+d9|u0m?3
z2iBd*sisu&(2n9pa(WOj6Yh==CSvDE%Jzb+wdF)0(QqQ`BtTWu*T7KpZD;nzVg+C$
zN5L3Pm2q#^<qN6ha8fz&D9%Fibt88TZfl0>Qw2yi$YtqCvH)SY34BJMA`luRW5%ly
zfH`Vg@08s0M554o!EAP=Iq7ysip>K*krWG?t!}ltWzuo*{zuA-JN_|+4#dS#2oeJ6
znsznCLRo?JrEpV>RlBkt;?8MtJ1TzPqkZj%f|3q>yYM}3*VHy)2*E4SE={6SDoSU+
z&!>vr|9b$nbI<UoU!65_1E6nZRGb)gS4af|$F^6;MhrauWYv)S1g#vLkX!4>F6FfY
zF%WbEtKY5zOR%@eIhh5h&E6~^6=FOK<KQ%yD8i893v9r`VP280+@&G`SYd)tV-nZ_
z*t^4gVFaoFsc6&qOsS}_1tS1E5Td`8Jf>#ky)%dlC{;Zo$jQxZix*@~%&wl1@|X+<
zxX7tU_QeC%==wdw(NQk}Xsi83G)%!yMlV2yONSSq*`|bwgyv&VRIvEoI(3BJ5DL18
z1uI9=yQK1>mKO=HgPkZhY(y4vZuHH<<t+|;0fE|6<!Y@dW?aj(UB8w!86v;0*!@l@
z2`sVg+s5Dk-=(I11#x36e^BPPlb8!wG|jfD2;Itlc_O;dT34AZ*l;_P&GC?Q$nmUg
z?Mm0~2Oi!UFg*^n5ZgRVwbU0Y{Y<{+N${gY_VH1vKQ7pLBZt#6@&&fZJmk+#+tPF5
zZFkrxB)$jLxHQ<7;OJPsIz*V}hN$%UU`=cr_&0yf=C^@Q7mrd{tLpi7EW9_W44e^V
zz7){~d~!9(zk~7oxMG`sN7v1E8v8HF^Tgc6)b&t84ON)TVqBL=R_qiavOv-1(zJ(}
zIRks(7>tUIS3p`*fJ-?l9E3eyrpNN>s{sXs2ej9nRJx4!L1th4`h21sO@(3qln$eV
zyoTON)77@!^_PJ9ghU%l*O2>fJIxGHMg2>%PGURFgS^ks!W3J)5xejK))+=QHDSm?
z%^e9PF-g=71?=&P`DQQK3N2LsH9Tun@QRN<EElyuadL9jLmew`2rFaAd(1zVme`EU
zc8FFSSdgnxgBjptdtwp#>bRPFm<Z|-jch#u&059l7tj_SW*1Gc#_+SWNP0kF8nEU*
z4ui@EM)KP5Ppe^gC4Qr7@q$)YK5^am<+|DjfcP&Rs&)-#cyuIOj^cR_BW63S`%oLT
zljwAB6VQgky2X`hVcjG+Gg?fF`4u;!^2@~YqlaEBsH?nXX-{U7DJgz}h746!a7MAN
zS0gW(G%1e1Hk7m5Y5ZSs6qP)140q9Sw0U_TIE=h3TW(O7Np{Yw&_$3!F6XwH<Fy7e
zv^|`F0nh9l=gKjwHubfsfT3t)R8(5*xq0GLg7elCri*jz5AbR@koL1{&v9@7Amv4i
zPgDA%`hDJN2!{ffa(w$S4R6Hg1vE)G^rxR)wUI$JP(e=XmMgc~nHgauVa}`j_1m|%
zt_2g29n+v!DtC9{G==Fo7<dq{IlqC*DFBm3_nfvhU7`LV@H}@|)sj7}0|X<b4izl!
z=gt*j>sB_jeUL5;iEvlSE@L5=JXQnio(7=AJ#2h|@&FoLCPe#w&va6R8>_toqHor8
ziU^hbIf%YG-90>;2VD)Xh;@krBTAsCca)j6u(+FLL)C&X%-(sId){Gq_vI1iAZm2K
zj6MPSGdaA&EMVv{hwMK`5ULYCsSW;P#|*4VMj#c*ifPLPX!QXGO^`5kzp>J(!)*3l
zPSSZsp%1p-A6`8bbu!0~8pQuM{OD_AgT`F1n0eQgh2rSGAQN}ecEODTX+_NioMwN#
z;iaLLRBVb1KIu-QIOM_QZe+E=xxKirI?N`uV%z)53WXGv3zT`P-kOrF{qIKYG$cz2
z!%{oq+YXfO*|Y*h!HV=#>8V1t35^BciZ}~@iVwuLo$?xRR{*Y-7U+o$2+_g|R2}r4
zkpW~$C=-2+yVQbKJ<R(o8zQ;T=ng5Hml$Dby#NqmQg~`?fjD(>d5C3!u^PG}J6S3u
zz6M`88paS>$l9hQr*{@u!17i5GQ=~d94n?y_EZN1WOZcO^WM4D`?wb@80V4GmVNji
zsD<;fK0o^vX_NJ9X&_vVl_?u{l8y@DEY03yYyoQvv3=`E92)6cv}fz9(?q7Mm;;r=
z$SD{;qn$VGkOjC@)@Quqs0%&;!3MDc`e<(QTEB(GQaUU#Twupgx<)W(;I+OvmWJ5@
zMmC>C4GRZfUH8p$<Cuulb1mHN@)CwKCt#fbJ!^DhA?R*AdG+HhVz|c{Li;YdMCUHY
zCITUW{adr?<}>}tO%Dw@gn$@`Hj1zZt<;r6>N&*qecF>d039aV?p#m$2^Vl!M*Wsl
z)`!&4a(ely5tr|BybPkAOj*ykk>E=+tpT2qh{B2DUGlke_TIJm3L_s{JXuNt7Vstu
zd1s(l*az|ssN53HYvEzpe68a6CjI=TuIfKrk2jn&)c#z5Yo8R)%)7T*cW^R$<Lu-3
ziG!i##|bj5_o6Owk%ca|`tbGQSw-Fqu&WpG2<A>O?aU>+b_><-q#}N^l841K)Wi}L
zQm7SU=M3+xok;Egd+Dqep|Q+AKsrZQwYwo9pD{FqZS+8#o8?zd_M25=5qG$l<rbZy
z?>KOu=Pqvn+Wo!l(wEn9!Z6;f=}LP={TpN$R~p%&p<!Z7s}&8$kDcem-J;g{V&4Yu
zn_N;F7@Q)`U*!i(vV{6F$B7*q6s9xa&`Wrw^N^a5veM+*7%!T^<a7kksh}OhhJrA>
zCvU6IE8d_*+XnGgXClfQ`f%-Z@If*1&HUtgixeOXAwbA6295yR2<08G$VPC6;{x1-
zO0*gYHNpf%^@p}S?d13$JDd*P0rbfxUmLeWo;|z{HcnTdCOKUT6-yTO#zOqv=#sq}
zOBYcF6m1O_IH2|vV*We^)_v_53#sX<fS<LLZQ1(yfyCS!Qj`5<+P)=Ek{@P{iNYqW
zGr0oZ^0_}IG?YaTg{|1b<@O!tp~fHE@mnhYw@-}2mFOKK=4pbKn~hZphtnvLcA1+?
ztFJk|lj=t}*J-uyl^RWM%u1(<E~&%Zxm$^+bp#DQ-8D{-OLBhy$XQn&6QOmh@V1_J
z$$u>~=<k3v@?qBV*q|hKc-ai}s1B>*MYbLyZa*~2+{PmZiGiQD)9&ZYubjQ-;9upF
zEwq@KLJK2~&jas_FT?g70cRt@In~jy$ZbI@iMmD@ru0BcOK7Pa)Z>PHiUdT7$%Xk!
z`PX5}TmXh~Q`j0DaS*zl+~lm#4E4uMj73M<l@?NoDpfXsj&}k2MY@r03XV+D@3+&|
zg%qW9(FkuWhoq8;Ng%Kt0^#j&XojB6a>E9bUN{Igbyh#H#6X-IR$`ZY*W;?q$7SvY
zrW~cF*VFsE^xu0<b{_$vu9CZ1(BR#Jcr*Xyb1(=wC`J1$9g1icvW4bz*m6Cxd;mP!
z_N9p0KI$@lJR_20_7Cjh(kSpSsO}6s52J7yKDLRzXM0DHcd_CKO~9Ip)w&ZdH0jl4
zVrRh0MKPW@5d(=%ZkTl%g?%=KyB3PK#=5-Bc5#@pQak%>K2`V8-!<_Q?wTeLUBCJs
zy1J*NcIJUSlqJa7{_e_t&12eds4Zb!qXpZwfe%$3zQ{QhI<g+%<WGAf--YwLhtZ2&
zQ1GK`G*}(LW)DuieDU!(rK*|?Q@`q|lD<C<-*c}+x4&4Co4=En>ekBRYG!AjrD*}g
zY%e@EIuFJr60zj*U@(H9Pu^3-U6oeg*D{n2_R5)^FbFb%9_b-KGxq7m79}Lgga^IQ
zCes7309I_`Wbw5gEl>)WiFx<`=MwEZ2)aCmu#A1KJ_+ol^MhXBXdinU*f^58o9ST7
zElJ1pPH2sIbsB}s@7Farx?*jNq_jL+@Q-;x-v^AS(5>JatzGzFFiz3CMLJ}`<~^fu
zGbmaSTCPr~9Q~2mWR>$odc)c<Q~!7=&xF;LbeIdkMM~5}4W{W`H@4uN&ZUYoC*P*)
z4%|r9jr)JS4?(mJ&<svR0WnY(1GhaAjpd{HHfVBkw-3Sav(IMm8`!%cRdXX;AoN*v
zj^40ai2$6E3cR{>@Xw1pZgD=p@ArN;(5~@DPVNY}gxnX6&%^wG1SurIB7vB<6bqon
zbKcOV6*KodeE9IXmHsNR`}xz7GiLTK0&nX54MFhKZD!B}_PS%`Dr1$@Pf|_q>*0@H
z%z$idy|xVIJ@q|>S55TGK4fpB%ld;*96FCz5}_{5m}-d-0(lt#9qjdnSaeV+)Bgbh
zn$(VzAm~my4ZF}}0!B=uoyLyt&%6_t!PTee+u(v2;<Gk@JV0iJ62ZL+)Cfu*gM2_^
zZa<)BnxG3BXfq&nK<v-~8T#DtLGZYb<e%V(Z6=XpWUvQ^hExFz0lqOB4+Qgp%`XxF
zvpb~MZB{end^v#32yG?g^@k^-aCDA)>W`Es)eJHyKHXZmJvi9(ee-ACR2lAFIPk)2
z;Zsx~9py;9k}KBjN1KFFx!BysZljm{&02Ipy(nsPXrcJ0q#yzU71OFBImRqJy2tFS
zqb>j1yOC*=MhHAk1PXY?=HWFEjz>`czug<E1e4jZmcxrOnX_Fvh8$>x0-2&7DmhNT
z@}N$?0XP}RREwHvr!+1*A4eD32);VHtr*6FVKAoc4P1HiM&90x_vP7_&tT@+00W|;
z>x7cbYD*Y_eO}M#)qNs=H{8S0tr`(wOB{y;1J;)_P(AE2_E7+KKd}I=wX^@JXXF0p
z8h}*#)E35|eCE>6yJ#e(Ya9Z|2|xwUmXtY^6Q9DEKnWsou?uF0e+o4<T0P{c>asB~
z7Rb>GYsd;<l5|=bMl)|n5?*OMd-bfSU6<BRV?)~?AV_-_W^RBj1HEuI3Zhdi#9O!#
z3J9Vf6J$0!h?Jw|SJ;Bn(m3ICg@N_n_>_?f8Y32^q?(Eruya!&1!WjG+`C)SrikEB
zH89*A1tvlSl+7m4Pd!^x$2O=bDi)}5U~)1);aQE^KMHO55p)8Jj~oEAT(GAh4UW>q
zd$w;@GUV7ivuYld&O^~GLtKg6DI_41&Qc%kwPMK-D~$@t)mug^M*-%PVE>tnZ}i=V
z))HE)?24?Bi1l63QT*|3l!Ywbo4crgYhG0#)lvp?r#(@I<AhVCq1au2<Qd+2OOp2_
z!Uu34R1&3E41yu&iQ(s4Ex{#vVd&w1e0pjLJAl)6aHB5>;Bs12bBc-ssS~m>*bLYT
zc21=^!wHhXzdz)widY<__p)7e;q?@y&;eAU%O8P|3=54`<!(VK&p+c=)|Z?6<)XD=
zM&P_}!4-)RRPO)WLjHKU?G4#H@p0#%0sQSTpt(F*LGJS7O*%o3&+MaD{OKYeoeBB}
zNumUq{lrPp1a#cT4;-vNBDZ3k=9&1ljrg&qx}Dib<!{Irn-*zBkhxfR7&QX;8o)8q
z$L%TGEMN=D5#VJ;mKcWXg*2Kr5Kfv^QP3-f1ZqN|c%_~Vv@2o(0_2rD2-#O2>tcS-
z9)Cpcc0H^MFCV~n!v2>B3!o;R!`J5{hRVyfdlK|=6PWr_5mAJCS~$QGCKvhlX+~O!
zA2GA@<P<JHeT5^)@F%twh;6NzTgnqSPKvw?zwg(_9p7JB@oaGUX)upD)qF@><%Lyf
zZhj4X^W6#(tHKXS&F>{RpJ%;A8W1=}=ZKxcG1?0;e?{|E65{FlTN@U1)i00%z^ZVF
z@b8?&>A}97eaj35k6|gay1*kq%0LA%=xg_uItsxKYX0sga484hWOcvqm(r`S5KB2h
zggzR2ioS}}B0AVD@ub{n1}z5uv~pvCmoSz&OSC$9wnk@jH%F;`kXvW#c+X(OwJ@jt
z6qrJ1&(@$I<kMo^n$WeP6W+*GctEFIGQ{$Kk9aP8i2ZsE92-7_v?~pJjb+`=PGJ8k
z<#I@?>rSn-5;!@5Ygnog^NrKbZfzc;V^GaX($V7r=eGUBmHfmg%o`3vu|I4uAD-`h
z38QMI)U&rnzP_!mW}YtA%DM9r2cr{|*Z^!qs{JtxFP*-CkJe(}lVn2ylkyYc7yQ0!
z)*f_#b`I^uP%8$#^61iFTnnb|rORMr-R!h}MV-MDU(MWUem3<p8f`KWTrf8R+#v}~
zku#urY4a2A6H0;S@$!q8GA{T^&xZbpL;r$}@x|%<$JT!27m6rhp-&JcD9gj3O3s4L
z6EgR>W5j8D6~(4g*t6O6e~&uiOwp+0!Y&1b^2YDm!xk=g-(BJZX{K_=5T1rb$%Ksr
zMh#LG2|a?^VyJ}~@T?BiYaTtI-_o*J`Rm<**=Es}-_3Q(zBCZ_8ld0gfk{<?r<@H4
z0UH=0v<mW$Gz2(xci(X^YPNxJ+TBE(UQ@<L*YvXI9oEI#q%=^v13LQDdD=cUdcxK>
zZ#~3$7*!QW^Z%vy+_OJLziu}FK54kfT4wGwmC#v#l>Ai8Nxw5zy=0Km+7GmI?M8Q;
zN1UZVOG~ZcQattO2UOQbN8a6Wi6;zo@tc8jv%PQTL80z@`x7)X$4ycLpwweEc~j2l
zO$=P}=N&^uOk^lk`|zf1MtOQkG7gMnCHTqCwri_<+u!urRjC>;f;^P+y$vn}-~r&)
z05HqRAq|wUW0)W$216~=kV^}+V1|WK2Pk-6z{nCJ<O7gz(}5c2TTIK)Yh=<6Ea3GQ
z=dQ>ABKCE&8ojX2CbHGsx>Q?Y^EteuAsq1L12qyGl}O8_OW!c3qWGiO_N?e1e<jP^
zOCHpFp`AsCiyb|M<H3C4buzkh^=A(Aa>al9b~c;p?*m`})X-gY0_2M4NC$uc9RRlz
zA4uE8JKpF3tkZAXM)pOzMs~B+g}&i6#81%wMl4E5bW6R$3%YYKwD2)$xz|k(P=g?F
zcauE1UxkkWR}+cxX~{x?f#{eqRy>JxBA__$Enl7+Yl8o-mGr3MS-<B<@PE5HvuHe~
zAzGAaj9V$;V<CRO7Q78ZM&*?pD=n^RA^h|>po$=dZ|RnkQ?v)D=KNVpQ(Zl2=y7m9
zpmxtSM%19snPqV}^He_<f%aX`C&A+s+=PR>YuBCu$oKYqY5l{c*l?y|8j4a2m(9*+
z^J_t2WdJB*coa0sZ&5x>q$UV44}ig#LoPsEll)-a=wQRY)$a!uV*$J4v7zZp@%l)O
zHyg|eKt~!Kp&ytpbR(?*=tovPE!4&KeR?q2m=N8?Hzm*t?r??$=J=`uN)yob&SD^{
z6ac)et|t`sA59>;CI~mg`Pujf0vjDbl^&K3`5Qa*zx)z&ZjBa%P6|;&jO<W6`<;XP
zR;&e@4gD40we(nYPxrO+9lNgmZYRj320Cwr(QhkT3?BMB2{kR%k6SUPCMD<x1L^iq
z>O_3S?s?Lgr#KTQI~BonHQ*CzHA@M@;#8->W|15_7M_TGQe&O1QGe~u3t~|t+^Z4~
zBlvBYPUe@ujvxpz069|~it}gJYx6-X4wQ2kC&WXU{&`y3Fb6t{6{UKU0|i4I81ayd
zQ_OecohV%l00r6N>xJaDpmS<~*~bNToY<2iKfx7J3HH%Xp4%;=zJ-Ec1K=De;)o)5
zdNhM@6<kd;M2Pp})V$U~7HGzSRB5!Hsj?_U^7z9v;GRRuZmKk;%{vtS%)!!P>)H$a
zoQQd7gbSW(7<;Yt>8ez0%=wPlTj~6Kdtaw`@Xhab-<g5(WkM>9x5h#%XIm&$l)wLe
z+3L~!_i`-;6fJ|~nr>)0Zw(AL15C|pw4r}FM>vJKv>P_`+=}<M`e1py6YL`p;s#cX
zr|r=E!28QB=FMCQZFVm0<q}Um6gazONqXfVT6<NyGtYz%AXBb_lp?NQUSpB)@)h{=
zkCSL4fQ%gg%=^gZQ*wOOpYTfYt6`Tqtc~+iq`d=n{D0yq>B%wKrAd)CDvM1eYc@1L
zeh!=5o%NdddYo<XF}Dao#;QEw_PRc5|Kw7Ci2@l%*kZ+vNBq*grt5HbiVw#BRWeS(
zN|Q2pk>MHFW}5ov&ATIvo0^%XHpP8-3g%J#qiB8tZF?L}f?U!MFhJm%80Izb3u?b~
z#Rzs65Kr9<q9ACZ2z}%~L6o^ZMj-K#)KayxB1}D!`r%^AmQFwmqE+SNE(vx)u6;;F
z9wsm&7eg1D^_%D#R$U*!1~OpnMq7pep7qH<%jD4D*46Mm?1}0Slnli2#eKRcm%doa
zG4@HJ(N(z`tN)llZ|%`EW}#6rS`ogTf(Pnn&hCIhO6a+G%&b^Ci|_9hilguS_;=M<
z!-T3vYnqGuH^TubCcys~sKN~hao^W*u4WsglI<y4oSVU1mg**7xEyF;j@01EcOO1H
z<30lwR1{!_AevII(jb`;fci33o82#|rVxOz%tsXk6FQIS{_z|l;K(fjI`DDS+)6Ls
zRKT6s9s&4$>%>vuK7tAg2d}|R7H*7WTq-kx9Ag7YuLv%mmTT6tC0L+B{bzB^CO!dx
zPfIcZl)4+nd)k^}pW0%JA4H3a5(XNp6b9rN0JL!qG>-r(G#H`3B4eN>^8Wp;5{|#>
zD*6Jz9ztZ;9(Q}|<!^m?rrpploJ~5X%OI=>dw1D_y_=vlTp;EP^@y<3FwT3SL;w2p
zLSDsmBrH*uNo53h<`|Z{Pgu>IZv&CMf(N*{?;lX+__(;2M@wD@cN?lnsn0j!5nKT1
zm7Qe`pl~UOIOwjD38Dz`u;=^q@f_+YB6HNBy~@5#c*JkoD0lcl1YyV}ypxlAH;;fT
z)wS;U-L03LUtH@w=!FYzxvd{7-BYD-PfVfdwKmQ6;^6lYS~(nuD`K_`@UP?7PmTBG
z-8_-c*h`z;YWx2}mjeuB)Q>dx{!gvJ&Q~LL`ASvK0gTmNxC`%0Q`Jb9<vAxkls&|m
zvX#r;cXtm`T04N@BG5+ZRf>u_<dAS4kWIJ%yYthajvbB-h-4r(8sHhH^~yk<rU{8c
zf!LD6IG!KYfAR?FbF~w6C6G1Iqoz-fmS;19L@o`Q)@8Nk$v&BP(C#lFxyw_H)ExgO
z`*AaM7fsw(HHqXu;F0{xxaqizL#$Eh*lc189Scl@f=$zCYXJa9`E;SW@)bssbF_&c
zO-sLGgQ_(bs%3;BSO4$LHy0-}{a9)o)<%+J{<q>{bAyUPuGB(>Vv-%0A%ZeWQlx?`
zhZj8p<iZGpg%wV(*@mN|puss89~`~N<OKlHGyuylI&kI!RM~h3j9|{woOK(nAuILJ
z1tVF++(Y{p<MQfTi01%&`$s)6M&SS@Cue{&eGfUjtC4bUX}=y8V3_AsYN;eZ>^^KN
zCXwf1VNCPZafL?r&#P6(Ww!=@!|FO;TyVe%A@-yLTyJmg3F@_Yl4z3_0;?mM^Yn#-
z7|9b^G5X2=DQQ<i!)TtIc(EDO{&|*EZ*HKg@a~3;J*ke`%m{E6nz>>4)8lh<g^k`D
z!axMX>CK1slKLh`cpY+;f%SLZ+wu>Or99OT{Q@>)4&(f9LUADcaRt`6N3PeN_K32E
zjyq0s7|}+2r^^&jiJ&LrF3<<y5Ka8OEdBiH5;ey!P;?tf0w>Wk&!M*{lOKr7Wvm}q
zV9F%zr2j<vQ@PiP#QIkZ1yx&+;l<!Xw3{Cn2z9x>8!?$sUrsH4&OG@FX0ZUKyO!l9
z#Vt%22(401^aI4uYTK_@A4S9E{BFCoEv@fv%hB_nIq-a|ar-A3vOOcAA)iUzdiURQ
zOB2Gvapj+rOk{gtaUiUDRf7@zJAkTD1WW|`xqJ1nOW-sqcQajU*Kb;#AOGeq!#62}
zd_9@~)3ucsO7`wMc`Xi#?RR#4jQ~qUOiU6rff#<KdewIxRvLSB-vLH`pmCqHS3%``
zn~H<YC!^ng-Uo>|ubb9+Y?woz{T@n%I3X-Y$`-hd_3_qO9@aJQdsQeRz?D)*JfkK#
zx3@7-iF;Sx_lqTSJA-<s`YVbqs0Uh@s=p0LKp%Qpq^ZEcGTi{rzb?*+^lbh;5l|U6
zqrUo5vj^2CL+#%~@dPjtm-qha5yNmZKv;+jP1#TS7mxFW10)u438Vp|d3p_WX*mHP
zvxx2lV|^r72#6t@!2ryD=zN<9gvjC=7-XY+G7I}iZ2L!$K(QXHME2|;#(~_BJS&i%
zDMH2#xJRfxTBws<{y9-~dulQD_T9E@UFXe0f$jfff~X7(%zbAKQGU48(W%YzDO3LY
zD-B}^|AsHm;dPT_r|^;yO{Y<(v5*ct->0_TqvMpy@7^pQKf&{NdT^rp>GDeM1~6{^
z{pRTZLBGBIuYOw*fG6-qm8EP}nE%lN0H?u`rq~uc4n)U?8gZ#SnQ@U>Bj_>G!dVrM
zVgl*HDJ-DHj{=P;7Uqv>Wx2fdbug$+2llg#ARj9hT02yxL;tdM_qE+@x}~;WlEFFi
z_jp$*2-TLDV){a?3YClH@{R`e%007?3!z59UuOZz4R-Znz9SdxzVFK2_V?GjBT3r+
zm4&9%{^zv^?tlvDgzQlbEX5Ok=oA_c)=ZOXr^}{nP?aFk9thd~wCM(9z%p2Np)&9~
zwhiFcNZSCeL3e3H5r|J+Rvw0IkLVbh0cV6v%J*vZb+{Mw0=;|jh`r;R^M(ErY=@ze
zzVoD5<gl%QqE9iel>tdgJaYCIF)CkwHSP(c8XiY?`qVAmtxuRD1OV|1O0gf80{VYA
z+T`kNFE-AuCfLJB>${w2zvsw}y#nD}MGyXJ0dQ>Gq&v+m6?FcuI}P~9<D_pWjEC{i
z9yg3G>1E%s!52}u7?fO#2mkA#>>=2FgT}QI#I&zsyZ9yrTEV{-Mu7J=8zcDfNGb>o
z#t=amRJo+LorNVa+9V=t2JNyB7kzN2SGa$AP#7;SdNM}>eGgl~vJC7psx5LbeUhTy
z)=9e~KnA-dy@1a~458kkNZRhdS(#M1M%c*sZr11j{!J%Fs9O+;@v=4`AFeG*Z7NX<
zhv5i#R!Glm=rZ<%Gn@HouG*HNF=uYz%8a!8{U{-`!4ms4G_VQW4PdumKwY0*9fA%S
zDb_x@!9wEJ?cR-e`3PNa1MQ_Wl);grSJgh(0v6TH<m@eaDx=M=h2Bm7#XlzYZ|-e_
zRFs}!lOY#qF?gg_=)WOm8|EEpcK1B@*%trq-|s2|7i~dDQI4$y2G(!T!yr9}xg9t6
z*I$Cs6&$$3`LyK#u#7sU$xg_R@)fwx7_*l|aRWQ>NcW63_&PWV3TX1UVYa?YLGyO(
zuK<-@^=XcaEG3WW;4K^y2x0mELZL$T|H)pH;vpo=*{27eyTnAL?%xl#`8rg&NfT#_
zC#U{%d2p&!FmBv4pvuAU6ER7T?t+m5dO^Y$*<b|%Go**0FnVug;UsiG>S%FdFE)Oh
ztpL;)%V0cXk&=Fb0siTNS2Y{g9|H~w?+iC?P+Rekoo4fjM$tC1J-`P;I3iWQRt>R#
z&PyIV<fjuYgB(Io@p-#0qNU6KL~{B>(quTXN;dmHO%VPEAQktS1T2OIvcN;$EV|J{
z<EOa^Z8cymP+(rZVh11#WI%HDkY2Aolq%b2q0{Xt#glHio;k1?08z(;b|$$yVK|vR
zop$^<@8;b9&L{FntF0>+7y!q5Rh&Jz3I-fg9gmCtO1w3dq*lcAxGjl7`EMrb)`wPP
zbx}Y-j!gq<5WI54mL{78#n~Mr1psD8ZgySNAW5SFYf%AUF3%?fGcGb2z~llk7Y84t
zxkCddih<qsL$IT00=72SQME^{y(S=3h#J#0nD%OwxmY$C`z@}KEqm3TuIv5uEBW=T
zLCBNa(OWi=Xpb@c<KOzIk_svdnw<?LjKguI_FZklf8EoENZ=v0A&x-z-<Pom+U$4o
z;-3{M{HUD{{UrHOp(C=Bnd)S?9FXX*zz*s3gcxcwU`M+dUQsHAMybe=91XeOMByO4
zoB$a)SS{uCO~Vqd2?#7{fGsSa$c>cvR&izd$`}8-tiJKV*KKR3|9?<<;I|54cG@ms
z<X+fA0JZNc-W2!$H=rOSJL(QaxU)!_y5<acn3XiZBOHZN4K2!$;nq;>750u33oqu;
zk`cC}ks#=b2$z7kleb-(JW7k1H`)KQ8(J6sqfkwbmfxr5p!a|`dFg06Zu6nt*bU#k
z4pL|u4699-oX6X^cwm`07cy!p`r>s723euNR<f_;I{Ji{RZ`$%hyeEe@q>?0Rv@oR
zD5WMuAIIzO{(rc8?|7{L_is3&jHZkVr6^^l5{fz{Aqiz<t5jxIl<m||iBLvF+V;xI
z){=%KD<chA8D(bQ&ll>G&-J~2kKg^fugCSr{m=J%e#H5Hzs9ki$8kK57e$lh><oLS
zmC?b8fZt=04y$>}c8b-siG4Kj8(O+!mg)8xCxb~LlU4kOXLU^0j-ziLRUX%{qE`^s
z2rDwdI~Rk)OTU@EA-zZYAME4KY&eCnnD_TZv<E-;8lr36*4xi0js(LIplzG0Ldq8>
zd%>&_cxLk6BtN8VpmL$|iNF0-6H0&k_DrU2LHLuGc>G)*Jt%WUNQEjIHC!aFVPZV=
zNozyWVAYRfZ^BC86y_xaql~Tu+zR|C&jGpMl9C|7km}rb2gSzfOs?me797gys*g?{
zz2l_*yWn>vr;XcvZY5D#sa^ZS-KHxCJ@3&WNQ&vYoqA{y!sXek<^NI`lcKlC6J=y`
zJ$JEB9+5Hw$@CIeEHfLI?qhez_4+LvOl9-t)nwTo$t)*BPLLUPR6bJVO4}DPG<<eB
zfV~vl_2u7{+ZYbYB9LZtzPNQV?x@k>38gF`eYUwj9y4#3LI!CKN@%7W#ii3dH~I7L
z9);8%phd;iHPBeSNqB=qWVu%*!4yj)Wxi*eKwB3nYD44bV@ziueXq5L3887!0?>q<
zjQ1NB1&Z>KUSk$@^R(J5uTN+Ftymn{jMZ`sT!I~)mV62wTr0PA=53fo`<Vb`Pkv+s
zmQO1G!?K=~1u12(0<3v~bz7INfc4JC0(X;=6yrI#(352<ME5RwMN6Gyrx+p<fpAEH
zbjJsM2Q+#W3v~iY;=<hYCR`iIetr~4WlZH9o?iB3@|VFZkmD}k6%o4WJ{-gTg7nst
zh8}dR*(GWhEeC7nH+{#5A#iEjL)&*LAKPUY)a~I$zTK-AD0G0mTOxjW1SeZCB0=*h
zXKM9i_t;#&eZSRHp%1J@UZKCG9*v(<oulyle-RmN08!{luCU7&+b3_tK?+vQ-pa?U
z6@dTm*c^)<Su+v|skovpSr~Ss<~FApNwRx-o})+iXc-1SW){UetGI(vz6mjZm7cur
zJcek|H(<Eso1?pD3HD8k4he96T$=Y`oE#kdB=v}l`te?*$&bhAWhUAWp<3k{nBumE
z7#v5wScFVQfzX?k09Eiu|MRPa%#ZD1!sfkbC_SBJd&ntm8P2+@s6%q?bVwv-`|BN2
z)|1wkK^~HSD>}tD2zoA|-lbxa-Ssa$w`Wq%t&wY=<i-c_F)*k*D?7B9X{Bls6wJ`7
z_iG89D-Z)LPA(vM#6d9vZNaW*Qlh9_V0u~yETFeG6q-tzB-<dXhP}y;^J!u7-_Mb3
zc@gj=-*@st+tUL1C@ru6ZV=qm)TA+oPys`8o3+792_X~3K}&t$+W|%p<@h7vB2_&!
zQrG@^DfM~`u+IlR2Qo$VbGjyLBLrFgcP&l~JW-JEQrkrqh;>q`zNSrgRsFsMJi3!E
zf$+kCNodR1fFw;byI<?uW8sfZMI&{OF?!;+7sp<^=$Vc~J_t8kc=C|P=x7<6)Q}e=
z^n<^KbH>1+y$6AQ=iFUBv@A;rA3XQAEm659B+nYZS`B5~!iv7J)`0iq>4kf4%vr#-
zj<C~j8X`rtXtmgeBc9|v1GXBW)Seqtdh=N5c7t4_`yyf@{!e&S0`FR&L@4xDQp_1Q
zde^bcZ1a97<U8f3q0!DoOxCxN+s8&U+Bw79exF&P(H^WlI<4ydI76SneZO!)UUjJW
zly}t1yy%5&<GG47a6PQDa<{}pw<>bV&ThMX7c*o6`M2@ezhX)&6fvTvps0~@Kzv<1
zijHs3N!s`<(sQ_Hl*IgEVqyc!Reg{=s-h+2h}QW-&p$jzT&IlE)reQ-yA!mBxZdMR
zkfI?!`Anm?BNJdiyarG#8<RI^adX$abzjVIb2^3S>b6&D>J@p6H^FJDi<dL=GlIAl
z>@u0W&wu^T%+5d<46>N}+1n2L_PYig=*&WS3K=$Zp_0l?_9K85q8nJ-!AxTh?_<O9
z0=3D)8EVbq9~>`Mfj!3#AXNz{VYFHEMwS_FdidCpR>rx{ABmqb$Ijcbc&);Mi7{)^
z%oec-^dTwHVsjJUlLCGaxRu@|qz@oOiFlQ`3I578_$`uGSDgMtL#mDlvWI&3&HW!F
zXg_063&D2JKJZvSS#^)gVz~}JH>ab3N)O=c?yKHG(|hn+VKbanAqV)Rio!HiMYasL
z_UT@2g|u7Fvsa?p;hl)_aVXzI2pxFmm3A%KaErurj!c?UzrLQvgUSVnb}c&U2Fx(o
zaGRIfFl%b}OE4OcPS&Q3x)Tqrc_%%r_(FIiO_(6$h$5|g*B;a^lZjptW~9GE&mCPt
z_rQN&+Ml?Iut1Spd=>R$e2wlS2?WkPz}^{R*0`2a>s+GB)uw>8^7905dM&0<o=&p<
z6uz6PYs1Yvy%<J_V`-q3V4;|GUE;mzr1>9XOy<U)!Q6Ot?n%*Q1}Rk$ZG=gY0B)mK
z#O@&IQWe;s`xmeHE>1p&<?5Z}^_{#P$)HE0EFcVzDCcmt`i#EGy9ZMa+hOu5$yXTB
zY;sfinE#SE@weC5{P-j*{Npir?;<ow2t|^If!p1jg*$;-2cr48WtdPLL3UeVe{>KR
z*E>`jl=9I^Nv_Ad5bB2MB9F&qk63>v5(ZeYG?|egEhkW-4W4)c>cqe9EzbM#WGef!
zB`+iuN(O{;!nmmcs768~-Ia8KmCd!008T5E&5*Pu_MyP2)20D^L)vR#7ZHoEZui-R
zLOVMH?u;SSq?2atZ^%goBip9={h^Rf>e&QW1zE%c*MP5>j_m)&A)uiMMB>xIgt7D6
z{LdV@iQ2FChO;#MVnugef4pqoClHIV^ve@6UQ?r>m#BPMr^)ncOF*=8+64F6hux38
z_N~ID1YI!vnt6MFf-nkx4gJ=O0#kmCZlN}hH;jA<++(*457JWM+-FM4AB`sG9fGK}
zvyMV035$tFjX#ema7a9Wo~>PDV?vl8#Qx&N;tJVs%cSf~Vn_X~aMMHqmN*FQfj}LF
zuLm<<y4NsG^}dHoVY0wU3*BAWo0T?Bb{Ww89I-T#@$q|(I!e>%htoWmi;~a6)(gK=
z(b#wycHq#{yX*da2Cf9;TF9jZSMw=dMYZku1>6GYeya46lyuZ5p?>+@i_uBnz)e{f
zVA*CQ{v!{CqPm`QDi^W&e?pN-g_m&jQ8w4ls3Q#_GZ?@Ns}Z>6S#j|ztIUSA!1=QV
zR_ma8jXUl}LA#&}nM*{5W27SB%}nZwnMDWv6h!Jo1+?N@>(|ySV@sCpJc2niJ@6bZ
zG39=iikUSVbu8D&q_I0!whNHDxYsOYsZ1<l4(%G(rR?tu8RhFU@A+J_0nqDy5cR7M
z23AgJY&c^>XD8|UE1BeF0JA;FWA*5)6)zS^H~N-XTXkQBGr=m>i1vCp%RFOXu03$*
z;ZEwApwi()v#qTy2?skH34sj?a{{Rv{7urIV|{|gxzE;~0<SfDKT9iUt<pSv1ZFAW
z55J+<bN+{$^ATW!AqrtN<Iw@C=OdI_h30rfPY#zx>5eOU+|MXi@*J?f%G3p79d-4=
ztkA=%pAcrVIf(j@QoZDdg(x{7$~4uYS|sxr_U-`0E?Q=j0OqUOH~j7n7TI<TPOu=-
zPYy2Z66O`_mH}4gRo3v>p_8S0%=I`J#}vZZ0=zkq0gR1!U1}vd6_F>8-BOr64Fm9K
zz_@q|6gz5>7}or)y7j-VN8Dr#Tgepz+~%xERQ{qWWGysA^&bTvjH1w3+2k?JGwxTi
zR?FIRK=ebb=Yomh>VjSL89&5hJ}pFkHvnCUQV%;pBCx`#Y4H%6nj_4dH^Kdx<d=hL
z>eCw8I|AM=(r|a8Xp8F!|5U;s6@tEY!dCiOG_GzKE4;|G7X#fEWS-2U?|=M}|EBA0
z`r<4Q$~B}CnoVZ9&%v!nW$9VSK<TEZ7nc~?W$|k#=xQQnS_9|G&gj;|yVczS$<N`8
zJJ?O|D^<$n>HjL4yYj+u3S*vc1;>4OTt^h6JZ-Aw-#&?-SOA8&q;3th0B{_Y35Q;4
ze~2jU1#I1|YPZRXn){@naI=A|Jn9tSB}Bh_IGv^6Ok<74+*y?b<1zHqo*R*^U5P60
z;?)IWYkz=&hp;Fdj<4JdZ|tx-g?_&l^=f})gvkqkSq)>};=5og1Z^5Y423Ub@!lNP
zEP9r6vh7P4Yq}F$9p6Z9-bGuc;li5iP_JoqT6h9M{AA|J142H2_TWr5W8gWbC*xY_
zggdNoSaS-YUGIUSp2ls}K6)O6Z+ET<>6CZ8>CxGdC;J)17v)~u(Vlau`_Kd&1Q&b{
zVLm|del2rt`*&F?Awrm}^KelT_8*lvQST4R03Bw5sKNa=3N}_K+k;`ef_7fC+ekQ>
z^F*-1Iv>aJ3{anQu%*f7nY3)BXL<1%CX+ONqHi|$`+b@>K*bYCLr8(>i|_q8Nyf8K
z3LAhPuLJNbIMiggu2&RX#U+tqf6`;I%b-vC^>8;ODKUR|AN?wI+2a+q-=y-8DIs*a
zZ%u;9_5NS`z+&X-NXP|y&FZj2ZcNo}z9b$JC~CA_U6w<0!xKuP%aaEEe(^6T4e^O>
z1ZFQke?W`c6pQ*!fmS2Q`=Dpl8|ZEnRk*UEa{NJ}jAotY50(CBU+><%o7^eL)qigu
zUA)jIVgEBsvqr#(`^!=nB{8IbzK3K$SV_4gN&H6o-q9pR#LtuIbRw<5h|8Ay<8@gm
zB?$}-jh5M=PJO3Yc(lLcmBaqGsTCdvK)F;eE@j?JXLn(_%(mS;mQxi0-pLMP^uL4o
z2IBf=uDB;N4rpzRF=Yi5$=|k?`h!$MA3Ck8upZf1M=F_jgm`{-;xXV&yXFTDoa$oA
zDqLq$!O0p4f-ET<L|MAoyW8gor#Q25)SSAxn~uH?Rp<P4YmWY}unKJ-_!Mhla{GgX
zqZ|;t1VQ)zdYyyOAu&1+?|1Q>zRBsV(<X|o*sBhh<ff?9aWG88U1?EvOUxNXQ(_ZB
z*i$NY;nm+y85Pa?9TVs+^P^{z-Z|C{#+PiP@cnM=iPWv;H+NV3k;{bc3hi7%+H<`0
zhV99a-b|FAt)gePWiJGMBhR8$G#(QXBw@}_dA0#JKMkSFF6GFvFii@*`Y4)`nlK8a
zEb)$8<pG1Z%0vS_l5{0SO(t(68?*;BzH2Nx4}kMx_5yV(ONE0eKzzb_Qrmwv`-ELY
z#V2vL)9RHf)<B!Ww`5s?_y2aH)tQJbmNh)MoNTxYa84{_SLX)Zs*TO3P3l@D{iATU
z?gmC{Cnc~vHqttizWw9kSsL^X{aeUzjF%yz-GG3%aD<}wUb?SLZj1GrHzXpjr>~H4
z>W}Y0l|`4}&V$Yfj?#HrS6Q0BSi4fRhX1D9%>hTJxcVJo908@ifw>855z;>9nXa~$
zYuP^;k;B2PIe^f2phH{tn*osQocR_rE(Y!i!vhlUpMU6k4eepgtZGJfufZxzw{5A-
zG`wQJMZi<HZ9dar{oeVBYG?8_DQ8fUUQ%u3kud+ukJl|$I&My?^eSpa%h5l=i@u@q
zIi9DRSE3!_GM7~OrVmy+^9N41$2_Pn=VC@fMKC;Y!b?ZU;BFg+>ZO|pX;mW{l1!EI
zsBGlw*UTcbB(FD1y$6kmx@ersK7b^_Xf@IcLra$1d?oI=i2nyi>>}ablEI(HIv*8G
zGvKDtNMeS`HQ47%wJop&-{7J2y-wu5ihh_((WLhNoJ`|cv2Ve#4*J>i!>NRRHZ3VO
zmMNNMySdsM)V4>s#pf?&vsAmH;cKh3Ay&kBg-A~2u{S+L+c>*rp6WIvI6Yw!h*Wv!
zC+3j5k;tIdA{Ydwyj|RjH)z2jOv%4?VyX#=aM@t0goPs3n4-FU=<emEVmnz_SZD?0
z+?M@<sjGf7LKvRm*~BIss6NxKYhdtx&PJ7~S`v0eE#WKK`75`cj0SDrj!#zln_@lL
zW8GhhLUAR=g(}RV&Z6t1m5Ls*5jgFWq%@XuV6X@N1cm;WnulC&L-=Kx#k#<2Quu88
zTGfTKKA;CU?YPvfGz0oIWXu@&9r>@lng0PaA?K6#ucKWIz>5TByZSbeKx#7XGF}58
zK`$mV+Cmb#<;g9@W&e)Keqj#g8Mz5@e&ml28W;n;t_}Hq#;O>IbXt<`oSX=qp6~OS
zP9=mdXV?zR){3Wq*ML5!m7uAZuHeVlVALP9k1g#M@d3VTZtx~du!xN7%X@*75@ZM&
z;mni8j}ZirrpeoeUiJ~-6Hd$S>uRVI{~(o=zdUwMwg+07N;el@P6{d2{4uw`JY~86
zm#r}ztz@7Ba6>zlU=$X-LSgGSOaix2K4tKO?M0E^h(_{-x7$Zvf3KF)oQQU*5y(38
zeKR^~ggwt&gdH(VYV7x-i5_jYcwip!y#i>f4#*HskK`l_E7CKmojFrct@lKH8Q2UL
zR&Nb&Yor<)$c%I$Wu~ik70j|(TjavEWpx<7_+L6l2g{q{nOPw2Coq@u?5h;+0w`i`
zV1`v~g68IzJ%{xkB#LX=?0EbAz1qORBtseXK??14_ldDnB)M@yi%laLW=A&)6QZ`g
zM8EZ&a8=Qt-$INg2S!*2=^@~#P8(S$K-ncZ-i$nlUWMBY_&<+2z3Gw65{`mvfuh}V
zxV=f7kvlvLQK}(N|Muu585%a%40Ob#RNIi7!jQ#)*k_laTfb(fl+Ff^g_RyjaT8S{
zdKvM`X76rB^u5_mI<V%&DGrWY+VjALd78!Lv|Hl$WX58L%scVMC}Ui7nf-T*2G4!i
zrWFSoRE`|L{gImYZ}kJ-8POJn=hyR~t<9-8V`Zd~_b2_h+EZ`;c8Dvh`*?43QWYm_
zS2W0(64H1Uw8up_AHa#4fZM}4ZWsNwUvS;!^K9;eTC43rwozmKR6Vbb^^(rGTX05B
z>V>S5o73rYL!gG5q(raNS`l+K+I`?J+F4K$N>qBDVl&z4W|a)}AJemqMWc+q|L8lQ
z(pv*>6rg^H3}S?v`1N(%u?in&LgR)OshoQ*BJ);AA2bA|UnUCdeO@8kc+%+QyYHvo
z8qe~33=-SK`OEEu;u-<=n?@#@1%yfvo#HnT+DUGU!N@*(5i(=sD3VuxYb8AxzkIq2
zD49-(58&Imtv<1<jr|eSpD|uvg>eazA91&y=r)p$pdv5&v0zwOIgEvor&rV{T5#t2
zZ)SqGDiy|=WJYq%oWbj2FM`^?|7Z`(z+cIpP;#U@tX+s281RWl&x#hvip^{z<Qn{u
zrytjums<%%jps*yeWI6a$SDzVpXj!1Jof{Nw&Z1ebkU|pD8ZNyx`S>+xWKs-eNK!p
z{g=~B-|b{R`?Y)P!BYxz`1cJTvqq81ne(0BQjqoJ{!^?n8li_0F5AhO9Yni??uBt$
z`+ZUDkt|dyeSX>j!#DGSOd>vQsL4t$D1^r%?<TvkM#)X(_@_c@(TQVWqt0g@Ykt#+
zLSt58HFM^YwR`=Mg?eD_Q+o{1n_acRse8IDilpRsKr<iM?6$x(Beqf5yy#}Bh$)KN
z`)+OT9U6GANUs)Hw~OaJs<8-RP7JM3G<6t$guUPwJhnE9T~oEX?dHVj142)msFdA{
zFXPaZc*|j!OK&IXG7!%S%6xPxf6|I%lD;*kZ6thOr1W|Ln=>ui5xMhAaUrUWXC;mt
zIAbAQ7e8n5{G}TT#r1YI`^Nf2dYk)y80U`O+E!Wl?o|7;x6oMTqGw}s6PL(x>5D_j
z(+CnygL@xuCo(MMMN)aG_MY_fg$RRI17ck0a@37p{ba*+SVIBNCoo|K5{#zfq(zGm
z-&h#3MJ0mBE){9gB&YJ!)x_%N`^S52C56`>G+<m7Sk&gfO3w8}tKh8oxAS$(oM_@2
zMKfLJmVpiKLbNASvJ47-OF~3%Y9OX!eq4q~+_T-Ly<)wT$9V35<qM+?-JVRFCEqI;
zbq8JF0es5%*d>KRwdqDq>adEO-oDSjjfST@JzUjyes^kD5eC$~#pi$j3$DYY#i5&Q
zcaatMdJPS!M1?c13=I8DV$QJFK7>zRA4`QeTjw)r8;$GaI3{2)!29Fkn*OucmKNFX
zKO+flrZ3lTm_LF~dYhIlYhSyC8(dsFmGWi|VO>6wbPvjRA9Ge0JwJD=%&$*pdZ0tI
zk7^r94FhkeFR!Q$0DSW;_&z)04t`XvE5x46JBvb1Ly~&Qc&W!T_nQGQhRodd)AbnP
zpcj^O4!`*dC|N@)0U{tME%@uc;fBTY3ODrv54{V$eWE1xgn6Og@==zOlIZMKr_i5`
z)YPiDxb|%h`e2_#SEBameRQ@zSP!AyKc2{bkh4$rYD}kYp*f<zzH*dSK+OnoS9(&v
zb6mpk;5(S27>$@f_R$T1Y*RfZ+&h(9k*5FYtg@^%Jpm1iNPt!33w0!%$S*f?=@fTr
zzJ7fCX5PrGt8wuh#S?jF4SDJOSr~3zxW%i<K<gE7TX8Xs@*=A8LXMr*Ap*kXENTG@
z4vQG9Pw`rmSgt&eX6H^(77Fc7MbD;!<L;*h3m$H>I@R*}#`U4sy-gpax}DzYZST(E
zcPYI-xJ~|H2MQr=--uVUHZ(dZ^vr**Wc;-0c==!x=M_OG5~Zblxg@9+sJVhxShP$E
zMTEVjoT&3MKT%{Z7K)HN`O{eR;h}pHR|*gN2nTQD5Y2AA{pS68nN`R{F;e+X%dzk!
zjDeP`;L^kjs~z{%-S4k>ambtP$d!Z;Yt{vwd}aQ$I(fxw{pIBL@%yf&c+R4sr88e;
ztnMOwtukN@js14zJGX@C7#8tN{S!Wvy{6T1v#ELAJiObpaAHX)64*j`(VxgcY3$ty
zl23V-*T&$(Nct5Te^e-AU9N@axt*Q+UbCD@^x+F!XfImeI<!!a+wR3GF%Ssn-iD*s
zB#H3}z2$+vz$8zNWRwCIYxZJ1`AUB^8Af~&Sip8sg&vwd*o7vyB`s+1<!T)3I@R0b
zHuTEK4}jQO(S(ZU__Y-Xg6$2152FsmgC~E?lW*y6)%*{AKb|bS;v<qBJ@%viZCq0R
zR#=+IkYLL`lX~D3KYzW{&M!~6tf{55CK2wZ7maB4(Zu~bb*<ViDpV&{^bGRE<Q&vP
z2gA}3eLJP<=5ss?8Wj1k5yrHMi#aU|H-}J|XiI0Wj}k~-Mdv>7@Mh3F$7H0w6<8Ln
z(vjBc2%@u8Ok@Uez{_8tve=fIhcs?V&r88mmK-hz1pMW|m(^#o%9^Fm&R4wZMde7D
z+`?y#;{%y~BXdh+y!m(xZdLnAd|c|%nXY}yz{#vcf}8(l%O0=oYpeT9aE*1uoO+t-
z(_Qu#!`|^eIwsf`CTV*ZmBM$+M9k<IPA+bBZ4?+9%l&+wGrQ?U<6c^oaR5)Hs%e!h
zw8h+~^lvy+=?K5mj^M+MS?xdC5Kd=o;bE0rG$(*g!{T}Mw^RA76ZYNTV;rm+L5tJ~
zgiPFseu*Mos`C{Uj9tA-ZVax4+OERe1FJU#hHDkxu&%vTDK`1j-?!#PC`D@}*Rjh=
z_6&;@m7;pR0|jHg+^{db2wy%sF#h_R1D5Ki*HF0v_a!7hW_o1(`FnlQp4n12Ad0r2
zlyTg?x#fD6qZ=F3ejALHd=A>LdU1=APCm8bKxL5+!f#)R7TiFmxgYYp&kzpRN!afg
z1jv{L;pr*gW7Ws!|D^4~%@s<Ed`gq}Y@VX*zFI1#X~aYb^vPYh()<x8_up`=HN)=i
zR((E(iXGpkC*GI~bTZQ}@GNSrpDkA4T3mM`$LhnvS!_a4b1#~u-{W6#E=(`3@yy+>
zomRpX_xV&7c^}?b&bY{wkEZ&9?`GZFJv`hrE!30xL1Au?#LwveV%~BRRVsdbl8;g6
zYGmV4{gjX4sNMWXItRr>qd=Z>-`q0;H#%dB8Ov-w91VHKQSj?o?lHdVugD^Gf5l#1
zY8rO#del-{chhooq;bNJPq6{XmPRWhm&YLE_=5Hr(d7)cWsG|40f9fV_`(<U1m)CQ
zyDy<`M?_1{Dm`~ICmjPTd#mf)%~GOP86S?F+mvw+T%W=mpTrsCzf|hdmP{y#=Jo9t
zIcR6muWyaz9HpX$r5Fy}HU>YLftp`!svfl|4STKKhjK!`+#dlCTymqI;PWoMpoJ`r
zd+!hLjH;sj);0TN-ibub5wOn&r8!UO59;|P$<cYA&+k1mvd*}4aB<n^8H$AxwUZjZ
zT{1p-D0MMU869m&=F2sW;|IIapRwybg_h*Be<QB_DE|6KBUAO$sjaumDGQX6sh9z`
zCFx2K8hJi>G27A1+Pc#9wOH{}9Dak;`ayTx@XI+}*@B{D(461dEmw)kQGJ_cuPgDs
zl;a*Mz3dex?xubPPQ3dhGmGtGM*gSEj^B6}@Sr|C5{wJ)uDG+pEi!nt+9kK3X6JaM
z-w2|@!%yZ*TqbNSR!>mi^~H1cf(v9?*yW}tK4y94rpl@^*#ik$1#^}1&ehhY%XHS2
ze7R^6aFqjsvRnt%z`R!?GnPdAq3qCsE2#V+Eg^AxjW8u(RH)Axu<T)Q`iiE-ndsUH
z*F$sH7X&Vwus*H|5))aboH>^9cOwU+m6D$~Ip@|`ptZ=aV7xy`rfWoIuzm>!y1W(M
zp!#xOsUVmv3}dSkpAGXwP<WZr(28|D!r=?zXQ2=0n({x|7ISNFyb|$dz0D`;+~@Z~
z3~O&hycdMfvM#PMk4yLhx%gfaE`5AYq=fial<Xek(sVNyr^bIgvgF0B%44_3S*A<^
z8}q>mL$RRCDyxl4B}ypmbEee6wBVZ<^29A%uxa4~DjkD{iZRmKV&)aml5KadGKuJ_
z(n@eSRH)km?r3J$)JXWUl}!Uj%RXBw9uq$wS-<Ldea-d4s7vFBVawf1ANd=*hH5R6
zzI{{i66r{LzmMDOVi9ofcPSCMkD3CNzxc~7C{Y$ne|MF#ApAtO*(j!vi)!j84X}+2
zJotEd1YDnX%tjv}`hiwT>{VQjHHU1`i11)w+y2gszUOx~?bMo00vDk!EdBgK$@d>>
zU1;faRdHb7eIKn@(kk#tx=iz_qaXw6h)uyrL$;3g+dd-80js!+ER`*qr@6O7iO0KX
z>ks_2-J!E&EL2n5Kpgc>x@nigt@EWE8yu={k0^I4?!D?F14PJ8WY!^;&5wO%QD4&4
z&XyYO2`RXYuHvLt*^=&`;AZ{`cS|I!*sqyY(Th$76UZL@L?KN(8+Mq9`37180_fIR
zd#tHWQkDL>i?8TPSZDvM#=NBx4THPQ8)~hKWm879&LvUz$7vMN>F(u;8?2$TGflY|
zoiinNly9``>DnjO9Jgl(?R_KW=PWyE{eRo~4;(Mr&phdPWwwfjwvs&*bDQ{lo)LRl
zzuL>>=L8^VjylVSDhJ$V#i_F=i{c1~3(z#*zWNih(>7&Gq0|?ve|CFJY!fAh=?Id*
z-HM!BPnv@DEdR(hCwrpbXy$VHhiIAK;l~J*dCOP;hY4uvUm098@nAd9y%d-%$88yz
zU&q15F{yzkdUs45=nFqoC#YwUkZv~=$irop8~Y*nqjTC|+(#-`2B>#Wquq#C%G$v#
z!tUkzpN$h^jRse(mhe!b4$<lJZh5U{wvJjscVq7UJDN7<>DTVNvD6~*(Xk@-MA+w_
zkC{bT>$N+d!u0SJv7N|nyT6@2<X5M)M>vOZiiJ^Zf9G>FowOa;mXq;n3sq~8BCQnH
z=}*+32&#*{m_-hzy#mSnao6v9jRM(?e$8xPtdaQwbldXdf_CKJ=`Jz2;>ra`H#)&V
z+GZz|(~<h9MOiQ*ZNGx<rre=t)<&1ZC&CPQyvGW}*X8HjL(|L#?Y>nx(>|JG%jMw3
zQ)1`94pmw@J8C$LIkZ=tUmJ(?moiO@CgwJy)uwd|4XrRw**fN$tJb0f7VX{L1fwF~
z0eIfYEXa~?rQ}WUvKUqsjCES4l&lRudDpFFTjNEbig(B%JzsYGDI@=lKZi7*lilOj
zla)P}MxIOLFD}Vz`<Nd%K5zNv=ue;BxCfi$E@G<jA$_<ikF33f4=U*$?y-t3$y60i
zE8F=jwNEN<xm8=7LLPRvSDdjbx$S<qV|X1=>-BDf4FRvkM6rm?w>>SM?&cD=TB~LN
z^V#k!kRV2f$rfiY@5F;oc5ZflGvc+=u%oLPPlpb~1yUBk`n#f*d{y<w{Ir&+GYekQ
zs|jg6pS@6e5yZ|Zr2$4+QJsge=?9t-pQ3AOPgN@9tQZROz@!TN0WyOvnm@iabFIl+
zO+n#4Kn79NH{>XKZA~kf0}j{{d+%dR&FG#UX&qLIvZbGNe!QPY9MIi2B{X-1;4{B0
zy!7^lr_{|ytKRYaeH@ES_Y=!zx&K7NYIs>)XH1AI5xFaM-6)>&N?9;NJ&wME_^y1^
z#ajyk&wYPz!){N@-PCS0LpGpFh?42ro!h%lDCH>&cN@HZ8-3@td)##;;ec$VMPT`I
z)An=9pwh^+=%rXO*Af^-#~1<Uyrd*5bFJXM?V*y69W|0IH|!6?E$A$9+OgynT<w;7
z_fp=4Wb{?ufw8)%7n1>GCg|tYo1=F*tv6;8Q9pR-99wqhRVZ_6Jag9W*TW2EE2(|-
zH|Ca0XvuA%*TNr!Zgns(dcMYE5HwQX)Vl`nd@7Pr6he7S%mf$<-A?8um7NjTn##`f
zsV3hZK*Z0Nf=OxC!%ChMUwR^72f<^<D<#V2vZf`DFRBGuN1vksAMesOjy6ls|1d32
zsT8KAd&&GdOk5C&-@N*<!6qrK)Q0grwE8#Uv56gidI9LDk&E3E<iR07dgH^p%{wRF
zTF<}KkfAS<-ov)nlX5N3TWwvaiEusKk#nhAf4d{BkfBxHDtAVB;EvF|z4ey_A>fIc
zOZ#oRtA5g4Z<$uFc>J5`I_iF($AWE&3GB%$ZJ#b~@LD(eY3_~-n9!UUc+9bh%%VI3
z-=L#r-$&Qa%tkNV$`b9rsCo>)mjr{`Hhy{49oOn%ok}E@V|u~Hr2M>rTEP=ppU>x}
znOpTa)aPXI4*8EKx%bCOHu=kaffxFw&fq(}M4p+lk&Jo2i+^FTg+)rsGES>Bb-qU9
za95Y4NuNTA_!Pdu87ztlXhGrDzIV*XIJ3dg2ct9ZnaUA)ea@z>zqt4HCe?S+ZVl=p
z{&MNGgDN+-%dn%w;5KvN)o0d3ZjExjo~EU9Uhhx0yOV2Hx|*26O1G0YpwZ4bDp}Z3
zo5EWQ2GWznokSh3!0A6n57xLWeM%R+!xnbMiSI@e?ZLKZ=5M)x7ngx&86FA~mD6c`
z?d+?r(_$y`cH~hO%yo59>lhXXkH^vNrpJ#kne?v+r!4T5x|tR>6-Pu<DJcu7{Tc0;
zKv#?r%MEo;xD|LD5QsV)X8S7&C0FtWQF6;y5S<Z6c_jb&+pqz|2y?Lf<QB?oPLg64
z&9xy2RKRF_U(mIHYn4MmbdIMn-)$TJv-E01&XR|S225y1fxRDoC(qG&L%(nAt;cw;
zB+^j510*sKgV^MznS6zm`WQHVSR?8N<8lQ7=lzb%J5AggpLZJ-=N%L$(1$D3W?RNK
zTTUXaeVdm!mj~xNSF{iE{1m9JLg4xoW`GNjTkCU|GTFJ$kW0zFrwm<A38vTQJ?3;W
z&6i>Yvt<4?KDdMNhVAoZw&hCvL6$YadMB1C85bh&Utp5|v5QESPIR!4?)UE&$f%_>
zIM%TVH{TCGafvVa9B#u0)=(+?V?vqrb^&y*Kg@-xDqBAAA($*k*Lu(&j6IH^b570Y
zaoQayJ$LxyU_J-YkWc6h*k9awU+AJ7B|EL!=0vE4sZ6Uj(Xy*PrF(SI0=8xkI57%u
zBt6K_X8ijwqJrpBhd(6;J*OBUU=XI-`8<AQR$)#vf)Qfe+d7ivGMh)ga1>lX;Y)#@
z^R2Tc)DooXNI>(lE)N_19G*^O$Ut?$xOgDFn1K-(#gD)dUSV)G$7%=h1Q1Ra7q9fV
zk(qRm#~VrFtyMb5BXV;~cC&oCrDlBg#4?xAvT9pV&4$!hBH_%*?;`n!3fwTTh{V~M
z{FNQY?-9e3;J%P0YftSkf-i+0VI$1!JtXm+^F-FF0r-_C+E}0${5280>2NRWT&=bZ
z)~8k{ZoM8Y=v{Zmr&sCzga!(-I%t>YikNhFYE(v8k{x1JiXFzedfbf5?MQvKno(LC
z5x>;l#GG7|pWUb2@{Imq8Kb4C_X!ECD-n*hF}2SGi2Tgw^_MovMu6&E{}Ev>f0e%d
z8+}SflIu6YM>oOQ?<ka9NGl@}r!`mKaik)vpDJ-6uFEq*>ykm{!7bBjZLG%=jr0}g
z;C5yw$6cEeAyYaUpIXjC*V0nLXq#Qd#e?K$_;qxbn19*0+j6>))t;xU@p-IVsRy}P
z>H0$!@O)OC#K?Tq9<OlyQFJMvrtj>MeHR{<ZWT=0R^t6FkbYHMeCi&<VSs!qz}|2P
zrRKt-ZeM*R{P<oxaqWE*67tdM2y9FWiP>P#-b+ah=Htkk!kMo|%{Jg@<y-m51!^)U
z=zaCsd_R|%;U%vCi8mc3b4={VUMXAFokniT9wmD%V7Qg2I(WoaztGRlW8q2R2Q`kr
zPchJ4tG23%)8gt-ReyTfPG)%)Q@)e|Vq19`u8r)tWw=trif)?~UzreqnKv5)hu?8h
z4|`N|2wK1e_l4~j%|}3849HANNbgx&I#ZLsejI;w$2mhVqo3lV;SIcLNXKAu!1%q2
zgfEG`KUm(-y0_N9eqlJUl@IjP7;#DwEW!)V!ruxbt~{`*yP>H*8>Ep6J(i|-e79U`
zx~`#R7?J_HL*Zx8efUr#&*L3GFVk^z3`MIug|VsZ?vj0()8;52k<`aia%(mRdrCDP
z{CY2ahB+$BnuOi5ZWE7#T231z!X&D-N0=jdqT~1uA3*Y{dItin*Kvbww1`5Z)Qsd%
zF_4h2a7$YqEqi=#f-1i)IFgUR9JVjlKR8*JiNkv~U!uNDSVk!e_`qTG7+gkKHAlUX
zkiT5uQ0NiAi`b6LI$KObP~clRC;sH((j}_W0V6*~q57Ai)u2PmFi@>6%LCJaQoaT2
zHPasDf-X#f?q?H9k`Ws!8`mxvu58VX-omIBKxfxFOvVk`uiVqIyzevW3;gU$r5;!e
zC_WSpaz1HMF|h_<TNJ+)x7gub`mR5pHH762x5pJ@Z7f8UXSz)as5lW&&c&Up)Le*w
zhR2EN(e4=iyzncW@NV@x?D{rEOSik89<BBm(~i$>kCoY)Vc)EayhD5RoiC95ZheR1
zku@Ru8%ZeWXJw6345rwv6wtA0uXLu>HgM1+kZeM=tqkB$3I`Ru6Kgp9N5Hs#1*B#|
zgVv~mFtF=C(F<kpM&?di!@Eiu<Zs}o`ch=3P%&@k*K-`E<Ky`ROqH8=b+?u}6zA&u
zx2I0~x4RCI7M`KdJChvzysV5?Mu*if`cArjA+o=(4_c*Ht{zg;udW8pU`JgPXwAQe
zb{k)#ejm!no7v-x96ua=L|JHXMTtFyUzRRd*E}XM$F8pAr(k^YP7zN^^}G_V&;zP6
z;#&H)&*~EiRKjugbC^Y93B6xySb<xP29Gx?RmDl{H8DO|Q1|fE=-%N|ZTCWU_ssXk
z;Dv5Q)sK9#OCp`yW8g)M`|H(jI`sH#{W#x`ojnQn&lv{!jUb0|1eDWV>sB)^2yo4*
zkSG{>6N->(Yxz~<Wd$P*-FA(mwf(a%ZR<=sK<F;Go$4C-sC0qW*V>2oJpYA;vhQJB
zW*PVPUpe7BTyR!*d_~{g1dyX&63(buy~4Tm)(I6+agLm$!NlTkWy;ze&zm31XtE2Q
z4)5$+%N&*{dVS)Ct4eT%+$g7@&Fdb4fGM*ef_Z;dSN3-;J4M4toPG_ufWBT5OnecS
z<8}lj@{cSA2ds9W{5!Z1FU5gdnv`0(E<p(?`EMa_8{)!Y<o72j(-XIE7q!Vr)kV#_
z={QT%XQZ``U&#rS=GYTiV)9wMIU~k)<d?(&i<h#EI_n)*WLuhl6AYb2y)X%s6m<Yq
zoz%*#i{WcBTyDD0wxK25(bVDcl&+o%=IA!A^{<@kz3fOi7n~Ull_-9n$IP0#ujo6J
z#+0JY+BNoN6w)&9V!3<Ef<jH~RQ2NGD_XE8Z&HUPd5nF(t%HGDi%Jd;5rPcU$kJ^u
zg(C~F_%8}OhSgi-vh&DAIYL|)kb!^O{wu_l=ZDe6ur~5mh$Swskjd6_CUVSA!xosG
zouB{XoM@Am`8+Pk4WRJ*8VZ${3+5ezhu>-?K}Jw>EZi!)Q<-)LAM>T`{6K3TH*j_4
ztZnhycDFKkk4(0B`vi;mkW1P7jI~tNK<npZB6E)T1Y+yU1ZGnXAtx!sh;X9Z)RdVz
z%pke2qSzqe15R>qjVwJWZd)NCqGRd;E%g*mKzPo+xHM04hNzPL0)J?IO}u!u{-)a5
z?Ch_30%lRO8OeCE0J<NfW~seu4qLSX&yciLj|ax%ko^?-0oBk#e@c;7`r1g{IMA6(
zPZn&8@7c5ewN$pS{}b-@bL=Fu@Zl7louVxrn+oQb+W)qhWi~)bB<lUmY*y009@<X3
z^~Q{oW6)Ul>B$h=&XV0&lmKJjN%hm!<`$1UWnxgB{ULzPn4~s%hmucQ8!`*0sp;F5
ztO=+e6dnvi)a^Q$)94FuGEToLcdTUMIC_n3yEqwdLfZV?zJY~f*;Q*D84Nu83gXc=
z$wUcjzwN`VB#dKPq1%>)5Pl?p&Z@OjFlL9D!mVn8|AA=IEz`gV&eqpA5;-1m2g0mk
zD(>hVnGn4dTrfA!%;In!mVR<%;G>CHH2~=&FjE3m;%&I*(K(2_vlmk5kWamo25-vj
zjeP$49PoROKKlN~&L^Kq+=bXffNSe<3XgW?<lL(Sc^*(7F7#`;WDZ|F%NNy%EFM$W
zRmI=?M~Moc2d%B|>CvCXy|mJNW}QiVSx%2WHFjbaxzupgdfVbNGHcFezWmJvuuwqT
zEC+F##-8o8dBC7#GA2Ub545{waw$!Jr>P+vobXeY{qf13QbslM0>72%)X79=khnlA
z=jKaZC~z%0=ZEWZ?dxzRdG>i%whx<5#%!>6iS~9M20=bzh)A>GVx^W|g5}47c}j&B
z+y*}IgRLfe;BMd!lxq%U!HGOlXO-!TDt-y)<sb=?ZmU*x@40#!6tc=ynwBk4EK3Q3
zas;S^Na#KOuiFup)~~+hMbD<RmzIvNk;MTr!cnqV-BPw}4yWZJ%=%F*`dV(TtSZeU
z9#4+O%7TTf2{d~hRigkk#oTc>9HvAio&lorQqgoSb(j&N^49Rp2R7p5^Z8`&zOsUw
z#1!gsfRB5#;7Q*GL&VW5f{qPU?#3*mHh8LaV@LnnCcOr<5!HGX^ULxZ0-o_2;hVNk
zM8R#L@)_3ws$xe&=^L53-<6X)&1>%>QP@)SGvZB1%(1KRAz=!qz~5Jt55Kiq8gW@f
zJ|B+*f|#(e98vw>!qLx85Jcyk=$rVd8KX1?-V#XfdA!cUt+9&ipNvkvljp9x_tEt$
zXX6D7hrhzD6S(l&WYV+swvXdoHWz!05GwA<>0S|0aD-RqtR427V;i5SjLrsNdk>mu
zH|4@Xa2m-zfpiz&0MItc=FlUDKRUH^X6JM`mOn@JnJDT|mLU1Ir)U&&?@o$*dh$+T
zB{yu_+iyzjC5<>z?xCv`IuidDJH|+$Ewi+eSit8PAE+l_aiWn)Fn$rLLEc<<?+tGh
z`tTMx00@q~mC|qzI$qeVlvxT_F*@7gdMlJi+9i=N5loO7ewNH4S;i;C%JCmn0ag7f
zfWBfBs<%|a#mTf9gCk;Z{}s6A8;fxq2%AK`jhtcFe5flE63z$jmeXUXKsX<@z02H+
zi;DV`bs`0$J@x575tM7FKT(A}B{2HS<^)Z+*V~`MquHM<<0F)+2*-Uc{h-qCGlG{-
zPi=bijMF+dfH^pWBqp`IL^a+w2kJd|YQ54mE=;~T$U#w$$2)?~?!~4Q*{IrSxG`Np
z*J;WJrMzdrL2s6e;bmhA?VZwge+%94I%2=vI)a}dEc7>{qLo4NBf<eic4yIyn!^Ef
ze61ydL&sj5q$slR9D;?tVj|{oFA2^jDSWdl;tD=^_~Jphe1bBYMTrJE{N?FI0w+I3
zDI;ge1i!Lyx<*$cJeOst7B79(Sm(<ZC6}sr?B%ufj~0Gg)94k}=X5Vf>k<J^8ipgA
zR9#uVn*1z5G*}lGd-bPNQ-Fe<meK@L%~$o4FVM`M^Avf*WlAeUbnYK-)TjuzRc(F&
zCA$n=cCLxS1kpe7nx(O+5q_8Rs8u}XH`tZhQR&%V1gR@6@~2CwwQBQKXeo;S)-7vP
zI}A@yzDM-u_g6PM>Q8l(Ew~R|y4fK}B}xo2{qiF>(o)B1OZtBRd$g1NwTJ$bvhG%}
zz7(U3Vg-O~t&;QAP^8Gsa3QvJewV@5xTpTT3!6Oi!!v&bio!-e-uGa~Gc@n6W-N`!
z4Q#Q5xi^kadA?jT{@%E}f$OVlzZxd)bG-OmW-zBXQ{wvv`^4cmQyS`X7HH8ez+5a=
zm<dlEo0=R{sKteZ*u9Y4>r3bOPCFvgp$PjTBDw5Q)$+KIe*%4zeYly;=Xaf9sL;pF
zY;F>;w4w6Z)S)_gB>)HIl1M(;m41?6E(AK6v%Oel_6qTbPsSP@uEG8!kGxj*ZU1~#
zg&YrKXl|`}+%x!(y|LEM3ykRPUT_}fq+7m!I#KbW?e4qj+b2vXWS#`vE|t%?^tjpP
zv~zzpg?tFpAvvb-5BiP&d_Ruhe+lGcoZxlNUV-3(cs!;LCfz6CDn=(gQtxO{x!^D*
z)*vnBlkp9?LcwLNw_2%=e-;k}`4hy|7n9t@lg<NymI~~P%q4^3ezokFScYKevUt@&
z-e%m<NG`*xxn^_iBrQXVekQyLj!`nw*cqicjjU)z1Po6jE7#cI6bc2*Q(24L*$2;d
zbkhF4-CC2mGv@n;{`%n})5C~=RHU=AcSr3+{KIt1gjbDX>?53UVeX34QH7Bgz(V{K
z)x6AI7OL@34Y17Hf(@P92F!ED>Mb9>ISe|GhVnK4tSkCKAvr)8IwyrCL=G|bWseY~
zmfI{SSI41J^DAmVUmR8Ok^(>>o&r0mX=nPTNB?B+wX9xi>w8*_hjI|v{)(k6h@lY-
zrG!sKw#LrC4a_Cz4nw-#cJHA&RJNsUg{Sn_I+raWSM9Qx*^I^O?!;A7p7(Gc%ewWS
zq}2k8_tCDQC$$yN0u)y0Y*0lD?N>x;f{*;k$py|yM5do0%dD0bb}JMIgrO?w2bKR(
zfoi}MtBrO<p8j&tFZQi;@9RlKHbc<O9{HGJlm>hZa6?LF652CKe(3jF+!=6}Z<MAS
zkNlL`TlGv<d-K$Lm(>uLaNw7j+grGVk>OHae=mrRp>H=iIG6Vv#sBOwLHZhzfEsx7
z_8ncAEW~HoZuOPzYXCJae9Eq8sh&Ix-~1WM=Xi&7Iy7HV7C`g4)SrB>@+^#nclYHB
zh=7w@$M>8fK=`q@k=z7wy662u;m>A9MHSk;66PRSRdGI$a)}J2DE&J3G?J843KC<_
zVA1MqU#Kf^v%&j0QS6x4sDJuBvu$<b#C6?GyE`)MI<ne04hR|n+1OP=@w@mIr33Ac
z`TXS!YBcvuJ!i9*r+laHhCQ_hRcP$<!@^zybNPGm&*)tx3$lD=X53g3u_tW#GOQd$
z9LrNn*3${FWkFC5BSkF|>Q?AUBL?PNuaVj%7Rz3Ip2k}5<+Y#Kf%oiEtnIxg>K0R%
zzlS9-imfD*>&miCYl)bhhLt`XAkQAtq~`RG<g>DpJkj2p+w+bzqGbC>kfv5!O;<@c
z+8o760@hEC-QeWA`o{0@4XhP(Wg7L%*k63^Vx29;^^yI>p}AD9cB!ZBCupXf9s0F5
zMW!!u9;se9b;8eFA4|6_-uG%I1e!7lfu^vg6O{GO`roYjv^+)Xxw<GrpF9p-)Y;EU
zgNj)K&k#qp)bC6LJbSmgp$4J(m)N*OfzzWvBDpb5T%*XX^J4nErXMI|ct)^NwD6lJ
z<=4F>N(Rgl<HTC#bP5Lf3q_9*9GB#P6*5TTDjy)dgQbhypM!~dK+9>FEGrEMiQBsY
zuv!X<0W1fb-XPmzFZ&8pE6({n3AFISGqR85L|Vi9CJ3*F!jvkGF8w7it#VCw64%>C
z%^~J4TyG+le^yF9sG&dr6&$*A*>?STI{Cnc<WYR+p53%6f=HsL$;BO6Ll@{d2QI0u
zNHGqa6L_LKAC<$oXZbcq%MO47VrW_M^5%B`_1U(Fo`bc%I`SZ7LMdXhVhIj<XC=qz
zBjQY|NHnktCT1Nj8jDy$ssopbn`@MI16gj*6$FX(K#kdhh@NuGe<ZQ>Vj+-yPHlZ#
z`QlY2jSo>3p;hdt+IUj9`{@$Z%Sc}2J$QCEaKdZCX;qhd>&=*J5v<O?PHFpbZy7rL
z|2eq-^B!qu;UWYtJr=>IwjURR?@ENqy2-<aa=Kdm@KJM^U_K&>VxpVT8RcV=Q}Z1j
z%0)mcYj<B>x)?m7LczkA3BNMjDnsYpSc#wRT)WUMR=UI$h6fxG*r{uJ>zth|P;MlE
z;Kllf3_@vKK^sFcj93selX`qSiidWfi#6<2JaIjzUruK6QW@uyT)<wqy@jAjf5NeT
zwqIGeGN9_J1ftJ<T(ElF-ed$Vn#$V!I-T=}bl^X_?8kjIvO*f3<HfB=x%e*PMi9qC
z*SM!`^|bEXh707<(j5I9KnUllj<lKK^m1jQIysfqE&dj+GA>lwg&*PS_o5iVpBld1
zQnQ=&7pnS>AZ;UD+s9{9B$5Bq_aM@j_7j1fB-?^<w=B@}d-Y+!egb*tklFr$<hP$r
zjCW52k8LT%ZYkC}B3t)#q5F?BO;++rsDtAN!7M8Q!n+z_LU|dFsdoj@IoALodUf<6
ziDIVqh@wvAlhm#LWUObIB0oCx*251sHmBA0nInnsZEFsBJ9O+D?*eY@(U@ZJR)SWl
z;($>DT`aGErHdu4l*Vz_pq*3lvH2GCsfuKDw#NR#8=@sg_<uiQ>RHiYlc8+F8KxO}
zmU;;}Bdxt_c&Yc<Y*v_}+Ad`^E%3sNLoGLVppik7PoXjJD*e%tOga+v9c7MHz}yOT
z0>O;}8WXO|&o?1fp1}R0@{)fPR@5{4z=DRWh4Ymt-la+0jy0>?fBG0bg;}YVdeku@
zw?K*y(bW3@Yp{Te0SO`}zf9cG?y^hP$-=DswC)QH+7~;w7-mm?DRtv?9C4lcFnnq~
zrUp2Dy;B`}=`GE!aele(sLS|AQBV@-NfxPUK|2EXKbt*^x^6_}bIIK1JVdG|Xa+BZ
zxw!TxCKn3C#j_|yQOyei8y}d+e!v-NtH0|9<t%{inxi2G`}NW?fRMCs3GDR0_~+Yz
zx5B81kn|CWp$!w|*>(8#VP}4osJgS1jMyciY4<riK~9n#Qm8t!VCy>3WVvF$t+`SK
zfQY&e%xkhS!DY9`)I;w8&7*>hQ#s6p&CLBlQ;%s4z@f@~34Fvz_j+jU(p?6BFLD<3
z{G@YzzVN-`KO%<|=5^?UwVY}VtIC++*BM4h-Th2YLOw;W`5L|AsQxKNKaM?fiGXmX
zAzYpTP^btFpzR|I5uKpbi4U$Mze@@x#lW(hT6Fs*``X!mZWDE4iRAbAhhLD*S=X8Z
zF!5<7@CpK`9VW>e&>c9wI@J<!s(F03ir3DQ?glnR^tQ;9p4t~%H`43z9wss%^FW*t
zk9UPB+ukU-DZ|jZbT#1fsnrd^Cu(u2F?WY?94bDz@NfC0@NkmkMd*C)gul0fVgV4)
zGT~>H9_=}ZHB*eD1w9qO(i&1`2%E$)T+zViDaXS4g`^^Iq!|iFE76TnUWTOCQ`E$U
z+LK}*mx_VX>n(5y>K{LXhLj>XqtcvQ{y%r9>^K!|OG18N;mi8<#phLUdc}U_XDG0`
z<1sNVhLL+@d>=8e0;iEu{dyR;n|`ozt5`qEUG>)FKYCZ5-W5E|kalf4XS4sk+VQF7
zQ6ouQyhPY3TozTY=hZ9b{(Yp2Oz9@eK1e3&&&$9Md`Om^vAA+2tqh0Fiq=QhdBYTV
z7)^e7?-*T$_|@o{(RfeDX*fp$UHV3}l4;sKXup0+3a)_9SgE%E8mdfk7bgiizs}8#
zAadd*I1msZd%`pLXx74@ip=)dxaUxNE9Y|AP)Idh5ZCI3|J(*c=26rD`;kN#L3j;A
z-noA|&v|00$HYh@FXNuzbJb1-*SS1SH+5EoZA`_r2NBo)Ud;{ET~i0Y0m%8(Ny;NM
z6Q$*SWCl<8KDRFSOv$|r;V3%vJan`79|6P%Vd4|5Ij_Udc#YL&#k!Is$C(-VWVu0o
z@6R0h177@70t!rk_je58$QrxG_HfsS)*}TdVh}^(BR`gblu-W!F7zDX@^WUOWJ+Q_
zI*~N*bwslFAR)IY(0Xko^c!C{C`LN<J$(?tNCpF$Ie6^)(ZeW0AyLdTD3J2^U%vfy
zPxz@#)&n3aXr@i=&xj)aIO=P-z9$v9Tl+*0+qQiC<vRNFotdU+W?tc)frH}tDWCUu
zhP$)y_<K!V4gbf+f`L;@HC;eFCD^Rnc8~}=L8;$6Hf=(TCf!9*H#?y}$3i~zkr=zX
zo=ZOJIz%GOY;*sNU^ijak1HOdyy&lBBUPVAQ2xz!%K6c^U?0Q>8yo>rUb+$^znjWs
zek*<fZQFExMVns^N}$MrQRk4lG3s2^_bzKt^K$8p>>sN?`5~%!GRyeYPx3DXjQ1KA
z!jZi_%fwUm0~yu%JI-$X>q5_>2D62N|3n%^TDHjEwv5Y+bUgeC&!w0YAMYjM=fR^v
znqN>N;>_jN_bvq~TZfKUFT?HK$6QC-Qp{q+TUuqbeXS%83QT3TcTHwTjUE3F$B=(7
z)^r0EXpc0=lp2UD!OQq18Mo?q{#;rq6ZZ?7{V6)rAwWl2$@F5=Vxia*w)P~{S=;A3
z6dS&sai&ojWbY2@(Wg(>=uMyWYP7pb^H9)>#PzrK!!_w=TlC^q{<t~6+WhHfEx!{t
zZ%Tzkt)~|<w|;r3GKK|Nd`2#kAN*_2fd&y<dC5S7b;kWcI%0Xs&UGIcHm4xgTCy-P
zapud&G2C!_()+We_`$uZsv`rs;u%9P_s0Y!8Pn3Ga&Fe)nfvg+ThQ;hu;sWVWu|01
zu-f<@RB+W?^5O!XzI)EjO*7s)>L|T?Eli0hNw{P{tyE($%!i?SU(**)If~|+#?8^W
zLL_Lp8u|(5aBWuUx{7o-|Ms~vT!mS1fyny5EO_xxrh)vO<`TSt_#XPjrUzj~&vS2`
z|Brm*i>YfDW6PE1?3d8hoyM8k%wl?{Pij3xR=KUF`Z9KWlV9xinOnzB|58Ts_2KaJ
zf4m9uqV2pJ5UUJp#A~5*X02M6VL&hC&H0`3f8RBpvfH#f^2Q(idbOncZq>-w`~qFe
zyz#)PeP2D*N%7y_c*-NB3M~?wXWj>HdZAB=BsE&07*(IQx`n!Y{#o;X?|!K{Mkb))
zNqF1%){k7%?wS2-(Y)!awSQ4^v<$t5{4w3a8*lr5m~zdyTbl#Oqg5(24W`d%>g%c7
znGO~<C{VnYn9icw4PFA^X}m~0fb7km=P%c#@OYnJvE7&7A9xRm3HGW@|1)JVcHO5?
z@q{k)q?xx?@t;2;A9#+?ep|tHEdEj0Kk+O%X9w0~&m6yBv1`?Mr=DrPc~R`l*9vBn
z?9FcZlF#3$lrT{|VVquLp0{Qi-k9>)j{SDGqb@+O<&F6y5^&c@c>2$q<0bsXsCFn~
zuhM@r^8s72ZUb`Lqp>kA@};xImorq(Je5~WH{RjGCy@((3l~0^Rw}JywESl+!_*t3
zFY`#B{)_qRP$)2aLD<?srNSHCOKzT-p<FyE&&Wn->HA-v@or^(wCpw#V~f7?P+z<2
zMctb5hBUMZt8JdAzkP0W(^(?D3so-56VO+M5*cZ&ysmUi9~0a4@%Ib&K;OI_Oh)@$
zvup2_loh{K@8oe%w$PEwzgsYUCf1e8qn5NV<k+q5T1u2A-OC0+yA`HE*Q0LEct<r+
zY6Z&Me<-JX%bwj{DbvqlimHdnMuIi>Z|?gIuaFG6gt?!Y3GP9Ff@<B2^+yGe7qGmW
zZz}lry<)g4XiRe|JAV&9k{pwFFBRa_UwRT(<&BWdF!o?fH7{J3Lv{3B$QO(64A_cf
zFUGWK=l%YF?;_0n-l;WGc0&=Z>b-1d#y;c47DZ<lnpI14E|iWyJ@?gGS}9A=xm6ac
zQ0cZZPaS@QLJ2wyruLT~La^=7={ln9CvnfemHiA+n*4qIWEoI>@7J4l-CaLpvnf29
zG;r3wX-ZPVrZ1WJGWVeAQ}!j7jMp!0%b+|inNQ@A8Jj%Q=Kp?>43e<kuU$SaQTTmT
zrv8QcoK>>F{rHj5v!@PNjFvnpV3n(0;jcTyw3foNda~JI!?xdF|6jU^Gxu$J;Tr|~
zD;Hireogt|@AsJAsY?2NdCvP0fepqQo^YY^FEq-7V6prU{jmSIwUvm_ykEz?lQ{aX
z74KdI>#=yZ5+~y|<Ti;K+jAv6Z~DcUNiy{Dna0}&5-oZsKRHaty#K{PMguF~VPYt;
zi0xm-Z8`Jkpyt&tMIN~!x(N?D@|-R6t62Ug*5W@soaYX$&X~W`?nj$_X0Bet;L|m<
z_)UZ#=JB35fwr#elm(hACFogNg{FXulPlJc?AD|hM8NUyKKt2983`m|vDaH?rhaE$
z$*Fr~qR11sa^u@Y0r&s4s~YsEy9m<$OlQ7&>NyoAzchaXB)6fNeI?Cn>p7qn0H3b;
zPeM3z9ew)bLETQOTr)db=$0S&)FjCNtZ{9Ai0)U_v7hZfE5$<;6I|(+@GJ_Lgl0k`
zrr(^p8u2U#HK0BX0{MPHg5L~ZVCM6tm$Qs*QXpU6)0w&YsCHUlJ`oyJI&+cf^|=Z}
z`%is`DJuk@^XT2Qhq`!BI$zLCGdJ^_rsXply*w*7CdK#L&3q~9a}He3pT%kmrv<-x
z?X+n9*Y@o{?dX4{)|9lRmw0$I8%@SR?+1qLEVG|~`M)H8k|_{&Ibu$U)MNV5=EoXk
z&*w}_<MPw{&9+Xh**_5%r)^k3#s&L2@O+ogOf~W!PLSG+Z}EN|u4m9a^LGm;L~Xt#
zL4)rduAaE5_U#mobD4%M9n<tS)6Y?6ncip~7_kOvQ*qqY%H8@NzlG&L9wU(>sxxeX
zTywvp`i_}eFgl2Wk@Tg%N<xl&UFB09yRCgyS>Uw8ro2xfV!BmZfEZ1dHBMph8LaQW
ztKj&{=bE{Q-wTdfjKFhM@9yOHyJ!9`Vvr>w9&1;)4;WRhi#1+$yrT}D%PU$b)~5lA
zQG2GJ9W>{l*wkt$y^PRzu18`O?jiOs-##$IVfg*HnI<Jxm?(AasnVRke?^B)a=ho_
z&leB7A5`UT8sui-bQ@y+zo78omBwBb@Ta_)PR@p@(z8vk!*?}I<(<pd9JQ2DxEXxc
z?mzG6j0Fb4pO9+~N;edlxwmD%Xp6_}4oeT$kHk1M#gCnTdOpr$)&C^a{ZBW^L}OC1
zt_wN#FMSZp%+RFfu%DhK`q30Vq1GD{wR7s#qST?lt*-@?mzpz%_ErXJ%*em~=e0&{
zq{h3y$W5(!^{<lh<wn@Z=PTu8c%~FwX!!Jd{-0H&#p8z$OAm~WOe&t6T|4-;{3_?D
zl^Elr@X${=l&j7aY4J@-5|ReU(iTg$%YTCg_@-s|-?^WwChzAR>zDsBoW49ZuSLff
zU)ixkCjEY7-o)~sL8ASt|9`R^lEW<%7x>qzswVKZjJ7OK88;vMpTtW4FKr3v(lz-t
zd8lF2F0uGcCV^?Jbh$hLfZfa1cgCY0-bh&K+f1)vrn;OyE&&9W-me8M^7LoQNz_lc
z7^b6Q6|<+**qKMa-IaAj$`>y!Q4vH5MrT?3(*O9qe~W{^IuQctJNjY&5`tI}n~I~@
zL!(f`e!Dtm=|NBN!+Dw`nUl?KOw;<4=}f{j>>njX1klg6!sjLQ=gWkqzcJ#!b>m-4
z(E#uc%fGwv+P^&;S>vXA6Bi!b2QB93+}YI^9JCk1Q*HLzvL$gEwa&8NGo7)fUPGv-
zy6oW3>q|DYY?9zxJ~LVS*Yo~t`DQ4P9VQ=}I<o&MIGEIYTj~6APwteK`F6&y)4TxB
ztiAn`x6EKVKd@bE|L?7nf}f*uRtrzmvwGSe^1oT&p9e0A&Ybj(tvUMt<s0iV-sOJr
zd31L5Wd|*tc&!HYh16iy={%#|*7oTd4Yi9f|3T|Y-tUF*Z}PpL|Iau7OSJzi*B8ob
z;dKeGh%GZU1=SohU&pY<{xjyRn#q@Y7Xn2|PNgiMx8-Ar4xHAk_h_`{O)DN(qIeoZ
z+ttnB2Bck?jt~B;6;sV8-GIuj2R*;{YwEQqTJ(OV%uO3sp?~BKJdk){e>p90qnxm5
zAl@{P@jeJ0$l@sdIMt;57?y{^sFHGVP4oM&o;dlcrrRg=`Ow||3p0P06`9*x$vzsi
zzI;;(eRH``Iq)s=ps4#;-v7hfcgJJ>wtwe}GO`jP8ChkAh>T0JiG=Ktky*%=xEhoa
z%1UJ`dqp9nX%<PDMWJLwBAe$pFPFaYz3=b+{C@ZIdj9D}Tt1)kJdg93@8f;6MVZ(#
z(SY=bkjDS3E5G*7p~?wJkM8-T`@eS^ebL;4S>ES&aG~!7wXWDq#}lWQ#^)I|{y%8d
z<$ZJCAktwosO<iQ5#H~UYj}D*<^Pih`;QUiDjN`RB|sI*4SL`Va%D(pGagg}4rC|l
zM9MZ80Y1e(U<J)H_FY>RhExZ;&i7<1vB@KBYs4tY4^@XV&8^g-yZHu?b1+I#+qk!j
zhpsuMc-7u<jf88Lg#+S@!-e+Sz^0@t{W}{_4P^-HE6rEhRe9j5VX5G%t&LoAAr+cz
zmF+tyyhfu&F+ko9-HCXGr0zvXQ`&8}4Onv{aRL-qLFC6ZLvbuZ&T!L!aM36M)A^p5
zphD?1g;p!7Es)LSfQHux*WeX$#oqf7U`m9`9!(MDZ*S<Uw$}7x`AH#PNuHV$o-3sD
z{Vb$wvzurZ*QSY!QAlcHI7{4u5a<x+x3Kx|Bm7lLKIj8KPj%bwgfe(Njax1T6MjIr
zhry7k5^vNwx^GS8hFFvxGF6*4DbSIo39LM}eq$2!+x~G<_P)4hLT9CUDi^FAsTg;b
z8*E`Jw*PmK6Obd)t2P*Va`5puqMZNyJM7K6kSt_=ur2uY8dZihBBCg5*jk58`b1Ly
zK^Ss%GaT>N91l%xX|)xjEd34hS#QVOiot>s>)OyOX$Dp2$RPFc<aYr>{5EVjrKAn>
z?Y)p~5|v<Im!7>zvwQl<f~L4G=O4?1Y_1q&D^J+%7QcXXToJ=GWmJ0j^Oqv%;6aDj
z!DR+ws$Xm_lG9(r<?>lq_c%)!c7IAMy4@I}>f-l4!vsVtK?~{O#m+XpcCB&F-POyP
zfDq~(^()UUo)~HNUi09Li`B`n@}{bZvDU2enSzqO27aOI|1O<H?7kbrc@ujPh4!S}
zJJxI>sJpU|V8M&YYqgz@K{e)k(IHV9dml(uUV$|E*-M^l!UtkeA=lcA+AmJ$BNXp-
zTBn*Je#lh*a7gl{B%}_+XP&p5S$S`ES=joP`}*!qq@?}L%zZhrYxc2O42sX<BSymg
zgjjK9lW~q@rH6A+I3u11HJZL_cVgBUSnO$1mTn0Ad~q!C)*@3WwuHzi=FW9#0@pzl
zoKW)Zh(%&`6kl}Kn6KGKdoA=_=T~o<o&Nm)R*P!S<BK&|3er{+z&m~9Ogj7Lk7?{Z
zam5>@uCf+-opHFRWPHbHnJf)bfCw*>COtw<xSq^LaOIR)WnG%ifdCq9CvjP;J8~cZ
zOSXD{yLe5(8BSliP+!?to@^Tw+;^X}CX~XPw_|`p{0i#ulY@#SB3b}Q%>Metu(#V~
z->Recv!}<M7xv*x2psLn^z@$m1n8zsD36X$9!Jg3p9Z2(Pd69<Cp5xGNKG*-pm*HP
z?b29;-jQ{u%i5y2AO0OHWm+}<IDDJ|qwAgaAu;SSna#A~mzTi=)vR;58W<<HLjywC
zYl=Pu{6d`y+#wAs+E$dqwV6cwWq+HTdaj@~l?AWCj#AeZjWL1}?0+_({Rf4?Fj(B`
zI=rSDT~A(mSU`hoBknG2oBbTedj<TFHMe3=zt0Guq<sSrbB8sIZuE7KzzZ1kvqKxf
z#Gw5SV6|DH$*6y+O^An+NRS`?M4@dU3#l#r!xz^?tk@fnQAxJCF}{{Xi1{#x&NIrJ
zo9HsiL1f>l(GzNZtvayN%*V6wc1)LU|M&^*mlXMdQO2{eOKsAHLlYGD@S5w=iVkmF
zMJqzd3C9vz4W=s@PUvs#41(Md@8JIKBb8orLR;YCG&Wn>ljEdO9}by0p+Og`PYi%b
z^3;Q9{YkGen~O=@<E}#vwE;P7yCM|ts2EMXTD`yXjc?^w1V5<_l*eTJYaFpVd=W>w
z=JsNf9s%@=+3Qs1wl&uz<6_Y8bK~9F=dllfhNDPU&Y0kkLqgh*Jp@7DpwWXCCSo}t
z40mCpXhYQZ2)qjl`6JMn{RP7wHwa}9Er~ZCgo~1q06X#SzMifV987b3UtZgq*U^D$
zDK(|W!_KvLT~z~4#<-}SyLmsr0>Z=$j0pzVgR$0~Jp@pla_<3asCrJS_%W6AVA6Im
zJI(pKE7c)(Uf1djUhDcDZ0UF1I%tQV^EC+G{$HwF*r^At?1O8BSCp$)davU7<UPbg
zTmT{G^?66Ih5EGU-G21d)WFEqpL1nEq3{BD3hu<(tu$LKe4W*P-YO}7nDX;imAoDN
z!jE+DHwt8?aaGS_pOL{fxhmd$!UW2Y!+K+i)SCT@AR}UHY<eyNz4(6|pC~81l9*yL
z2k{UOypp8fYTjQT)fLDug(78!D@R&&nxlCGCIF>1SK8lBk8NA^vjKw(x*1_`@4|t$
zf+xP_(ar1tL$GgWKNG(i<KqA7T{_S&9*S9t#C`&R!L=6MOt7d-TVVnCeFFJycESM#
z#xC<SKl{wtEWTG;u#}zLV(yjf-Z8?9O`;}vU~HS-C9gaadm3K0xw|jh`x|@_QB}yi
zVr2-cPWl1K#^!<_TqSXTx|#rrmR<{rCaBIKK``!>D%eHz*C8Rf*yIGBpaBrUpOtd|
zVCVp2%d++w)U)dl*sOL-947!<{@mujUO^AFhEQgAYrR;jXD(v$==d*O55v+??57t0
z(L2P9J}(|t4R}~K;i{SAsP`3mYMaFu`Pw28+6?s}oqD|JB>-eQi2LgWg6Vo2?%ODJ
z23VEK%D0X^_XRxq>+I}%JAiRL98p;AeDGVe-gkvLK79Ca)D>S*pELu#Rkec;hpi-7
zSHGq$>*Cxs3o5>fl);0>4THP?{C`m%kSDzC4Y7kfZT8T|i#U#)82Ut0_oL2HXw<JI
zakH1VK}}bupwrF`=mTGF)ev}-&@C=65B*T6OZqq{@E@oX0k!7MNwk@seu3a%s3-6A
zGNM}=1O_-R>~p{iG6VwWrn_e)pbjI6!|ekRwbNdo@!syoh$a!Cl^5rxoO(WtegzQM
zXT-b-OnF~^RYC*t4`|Q*GGGG8KF&Sa!~C<i1f;y~zmga(&$fQlIXb$#eFwTL8IP*Z
zewC#auAFjn9Gw}c^|Vg6Wl;Z@LE)c!@%zvnA!H2+-ZxaW4*hTi6sHio1X*zs*boY;
zav|jQV84*Sw3Wzq;5o2{UNWVwxfjA`_aapDIR&Cn)~Fvlr+WeeN8t1k%Vv?FnRVDc
z1k?`&`ZifFfLT{8pgs*DnnUKt?}b4F@7}}EV^#{pR?_E$JUY!LM~$6w=}}{^Gk~;6
zej;kDy=P{|&&~6xbwiWJ>42el$=qLFxivu4MNHbNZP|00Ej)HvinC;s>DgAly@OXM
z?*7gDUiF24uNl5*p1{b(^-LVy@aIJ%o**{m{o>Zn4!ZyKMnS0&X?*7q69YmQ9tYOh
ztAHB3@668#P;oSM;5c;h3*C|&X8I1c9f4>O2~looCbVH%a4GCSO|cc$>cD3Kou<bT
zwIcM>?jDOvaPEd?ze3<a@;YEMIrzp6I6A#F^2M!yg{XlfcJQ|Rt6yGuHTYt8Mt-$6
z*u@I4ivgPzMu}1Dpuh9--g~3(YO18xqQU=MAJUBEcx`g-ZUdn>NZ5xxq@{;7iS#3h
znKi+)e6mUNG^)EKeIr2~q1&OM;Kd?(zdjJoaQEeV$4o7EiXH~KFo{n4)ES`u%F^}E
z#8Syl3ARGJ-=-I+{Sf+Qo4W20r_Q{8SQQ=n{ze#}0$&2CKGs_`e)Ls#EuQ7DRiAwn
zOZ8aVvC{Cz6^i#I;e5Q!Bbb3D*8F4pNkByheEG)qFGUo;F036|jffbQ`Onqh7?7Sm
zzA^6+l$Z8XZ^K5eB<F^>psSPZ5cEBz-n&wPXzLV%w`3m8o;M$b9vLPeT{u-vUyOjg
z!176t8U~^bjQ3nWdkvapP{Z#@-vhi+<oyKz`9!>DZzE-AO?GEm13Www_0j{!*=>qu
z5!?DG)YIGtn{o{XPvvt4&fgMyy`S^&{!bTvz3Og4?u8Pg*)}w~$h`~rLQK1%Vb0eV
zF;BFEI3o^U1e?TdW<~TMja{*J{y+JO@X1O~`k6YZ@EZP8O{%$+kEK!LeGB;Zn{q5M
zE-QBB{I@?X*7}`GQbu;oF!g)(s&Bkv5X<58M=-~4U@6G=Qy2Fp4U3{r+86+>CF><1
zkw~c-eFU5=eL$CbHaxuHDG}1|h#2Gtpyj{rI{{QRS5O<od)TImNl;cV5mPMCKiHZ3
zeGSvau>iL~VkzKh-Enk49Lx&j@5hBy4}I-B@HlXO^r1D}4b)TNefS!1sfe&b$4W(U
z-ftVgeG2Y?8!#YzO^@!(u=Up{c_xRnoa!;lM94tp-`!sODHOio2kZHrov(6WHQY*!
z->^>n|J)Eqyo*{g^q4@cO86N&3bLv~1Tne0!N!NC<~1Qh8ljH^zBxhY<Bw&P90Lww
zVW#$^8j|<pK?FqqjbnXqKNt~HnEIxYQlPiwBzry%+wf)QEoYFLacD|<fsoOt2_8_m
zg<XXgw<7X3)0=muft{TlSkB)|Wp>Rwf+K9}|Hu=IK`+#*31%+-NcJTN_E$)z9ToG|
zMf<Bx%}N+I3r8G&xsYa+3&1|EN8JQ1;rFlU@DsINF1c<%jT?dEw()i@#xrRcYg>W|
z1F{JJ$<jX5-G+!oW##fdZiY@U*%%|cn?SgwY<`pqT7(c$>_iK>OSkJY${)aLA_7yx
zZahR>C>2-fPjU?iBjQSE5+B(40~)+v0&m6b*!RAv%s?s~d+gi%#|IBvCI(PQhZ~|;
zm>H?*D{+`zW;3`S!yj%Wyh*6Ua^Dp_zyg^Oq937+wyTiJ->dlJ3#>5`>o|kJ=s$xE
zB1}4oDR0cL!dS>wfmlD?J<N%=R6MjAOcf)LSo?+Fny-;x+KY(RSlNz^ZA6mS4nzx&
z_VvL&TtS3kh-&NTVJU9Z-%w%pj;cRmn;gu~iW7IxHJ1h6fJ^A=3+Mh+t^~3Pmq9K6
z0}nr}DCyklk0<0A^0nu=O4ReE?G_EsW|HWxbOx@2&ixU?{uWRiGd{!(E6-;}vi|$@
z22mVNaY+j!h{W^U`Md_acp9Ne7Bwvz!aEVk;R~1BFcS1L!22%&&2z<P>!RyF0QG%s
z3@8zNU(EtZDhq1tF*nW(1GbA5BP`%R?A{LO=RP3YJ8CXVEjy{>T>M7a*Wr%j5L0ck
z<mhO(7aq9+7+r9g-<!A)w-9=hTDH5`u1S)uv7e+gyP+x&5=;`$w-)v3$PP={RGY;f
z;RZO<tJD!+kcWwFAj6YMfQG&Q`D`N48@U15)v??18cZqx*@pQVjZQ6)YM?{iXJ`fD
z04e|hMZNL}gnS_DBRH9#Gw&q~0f(RsHxTu8-fEM=XZUrF;wl5@3@esi^Jc#DeOEYK
zGoc~$Z{SvcZu;u(w*vfzn$oE?U>!t}5iXP&)_JSF0xX1q#lsAxi*St;BS4Yo2F)Qc
zE?=QPl@`pwsYZ|&+n;uOp8;R;XP8m?+Nl%BL1|9hc2h=+soLa@dNMx)g1=FU=gE%i
zfxsc@mdSVE*E?vYjUOEyjFUufrgIwi>4Yr9sft^GjLQx;k&oE{ZvU&z1{YD}jWRyP
zt>UdlYeZ*t6?kF7`2ny!*-BQcmC#7N^1lWQU^?ql^T&vn5@E2=&6QhL#47?+ET}E;
zwnvNFUPy$Wq}2S2dH5x4eVa>QDi<Q~L%gS)`DKXbodpQ$eI3#z{B{79r8hKL{J1B-
zxCadw%nD7JcLb?o7!~Y1g70YuenNoU@fr}}S_$gAHN+fI2d#PQ-`4}aSV}|=Ju};y
zHD-i^i#k`D9gU3v?0=CsbLWa@FhRnjoxlPUT&@5Go(%8_Gox;EaKMGp<wvy66JnCR
z?;)DChURWx!N<Tuhghj=5gRhlD>9mS9AN?NkRF<AZ;=b~pH)fkY0QJ3t*X?kf{vnU
zUX9O|7wFHr=L$!H{^eDlS#EzsH@-V`@<-p8J6AXR2`mi*4Q@b>R;O2ifq2$`Ls3ER
zlcB6jKJ4zhW4mj)Loq^m8A&WDm|gb-&eZ}#3aR|Hpxxg6H1ylZIO-DP_qqAC+_;(0
zIH~6yIPdFmH#;jkTkf65y+=n+t(|J}$B+}}t<6*E$yy&yNUj@%L4@h%-4s+18nhXP
zNcy-oe-m99kP%PnG1RICLqhEAUjYlT?O=0qaB;Um4oE7vx@q2HAFtoMc{8>1!UbZB
zpsS(u{GoICFgB@!ZURo4llSm#5oc=^_RR;noCaC=@y~of-Yt7F`;52rKD+tb2l|;a
z4e#nS4&6IW40d*7&FWkZcKkS~zyFiK*vt4+Hjs3~+8=EvF4Y!yEjSm%i(cQo{hRp8
z>l>58w4GyiuWvipYdk)?Xam$^_}2$*2NS8p?VGWpqM}>$V}#6%(Ns8aGZLeOEj~uX
zmb4UFs~D9y>c7t#6MK(i89AAsd^!jJR_%O1Z)8%|!GzgM+dRe8#U`XA(!lgofQMKQ
zKhJyZoBg?vpob(L0S<$sN0O)4Y(I7ZUjJ&MtiLjr5r#CCUnL&J;%C!T`(wb$z`T^;
z?b071B4#0?+d_gd$RTz|ldm=JNWarjjNGMU0}*3OvBObCzg~42m?8e*W3}A<YiKJi
z6F0!S-?!5yeXfpeJK+pt{oY+V3iCDpe67FjIT5D3N_>kwPxiWBlOS_Vw5Jn)q=2=C
z*JP%q(o$S`%>fLI(~35)^kee784<zGtDO)9Gt#!VoX<tPS1tFn*%ZAVp8R0*CPgV&
zyu=-pVtp{>jxFC{KtM6jS$~YMd$2VZOO-{*=Xz-Wl@Fto_q3uE(4W2(R&8eeu>J>|
zuJzGb7_LKEseE0XG1O>L55`CBDwnh7jM%8c`yZAgK{hzo(tP*({T+YY1h^u&4MgAR
zHewBxs?TF?Z<g6V=mGOk(ACi*mN?nHJJ;ml>o@T*3cePozS>y6;m8sLe%*wiT>15#
z11D<9A;%Aa?Db0jY36z%^9sXquTe);F_X(MDgo|G8qBi{9)bZaya&8SMAAr13LjNI
z=~R$L?}{BE`ChTD_7gFV<v8^#R+>x6L3vH>FjD%N!ac{e@3?+K5%B_EgnE>#IM>Qq
zJ8yMBO2GROG9oj$0)x_`#S|-_ou){;cTc0l@m)W#i3GovaOu?3iwJ@F#){?3_1`}{
zO4rXeaEJLi%QijJmcVf$ZXdyU>1lV_=>g~udwnD%@F^?;jZAf4TJT-AfzDlsVx<-1
zm+L_w(_nthlA3OXkDJ9Ux~*Hce6!>Cdf1jji62}_+RjNcjHNn7RjJGuC1j(0qMZW6
zN)K0N2aZL?58LAM&r6+0E46=2?0*O^WWWE>0afMB0s{P%=-TS`T)J0{hOON^kx=*M
z0t`v72eyJ}PBZQt;}Uhh$w3bJ#mO5EXA5BbkmWOIfOBZS3zLA0yhb~drj;QS&+A70
z-tqB{?T2$OKy@M}<_m>#<jtE7?Orek#~q@S9W9LppaYC{sNzMwK&AKf-o%E>Y{Ji-
zPn@EBWKeIxAcNIy^wNAm>Z7`H)5e<>RF4s6-~N92m3zJV?Pva^VU1@qSq0u}wxaqG
zEYZ(<<eL=62qd$JYlg4|I3H3N|5*zW1P2W2Y%H9ID^~}cIOrIvDqyT(EyUja;Q2*d
zy?T{p;QrQgkV%(tiJu(nk<T{#bpokC=+M+nQ;i%chAi9sFIgCXGE!VEoD6gQCe+l_
zQidih>ZAF2VW?pYgP`%uTrNfpa#}ALyOUwaQHN6Bn~x)Re&T1SR~z6GVlh&;xgyTz
zei_+<v!@uGyBdb2dc*24w4xVQm-#ny#C|C#Pnhz&c&yEub#K6Jr*KW&Zf%*RQcVO4
zXAw~qr(HSXSPFDRqJc{bNtFU74{~5yv;tI)=!&lfDx+hiIRw+1xc2RP_qcS-|3xT$
z0FEBgQJf)EJFj~Vzn}eeGd0QL;C>v7-}HAups4!QH8h<MS9>xyEj@i!ZmfJ*=%(Z;
zb1$4vv84akGkD!opbr`S2k!U9h>!0|8Ww{|h0kqDGq<gZ+}e{|T_Wx_(w&VRy1J|X
zL-_p>nqIj#RA@}|M_Ab;E1Pi$Uhf7m5}#e1ItDqxrCt06IR&j|Q>PnYR>%Mv|8)ks
zb33{=QS>K~S3u+K1f*kZV=1RKV1AntiNUzVQ=nu=gMM?3OVB*HrCaf51A(dO%;?Xi
z;zyNB^Ir}ibrTeTkyQ>9x_7pm=KjvtY-e+$_@lSqpUd1%M1062;H5JA=4*%7p2OeP
z_&@d=n8`D+n~^zfyd^1|^IkvC6wT%4gcrX7y_Xkv)+*>PLB4e6ocx9RsTdOS+L<4H
z=`i9|_r9IhGoDYx?m9cJGViOOwy?9iYc*dTEf+9X3*0UnZ@8Q~b!}4wUiZ<Q#y!3^
zXIdps_XtZxp=`=;tFg6~<IBc6DjC6*1M`_HNc*|Yh^_n9iyRq;!yA9WJkmdZH2!GA
z^ySAU<CH}C<gTpaL$z-S<Ekg=bZ}JQPYQ_HHAX;L@uoxpj5n?dq6RwO>$>Ucv-ObO
z7A%4ZyCV~5BDoB7o_7q}S-`AU=n}7oNWZZtAn-^Ts9jO3r^F5)E6Q%m1sJ2bzc5Mh
zXSBAaW)$3HRnXX)52-axAD^wv|FD^wbxY$!E}AnPBVe=(2AfK}dibdFZWufMez4nN
zz_K$5(sNqgS}`S-T^!Zd)`{S{Z!3gA`6%XxQ!J{ixu~DHs{@W#^mhnj=15&xn@<SX
zlM!}i0cL%A=-?>OSYnuEf1|ZkFVnQFXyHeAU=%d@nv*jK*eB@@gtJI}rYO<@qkyJ{
zamYW6236Al312}NcYXsVtX%0dAI&zpZ&#(g?4nN-5*4-W$h?NS`gPc|*%KjCW+h@$
zQbxbNf8cj;aL~Q0$u#>mafu7YvQLtPlj>gv_uw7t1YWVGa6*D)aKA1E=Et@byYksr
zuVDRL@pZDTjdur*p2wb8tDM8Y32U_%u_wp4ZS9qz;q~#lvQ{L5s!w4)s4S@*PSm<`
zv*ziEc-5O*$@XR^hSFlO2R9AqKB@j(2&Ev_p%J>jU7$brzWN4c;dm8q$eqJ$ZMXtS
zsrD9i#Z8z5+FqI7o5^=m{}f0N`FL`gT7|c`WFt~Tnw8}WY}uu>b`|RuOH5OH1?k$K
zExZqu)*Lg<CDWk!=e48lTZtaRIlDHfwotG8@I7*m9kXn+pPY>w;~x*+<k2lZba2I0
zr+I_7;`^ol!@~>Iux+uy5}t`4);XNSq@d_GJmtf%ZiBTxQMlKzZmq^-3j;ToYs2Mh
zgd(1Gao(!6eL>cr^e*Y`_BJ>_WM}dYv<}j&xwC6MB2avJNEx@jyW+*M=0_(PK^Kp5
zYA(}|3IYvoF)jC>$Afn1Qabp}$1^oPfDf~xO4`MiM(}2D?Iy0V0&k|$p%A!MXlU5L
z$?WTUZBE;)1~K*FR0WfD53AS>@r3X-=>RXn-R*tA8Md@;zt(+gj~oX@mnZhy*K9zq
zJ3P!xuG)hdCqlfR`U36{?Ybi+KEAuz^ZAL1x5A9p4?<p~=H1YJLH;u5ET8fQqBp<7
zhiPhYssAj!8;EWZ&%8KuE2)EW?Xzj%**isb#OZL`L1SLx5Yk^O-Y_sbxOyk@7@&`D
zyY4mxtOb!-Vj*p<LXUND%-R!LCk&*enn;d>tnVIJ^OGQJxU(LOI&XA{kMqDu+7+X?
zV9l>!4KamO{o}7QhSO{c&smjRJP7Y`li+W6_!QBv{q=R9-hhe0d!c2V&TAwH{suiP
zH}$D1H|c(3IM>vW@R#eBTZ&C;7;owyDZ8?Fxb9JE{*A3lFsoL~l25t!`pWaGKVAOY
z^VYn4kQNf);>`IGn9@oR8}<v;F!r?{sJS@QBhfeo4VXQO!&cLS&sctZ`!`~YFQ~5w
zUO5Rk($x>)zhZB~m<5|bHri<m0ENo70salgG9ja*7lakU;A1I8q4UZN|6xPkfTm5i
z5}#K!`|DLBDHedCFk0yfgOWK;<$-f&aV9v8%IYx!-y&Io@)o?h!7p#pPmx~Wg;SE@
zRbPMFG|-eP^;|v1kG4R2#d|Io+it1-*AxELSFDUEnNbsR@%!(c?*uY(j)i3QIK>%*
zs8*Ap`9qa8A@4aXLm*f<5Y+)Qqt{-fCnVC;UAyF$<5?@PtfxQjc&iu6xF@2<xSfTF
z*qv@g$GY#>H?i)@uK)H@3evHu?Mhh(fA4wic~BuapJ*GO#51N(ZR?dg(&Uz`Puji*
zcnl7^tamqpu!nJn6(80Lf3;?;wW@2stD6IqA@21z&S58s@lIOEVA$S$>-TM)P)1~<
z_c-k$ev|aVo&u*fAt~Ppo4E8!lM#Ts#(;<D&lr?|->iN}VuYCbYKoD4v8+Fsavt5I
zX3IBv#K*bdgp>>P*4sX89Jm;ybfL#d?FCjeqWziXX7M!w_IvHCTG1PCYR&2=dCW1I
zGL+g0b$35W9U%v?9M9U&_O@0zVc5y$q;q-*_*#BfH|Ap56TmtCoY_>h-mcJG_GAx9
zx4xzR<|#>%G6V|Xwjm?1m&GG%y_~OmjcN;|JC<Df4`3q6MSh5%k>a@(d`d|TUcK50
zW2()`=9ZTq*;QzJ2JG+Ee0hHr>}E)BNMFUctgeHw%JHJ=QN;(fWv!EOv;sT5_e3ai
zfYfe@3#BdPTG<m<pIz&YWZk!xH_!>W49PumE89S)ol7ZlLo3(!2s7G}P}q`+F!8pw
zirGTXGbkk#e!yP|%ZpwnY?BK<DEaG+##2DPD)tXecDSru-t%48#OKyopLI`9%0*8v
zAxX8mw$<`;=G3R-MVX#r56~a&7OI{SALDGC+BX*5YMBoQ^2_hOB(VpW=cK-jW}bBi
zElk!eVRL06KA!cyknoN+z0hrZH2>R>LWt4wvxCjXMw_>_-R(#?T(e;(2CoXVwSjMU
zMHUoWXJ4Z^O|uaw+SjRyUF)u21-t9;+r;`=3Z}ioL-T#N)IL35SaHiBfMeOX?#i$K
z*2g@^qH%hGqt|f*PlB>sg{DMU_ccKze-FuAWy3yRat{bg;NJJb0KBF4Io$)8!!+Qr
zTsV>vw(c*TWMGdte@i8{%2x@gu)x}iVjo+Y!T*$<LY`~7m5f&VLx18ppZJG`Hbo8}
z7uu0TxLyg09(EY0k51RTDSGl-xpPmIS;}Izv#EvIX`j+^PEh!MKHpd{R%EWMq*pG3
zWm|jR&tY;C)!x+f<HXJcUFLj%xSe*r5?u4Kpf-GhY052m`3-G>=nrvee-8^mO$pDQ
zxE4-yWwPL%zJYe{PYz)%DyML1U=%Vne4QxIHucLbAWV7>y|NMP)b8`TJ26zqx2sl2
zty}Yi4Vb*Gr30DT-1`JM?E;MDl5cKPuVwq+3lF?x>%cbOA=VpOKcCmUgl7fyk>r?a
zgRxz5Jsm*V6MFp@7$Ez{QvQdqk|@6#MmoCwGR0Rf?^wt*EmdXToF8^z(KB|%LigtN
zUH!OjSJ!=u)S-rASe<?RdEy*R8Bd*upIh{JZoRyC=MK^Z#-xU230rHQbVj%ZlzdTg
zn1_2^+!Qa%k(3QK$xJ19oo`C&qr?vRW;ve+H@O%Z$LJ5wUB;RcEI=s>)yO~RAF8h)
zhhg%wVe{8P3d5T|#%1odFdjy8atG}_=7*<xgdkae<imj*TB-GFwrHoUiSavS8DLme
zWh`lZr5z<;{H=cl1$_ri?7_y-xss;%?Wgp5;jW!u7Ia`tov&)^aei_C6?d|w+%gyB
zBe78+I*?8X*-b6}?mt*68oWdIq*Fs0V*q0B_mTXVq3um)W(K;8=yo3?^O^NC&d}jd
z)6$}Jk{N!V%m2u0lswznk|Qr4w=1UxO#5@6@D1}S>`TaE$ZslMul2hyJ2!)~6T$in
zOGPJ4;*PxRQ*5@Aa_<SgPL4{#+<G&EG~plGWkw5`jx98pH#FlD&)+`1NYOC+^9^->
z(#wl=ci-%n&fJJ@={3++<9YY|r@x~VcpakDdeaM9<5E}{Skr*xS1mIch&IC=b;Vm)
z{NI=1a{b;drFQsm3`|wW$HpY&+ajQTQV%6UPD*C+@d#owe84t-4(WN3`pGw=t9m|M
zW^Uc7tiAW_Nnv<BPn5!OjRAD451lqAlMT-;<&>8a=+MPTN)v*;Wwq4QB2$#Es6$jv
z_NBK`@s7bfB=`qIo*W~V;R7!vb))vfl?`gv>xH$89)C`HQ^vQ*#>OV@K@p(U(3tl<
z<`tz5a+g<b_<pV4FAB>sSH9ic>-4Ms@~d=9+!;cEd<b8C=p9^_%w4zP*lHdkGYjU<
z)&Q?QcOssIV96iiEzg21sUqtOcY@yP6(GvH_sL8fc$w4ExTO4l2|QuB?Rn{~F&%j0
zS~{u+tF1vCWF&e$7*`H5jD(6+X7?Rw&+Cxzz34oUrJtP!NgIxaBZQOYCH2@ob4so0
zk{Nr5{*8DI+yb$?>(fH)jyn)yp`_pb>9~}EmF2Cw`M<xV0_~cf{A{~J-d&7&g3oWc
zwHf+DZqG>(hZdlsOsRHtLmNSs-b31z6PEXV-n+Mi)h%(bg!bNJ(tgmJ5pOSSBZ%Uq
z@M=I1?)mw(iQ4HaRD=tQojXKxinguznpkgeod@L(7E|Mz36ecnm`ePQN`3?82$ZhU
zt@NZ-xD$}|1qqpkd<zV(zYaAS@f7P|ODfvJLS!m>x36sZQC(fGJ$v@}o05Hr(0anS
znH)p~uiFoSS;j^wWKt2X&aSO9FfJq=lIzzK5)!QEAvZny{n?EhHwHr3&W;b{n^zil
zK0K9}DnLdMnj=Yz)vvBc(msmGPw=ShZ(0s%9}S9!q{>QilIs6HIf?fqX6Ur9FbEFK
zlN>uIRl2}N@&P5IchU6vF)vNE%14|};fYfu-pq2#o&zc-49~EfG?zdX<4Py)pC)Yh
zj8-oIS7IPT#U!I1wVOjDMj#UECNnMi5g}8<Z4JLhy9U7eZJo;LdexBsc!>`s!-|W=
z_ZStW>E#%{dG!H89k^%rNII!Yx7Bipc@byxT7WJJ$5Iv2-L)gW0DcMaKg$%Uj~Vhy
z=4gArFNytV?R<We=T`ZA&&EppYUq3Y97p|_2f<y_x!}nQ$&^97NAlbN&cZn$b0^^}
zp3>RBNxA6!v7$rA2POiD+BdT+#ve04xwSCneOu`pIh>9>5rQ%DktOwbM<wVAF=>Bq
z%9Awy1XQW%ViOe>OxQ^;jjtxI(=va28x)2QR3n4T$_#@p-+RxSAllLTVwruA86=1%
zc|+UL;kLI}N4)MH7J_~M&Ut8_=29i>d;5<2n+f|)DyWldyidw!JSn*sa;2!M5iFjc
zgZQ^wiflC$y#d!^XH@2%&JjkB|A;qfx47l~=k9l>;_KQps?G$uZf2;&VviE?3U4<C
z+x@d0Durk>Hu(FEyDC3ukP}L$4K^q|H97#9?m2Ss|3>_n_otXzY*aFA1Hb0b9hHy=
z9G2l=y8Ul=P!KK|F{4f(=l;EWkF>P36j5+<aj7~wI{qqetAa)>C(grxwAkul%xFS4
z%h7WsNAqHJrkD4+Y>4LUCBo?}zie%)_+bdOKWfik|C|lzRYR)L&s}8Dqm?Aihj|u>
zgz}<m+CkWh>OvIBJWkj|CEeDBrG4*74*_ReWU?746b5~(<(z(9Aeitkn3{05SkRt|
zca62x)b^(NI$Xq3nKQf0q3ZP|lmOa5^p#M{c`g57kmw&(NsuBqCk}%&a`Blcp!q5B
zBdYFhJX^0gHHct=TF|h=$Gm&EXV3A1x)UO-RyRF+I?OLf5f!xEsw5l_M8AOBF~~ZW
zf3h}1KRbqtmv=`?Me#vMjD<-AAUk-Rt@hKiOv!4<@9bq*3CxenjO&H}c1-?|#=rg#
z8-ont{mvR*j4v%5@JA$=;=(HIUFd+ngHni}9(L%Btj53I2k^~WT@Kk(W58i#E$F!|
zyAsV=>h}V0WE=fRl6o|s?gM9|>|0J29|a<w!FXzJz{b>_?h+Fd8vqbw><wORbsUBy
z?ab_iZjnu$Oxq4a1!Lp*+xy)Pwi;&c&C*FeXi2Lve3Gi)bz#P>$7dVAL8G_k*QUgt
z>tTG(ZKYWc5~MxV;W6Cc_TK<hMT3j&MmS$~7M#Uviv^3qMvbqf8a+a0RcAEr8aK=>
z%^&OW9S`JQI`-?~u8waVUCS{?mWrO|d_J=)2z!;_H*Tlg|88X^|8kSicG2ucmB?zZ
zqH9<#2-U-0<Auw8vJfTtdRf5)yeUiA#7etz6V%X6P(J^F9kQFRH~w%tk-wtME^M8F
zL6kPod0PY~%7P{&XCj*aSPTq2QsdEz=K^$Wow`@+SD=j)2$=sx|4yYO152?(!g)NP
zX)bY&2^D~hvO5x?R8WaZmi`|`peztt0qD~Sz#cce&ovqN4D3TSO9rK`pPsgqdpeZ?
z9(UjrplrE#j1#-F&PfOu6&-q?Zy|h3!sR+tc7^;WayqpDcaZKkHB_|>hDR99`08gU
zz>>tn+}p&oyLab|eaCwhEV=@nipCU4<a9hqgzD8@{GU<2eEcBc_H6-ex!JS~OJcrp
z9J#WHO)ZTB^t@Tp-yK%D#uAB!$+VtORt}EZedOswwgI~b8H*WbS!MlNV*{XLpaE#K
zKT{+mzw8AlUKB7R+kMSGn5BK^fOibcjnhX369X5o_?>;_kgT!T;r+dPPrWwDms;c2
znSIXf&C_FJu<lsqeU3J^KYnY6OL(wA^^@QK@!JtuV^aP~9=;qvNBR+OLKfI4@CMPH
z77SwLumii;ioT!)eGwa(w3GC?Hkgowkq)tyYyX5;Zh7JKw>POFBR!Ah4QbWX)h9^!
zMQ*jTd2nipzRxprVvb<H{Uvh)8Ps44lt`*(04MyNf{HfhYM>uLt_Gk!856(HX^waE
z;u3sa570Lo$XIqLFrH)h)sSKFxC#J#9~iMug{9_)0d|mnEZ@9i*x+$S&^7>9$6JZf
zA5uWIPQ49bvF3ht@_PoSK~Cn*3Cu7KzR$fSU|d`)zh{2F+;cb$Rj7xDuBbaf{xM_J
zAlKNo_NPThp>;@W_OT}Ux^da}eOc!&L*t3q1R{Y|>7`awJZrK?Cn@Oa&I1Mj;dYME
zLIw-n5+5Bx6~Fh1YV1FU37?@v09AbZf@2TKAiq?OjL6V~K#s}HpaMy?K?Qo`-@^^9
zTpb0tIy;qd66`b<=;}<J7+ZNBF*#HLJ0+84xw*MT_-o3^9mipMuNqis_x^9U3vBAP
z0un*vxt6B3b}T3trU&wYOBGwTi(tWcTd7i?NI!IWtEm4k=Wib?{qtT(I=8dKM42mq
zz7Z~qX%PGg4rC<4=Bd!qn|=ap?yE@FlLy=K%mSWpFEl{bUxDP*cKNdvkh!gU><~QW
zC@?+R#rrP%m}FWu{ehQL;QhP9fYAnM#E7(BF5|vH&Ud}*K|euz2hFc2n$6QYohC^M
zuE!r?)E`bhvdtFEj`d3SS$L|Sza7+;u`SAcwnW=(ij*8V7_rnqSxOd${}AN`67f%e
zaeM>_g<8<e0?B#bmgX-_PlIAHMLiam;*X@O(@S{_Mp$(Ll0p6FmshqBLk4>+NAFS!
zojHer5N3vVN7Z8hWww-^J^^T!(07GDdL)MZq3mh{uCQ)~PI3-c&sUi5$Nuo6UpB|9
z`cU%Y(8FT$j7J(k+DGqQ%YPgkd^0B_hzp`Mk^c65c>pSmf?#|5uRA3v=gwp-!<2*r
zM@Gp^!7E6^;xM?n973KNKz0WgWfE@QWMZU;1R^0*u;%jqWx|jw9zkA;ISa)cO8^o7
zW2}_ZdjVf)5lQK}wEvx9sjF3wa8mGn|EXi$*RNlHx@U<0na$gR%-xsZqooHy{6;oP
zWo*1SoXx*fR(h8XvPO2hzmZ~S!Ny&RyGFSp+?UF$0lJ?C*uCSP*V_Q$lz#F<`H@rF
zIedC_pYLfW3W3)!J$g{<dg1}^JS%T9GP0I@3wgzOV}}prC8#?j*kh`38Qf-xw9IHg
z3^f2ACmlR!-Be}DS-?^&8NEu7HPMSWU#-OSy}eLzVxWOBZop)$y)YF(I(!GT7g;<3
z7pC5nAe#<Rbk$<!21=Np8T;y%WCN&-et<-D4BW3m7ZuH0C>fg9bMXg?^L`Xm<nr?J
zHWb-5Y_YA66&4Ph@9Np&;K&VN5d=Luq9C-uxDz8a^H3$AN%W}8hpK=_XFxYs?LrE?
zLHKdEPWLr8S+9{ey*yK?n#W~OUI}Hsmz`H$-*$I$(xc;n%Pl}LXqyL2>CZyTL%QXY
zbMBv>k|^FXTFoMj5wg)AWqAOz^Y0H5VuDwANBJi1Lzx~p%7wRl>TQn3E~O@$z^8&P
zLnfi!OnYNlS=FY{0#;!NYWujG&eArkK;~1<T0vvwQneR5+$eT@S1<Y;uuOYkzWBX?
zQIC#ZKyy!#D2$&9kH<bf(ZssU^>%D%sT{@`xPN|5rg#>@eA_1g2!bvKrk48#aCf*+
zi;8fmqCx!J8QA-NXoUHpUB7g;Ct$h`1}J}_xZRw6OqfZ+IS!Th8(&i|q;$5XC{c7k
z_{Nc~c7taEKq@s645i?pC#p8Y2vXMZ1-$_H5T!#|?jQpw6-^(Q54feKl)N|T+_Bbt
zi=JlsM=$kB8U|Z4Z&>w6e(<=FV_w<G=<zh-M^`9w8??u$f~~e~PExQ_T0vTY?w}^@
zF00J;Raol3*<$P&<cw}-qx>}_0oj9>UZ3~e(E9L8I2d0}{Q}y>+FzDZ@9P}%?x%h-
zD8n8iJ%0!*44T+4Y;!LDIL9AUU+30H16L%57UsI{0vmR};*QXce(s28KpbKVZ4$wG
zt$Ai9Fxs30k~Sh~oAYE~My06R=R=A`<c$S^%V*!JE$|Hn#rO8)nH|V9F3GZbWXX>Z
zRdA)4ieR7k0U;@#Mv@0JGZk-&+8&xx3)9+v#jH;nGAytMSCiEs*pCG-Vtb3>s*DjI
zB-Fi=m9L!vWw|y}wf5T-C`V{K+wB?dC?f<~wjHl!YnCJ51VEYL0unL{?Tyr2cqQ8<
z2;ta=&m+`Ci%>Eil;kPKYjD7zK@`2!^VCVto8k`Fp`WvP$gVRgTs-1@`Mtvc)&N8x
z(_9MXIN*cp0Ke%zH|cN$w*xR)*0plzE<qd|!SGET@H^Xt%}vkF4lpXxXN>|hODt_R
zzZ7aGRcB^xpw9|LcFRJW7EgXH1zd>*;A8CU8=$&rTOC9cEN-y6P@E1Sxsegc$@Y&X
z2IgI$vXhAVhg9Zp3XvbJ3779_N=(hji<d*RkjDt2bJLXgIQBD=g7Q14x{&#*il+G}
z+_8kdgW32eR_uuY&#%O(?Lqe}wfurk=K~if-vee6z-#^t+|YCo{@|0lA4BqPM~k~j
z+315Vo7sth;9bb=gqyX!WuizI5U(sQoh$B#uzkr%7_H$mZs`l{Q0IT`6M6R94OMRk
zz`>IgSJ<hD5<5}+$6iOgCSTPc5d=U$L{~=SKPUWxu-j1De)|g{-RKYq4jp@tv*$O%
z#%u=Qo;C_s*^)6oW*-r^0LpDvAA~@!j9#~sCmrOk_Qm!TCZkmF2Fypg+S=L(a(cd6
za8#duKXd$e3d;d6P4!rzoxdc?>94}l!4F7R31<!d-VJjfMNfT{Q8Yfu59s3?K&tRW
zryuluEJ6Xp>RUkcjUJx7lN<`8BVd?29x@tg0{N~aQj378Xf7_U8E@gXF|TXGFf_pe
z#`QS47vBSAH37XU>fm9Y?vMN(t;7*T-!4^Y)j#{!J}(4^fQ(7E+ET6_&AJUrT$Old
z>`9;v!uHB=^bQO=Bp<v8K*4=EZf5~1mu6Ua+$y?6cM!m36W}aei7%zAOW|Q8l&_iU
zzBu<Cf$fr&9{1QmBoB(>?DjJ&KpN~uqWfwiW&4~~5-UmFq4tT44(uU&tzXuH#oDr%
z4Nig#%qK&LbG@L85D77d!pt927&hUeO5r^KiQ?-B`_$BYpWz)VePeat514~}_>OcP
zD4>N$tP_v!^u@=-@W7byTqyl~C^j4IdKkO3IDduwCGF5xySL=3tJ%~h(hGA&91w&G
zF#r&Z;=qA*Sf91(fo2?EuddwBFex>M&sc&FAr<)XBbLgMnCuJ9VqY9Hcm+5gkQ+*n
zn=!X~mu2G9wOS!EVz3vl^i&nAO(4%&vG3y~1G<aVOc;QyMFF7lg0pkBL2f^ckGc+C
z>P3O<(>T?QQZQvj7Nn&j!+R7Z)<DHw$eD(rT%splv-E7w=%pNb4C>@!_hEYIV{zqq
zRO1O5wHCm#nr=K}kZqs^W2`y8eR%XRicg=<r>|~RB{mTZLxL>VxBaV^=C{!NDIq~0
z@MgCQw^()*%<d;G8U4qq@v!+N=OOzF0NP!!3M>VYu!S}SJzvzY@QBl1td+!P6n!Cf
zlg5^9Kg*#eYA%l%l*eqH5T>XZ&L<H?*>vQe8L@M6DnefgAAs8=E05{pcA#zoK%`y=
zoOR7nCFa^>fP;gB5|c5#49!W`$>?~9bG<u?Y^$);;F#V}4K#p$ULEtxHKlowAB9+z
z8+s>5KNe_6fyTT6ZF!nNUGgb3(7((pMw6iZ3qZ@?_rgdSGAon+MnL>)%Scm(@*tdJ
zv?ja89|gtoCNSWOcH!FWoA+UePtPpw@81C%*z2*9>-A*&3dryVxb2>&ct+nIf|x~X
z(>XhK8Wp$zZqHA&tvuxdO*?o38DwA3pr!%vMI!n1vsA%grbDv51?Fthm>l3e(1N7`
z2yQf}N|ln-p%a|+Hp!qp?@m?0v<T!wzLnv>4A6o+`fxY>+*tsfr@>U237Ea2;pgW!
zzRA2k6Jo@CQ-F@17F6NEkUm%Gg)v;QATTdZe5LAex=2oc;dGqhrYm;R$}2IOYa_=Y
z#ea_3ZsWm^_a3#Jl_jo_`O#W$L4!xm2JYqSO!fNG$m9ziqS9h|scH;Zq<RQo(`@Rl
zM>@Y=Ir1Z0$0w#*)h*L@yVc|{gAaf8eKh^)#KC&2^rNJPfWNW4eB~+OC}=jKD^Pr@
z#~ek=_HSU|8s71OkU|~iw5J)DINrTg0V9%I9(kXtYiw+cOpJ)=rN}!{cwFhoax|a5
zdePa}@nR>gK@#a{0VR%Q3rjHlqndXGKQ;J=e9l0pwe9!rvI!W4y2<WDz6H$int&vX
zI`~od-_D%dfbk<;_@*6r9fIR;Z?<chM8?JK1{Z|R^R=zQ2HYU2&Eg1Sj>P>$j`PuB
zcG5VOluWfhkUjrWocaL|t?|q>UXuZ1g$(R$pu%lSDl4|{+mJp2B-T{xY}-*!SHR-f
zEgv#9+B<zYkHtXxVh}foPfY3VRgBhN8e~pkM2rCJgJ`imjFq1lx!=Uxl2TZ7`s-l_
za09PdzJzE$1|esGCx>^SAy)E(J}3v<cUk(u6+X~O4<o9SNYY487?pcudVKHh&?z=~
zc(Ta;%`Qdb_w|IVfKhf;6G5+TaY!~8VQK5`MC}X~7I4dUKk^=v5QNN-1tOr({0T}G
zHv@z1Ib{5&YA|{to9T%!1M+VIBnH)k`U9LZiT;BFQh;|u4f^(6z>*RPYlE_}xe*`F
ze6W_O!{F@YRwmh$XxP}G`@|JzWCcx23o<O4y`Mp{Q}Sv`KmcG9QMwkui%@{w$<(<5
zOb+rU-bL9j;yI+p1g$?rStptjCSaT@6YN#Wuhd^(A8*^Ei%)#<sk)uisK7`G7#3io
zLY7-nRj6;u2UZ2@Hjn)v@3;I~x@ok+LK!kv3qKQ=qf}UZ4?%po;C=`JkO3|)npOxh
z^Mer-@r-rSM=>-cT?+H#g2hH|S}eQ9$_1-s=8*g6`$co95NRCiZkMd-OOO}7=`$3m
zc;_U_Lhgm)Mg5fc-Obyho%Vq58T0I5Dh+_`e+K1tk8C0AtR`A5A?tA^`DJH`Qv-x$
zT&+#}gTtO0mO+QLE%=VXs9N>64x-YAmti~7LFWX2h8bTs;8$8B7@P+nokCB4rX%p%
z+N_bgCh>QOMl#%rirS^}3_8=794m*kd)+7c)%KV?aJCd3hU};c)YuH49aFfbFB0;=
zrOUW0q6M0(Xh^4io&X0oxx$Q}*u6j_e(_J6vnGUJu@mMfj!$6Qz9D5-H8^Mq$^+Px
zt;l}%L1<Ez;J1(%#QUfzAO(*NwgjAsWv}Hb==3u?4LBFz0cgO&PZt}KU_@@YCZ;A`
zya17^I#hI=+K)qCt);~2m|}c~Iu}e4bne&%;j^S3WYlC)8jM5mA^kh^cPE#XwBF15
zO=g%vNBdHpR8=7Op)FbZ{J_(4m80nzmdfi|cJgVf9K9?(HC<iZB8{xXbd(R;=h#|%
zxDTox&n*C0^1cRZ#ug)NAr!duGcos6BLmnE0?cGT_CH9+U%22>Q?SPXo0#P|JdHeW
z6XMbYciEBwtTnGORHHNAxi)Vn{=5r7X+k5v!4C^aIwIoZ`5ccg!VK(1M>^y`+MpCq
zaYY`q^|7naTr4~NwI0zDIYAmLKL#=yP`iJ5f(kWpnR<WlRCp*0`60br<987j09BXr
z9Et*;6!om#mi0G)vR@>pKe!;GuD#VP-;BukHXMRKPf0{kG%-Sou94joe|Hnug?ezv
zC|~GXZt-L(f;$c51n!rd1@uHhAkU12enUDV66cFSH7n4F&Eafu9{(J^RLkKj;Jnod
z-2xyaeZn|t&m2%uLKKWb9*#d9B$1u=0XuD~y)%&JEl~pmFvdLA8-b_<3O$~R>LDv3
zlk0748csd*%hOdZ02VzJXd@8d6H(|*NpM4rd8k5BE%0^<rgKA8qhlA!7(xPwTZM_f
z;L|I~9`Uq?E?ov-wfLW$P4I(mjsch`zKuSP-IWZ+Qws`eZKb2*78840fKrnKV6i3B
z1|Fo$!T3B{#YhxKKz5wVbEtK1T6Q$FEI=S(=1yEw%V!TAp(VnQRs=+1RS<o;dh~Vh
zg^f>^ehSVx8GkmbuMY~yl2<Q7)&L__0GkS=S?-t4$b@d&^O@p{E2Juv&B;u5^&Y`l
z%G24o5igm>MzaoUgBdoC`yW6$EI;p|(@u3`If7uJy`ISV3!!Eca=d;auK_N(4JZ$*
zec{!rR(p>ON`l`jwT%WF^m9Ya>cxI;!Y+(sp<<I0K>kf);F6SzB;X5pV6te`mshv0
z#OI7X&baJ>q8(wH^W@D1)-_ZX58|8U&o1~Zr&NtWkaGis=d0=%)FH0!kUZS_ye(${
z8ev2oTX)~w=VV}@d#y<a$I`m#?~PfP$PXo9)H1%3@Rce8GG^h*TvawZBj4_Wy(qs^
zca=%(5J{?z=l7n+7Ql6~8=AHwCnqPXKX(j!k+>IFn?9cXvLo|Is<Ky)9!)qj*QkQ(
z=d!3e)A5aoiK*z!th+DJe;|-?$G4#??#G_&GX(z_m`J7iU1VDKxqwqzWhv>KeL0tQ
z97=4cd9gRQFXH1rSDBvU^O$a2R&7GyKds_v0B`Z;ljbpyT3jiD{V+>Tig+(Tf?g5a
z{^SC6x>^na9`Z$=>VRD#7!0&3^lm4FGZ3IaR_46pe1-2%l)S1d*#Yj-bubnhOAE6B
zPiXa$f?x_IM~wtjo1&q5ut#MMngi2d-O%h8&zTPmCsoztP?m}WOh_F#3;p2oO4ZNu
zUZAT@hTdkBP+xvk^q5h=uz6aRfA;`zu??KA6&mk~1%fy=WHP}HQ#RI{L)VV@XzZNS
z-f>}-F8W%9eM0a1;&0y`IlxoIadtlCoM~&8e#2KU!2EPT<1vejk4Qn>n4xi{Uy1N`
z<!hQ-A^DvLLSB0y5aqw1_6PbnyW*HKC8^ILLLs7-NLBKm)#_(qd?BKt86!#R<NJ9J
zj-}g`_${igW2upJPi@Eb=AsB|hx-I<=W#sDP!ctNJlowuD8k<sjA(;S+-aDU7!44g
z0ja?sK<Z-yG49<YAbv?fLy+(KmT_lo*vdz9VR`Fe_RRw+7;@7JpErR=-;RaLOkB$N
zbyE;Jeqy0NG~Sj(>nM>hYW)Jboyb&`aJ$(1ViD&%9RzBo@HN$9juxezyhA=#5Yz!{
zSwGF4y+@v~K<$QG(B#3@!xvl61vL<KXDpkK2dXNs3ugcL7R!co=Md*1joSjf1DEfC
zXiY?FbfI0VM84KLCocB~K!+=(rF!`qgmZ7)(in(u3xM9NOI5-D8o;?ius|xY`*MH3
z%`B9%fX(UDHPX8PML;D;4QTQ@cXAmz*J|aLXPc5{mzW?0NJD=pnb5noEI`H!%d>3F
zI8wDNDketRwK)$`*DBD0&J=gMQVs;_$P-qC6L@<rpzL!fe?Pb-(V%VA;mrG(YZXfN
z#`Mhr#Y_Wy>im(oyGZmHn-RCx-B~WE=`N5!_cvnGNrO`Ik5eAojK{&q)d3q{s?OUn
zeW)jE8swR(iG=j1%>-k0R*(=JgAtLqRkg)#$&-{)&${0gTXa%X`}TegxbeD>`Oefy
zJ_pV3=B*nK(JyL;D9P@ZJu_2~5z;vH7&<1^Ax{bTpStP9@*l^JI2rwmQ#gRUUVD|E
zj*xj0h^ahg#y?ptPS{plI|t!l9hl0eodHl<vHkS)qE&P{zI6vXNdm_bR%Pem1Y^=Y
zWHU#-LD2?2?A?Jerw8!pyU2SzD<Q}rrz>O_GtDc@7bf3ocZ30X&aJ0s^KMtZNf#dS
z4n5Hr_ld%>0HwVJAnoma?VSbl7&v+!gUt$Cw{F#JnB~o4--0SMTDK%-c~1l$led6U
zk2=iEWCtIK@!ewxU2DN*IAaS&O{oMrDjhqz`hVF%a?2g%!=IertaT-G!#0Ek0{8OP
zmlND)7k<=Kd7T1KF2G+Cqi-)}o@Jc>=sy+>g0FsgEBBdiTJhqM&_-?->-+gKZ$Hd9
zZZMx-P8rf45H>B1xU&H5oOLjEGNO|8Tdd)=Vcx_upAXm><QRUMmmRZo(IaAM-4Da4
zpHDv;dQu+=2%H~`1zT&api2;?7H1@-$pyDdxu*%wXm86B)k{7n1-W!&L^ml)1J!~h
zp(E~U-}Aq6W<i&O{6ro(zT>lqGMSo%kTE`LPF78}*ZJ+;Fhull^3kY-MfHWA)472f
zP-p};FT7R+9?D;`{rX23p;kZW?|Kg-Ml6I{YauhGh3@u{KU97bbRxX~48e3H>M2yn
z>lS8z2JrBy=Y@XE?t~nUAj*3NBtH4&vd>J`a3|ChGB!WuV^?x?%n~NOSibkzD4UE=
z!8^V^e@Apk02rmE*>029S6xWz4;?7blzHdvY|jK%n`Sql)^!dgdgk4myLz$9_-Q~&
zmvYn+yQLX~69i;CGAx4>89hgk%PUM8G*{UQMIV`_17KAnQN{3Zg3O+tz~xF|AB9Za
z^fc#=qAGcI<TGbNa`5^5k-(`1=x&Sw8#22zXaYWVy8A$VOOBCP&%MAGe3#BUhR<zu
z8w!7tDQ9eUjYVo#U0q#~_n5hb>xrrP2FIeJxqvS^@^5FyG!~D@?sJ>HB3wnIe@_(J
zwu8r}<MazK%Sww>C3EM!hP;FIT|;!tf!&}hTEM7fT4iow_{z&^!?0|}`{bIrXWs2i
zJ!?_v@vWtbos+Nm5_v`PS(RH5A55ZXZL&ty?4e%o61=M1T=Y>+j1Vn^CvkI9BCHfR
zme0wr|BBc`NHaptzP+a%TWDQV4^_WijwR$nB9_$~EGFMVee{rCNT{F5y~8`-y66`K
z!<ge!9llw27U!p@N!)baSmX)}3BStBxKGcgN9r#&MPr~ew9mx2*q&CrGv8=d#{mFw
z<-#p_Mp9F*6Ht=c*RYH7Z%SJn&qJboNc_bUfF`{29*^_M8f$MB1?A4J0~O1pOLD`e
z+EYvv4@)l0A9@(Jdr_gf96(pYcFuq4F?^;kDIay}iAnHTMv{MN21lTMB*>W7wUpqL
zjBmX1<#%qZaH}C0T*L@cHT&gO#>hdX(mRk>DRydm?H;f^9k|R3=sw90-=mM2M*^6q
zC-$M&sBw=sy{|-z5{aG8lqn?at@M*Ini*KSCk*ta3M59CUba*P_*--}4(YzZ>0dBm
z_?vlZHbFt3U-ld6d!!;i^m-zXf6lD3R%}3Gi3K+vw`(rvwCuYR0^RSoey95&kpi%R
zVF>+mZWWS%gyR>v@=s6|a)Up#XIN4>R%t-W<oH86N9F2d0xVdOdUt%-s*-=MizP8d
zRCL_3va(r^)Kpa?W`BNd*y-JEmHE*X13b~P#tFV(m^7v~@tXposiVN^e8y(d<QPyw
zy5LHcCA~-p_+`W-16(i*u40fp;il&|h=L+k{K=MF6G@|D`$z{?0_^vXmrM#rVgD3u
z3in|^9%oji);vENP_xqBhl3z@qtc|&SHG+#!_r)CC`7v)`bLd;4}=wllGYUKwl07K
zya7MwR<>;)D*e3&&+%|0EGF<3aKxRzOsV#ffC8%sFm3T){V01#HO!qKfD~(eBx~*k
z&_NBr1i=ltN9srQ;7X&;Wj@P;;<13s(QQ4eKYqD82E3Ysc`)TYNr_qVqyS1~p8pbZ
zYcIl5z~`DjBaIL1Cwaw-(B6M|8<6XoMAe35R_WrW-|%uweDlAO<AdR0bDVdPKMEON
z0S~JiTlwOifjgTH3YDjfi=$U-#hyR8Rl`~NyX{Gn1ZMD>ed{dAc1Q?3UU>XPKO~0e
zJ<sQ!WDRGP<?DKQDt6}quUg2VaY3;5*yq~V(A^^>Y8lY1eof~7g>I<HJXl!x*<^QX
zK+5Im1u`p=5}h|R(<2>p9)Y0CbRnUiH2cYBVYVqS1}?}g20^p(V&I{;utnOceDB|7
zz!Kku=AvYIB99))fMUacD+53+2mF!sU@O_d9#d4})GkAbWbT|43gqz{BTLvA|C(L0
ze{75%d(hm}a076Vo{G=k3Pn|6<d`DR7CEDkv+;Kbw%mcqvmTsn&7b$To+|&w1DFM^
z_l|jqWhCykE{gOv#~~|;idjdVFuYEX<p-}Xu>XFO+zt4>I-o5)-DRp5LHaHc^k%>D
zNhl7B0y}AvEsz{=b1GjYjc4OiEcpb5rC2}`yiGyavkVIs>(vUzU-Gnw>USBBSp+@K
zLAq>eCpWTm62&MVM!F54TDw!4q}z@LynN~+_(|@K3#WP@O@R<NuH{T7e_TJLC<ghV
zwXfOw21FszC_6lGnKyy7GE$9<zVQ>}TB87o7uSFNLeN73Xy??<kd-P9swk84HG@$w
zJue(3Xo80bxO5hi8LJp5Lz)Ykkdy#>!3SOWMs-qh`UT(uNcXE!YZidtN^%?VpV~mm
zRyXC)<?5j`!w9_rH31_)OP`2Oj0VKsK?F{EpY`o4kfw4%GD+ms(aR`w1~}ZNSJcr!
zEvESmq_z3@D#+blj%kesMB45Dy@pu;cuWf+7!Uc_oOTGjfHO~6KqnaAFuhJQevM#D
z?pQ}8{0?vxNC@dQ>iGPVX~*(R_aI8TW>Gt)hFIZPpo@GZ&=C6`utmNgU+2D@ur(y>
z*U2&5Yf()`AGfP@VsW}h0|NJca7DHli+4+W1Tw{mvB2d6_Oj^0g4Aw))g+h-+f(G$
zr~Gyo<h$wURd=;ghJ1SoalopcC!BrIC?@&)C;cEz2xs^pYZRbz?ee|3Ym|ZrEslY9
zk4>S2_g_!xB^J7xXh5rH*QFLV;v58WJ^2#yc5D9J9Iy;<Otuqma@R2diYa}dai;OJ
z#;YO}A;&xA?XBY`BRWO8(t-JL|I#4$s-cR=#sYu`!Auky086HZcMIenUWIt>a_2{9
z1hr$4J7WkirN*NBZ=X{3yCRSzw7sN3tl7~OJXhp5<yZ#B`8>l>7Y#KvG7y}-2<<fU
zDdfJv=`&PJqKClyj^&iqdkS)z20+>Ck&#x$;w^q!{G0@?n7=L+@o`k<c^2n=d#BGq
z$TQASiJ`l-!#EiBj~L+Hl-u5%xHFt7XZ`6x3}EOV(5B0i8ZtuO^}|)c^{O7#CI<`m
z0vXGLcv0I>#Bl9F!ow}}XhhUPBm@{GJGo-Hh`S*MM`eQ2<@VH&)t$xWVfa)P;0U<|
z1z#0+0}C_{lwibOQSiD!G8ge-Z4#7*T4)&na?XFrz6u69<JN14OEjQ!#XXl_<mh6w
zMv0cYySt^quQ|wgD!ohWLZnSVu%W9{1NhKdz|7t;p}*^k+&T!XyT!G+zju~2xF?=B
zP~Nn9JipKRxhZl=_Kxj80B{hrC3gR<x}g(*u-6?WOo7X~IXW(CHy8=wI7O%iYPqW&
ztcC(CV(U8pDExrmZ!E)wZ3K3ro?=tT*w{j0$CqkbGvInVO5IYw$)UP0B*J1G%ngob
zV5y+ecO8K-K&m23E~g_l-CE9YnXMkT0HS*cr4N&4jhOfbUjVoaNy}N#e(s|sZC_NI
z1Qn*#NHRoB^wOS#y?-ij<?o%#&3b50;@#f}gaC|FuZPI`zjD+_WhKCb=$Jr+eI&te
zD6^UVW|YnomL+JESA)v1Cfbxr!lw9#OZg#@#9zb<+|7`~_l0z;Rs-U-EOH-EM>SF~
zClP=xh&#FUu*f;^Wq>?;XBqI5YYtQMx&f~-Q0Kf#Qi|4wEGyDC6@ZLx0dlgOcPxn;
zH*JXkTlH0yI>vQo+zLUGZ0zjf8GM1uydTVi$g6*GcP3F|YI1^calHfHw|V~G`RE73
zwcrS|f}!+X3ygLLkJ;&_fJLdYqylCaaimea3FZ9nG9L$j(=yOJz(RhQi<|pJaYgyJ
z9g-(+J0wIu2fkB{v5%Enl6&+rA{HQ9eAn0Lax&bSQGMA+A(YU_F)U0P1vXbas*TL3
z(oqGZr+=j5{$1N_NZUasJW5A$v(XOiTM}Ax(_@Jxjx&vk@{cpH)Zpier`?Al*y`;i
zeC0=R!B2=rASv}@Wr`>1f;Q48tYrV=Saj8gY(54xl;j$o+psAykiKaE`bouO_PPg2
zP8$N}4-_$~@%n_9pijmOQ&m>q2-}Y7IXj=L2QjMs>_rMjbx5lqr?n6w`)@$g*!&ri
z1UDd;x%pMNC*<1uAe&KF6)>;aC4y}1+k4>-=9Zr;r$6#Tu;yp!{%a20{#}yYu8P0>
zPFg^wh6hzi93r8Is}9^xEmRX-eeYEVH~{ak<Zl-|eI2hs1pjLda*}<s{*Af598ifA
zW|?Z20WSeA9zqS`H-ek65ySgaC092hh#QB>EGL0YKdB3N!8D=cZil#hH^4-_RMACX
zkpDmIy?Hp*-P=AKQK&>{GPE;=%qc?yMP?a7#xfR)OclzoqcliDQYmxMAW2A)A`z8P
z36)Bc2ALWN@42?ko$j9ddEejncRcrTyvOq&(YE*ctaYvHI<NCQFBJXgV~M*I^TXT@
z34T+a`y%&J5OrHYT2#0{Mc)Q2r3}iHTX7^UJY$$BAh~quAjm1*#2t9&@-J1G2QvR#
zvyUOI6*v|`M=s4I8nI&Wrc{4+=u0w-C*Ju_b6voDMpqBbS)@s3)f76=?p@MH>FapB
zh@6sd^n3TSa(VlxF%!kk(heMeTCg>%npW=&MXz56H^*-C&03UjLXt*1!k|eO>}AqC
z#<O<_2q?7iy2Im*gtbKSWq^rEbVdme7;KB~Kh!lH;8SohdUEd7740K}-jJz?n_q~Z
ze`-W>^~>eTQA2Z9=&yRTY5}dZHU7^^TSQEAq+)YHFq3k-t{n%Ir4o4%2uZ90O-g^_
zb^zBQ_FCO#T!tDq9S$z`02bHr^wK=vg-hXq>+#$QK8`MplNM6>Oqq#_V;LUfnQfV6
z!Gj+*)PmQM^g|*QT$G$`AI$6I0S}QH`DEJ>LYy9-69o_m%4q9it9RTq2+HI?KK9W#
z_D|4=sDvw{dS^+R1{Oc?7GM4MY)XqJy_h+q#Tz(gC~}+INHJJKVnfj}Ak&#}53AXz
z1Pl0M8-m0kCD~RpwCsCJXiCtdg$+4ipk|^x$=+gl*<p|G#|F7S=FyjEl{!zV^BUp>
zQJnf8?_Bg`liVIuEu)~&AOtXEi+eqUML0a;=?Qf9#Qkk!sp}(dKj5d%{k`uWU8wc8
z1)<y<v>P3L{i%f)>k!E!Tfcru#s$$0cv}aOD*b_FuuGdG@3hbLbfMRZfL@mxcLPPi
zzvq0pMB19XL`|XW5$)dk{vX_1Y92n~jMg_mMtRD_L%>#;(n(|=JoCXHW>gYLWeX5#
zp5%Tu+7C4B3$qV&0c$c+T7_D~=#S7iob%8a7cXVum)01=d&GeJxvo^bd5)9L99(Ds
zq!gcCOqg9^5BFhtAl6BPh}TE)#Vv)KUT?~C+?$rv6(I&Fe7>cq&#Ni;`|e~3Lu}M5
z&7NyGpJ^3Nq5_i<-<9KjbG3GDCxDOz>PI)u>hjk%hVErw8?p)vdmBRPY?<`I_l-}A
z8Wjf8w|t8Qx_n^*g+<8!cg_Y*yLH8y_8V^{kHp5dPH8hj&603DjnYEzEZj<-%zb~X
z;7*w@yRmxq961M#nsWpf2H|bmiNs?OV8OXQK6jK`!c@#@@tUcv=&3UhD4!%DUoQb_
z?Zy@UuA=Gsj!QP%9-ArGUYS9vD?s)4t=|6Kj*3qdA6GqGuGYo?T-S5&OKjX`{-=5U
ze%K&hJqD-UiZE8LRh>0yVQg(+u<f;ZcPW00<_yM0dUDbzAg?e=Qh|WA1Zb|<;pJZ~
zYWRuXs}`tk(vR)VN^()5Ij+xg(usEUjJ}9oTr7ZxgKRFZZeqijV+_LQ#Mr?>7=3f|
zv{OeZ9)DFF2{eDNcL!>{Pr)7GT%F78rrTv}Qzvi6Lkf8S9IcO^ACPzCp*rjGz?3?z
zsINj%J3y}TS4^AqYlOyS9Y~sL0g@?z13taSNP@UciGfRc*BI!6p>2#YH$uCW;~Sz{
zm(e$@|J@l<e?3o8S#W&Ox7fM+$v3pWsxIcgn5*zIYwFoa-Yk7pHfVYs1;qQj1Q|=9
z@!(=>%}74+j=iV;Uwxn%9hT|A@98|l!I<wOhQ+DS3iDo;`h4I+=4#mk;UK3wG?;R0
zlw1WQYO^%gcWmuxdB~SEpuel8?^M9Oabx)i3%{&p=@M~tL$)+zuSDC<Ia+T9;{)Q<
z4BOcRpQ(>bMi+?E!3f9AC~J>vA}%*b(;j_rru+E|OEal_YZ6yTb)5N%6zYs=$E>cI
zk`_^?@}bc)o-VbUW@CsXX;PKwsLuMYC&x!2dvkKlBaO+lJ8|WId?!9)Z%z8D1WpUQ
z7rfg~os7+ZtD;PNDNt%x+GnGC0Va6;tkE6&Y@&?dGQ=&+DiCyePA5sNa4EK(vj#zW
zm&&#dVN!smOIYP*5N#@9R#eR!iEsCEx1f?}#o(I7e<}+Sh}up49ZRjrDRTHa+exi7
zLLpa6g0w)5Gw{i^(;bUudTeI>w%>!-xIZ*x8@A??`wfiab4d8Vkw|mW0vXRvekIQ&
zM0*k6iTb)kEr7b<XO}S-Uo_wRk-Sgo^_S!$qt~+9N2dDwPk;dE@2JUhxBda^e}kTn
z`o5S;204t@5QZEuy(ML$?}sP%k8Q~Juzhx8XTIw*Z~%1rzkb?$LhJ8AR7s1}e-Aj*
z(p5leG=2HV{{MUZ*TPMO0+=`>6F*iW6((UWtpCTutYX6U_eS8;L1PW*<$?ZGpN_-z
zmr0`GPp}1-TgcS*^~*Hs0c$Jz+PJw`RcWrybGM*Jgw6&)k;XnPGo#c$wCH%LwQMH%
zF6yUfN_E<@%w<JqXf^CX9U4=>y{-geng)WynIGFlRFtDik=lrei|YjSzhAI6i?E?E
zZiaYMM@^bGH$G0>B0EW@VQ)Z4)!z%B2SD1>ldqV$-jF62sxx}%q5IFGhfYvTU$FVk
zeW=*yxg0&YgctIkxlXOvojZ?O1aPV1`uxRXd6WtpYzQo^^0tT%#Y_Ug=jD&;{LPDI
zwE4B6&zttC0OOCGUKII{h(4&=ByR0Lv7!m8EXyTZ8q{Pb2ks4v3lksN4q?h6+yvaB
zoxElVIx$kM=JT`>hCRQ+%gCt<CVLyLfV|+PGju6uObLE~FkA&<Ik7=L<Lq(m?M%z!
zUDy89$$-@exCAPCN1k*LL=8?o1W}txL)5$h9o=_L(PN|X`Oabeb~SWB<J%$e<*Q20
zp+|MzA$H7AHDP|X?|T#D;4k-op&d%Ue0tG(0r<>|K3+$@uyR-_DVc(fZ3xJsI1m%O
z)wTnTT+KXazp_M!ju6B|3O0_p%QMkjprNyR3qcYg{^~+8KvT_y2Z4^jBIaIvt|J3Q
zKFp1e&(wyYEE@1S_=5&LpR|23<&c<IK)UrV$rLQ=qeQSkio+^G8LgN3MB9?1Cyka0
zhxCwQuYH$gKvF1*y=}9ZdzOIr+q}mV>#wGTE|yB<z2U~PK12eqfGPlt{VcUXpv;1{
z2RpH`nyH<b@bGKxSu#H)gdfeUY<1<lfeh>XtdGst7UeBX-p-UlH<#1!PZ~26;mFo&
zL9H5tL1EU|xW^oiNywr`<R#2$<?iYk0bf6gOIWC7rz4$C5xg`M{Ppuhs{?`?etz$a
zN^KQew|_U^k))8aJGLAXb^Z3~8P#KmriQSX1WO9zJwdkD=aQIs%x$9C4xSgRVB8%o
zFg<Frwq|r|_1mp|!JF293oCx^Gw`*(de;Eo1K)LK>&$}P+KZnDe%Nu4XW`6?pocHO
zf&$#XhB0tXFN*^?PZ8s3i6qT5A{6XKJ60E)TF5zO4B#0%609VTDP?j@yM6-3bf$IG
zA2+ONg+)89w=IF*wy8#<JLzq&&glyqD0AX9j-1T!4Oe0F$;)GAhEXh7N^SUm9A=?J
z@emdfn_2J&*mB=|k=q6*AR0UC{&%?K(jRKe`4Vs<0jo&lpsAhl>1+EVb#w)|7r#tE
z_lV84Av;-Fe2eY#eSuwfW<`dohnaoZxtq<2oH4j$p9Y9>*J3MpN@*!^xf*aoV&B+O
zLM17jb!1hgCRD|5(P!{AiIpF>iO5chnm}|I`V&+N<0sIP^Qe*Jd=d5$#T1R-jD0Jb
zNaOdQ(+Auhl}l3irZav<f3QH&ar)TZ#);4n#geKq4~*o6<ZGUDLr-)llY?)#-sh#~
zF(<py$9l0P>!=%o2hG4gbZ^>-y0{epX|9yxgCcO7oxgs3)|Qstx6Hr0BF1oqIB{`?
zDbn7zTDlFmGKd2=QK^uAr_D8Q5sM4RRw4B(YuTQA)QP1vJ7_R~6l1yS5ht=fo>1H>
zusCt*=o=jkPaN8`SRxuHc;m4dV57pw2o_qtN(24E2oBBjXwyl8Lm%0){M8D%XPU8Z
zR+hjWh2h)Ak7EsQVd4Ze!#=iMh>^u<55CS>_4uU<Yir9T#f$#9l?m+Zx2U*&EP{WR
z$o6=yl4*M#3RcooL<_$uPHvNcYEUza0<nkm@h<Db)6|3SPa4cSO_EnJ6F;Zcs$+(c
z{7Xdvk|_+~Vpbz8U$E*AS%m9;1Xp?&X7U*KOq^2*-8jD<B<dRc2kX+oo5%tGrZP_I
zWSy<p7{)uc+fUGHXrCp8$0)Q%pY%WU=#hL^z<9u;X(MQ=tC`n&Udbq#a_$9k*K5zD
z+H!~&1At_wg5VsC)mZEny9|?z=Ll>~DnSM$MZ?Ep4FCQ_8OiM+${-b<L`6hgJUdLx
z7XBjKv!e;wY#KT)Zc1Xq?`NQ8cUSb$&hkKql*5!kElLm0uKZT*X`?42BogPf#>=#w
zi;?B~d_+gp*eiL{7?qta%3R$<Kf6(j-xypR{SRt!Y_!Do$pzOYO<ZVBz|8ZU1)YO1
zq<~Nnl(&DhNZVpq2gq6c$A=BWC+tsEf;5CfnNt21y6iauS0u7QYrv+BOU<EsNR4e&
za1zRB0y>BGM|#g``5lWKEx+4@H=ve*{LY9oFprksRX*YR%oXFVp)A1EjLY70Yq`MK
zn-nBItel$rSAel*PQG4fI(RWgjeRJf?t~)!N>hBJCpG#XLM|~9F*b&VG~_+YFc$iO
zV#B>kqpBG@7fNScg5<lbmOpiRfuR7ou<J9|E+{RG2KiH^M)l{b={ZP!rXIFjUSxFl
z{@C^(+j*6=^0?Kc%>~}(Wp#d)o0_G$3>Pg)Aud@-2T^g@Y6=$3voS0+(C*TN*l8<Y
zCNouICS15GQng-=k}j$3cvky7)Y%zXg!9Q!4Bo}s8ui^V{(#oV;Y)j+f5eeqJW<m)
z3NT^lEkaF1faIm({&U24!nkMAJ*6F=6EZev&RrgmOL(BEuOoY=%#;)ZVyVz?juWUg
zAhMPcj+XU?!X2HbUwIv#yI7Zu7qnMq8~a3~>!#*Fle%8!et$8G9I9<jWqwbZAo<YZ
z-vt7%6N>Q2dnO}--krFnUz`ju6M~&wblPbn5be}gH)vA&00X4Gi1#U@z3Bc0#F&PV
zRFORBfb7K8Ibm5x*M6=A<Z^>8!hEPAE;#l2L|l7@BUA}KEiKgrmLZN}Lb>TmAKL6y
zBm)L}*c&4<XTLcrvth-<j_IbpeH~`Xz+)f9og2nPyrT)qK9h0|8D5&G%+$CHTtiGK
zDfKjU<gDtx6g;Sh^xgj1RbT)l{_35GIYeT3{XB6=2xhv%*?ivKd+Vfl54&>4lr%TW
z<ZT{i!@jol*8;j*D<ccRFxS%OuGOccCK^0`XhL^bY8->HmvLCYM{3#7_cz4IF1r!3
zvlTxuUX>U%*!z2MnW|0Z*=2`|n{c6xrz<^|?3nzDsN0a4y(Jt!jCj!n<oevyewqG7
zL|6c+LR)SkE15gUR^@qI*1J`Ej*4$a*pS_6)!-n@0?%Xx*V;1YLq`sRWo@al<Fh14
zm2u!?@7se@xv9D2e9|n*(Z@Q5KStTm?n`OsNScH0p<}zsKf#mJQAn3;g+dl&-Oj?d
z2X6H#G>36-g`mB+{D>8YwiD{D{mTM+Y2_*9I5V<eAu!#Fx(2RD>{GWBUze3Y=1$53
ztO2%bK~c6>eJuPFoReynnk4pB%)(^6X#|DA&6;@whTTPzHZZ6&m!cz94@!T*m`1yw
z(mjCqB%o@X=ESUe$>y5F&*G1O0^EOQU20^Pl$bze<NE@lCb=CLE3;J1N+$I!y@q`r
zg>bx&t&p=YW5Prb>=y-v{~(f3T)$y6y>c%Gg$Do;lt;_gGH=Uq&RD^0M<79I5YBg_
z)i-<CRCb~aqcR`vq!mBEs1somQR>!r&ObTpZC2z(w98W>@4xSK*(4fOI{fqDa0MA!
zg=a<}-h=+Q>(e1IE+0OE7t}!lRQ=o;Yh<E9F`Ebvy`qkv7t6~lA;ems8~X^}k7Qs{
z&8L&Ag2(FnV~j*hLT3Ij%p@*NgcOBt<Mj)7EW04ks6Z=Q<7#HgwR74z9f<+$nF2pE
z#Y(Bcpe{uLiy)wU#J;=-R@pLrH-`sqYVMSfxD@auLm*_Q{{@NLEs!`()$3C;Tm9N>
zI)`JR-R2z%XV;vH-VpR-HMl36msOY<`<X56U?|A>JA;JUN;8u6cJna}BBH#lUhh81
zb=6<vfo^L=%6sS=i^kSOxY6)>%VS!QL)*^=UaAvRMp08!E97XgMfhY$|MF+J`|p9Z
zkt$$?8}ktB*N*84BHc4wSKZVdz$6AICSOAJ&|torZC&44OhzBor#&o58S-lioxVbB
zc<9^Bi8`FBkLTG80PT!WojL1rh(St!svkc544<TFCtqtm!H60$%+HVwx~ZI+)Akb*
zzXNRg*}yrGk!)M6^A#^7yW26>z3GX`8Bp^zDJ*&pvD5mrzLUy+_RXdI>H9=12OkyP
z1YOGn9;bQ%X@|7mWHD}rjC+BJmyp5WgYQ7^G?ojoP7~vFJ<F%;y{}(=XXxbVk+H<$
zNBA=0cQ2m}oD-F$d+#l!Cym$T&irn4{QJ$ng;YW=O|J5so?<8q!;QSe%FF7MlInzU
zmv8{Rs+;|4SHqzXnbRx88=1cC7gzbhw}ep?2U2vZ`N&ODqt=35=?!1>N^r8|KOSOR
zMbgMOhxa;LSky5d5$%<MsS(TeBt`)x+xmHO*0}J{=U)xYuGsWtUqUie1KOO7gLk<D
zE+(4M1$RL5vjsrl%*5k4!-$e=(s%bMAVv?S*+%Z0atQ~BMqmow_uzfXklJasM%loA
z-zw=fu;Nz@UJrf;g=@i@+>ale)CW1Q?|2#b^h^H2d9KZ$7?nZmH=es>i<s)r_f*@f
z_Wm8i73dz8*6LtOwGzMM({J=Blb7G!v)-THr}R@Sum25E9xMOo*H;M5h1z|ofhZ$8
zQj87`3HCejesd@Lo}kG`3EN31VfuBi<$Vmwt53G9kXt?Zbg5<ujLA#FZb@L7@q%+%
z3a%-xd+y#`a9Pr}&{V$g%w|s)&g&emtAYsu4(2nQRZSu<0JWnBzC$R4dg6QM#>%4$
zk9};ssz(Ze>nGXHc)mNe;~uz3j@A!!+SI2n7e+OHRjLqto6L2WPG_{Rt(TmG7Cu!R
zL$N3z9`0|nfoVm=P)5kiM{&VF-^{B8ao^?H4BAqG4`Q@Fc(89@UuFwu$4l_VMNI6b
zE#9c?-MX(p1}wR#1d{6c*r0RpuApXNCzE`?|4+pJ7mM3cL_ruSNf8_QY=652Mi=ln
z&9rl4<o-ZT7M6k^Z=axdr2~o|<Mr<2cX0jRcfjN%<Z{cfgdOwG+%>}z&W`Qa4Tabw
zaf_7{XB1LtA8UMCMwXB*VVS=PcyN*)3;aRz(t-P4)+<iu=5%eYe)VrA)QOOxJ2NRX
z+zh@+_K+qeq{jPE-br+aC0us1J;?@DhuFcww@T&0cXY3v3E}^J8}M^c?Zg#<)jGHO
zfM{uh>_Fl-Xkn~`WeT^Sd+i@k$A`XUT-fnWmPa-j>`H1{_CpeOaDGc`0S7J2e;ITg
zKBp!UrWururLr(rd4{cSVd=9<6N;z-E@%Ei0$`&_Iv)dqPZFsjUUO3}5s=>lv^C4B
z`TL4<6mDqjr&O(FJyO2h3R?Fzs4S)o1V2KeJO^keQUAHT0g#e|I_da|zyQw(D5b<b
zLIX5D%92NBM0;cfETfNX#F4bKJpRPkjqrPb82Ldq+%waY-OpBPlq91*;9qqoc{Z2f
zn~*gzSbMbyAoz}9J@$%OXID_py1<g@OnZq>14}&V$jDlZG>j+@7=h8opiNo$`3?kK
zX~1f-zSY@><MG@dbVHF|ePVG{e8oKNbyVroySjD!;kCnv$lKFe>8Rh{i9$_e1G5i-
z)yR2>>jmgY2EK{FWiGC_q#Pz6X=t%}W2B>6bl36aJGZCGz{MdqZSoS+8Wx_k7C;y5
z!py2>U&8MkoHaTB{z8fG>d;qdFpJ5`5GV~%`yNz+Th_#pn@>;XhA@PLIU>)C>}OTX
zZ|usn<4oMuy&t;dG8;2b+d~gQ)p20j!6G_TEDt0>C}gj)RJMR?b1G((0abW4t6^Q>
z3uGlkyS@MGl?!u)cp3!Z;Y3VJd$Qo2rN)LzV&%h?q9p9Bx>Urq@2iha5_A5oMn;tJ
zs4kB4I0gUtag34{nE24fsjad9CcZc1BA$!8|LYT#WpxGJb;LiSMK|O_=0fQWQ+$^9
z9bK9}_st5q^t~rTS5s2pQ@T%*PP^lnMTJmS8;vlV{N=C%Y^of{TG;DF9?Qi0FR}GB
zIl{q1pA;<#Qpw|JU~Sg5jJy&Jx1gesh76Kj*mA{LgOj3Y(qB!>JcN`*2Z>N=t=VnT
zOM-($e(PIINkz+nBGEv7p0ZmKEnMUJ+2PVeNvME^jDjV`n7K0v$!^M^a7la#wJn*>
z<GYf-2@j<D`rHrOseJ4GUNMsm?wp9MqZ0VB2+FIt-kL)g%dJWLM|85F_JOI$a)<0a
z+HVBVNX7v^G=Zn>9zwliX<b6O3KoO|Gdb1OR$YR$vVE~YZ~0F8JvngsjOKr{(1K*4
zWgRx@Xq2R5p>22nut?Z{c7;P7H(obN2ers?rs~_zubtjF-{Zi#`+*Jzf*F){F$IBq
zSMrozj|MU!Srh4a(fo;Is&mjGXnLM2HV=tNM?9Np664H_66Qy71n>t0I4ADn*B|dx
zhsDVh?%Umrtb_ss3duciYbNTHpJf~J#JqKj+WPWGsaoj($cbZ--V2hlNQi~S5Ga*!
z#r85=7%?ntdziR26Ph<JGJeW<P(g-x+@KYEJVr1p80bez|IGeliP*VHkPXLTdMHWH
zG7X`DVVEgzMX(bI&r&e1q*j0He7Th`T%_w8@)InxI*3}Okv}jRIEG>3q%iZUC0GtS
z+P;=yBq_<PfVZdUklpjyYk>;TFU_%%VDvJe{UDEKYw`*6K9q>tje3R>x8rM9z8cpV
zK*skO15k_bT=B8IuYPhXxBKpaGH!l9c3y@U083nx&X2rwnWY_s%_cD3eR|_N#)Oc;
zm7`$>?J+Ujdl)f#pkcc)GpiAmbJZ~<04V%ySH+a(HC^4jK*EUXZ~W~gdZ6*jl@=R-
zlM!zq`}dolW*r4UyC+>s&*f$Dja`uc#C-(M`W`ALNmq;!4<*c==rc<oqtZewT4NEA
z7KU1^65ctImX9F8>jZ(o>fX7|*I=(mxo+nr<KzHQmzX#qsMi*FIqIEg+^6svzp8(F
znI(RCW2SnaJ5pA1)#$14;`4>fAUVp*iT_HTx9^G?Xg2Vy(S<7Lb-@Ct%VzS_B*Bq6
z^4p`#q?4?J$%E$WJHE?oDBe=pcxBUJV*eC9(>0YSPjTVtk%`3QQjJr%DhKWZ;cbcB
zz@4ut%b@2li5F2f7MOUGi3I4!-jJ38db5@dM3Lh?Jos&L{^NTj&32n{p^g}YAXl;0
z>vEy;L5e62upu|r?3S#l(bTiF!MI2z<6c86o9z3k@i87sbHo6;hQd&U>sjy<4_82B
z8dLo^&EuJvIJCvlr)XDhrlm9OM|aQxogI$)2LHTWSsG~JH?Ph~u*C>KD0}j2qlo=y
zz0}U?Ae>M%5f$Wa|H$Tjx^`aF+FaJpTx)$>l(#VH12og0ePM(;l0;QYW9{4ioZu%@
z=PFcW?R(<yA=;~@5`~LyjcLoIp4TuH0j*k+RJ{J{N==UMl|HhYJEMf1*hjk<H@Wkq
zrNS-K-Wl{Zkq!&SDpDhsgfLJI-t@p4^C-@(b?wI`_;_R*w+8cNSsBcw1;y^ZPTmJM
z$(ujB_4JOz@M<NJ4||=pydLF)(7$`*_9atoZ+958qNf!LWw#z2)Pz$L-G$86&~6ky
zMXNbXM)x8$B8dH3K;O9?hpuJpS7VpX78`5u{Q1nQZ6pHA_~VWa8<-f8?oHEf|5jPH
zcCb85fEaN;f^&jq%>H1Nr*+>Cugv+12?<X8mXT8!-eLI?tc@)O7j7Q)0wCc5cj8SZ
z(h0^bt+V(3Vhn~Y9{g%eHql<^ZBi{8R|Tgz8r_<N`rDz>dv1k2T4tIPCiFZ&6%~h8
z%kh|qKJdCL;R$N56O7P#d$*JYn?{2X*LOfLLKsk*+j`Dx%aaS1vY$Y;n>uAi7kr=y
z-zn5t-pY%s;uSp9qp_(snWlgE3EQwr#M{!ZK1)A?<r$0Pd=JdV+Z}AeM3hQ~xRPm#
z+|o8mQPr?2PfMb$rtj!#{?Ar$8WdDq<&pdgWs1J6D%<}7(H@LJa#o|4?(i@O;I~$6
zUei9^{mvydcF?|KAOf{Ik*Mr*hWLTSDJ<ZHhJt7W^^AFZlaT@3%7wpt1+s?SO>S>N
zvWSLIg1!e`YLciIY_8R26~=cH#g;!h5~W`cIXBOOfg^H=qo|MfJl?cM4BUDmZ6ofO
zUxrny@7}=;t^{RZsLw)G492`==-zlBeB^53IJ|YD`xq1Jk7qV5jnKo^ddn)gJS0V$
za^@x;3|d*FD?u1(IX1Xj;<)ZMquz~w-yk;;VB(?)g1l620w*wrR?nxsy6}Y(R7e^I
z>xV%+AxL!!;{;-zh(3w*;>}O=8{Ww8SQf`R*fu!G)@p>PO^T&!uB_9F?J9GhiJ>P=
zOmR&du!m}aDB4-Yh#Is)nK#|z{84)CBZ-T_gHah&SlcCAUu;=->cg@=pElexq~@?6
zC|CJRsJ=LiRWFB!c3cCO+TuoA7#Q=j-<oo;qeY}GMGIyx{*_=B64J?d85_m79vel%
zvmEh*@&4Szj`f3=?+RQ=Q+`%)OeQg5>5c~`zOY>IZgS6Ou7jtD5Gtf8SOX%8LW#Rc
z=Sq_&91B3V%JJ)R(A&17*(-b*0JAs+McxiSc%shpI#wMyfCA|)84gV3xyXuy@{(r|
zE;PBRp_J2-!stmBVKOnimv8OAa<0@$c2KL=w0~O|r4E|zYGzGY8R9uGm3K*XLhP%4
zlDPSGy|#Z#RJK@;DQ8r>%D9MC7%n?q`f%eoh3nH4G3XZ-YS=9~ri$=foP)NG$o=z&
zsv~zM&4bzcF(TeVP2`Mqq&iid%}J28-z{=z8g_#RZ8sGEYrEn2shZAplyGd=aAxE7
z+Vq3O9sB4r(2dB3jtc}+M-T_;O4^P7j`Q*;KlB>Oa1i7IxoOYb3s_ra(EuVW-}FM1
z)s<ikI_dPd30-+3(0CYKYJX$~>Buj<-`9%hZHF`c#}_x6qRYvHCoPR$z77BKgD(Yq
zkY<>;VrC+l3JiQqBT&WGl>eM`PYUDGpWykl2_MHqYg(ETw{V@bGWrHKKKB>L^}6tO
z7WG8-etcFJ{XAKHGSQ8Q7`Ova&;o^hlg=gcizhO4^5+Z|xjd_UrMdC$a;rsn{4of;
zVa^NlLDvaSkHIxsDUL4O^1DaX*aIS`+~{6|4#IJG@^#ocO1lY0g-($?@Xvvh5;cdk
z1~<0FH_W1jp!AtIqd!o-%k-vPGU4y5DIm#t<n%i6L7&2qW7LJ^z_%>5;@+$(xrVI#
zoq6nNi=vZRerfVcgax3#>BK<MenLkO2@6y74#$20)UHG33Ml)OM?Z%im#EAnU13yk
zN@4SpY9w$gezbBKKin#BMXWRkR`nK*-nbMb%cJc>@$*YeL&CMGoVVEu^(#IWG%a2S
zzLHRhW>Q4I{|=l#6X@X7W+b9qYo);;A!EqZ|0E1Df#K!oO29QYmW)V39FWokgW(wL
z;Pl`oA@%;q5@?o{VSTqw(h|iv_u|V}0d2(fE?k$T4Pv!3v?6i6;F2F?>yE+&tyk#p
zh-Y7$>$;ZQILgl~4$rK>3r$I3q`>mDI_IHOezPDdpUY|fmybO2%Ks3W(+u0_bi$K~
zcymQK2Tge$=qx8M504w&D0G_^kBCgx@jN$rHTze=Y)v*u<MtE!fFpp^&o|{eAlIe9
z-mX9#{mY1@fu_>D?Qa4)CVC%Oatoy~`?ByuA!K|kYp5ChWxm)_u5k6^+Dx(=>VJCk
z9-7ts8%<%h4W@!|y&A*&_wVKHN+y6`Y|~Io=K7$6w>SNK7^jeC1(5OE-6Zpy7uaiT
zQK$<G@nnKAtVvMflv4#xW(Ez{E!5oc*@N{xDWt%Ln-h23EOC2~h=?AEuzArRTrDBE
z=I$W+AH)qo$fz8;STa}UfW-!i@zthmdv)ZuZj{Q|vv?{9$<;&)l9zzU#s5m3OKG_t
zQpY;@NTqEPC7b|OXk$^}tHu6d4R**OH5LFhvv?Zj11KRZJxe$R=$}xiWa;H7TnqZS
z6YR5zUsjEzqxGlY=i8p{XWPh8h%jUZq+Oa!tT=MjxoV~mODTHOtC=JB<ux#SH=w8^
znF-jW{CU?^iUDd`1537JGLxNUw33U=sUWt{M8^h<=Fk=%zgJ*&%FUXVwUqxGk+7ge
zwR<45g@d@OF6}H@3p`_{{<^zGcDERzrM(UUx>9_WxmCW06u`l&X_v7rE2U`h>)5}w
zElb{q4Y~05z~rYRlr8oPBDU6t$Rno}x2`u{Id|1Wwy@7>yZJKgCU4gRK=T&lwMeTp
zT<YFkuUC_4rS<Y{Q!&o{Tr>P{$dD!Z&KY^@36B!t+v(wH+n~X|b;b-7LU+jaHeA*f
zrXMdmU(!}YiT@jt?78lH%uUj?IfXJbNpysf8^XIVpgK0GjlEu=EY%{%&R$0Bmi%Ea
zko2x$1`tKB<aICT4#O}0`ys1#G;Vg&nwb3|>JWpdgdI?qgV#92DN)jBn+@tmIm^3J
z4;#ZkwKb1K3k>xiQQu{A%3}VTtbO)qiB~*D=~T_Ui;h8)y|oPuXYqvCqjvDdMs#_e
zWAG3{v?7oPNXxS!ZtZb`WFhO_Q2hP%!Kqazreq{`K^SO2?sDQ!bL&C`^b)%r??xU4
z#JHZ!Vr-U)J4(!3kmCq_?(l6sJWo|x!bSbULAMQkq2wg$(M}@rq<;Z%`s*a_poFWb
zBDoi`E%YTuIg_IK0%132O&|)W(Lz(Z&qcA#?m}Pw9<~Gjr?s#tf)CMwT$e|y0BpWh
zf%%sP|I-aa-LnWQBz1}U6kD`l<b#KH5fQQal?ZpmH$$y4hI3bn3z~C_$ZdL%P)m$H
zG0P$~mI6!YXiM5cNdH?4Q9^i(NbIov@7OiDF%H3zAhc=wLXyD~?%6FsZ`?;n12_v7
zhjv!EM~U<i!uwWqj&2EZiRwo!IPE7`;GRzaAY#?e8E+2>_IBF9o0QYnrwS~yz*XpM
zy~N~9#~Zl%i=x*TOU!sB1QA{28CL*A(I##Pu_UPP0^{#iw}jYjh~7Ax^Q=KuR&3mf
z!-m>IYWwO#MFeCyq2THe5Bce-yu0tWH^U)xXR1tx1=3BPt1=QHp=<$i$V-huU!6d$
z=cY9Y;S%;QGznSCR<|JQA>kZ_A4b%gR<LNdoDm|E8o~FUO`V9=&jL63n{{siJ$OSI
zD-^Ge_F)<zx-~nl#x@oCUm+@XQf@$$#?!aR97Nkg!DMa$noo~RRy>);Q)7<;A3VzK
z&(J5cDezobFSjnkXeGL)v$uadSJZ7>4qNP41n%<s+_{O!T6oPnzT=E;WjZJZt{vtE
zt*V!c6JJ|y_WW|h_5R2U*7<8dciNOHgNYl8<@!I)1tZNl27*LE$K*X=JpKUVG3j0W
zzZAIWU_sOuT1v?BFQtU^BuLwjR;`8oJJs4KBrjY<tDIK-JC)ODb~$4`iDc~a|D8xS
z+Ld6Hp(Vvp|4#NnZ}U;Z2<&xQ9m9P8ji^IMZ`E#3A??cfAYzZSD@P<qMZAt~W1`MK
zcaclXi4rC6lqeTNIB?!=EIi>37~nNg#ekHaa%3za=Y#O)1*iK%5>gC26XR1RKC~CI
zBevtW?Rdr)ACy<LJcr|d!v(;d{|y&ojqv{s7i0DK|AvdvI|OCX|IfpPN7h+TfMTGF
zyVEe^`ZXRU(2+krzy9=*2VvG#Cn(n$Te4v=u;5FCQBV<f{IxW+uPDc9*_X?I7)}+r
z3>RP^CM*X$Ef(F8Cyrg!Y&ojv_O~R4KnN0iBSj=18J6e5xW>2F4D~;Cy=&(*LLcwD
zz;C@l;TT%-Vep?NA1xb+TQFZrfN6vpyTD2l+SJcb@=m5dPu@S9IPnSD9QX_w!=JFI
z|GU?2v;H#h`qfPjAK<LGC8T|hs0?>;c!HCYyi|cR6(83jOz+Y2|Dv$|znY(hoZLUU
zU6d~zcQ`(_U5ppwS}xO=j~GI!VrsTAymEodG{KdyT153T+3?;nU$xnB?h}ImPH@Wa
z7I~}^BuSbKqx}(x`Tra_jBNvR^4yOJkCx+h3&5YRlq<6_jB(@}RAKwfH5h${pQ#r<
zx=HZXBoeD%H}f~Ly|%X96=3oPk$TB%x~{UU$9iScA3K6$8&8}jr-+@)#J;2o>hu%N
zxy9zyWotKFbEUb}ae=?786CxHW60&*@cN1CXRbGzhNUbnLYOsmGC4`;w-*-x`Q_Hc
zGRchHYV4LFx<{?(NQ_mWLHpEyq`#=i6Gf5i@%ze4hVec+61SP>m6g?9s89DG&&3hT
zF^5-V6>HYn{j$3Ky+;N#0?T)wrf`sPZ(Y%6ZO1PN^#UO~nP+S)1~1m>MbzA0y1``c
z(2JU-<9~O2iZy{MOzceUXK?z6d4Y!G%Gb4g%ZdGG1sqI>jS)vEAqCecH1GM4xi>V$
z>llU%`*~tA_6k5kUsz9U9DOcSVQ%95LuPiOV&@^&M$Oxs+@mq+Q+SyPJ5c2rIyAIm
z=6{~!i`_p6T~7}~%1*%hy`Uq;IPSZ4TJ)|`|2FRC1^prhOUH(T5#<l9)PeB;@bp;d
zg)Isj()DJ6Fj+`znyHm+MIH8C>hp!=jLJ$rDW<RKW(oTTPNlpX{qLjY(%3$t{orpJ
zNUvvb5-Ma$r4jNjLz{DW<r`D}r*%rV7Pwq#ULe|D^2nVqT$i=X7d0ci2fR$kc1AFg
zqG4rA7#OW+W)j}kdoUmgci?)ls8u-4cME(B$>$#MI!0IMZuHCsJ4wbF4^-uRJXY&4
zRpUI!Zf?H<sDf-0r$33W{`?VG!e}C#3#r27S#udWs?~YU5+wpKUpzAtU9~~Mx9DD*
zbi~JW*}+Xz93X&Ux%!*<95rw>B;jwxG(#UlcXv;l{e09bWE~^1`vM)x@~e)XUJSv4
zfkH^W-%`Nk$y-6a?D_7zu4CJah~*qKifYeJ8tD6fPQ!+OKZkCQ0}AO~A?*V6V+Y2{
zxBWVbX?z@vSww>=>Nq;7U7JEBVs$npjzXAv5hq}SjZJ$(n$*}M_DpFW=Dt&XW6n-W
z678=1M%%NbmG*xY=MWy|BhCTz7>RSVD5k9f#zBmSwlamuhKSD$%Gv4rEJ}?LNd7T!
znQqm-0s2!#9RJ}g5a+w3nY938Lk>845gUGj))e~cps~@~zx3FFjh~D6#A;G<aJ>j#
zFI$TCUzY$w<3qO*UdjCPTBrZp$r3c{+c^}bA_S7qSqHi<h2ll#O$->VjWb0cx^O$B
zsC7_H7$N8g#;A=1WU3Qq0cKS@!8Xx8+y^ieM3{aSDy$Efn<sf|U%mXDOdUlZFJ-qf
zQyiTrGx6k{pfe{d_fojkz%a($z3PVz?(P+#bqSqxK33-j=DDNY#6QO`Z8Oq|I@96j
z+5CbnW5Z7+Ohm~)YI3rZkvNtBVH|iNJ#DPrnJIzye=c&_-9shA2wmS*r#zvDE?`o$
zDbY2_p!NjP1;F~np?yF0(LVw7d}L-~obfe6M+Q5KbgR=&96!bE&r-mN_2QKpz6X+u
z>Sn70SVtmR=1A<Ecya8H|1pZBCDBMKo?=Q)<31oJxqarkI(w1|Rm}z)x9n(QIbFn8
zGMNhevjPmwQG*5?ub$+`ty^!qQznZ<>pH$#O$j$$q&GCYm@jz)V!{`_M%$&hD&MVs
zUq&BqGo$mm#$oyk{(7bU@*ZZ;%mi8Q>Ew_5J(L>g_r3~`ny5ypBT`TUZRXKWrl!kQ
zHlt$x(WD}_)+pVLd<z-ZV+z%+g*sb5%#RAqKV5$&ab#hVUhddKV|<PN&%cKy;Y_kH
zF|j90$3qRd`18gC^HV}uoX~EjN-j7vc9l`Yk>%N`!hS6K+x%J73jl7_%{$81qetM7
z@A#0&LWzUQj5h=7GUGne*ei@KMYI}$FLczZ=KmJzG407=6+7=K&62g)|EIl8h|C?Q
z!&2}!&Eb`_X%hD>TF_NHDx?*ujm|6?Zyk->Ct^qKg);D*TS1LXDPG%g9mN04_?_`v
zAZpAt&M2JZt%SMQUHnS^-D!jtQ2vbqy`5a}^n}lTnAuBQK+?{h<$bxdZ+IUw^#bUj
z>gJUjv)1Fve{lFl->i`j**pBXM*WwkIeJ6?^--z9G%ZH-7w_5Qo`TwK3@wWHRM!mW
z{9=QK?YlJ75jU}r1&nAu93c%F&{;Pzd9j#Up_@Sr9G$A;<%nOQn)j+ekLCj67AY<i
zAJXB6wj-HCTZ{hwy1QPWE0Zw-WXcg~mI-CwhKH;SSem}sr@<QMrO_FW$!u7%TQ(yQ
zl!sZ8T<)t4lyG%iAY;8<)R|RTB+)%M1}@25$K-xn(%v_3INOof5yQsh1#M>PbTXtv
zb*SsGy|a)A%A$49;t6%EHQ2o_rgNb0!veHuB2Am4&+J}8XcaI~=OrpVs90;-)su|i
z;F*rXs{AGxg!dc|FBk1-NRafHs9Z4TAO@-4gF~tt-=&!1_$uf;)rnvM1Kz00%3Gdq
z=H0f-&BnBZ@oGkAGv1x#r*oZ!-t4yZ4_n<{Lf2%O(;RQPUBNvgWEv4o0#C1imSOjB
zr~c}QPA8bvL=-OvH6$A1V4v;v+xkwKzdR0OgE`K`!j~{3$n3uIypCZq&6|*1AyvQA
z{d_HzkH|jXH|1Yy_O6mLZlX-2THwm!IB4~xQ=JliLGjF<@AOQjYjAxHVZ@Kd^1pwj
z#zxBm_UsK&AHBp!c2MpNoVoL_k`|#Ifm`0D!IgVtA6Zk82<U;#<BZ<2an(_2+s1f4
zepeH}`cn#(euSulMn&9<jQwxi90GASNWI_A>vp@~goelZ&o{oUhxjo_w$eSD;p!U@
zBAgqt?Yr}D4&G-O;<zA*VdZ=@2`c>C%Vr$u&)ls;`lJp&YRbSu)&;pks*+#DHwru`
z%i3z*r8_>yB15@{DNJ*xO4>a*oVOlK0{@=-imMGe9=dpK{&1#<trtVrwmQ}C$@l>m
zy~9Ki#MyJ@-bT>TFmvqbK6e!zQwDC+nu$HH%d3kv*aeayHjruS$;PX#&wYq~k}(Jc
z^Dz!;zqxR{KxVMSC&S==nwNlP)Lt|%nX9y{Do-uVFB<rKy}L$z`s#b4)2Q*}d2m_B
zb!t$;5v{07NckfUsj+(vIn*rs=jYJ@qr_U-ZJHrM<DW+_c^;R#k`!6W)bTu))w_;3
zj6vS*R@+0SGfrmb+qb@5RT-8!$B;|tkv^H^`Lzs>-w{IlHXVboU^wfS)~1S(Mr3`O
zR}bmbR=`tox6TpK6Zx)nnh;)dNjk2-=GT>?u%*F>aNK>o_kul4r~sD}UQ+pB)FX%~
zBhhF?X?#2!Lo7|Jf(M@lKbyIYw%j$z%?`FFmbgoR)MMN5+0r$Y(FaY?1s4p?vSJFH
zBIDNhs_Sx#I**+`Ee(m4JYzJ(Y0nbgx-3VWfeE>%kdh9lyD6W!GJ|3DqUwydWWQ28
z#k@N)f{as)mvfYt&|JY7i4|_ySNUxfuDVkRlLF~g&#ocC8ovK#foT$tZCYVMV}3WZ
zK>2zh;{2{bSAGV5=$YDO*?LFjB~pjEzGLPf8TQzJGi>S;;>L;ab~GaE{T$ObJGaB(
zk`VIZwj&qv`0@InjHK(MC9oJ5M(+_6^el@5l2*Cea1ffKe^B93vvgGG3H>3sj!gch
zc!|XNlp+3Qenx@@kvDzcM-wf1(ZS1~cTgthzf8_%A3kkVyvyHl7pTcx;!WB<SmU1<
z0w4%Yh)sRoqI=)p%o`Si!hD;><qNq#;D%3B-qSYMx!w+8nAi%v#ue@zjx!sK1E+&!
zwg%1Rt&BQy!7F-e)FEc7&lu3^*np8Qtx%B|f4f}u{oBo9scCZ+vKOur`1=Z1w7=_u
zjJUr!b@(M8W^>$Lllu2~^6RCZDMpN9O!=lnqGkQdWmK~gW(EA3)D;wt4Wy{;JG*V<
z$JLC4D+NN+!6u#cm`Oc@XZHLfaqx<?JRLmbuXhY8H&-SdF$nfWme-bpSt3M~UruD$
zD1^TYuotJF(K_KmbU|Puro;KfeuxYo2H{i}?ao61W?%M!wp@t~>5LVn4gU_j3GQjO
zZO;L0Zc{bX0kbAuPYQZ*9X`bAkarNhs}4w;tC?LIK&6u0P;5iY<PrX+#7)1D;g~n&
zOiQ`><M(HNVs}O)O#<m*itmEgXXO*+opmi<5O0>PMO~T}0p^Vv#gX$=*MkD0m<sM&
z{ylIyBm>1PbJCXK7JyhHx%f)QKtf#H#!4SNGB1oYs6A_#SaRm=6_}jBqE9FjmO~jK
zQ!vdY3f?Vq9IwFOQO^?pZz)l`fEg9`HZx3naEeq$<VEbzlt%M+d9bW9CYp0)OmA~B
zc5TYL2vXfJi{1hTh7qK$&Cnl3@EiSwUGy~Z^|w<=&c8mcarg`Q0hJrSnr?q!+QjgK
zzdmmQmIS6OQD3RWvF{}Ew1P)<vLk2FepC~<H+@sAtYi7q_s!RQH1FizTNhKl(Ci%d
zOof%q%iA%bd~cu`m6e!*iO+KF4ochK2IgOe&NqpNMMU{eG<J@wt}8Fk4r;HNe_Y8c
zrS)Or22}Rk+IhSB`72%#9R!?pOIX;68MTldu^N#Kl;4`+=mN>@mZrS>30w5P4nw*x
z>Do3;G0#mk|LuME)z<UCK!}(vcRTtNxS7Vc3?NO}qO*=S=vhWyF?a&5wQ2jUrb%OJ
z!?9hkOT6_d!o2webm8}uJQkm6gE@;IR1~*&Co<j3uoLdMK8%6llcRShN=zsPKQpwc
z?Wp$PqIXX-jMa%SO!0+Zs<I_m-;(znP;<of;#w!_1>y>iYLCN{5cKjLhGH?1OZ7qS
z3{`q97xa?Vitf@e?lk@Nc(mncby0(?XTi9G8L|XYTDL{qzDyWZ=P9s{clorD;W8Ye
zvM{M|#k;Tw&}1c96waD*szc-3^DPV*2$0`<9<r)#%=)sdU8Tq_uuO~$^+qOYonwA!
zy;S~-fb$|!wHiH;=a54obE7xfc^G;J3zh93U37e|zXDT+m=4@pP|rk$5xp?pp*q;R
z_B{%&^ev!{5cOJ|LeE}a>#K=P-EDb-7s)uAd{EJ}q-_hGDv5teX}O$Hi45=K^~)HQ
zi@+ozgDOg1&4ntVjZE&=2Ejw>eKPMXX&c$%4;{<63p0jeI)*p4&v4bB=xoXnONb}P
zLMH0OJ%sClHXOL5T-{1tWsha^>!_6oW39RNYMuvz;3C_u8%K93wv=x+9pyZYTgtz-
zyBcQo$V@BUsXI3Jo(}5F!(#U*>+`a-O949U>`LB2dxcDqq#^G?*q3d3I(a!=7x-cD
z^^MZeFy<?SF;v6^R~#V$a@0}nl(G(F0iu}amIv9THrIW~_u60Hyot$+4@~Y7W(;i7
zv(v6U|FSeHv}q`_8%|G)MVqlc<v9aHEDLLpxZD!5#)q4_soMnu&rt*<SW57slFu*!
z3ld_IzKASe#Cy3R<LgiMR?pscWAP2eMMT8lU4Y}1qz^~a=$Y>7%%6=`nyd<SY=&GE
zY^9&s&jk-<N?>T5kYf?#yIgsPm6K8Er!Z#6A0v!SOg^z&<G|1G6a}fLL0c6&F6%wV
zHDBXiAg)hkU3x3){)^?MFEmySS4jO_=1AJa_R(66qecr?$e(c0+st_$&)t}WSsq?C
z{6W_Re%+%`ZN~MmDps6piK`8>n;+_tl&=LS?8}AyeW=7;k=!pn(U%2WQ~Yz)*kOhk
z2X>QXhY^(_PTE!iJ{Z|)*Pm&Jsj+wXL2oWFbE+dI1dzL7^)eg9ymO5CZ-tihCze5;
zKy+!6tLq<LC9LL*&hV!Pt$808<8lG=B#a2Mt|9r7?R9hTz<A;Wq>_>FcR$;dXqkg?
zH4+%ydzn0BM=KcWAjv(%EoG%}VVBEo__Ma+x2_6*8}U4I(s7*lKO4V6%r$`@Ie5p~
z7VC4v5WcH$#qt}Rh&0z)RtJ|YDfs8)`>r%7TX%~~kmfa-o2x*CVj??<f<`ndTpCHH
z)|A#S;QUsCs64}J5AjGO*)_7&=~X__QDRg+VxOa5tvIiaXkoCI+R|DIv2|mbx^L}W
zWlp4TZqZ+Xm`W-WT~$%fgQal_mzFd)W^5mX;V5GguJ*g$72Ai8NirLZNeyEKQG`n?
z!@<a@D(-&t>1T4M=wy78cVs>{(zh#Mp@vuk4=&*8IlGL?_tWdl(eHFjziVJkjr>0+
zl?P~5)SbQ5((ze)E+?{fTKk~P09ku;yX$OP@TTr#B-X6}bTq3E#-g3)G6&}6X&1t@
z2Vg&w4sldm@wr69?VT^AP>@_nI}ZVjpE@bA^#YJKN9gp#l?`txIPMYzMZ`6bF3DW!
zBj=)VkLW_HOX{EYO=3RGDnJ&zUWXCq8A5dX=td?P67r?{)f7ic`1Q|)df1mv?>9l4
zy4rw48sNRi^0;X-TTHKm<JnFoyv|d7;awq-U5cSNt~yny$D_rKQ#35(c+}n$e#p9#
zcKye&^Rq5hUQmLU3~e~xL-Jz;<wXlFM}}-B3o6><JQ*p*ZG9mxZsd@qHEK>l+kjLC
z>D0lK>Yr-Ud$dl;+sTtAjLSzTYY<|~Sx;}|tkDKop&7c8jA6yPem|@R;~5YqS5(%d
zLN3$|12tV#6DWf=Q$aP>fy(wR{GH;YvG7k=PT1-E3`_-Dodf$oy>9qxrhzX<f@RbG
zwqMkIjAmhh@REDs(%E%)-Z7-7;cvr5%a%r!!Z<*1?HPT!;)N1^q5V4`VXg7dY<`0D
z6UGOvw<gb<`r!j`iRnnY+{q1V>73fww=_x}XT24w`E0)KvH0zrn=E>~tujUn#xySC
z=N>GExc*AowLz@yk8g)isp=x%kgVKJwUlIjrdFZ}bb5)BSG8c2I$xJ-<1D(4@Yt)f
zoa2^iEMSgh3Bsi}*|{&ZcSewA=`Xl|+Ab#F!MC;FU1}s4GsV#WTzvObDlOlru_m%X
zQ%Werqw^iq7<dZoVVp>6HF9yhnEWPb$Ke(I8M{zi#zmtV;^NULDS|-9+LVL2X8<BV
z?>SV~12d)Er03ckc24;LpTabVw}^v--SQ*mY}{fLL_sosvN2M<h*7!vq@p32{(G-v
z%gpx{De3RID_$W{Fu4)1jmRF-uY=7W^KivLb#&Y#P4lLVuQ>Te5Q0A|+;;wL4a!Jy
z(i9iDUrET+q)Tq%ON7K2&!T4~IefF4XH-hEW(LNiD>XoHuEs8R+qugF9+dDRIeeXt
z<bJy}iGOU}eyLyn`ZsMJTE=xi^)Ass{I#6zm{XhLAp{4eFB`B!hud=LaN1fbJWcnr
zNzj5IHlc<=7TlBah&{rV?kjb8#NUuS^4{fcRNPig8Wf`ETV~>dXMt=E0E1zD+D{j~
z3tac7BNu}&;xXbRgs$6Bm&d&C$!My3nfrT>#QDJzP&KHi_rp^elN~l>0h6)&yRmz%
z`!~P~?HL%NDErH!J@Q|LA98!ysm6{gj%A8Br}+2Xrwk~UU;&Ea=zYnVE@gOHk$g==
z>dEj1sxDxSsFpf%to@`q3<7bV!?~*$t|0_u?HzpZw|?#T-AEOtRSwZ|K1#o=7UV^@
zS%5#vs;}1KtdK`ZXsM}phyDvKM%aDNSxhb?GpZ68b)>tM3#D8op#IP%x7Ey61+c`F
za&0?Qy++|BJm89e`N+FBDxl~oxaL+BfCAA8n+7e(Aicsyk|?>+jxws*y?r8(p=)w*
z@O)PrTnLXgrXYa?aJBhVd@FJegJ4|@-88WhmkeCX8$8hTvUNnu6xVWWWZ@tP<>r}v
ztEixl_2Wr6#^IK+v8(%TgBb>-4H}nQJv0mf)}EMY(}7gC%s&%u27-XP{<&Hc%9llE
zxer$NKa;ewl&OOwX1eG$dTVgpZu)CqQS4|o)+eQ7_Kg3Hs67aY8|O=#2}Mx=(9dv6
zY8M^ZkUG+2MCKg8sc*#|m-V9jo#QHzv<fAy)DL2Z0$H@~>1F<s?`f+yy?)!!Pk$mp
zBcm2kr0}7#S^>GDJe2&zxBy!Tm%V0y#VZz8>?tw>*vebQx81JyZuIwoa@fAK0%VxU
z!&h2_<i&InJD(T{4zU;vn!sao6ML8p(k0gqph`@AI(Q-$6@rJ}OM)4p%g8z@CEH1}
z;Y*$7;!c1def-8Rd3w5ms@#M8r&?!D*1Y8brUz!p2$+9{Bpdj%oL<1j;b1K3?aIC3
zr18N~S9*>tw5XXYogCouV_Wjtar<rj;vouF(x3{b+_v~}=bWm+fyM{T5i*Kcx#v7N
z);QXdIUd5+K;{Z$-<EJ$1<RC(32YCS^+dqDMkF5oLLySWLRYx%tAAL_Fylq=a3H2K
z=A(008GR})ZsyD_YCPX@Po)82h|2-i9gA$H0V#H=0VQ0UK@Qra1uu8?9D>(`suh=J
zVHTAS{@)9?-sp=<?*tS|1{&%C(ID7rP-8~Wcm9gx;Nc;8k7lJQE74PS!r_g}jaWiN
z3V7mIxfhzG?IRK~NbKYXh7YQ-+ouWpk>cktWqB4xOACaWoZ3!rz-`~aqw>#QSWy>f
z{(k0LA=Cf$fY;a}D}S!FZ)kp0A6cyz@iXbmS#rFDK$O6Aa=3#))+e~H31+ogUf+IE
z#E^MG0pJ!X*JR$~r(6WAHrVZig&~=AQnK+&4+;H6eqAkjRzGUyRqq}30ssb{h7>t2
z7eng|;09=~b1*M?w;6Fx|5ZpSb{IWDi&_i@vZU$kT#qdAo!XidL9!HykiGuIZidWP
zX-k{>bTHTW)PC-Ym3=D%x1N&rD85N_3UJrAI44aez#)gKKU=r@adL;e4Dun2^s-wc
z|KxeE%$9dAxeyq2_9(~@RB}(sw{7Tv?Of|tuQ$>=uKBQY)z)zvYuvJt`y|?e)_u}{
zkUTyDSWWRv%;@`4qs+966Xby@iw@DFLC8cZ2B?_rojS)s+!D7RF%@jaG`_fA+y|Ao
zRD@AY5bqqs#OW6w7TH3L<QTn+VlZfP6^?ae1RJID`%luNMJ*DS3pDB;WP)iJo<sf@
z*<i#pl4MI4P~V#_J-5848+v$Mp#JxDIG=8t(&~wR4zcOY*nhK)IK9(6y<ysTzj992
zfNQGC6(YGqvm||kgavW)A#NyG)iGu0W6bCcjv+HcC6}Ar-g<LT6#fEaoQhr^n{YRP
zOv~E(y(fqEeEPW)CNqMWQEkLNxX7rtm>u`eR^3AZ^aMR1(6Iftbmq4c4!2DLtM9K=
z9zQ`%Z8K@P3J9o#hsa>Z?y^5Knm>6`k{B16LJ|QLt($lt`Wgq>H0{;{Y!b#i_$3Qg
z+dTZJ(X;R9(1OcYaH6Ep<K#(L$?PX9L(C61XI(2#aJq096Oj-DXU7R>8!^m5lV~IB
zOwNjfkZmGLvQOcb#VzHJ#A)Jn&7U?fp@DaxIZ;_uV$Ao^`t$1^g6`qD`9t1MoPxJ^
zW^I_NuxZ!15`N-Vm>x9PALqM(oc|$`ZDnC5mflk~c(M{tT4Y_8wW;m}a%*p`kboZ!
z65Fe$>ySbb#S!c%8lnYU-{4-0S-0N*PINls#*E6R9RmV|1@AYs7rUKQV@IFAXB85o
z^<n`VCS%nV(=~6<M{nLJUeG(~Z#($c{Y}|V{08I8TKV8w_{;T`bQRPLtLf8Esh7;c
zHD4@S$3p+P61PiZZYE}7NtV{SG5SG{DKGiXH{}OJB!`k%-Es!#BF&tOJ4uHIB2B#-
zR-TfT%-;KQA?d7y^MsToIv&y3M&=$sTyfEVCouwE6J%Ue_OhcKQc<C74Fr9xT7e;j
z$6ybpmzy}SV&LN<z>!xVg1MEG*ticecA&Ed*ow&Z)=MOD7w-G{Y0A%WDfOX(&?9CU
zRXa4&AD&ve4kE^j=8J3+f`TXV3}Q&$Q8c4evgiCX9bRXe9a%#WqB8!XX{CsNe}-vS
zKrS6Tn~GlN1-Au0R>3Hc)eT>D$y)0KSWm}V_oXcvWJ(#lz|-#PlAWU#U>=MP`}sYY
z^Zs$sCs2n;?<k64eDj=R-(f6zm)BV{CSP#vlj>Y*X3>MTAH0d)A}h?_W5o6I>7@-@
ztwp2WW6_bAreVx3^?4d_=GrTp46mWi7C}6@<-7P5>FE)7LP*#>Wo<8JRQ_;_4oY)I
zH&`C}+w2RFe$9?)HOt-uTjUDB^3J{OkK~>2annpJeyFVizTzergd%m~o!yBjMPmrQ
zj|*>>{s2;6Vn!klyw%RCT^F?gZF(=l{GRPvWHJu`zo~lD4JrW8Sje)gx-$-nV_u`!
zrk<4YAAl8(BFG$MqonNK@)I2!g<@4Ht6bSxJZE3NELkHl`|hZ87VdBWH@EJZ{Nd@P
ztxPL1aw41rcR-r%yC$0^h1aA5Gb`T#*=%&OWqDrH&>r(ulyD=?yc@)qX<hV-C+d{R
zTbE}V(4pCtH#^yu|F-CV)d;k&v`t8%AS$jHK0a}CkwKyGRb68;XN5P(d)Mes9L;-I
z5xY%72tzl{q5g(xQYfU6tM4mrQaa^Q3}09xE%jgLXv=J6#Pz1Gww6o@%yF#~eFo&)
z;u^j=c!K2;qr8R{^wIY9ULEzB+RrqtPzkLzD4UKZ_2%T@Tgq0HaJc?xqoXJIB-O_3
zO9(pQbGQbN9y)cULksJlQNoZ+m0x8m-v~e8VuJhKatLZ<nrpUF$J8@gTMyCxP%C;3
za}tRb60SJb2UG@!PV6|Lvh5scYY>lm+$;y!YNL*H(Hd7P2DUT|QNt|yiE$pNGZ$_Z
z9%Cd{C>O2FOux*EyD*DTAu>K(j7dj|U%vRMj^#bLUDG)vRf=Ijzr>SY>etbLFMNZ>
zArUw|aXg7gfd@!T9W~jS<vZxDC0bifj6N|fM$}|SkyU9*o|Lj)A{4a>TUD?vTvX+h
zOAuV8jaTwI*d{O?Gm3eg2)X_`gNzBBsQQnPsY|#qcMoV{xGEb_zCxt=($4biLW!t7
zY}@VDzsB$)<~d3}Z4uWNW<6EKptJ0dNkPw;RIa0OBg9dG*fvqeSQYJZ_lQ&1Lh3p%
z`6t5c!uaLYSEb<xW47NPnDhjw+ScB;N3;Z3g;)QKNDpp10J_T)yXSfgV7L@*IaXN2
zg*~|=uD+E{vuSy`BP8zESo}K#KpnpO;#7&@;?L9hhF^SLez8%}TFg|}^;)uNnnvN{
zn<r=LrQ~SDX+OHiA)&J`y29HmZBJGD9+O;KcbBIr1}DF^slN2-{F$)mUG>+KpQ@|h
z6|JxKKi!kC$zOHTgP$AEyj#lRR1ji)Q=$5ctNr(1gMWxxEt6kjyaIT(C5suGC)+II
z@nxb%w4nuG9tIDsZOq@=+3xk&Y+WUS!h64J@kP;l+DME5RbZ(Zm#hp-AT?{kP>xC4
z&FP^fK)$zz;CyxeS>yn!yl7(LiAJr)#C2i36)P8#F5X*2l^qwJsxCsESAKi=q(`C7
z19VTTUO6#)i@{PFmqTgoF>S87xhpRY=4&*>H`Rx*^(WqaBe$vPYiHpHX@9hQq<$Q+
zxf#cGV?Rhd@L{!edS19~D17&s&5zHCActSfXv8GdzI?X9a4xoT*o-#6?I`x5Y?Z$y
zz|#9$HS==T4J8*7FLPGR|EZitJa_-FbI1P8n6;o!uQX%Qoyy;S@7c_jK!?2dN}N=J
zD@H9@Td6kkq_ugexT8qI<_xuXoZMTq#S|(C&LsKc){H`<BBIrWG2jIUg0#Q|%Q^l2
zQetb3Sww_BApvQw{Jflxn<s<|EDxw49@%p}=SRCE4exFVoj&(q!(slV8<$>1(nvzT
z|HsRBNupTp#UCnN{mq?N+>wcf#c6L+C)w9*hYw+=pUp7)7WB>Ff}>B2XpuI?5El$I
zB<*7#o>aVa0Vp_FP_O-h((;bP&QoxSdUVN?alO*EH3P*Mx{cOFWS#x~{6oR1b9ZT6
zI))p2V4i#}F?9p7w||nc?A9ak#~=25A_ZZ+$;3rlo)<no_UU`Z!7~^SBt~Y(LVT(e
z*Yz+WFBte+48VdaG=g)^eFnjn0LPZD+c_#Ckx}NPwq^MsanJgZr1%PJPkIw!iovm?
z0piyZ=a;xnabw62a-H2*lJVOSqgXQ*8CK8aT`)h&9Q(Z1O=S+967DLJ4f`xY;zV=B
z%GW*!k>>V#JCD6JB}{nNvK_>XJ|Xbyv$oR6fyOr6iUh39L4o`OtSRCmWP()xjp{Uy
zC_oK&eM;{ygOlUW?{5y1$=nB}U1Z>LpWyRAnfMGoQsZnh7)YJcwYXx>q#ZRmuRKt#
zMkfXjS(0f=sgVFO^?(4RA&)Fiy(W-&^zNG@xTv+0SM8||cONM3ZoKKaMM!GvX4&#N
zUAbNz=N|OH-YnvSZm21d|N7qgk$!Bw1a(OIFwN*E;)6RB+$j&G+%I9QL3uENqp~)i
zT@iloU|#UtYT`tZA`sx-(UUNZ>Vnh3anY{(;esgj?sp*q8qyr?SDV+x3<Vw~G?_}H
zvj+dgqK_;!HG<po!?C?{l;=#HJ1!^kq)zDMvXy1^WuOXA@tmjK&Gb9pMopjm?X?fm
zOf@OtY1wryv5X#r<c}J!kvu!n7X2^JCPx+tY*X`xW5;GVjhw%I7x8BuKj^r#q@_&w
zMm0@X+#fM|A&nes4sGOjR40=KiE8Ybi`Mn_8e;R5<1MJZ+XofIFWc(3&4h1XOSC+Z
znf6A-MPfW-g@Z+`WS}fWfe)$yrh@5C)uxM##=R6p2=!=Gf1N%hygsp1pO0#-iec93
zqU#SujAvy0{^u#;Sd>~>@!1N$bwMdbhOL75?avc!MiZ6M)p`#&0e=RPp0pvpjjvgn
z`w1o}-{t01{aYCO?OPpA@+@_3B^i%<^Ai9vdQQXYePhTP_l!Y6$R!NO4#NdhK9gYf
zMP5uEgNl{_Z_Cafk^ARID7?DlBJwciP|e@SpcW<f7vqyOPcE4B+ry|OnR%w_DGVzb
zRIV=LEa-dbggDN1NwfU7XFR$R^v@T?P2xj5_td%LdqNy`42n;&J3QyL%>8ZQwHH#!
z+!utYS~vaUOX7DkPtnCqv&g+6Lh%lBbN|*HJkQs&W<LLIBmDj{{ItKUvb&eg*`qJ4
zwhY6I1!tb_lAp_z1F8mPsarzm_-#7!5lRy2XCn9kCzKYyeXjA}mYI<iE#FpSN}GiE
z*AF?2LcZ;1a8(LXE5}Ezh|k=Z^rx5dn&J6@3@q4J7q(XFzquHP$gZrpnV9z5!<!+5
zb5N#K$HwR}vt>m~+q2}YfBI9Il$Ixiq8-JyZhGMO?C`f|&Y?Ypwt_PYrjDQfr@fM_
zU`yH~XC?<ptYCSP>EA~NL7U=5U%)v##O3XN+t%a1r8?1Wdnf-I>2aSv>OHwcdKZRt
z2~lnbhMfO<SipOxwlj?!vy_wT9{e*oM2)hAe5Bc5Cy4zvD41g{K2v#Y%smcm9UN?N
zWq054L3Lzzm8_<JCh0=$_TNYkzilG5fGH${={4WvJMKAz>}2*!Y=3brc@vAUZpeUI
zN}B8B-;Or<mST&oPJ%~m-vLLPaj)>(ADY0J#YKWgR%nX)xZhLT%kpJ~oVw@<dqEBj
z{3Q-P_VKtH{kJ*xhK$6-*)dw9$j@(wY3y5v`6LWG$}Q;_`=^m@v(E-Q{m4nLNsRB&
z_n|!B7~A!yGb3z8PXQmTWBl~TcQbZ)oG#EZ=mu8%5!3t8&!!fQ1&nWHP4Z{5p_Bxc
zFvqCP`hDw?H)k^Xr_yB>svQ2?qK&>B`9YK}IaIPor@mh}e*dk({?kAE%t?~P$q*~}
z70~h@p49|~#E}&Y-MZKBpU;#Utl`O@dQf^M;I}g{0e@@i1ob71bA+*gp`s^~|D>QO
zL-3~M(9(4V>A!x4v32`v*&DEh(vxqTZT;(?XkQ7h!n4|MbY>b;GlsO5JnvEO{L=z9
z%Z!AImZ8nz0)Kihf4ywle)6pJxz_I;8+k_iVOj!mg0&v^+GQE#pkLtdF_HMw`c%l`
zSdv6w>UEAe<3qsjA4{zpr#;=iTf=4i?Om}22YBPxpNj&ROhX1HJY^X7AV**25k0fj
zP`s2t<}3g83I6t#S_ADtePl4=`0J5Vg;~J%Uqzg$Kue567gra#DVY|i)2{RSWOeK1
z1?!h8OLec1Q~sQL;QyiRyTiF|`@geEM7pv|qG=^1BT<r4DlIY#MK+n4X^;^rQg%vR
zibz&g%dD&<WRxAU$$nm^>b}&~ao@jxp8r%1eSJRX`JS)gehU>zy78ut|Je8a{2cg?
z<=4*tUS-FAu2PKrEtvoII@Huiez_~KGJH0^Dc_vubNMP^W*Wrikxep&BEd=aun=9j
zFT%7)r9g4zl`XvO8_I4tL~<x!bfy*uQb@k-^~T@e)IV<Sl^@sNx$V}m|Nc2x9+5b`
zU2|vA_uG$$=5v`UF946yitSr<?zA8}l(ctl&3=-_DaMsTj-59OTax0MooTt)dfBLo
z0<OrnkvtfdaWeK#&%Pg3&}<$obw2p(O`P-Ke!R9#B$69-t>#>_%5tD(S15)pShK5P
zY`2ntjv<(;f@WCFT<8qYCOGFYC+tfXW^d=?r4b*?xmu*ZufRm=|KkHvmhvF8kthXE
zp{WG*q!_Yq$Bf*tXcf63iYxMBJ{t{fW``c{ih!cg@I6b<+9g~+wB^#M_kC%bA#RTk
z43^8h!UZ?}DMlLH{qe?5>Rj>tQ=0lGPx3L$o??c@5}f)TV6XUy3=K9)1Z>CNZ0hkY
zQYY3d(K}#8`e}TWH9tPeCee^<vVL&DA!*R##(aE~HLI@w@uNij_$ahGz6*Z6k-sGf
zc|NJxk0QmD!lmK?cjIk26JPO;2!;`}=G_3_2#*P{C>wfUwLm<K$l65>iiiol0((~C
zyb}WsgLL_ojrQV>-Jxe{2IbaYJ$WmMzX>%+Ud;uqU&HB-JNx{v?<b1ho_5Wjij+C$
zN}e=96=KE5tuh&@iBd1Iz<9U{I)lNWtkez^$@Be1Lv;T_wCzHT#ONB|!_|cBltqj5
zDXG;Dh0Qa#iQn1h$0rQ-;d{e|>@mo2(t-XwbP4zt2weVGlp%+iMe#>SFO0RC^?Up0
z_5J>phdP3W$_sV>z|5c(Vm57i2rR+SV*<L<YeA0Nqk75cFOSXi#0?h9J=6Ocm|YV`
zE|sO;-{wEMw>c;09V%2ik3BiP1_K{@j&2)oLy6$h9vrxwB>_#F<()scKjNmkKmL*6
z2Au<cg**T2H<Eu%vd;0>bh|I?V%==n<W5dL#;?n<vLM5cm3Nj&T$axw<Pg!C!Y_sg
ztPFInDWQt{aWAh-7w-z^LWY;N*|Cq+D+Hg)p-4^i53ySQF8%iiYEb-F;pYDyxq?JU
z-Ex(AI~{_ahZy<aY1Jq%<B?DlO5jE=`SD1E*%fZC!hhbz_PURSemlY@hw@?~WAIx<
zMuFz_<KzeIYG(Pvr<<Eyl4d*gB%T$s(1w1}xMdlvh!5HED$wzO^b=|V=_BWC+lBcA
zXD<l+4(r0a=Kex#bo`tUHUu$pBiTktYhUato~^E00Ce%=y8^Kh(tIo3^*}k^(rnD(
zqdf_Kxl<#zK<wI2^o9u;6l%OL-!!42A`3e3Q`7r(?~T&r_`lTe&dGE<b4C7MmqPD4
zQ<OmsWKwh1m6P`mzmNRnkAn2`*3pzX*JkeT`7@@w&sK>C4V~0ELcfJxes%6hk8KRQ
z%6ROe?0dsdTPN1KT#R-lkZ_BR7|e0~(v0z?Az{aTe#xL;o+vWpI~@F_bUAh(qnDLO
zZ1ZD2t8d+KUzC+{+{vzh*Pdh4$Se^)AOo-><Q%QT@UaM-7y}STGzy;h(%KkRCxnAE
zrC{Cu_Mrwv8n6Y0P>CaomBDzBm){#V478Vu!k>^pH+H9)*9%xQWVOAZ81)?1+*c9B
z2DgE`<s%^Cy~_HaNhO6;rbJd?V2dz($OkC11)=L|quWgV>f6m<A@oc@E)=C!(v~0q
zXApP?`qg|B4M0j-f;G(p>6SX=-FHxQOiJ%H%Daf2`F<ktOOUzA`6gezZg$E(F)Dw4
zBM!e&=)#>Tgp2`CpVWK3(6}=Mrco(?>7im`96pZ<K!ta(#w7Z_?zX4wy9xY+%rA*P
zIAEVj0hd7_C;{(<GWC^53&Z*GTji4<Rdqy-5t~1Q#K{+t`R-4JPXNgiWC9dwIlTUV
z7{naKwNu%Cw28WdnsLRB=NcUM8KrFdTaaD+Ufs@QIZ#PN_HLU#x)7bDUdC1WD(KmG
zf~ViBm?ImG-5Zhk_B`FH*`OkJSH7s(#M`PJnc!9Ce1sxlKL1|rmtDD?&v$Dd8i$ZV
z31)MC7lL!d$Gdt6Um$5-DyRcF-CoNMj!DJZ5OUyU+Lp4ti5A*?W({YaC5d1vG2|NG
z4#)Abnf)J(wS6Bj1`M;;*(Rhs1n_fLqE7bngzb!{`Pj7Or?DyH^tnacJK-Xm`tIuP
zY7IN~zjmHiI*z8c=}i6UhQE$#EeFv~Y{^Hu(`xLWL0wo;c06>w{zd3i9~}8w&;7`E
zOdd<47EB#OAo5%t)|0;MUSqYEjUPDcujnftSu|oi=@L5bG=x{Z8w>*%jo<<>W3|n5
zr|eWBXvNCT>JklPjHkIcw)fh$^<3mO7dxkUV=(z~Z%b5SAZFQ&@tI)0aE{i)e(2(1
z9Y$@3PZx-6IyVLTbGahT{dHwcoIueo=#pS^h)!Gw=>e9wJk}b{37m#I7<32cu1nd#
zu4U<Al(%rxC_|rIxr@!VD=f+>h^vY|l|S_Y=7w?kOosH44Vx@dPHy^l*!wjUD!aRn
zs_ftQXl7B_g=?puKc#<jFh<ck>g36Nz0Z}e*himH*PjF-%WT+*y_mDziM_`WuqAUD
zb90r^Y{nH^8eUCM)D4OlrBr>=j%x3xm7;g>ucYYxtfIN4;X+_=jB0bgh-J#P^{c{)
zyS^>tm2M$Amk+lLe#mA#_U3TS2qK;4K9n+^?~f<!8_o6++=5G){ZJ&%SKg&MIq9a1
z&C_SKl^Oja#S5j5^or)5oh9=<i2(ZIl?@c9FUlV68l8UrwDqGU+x=;tiuRD1CyRE~
zPQ;IXQj-e41!eh%E-npc+4sr>FTB=yMzn0=x}(aSuGjYypM8~Vvr4`qpG+N=PzQR&
zS(0wgM%?P%0C=X@*J{-%Imi<ga7(P|7;DRXJpn1uepqo<-9M%haWEGIsV;~BpMBXQ
zhlN;hri(Y$U^4HqJ+m0Qe17-?B$+k0)RwB7sxp82XrHCvAPX?`a7*#^lJ}B>&eXM7
zx~H`-oB4KWXBWe%L0aEF_!voRnmJM}bj*5|V){l=dIavrwYS|Yr2SwCemr%&$nuhV
zrE}=u$8r@Xq`UsbVq4q8N1AIk+5T}>7{E-a8W|rt7_>1z*XTuq|D!+Ee!rZVFd8lH
zsV#4t{ZpK#hR2C%12(rSwj9o4T;Zo1K3&&1y<|u$PFHe7#CLIT%x*$Z^^E9pD|;%Y
zMyxSrOf>)&)UvBFOm)W58JlRFF)vwotvt)orhNLQB>u_~O+|8YSUy{K$lu|`XW3IO
z+b6}mj@-I6UT224_uP?w+7jZ^Ikt?@G7A9}F)WKx_^j<^B(rqwi>K;p@ZlB0LWIR^
zRax7+QT5XB470cq?9=rpcc)FO*{5qFq#EEhT0<R`Gj}qf`kthvl6qWx0pT4$qMRY#
zIaBtMj#H-~HoW`P_3hTi<|<7C-@KW3{<>0b=k2H8-8;Tc1{AZp#J6?xL&dXvZnXm7
zEY|8p0tWFHe2{f0^KIBTSGVicjmmZ`$}#Shwp~p80${E&)U0#alG_rbBGfu$n}B`}
zw#aHKIZ_9r<SEK0l5*r|Yvsl}RuGy})QkMC+-Y9DhM{2tAlcg<JQ~dfo%?p<^eU}0
z<vTX{5S>>!P>g0`{5u<6C-xGh)CPCD;O23snwTAoDmr$HR0r9=<k+BXGOkI6p;l{1
zjd6vwDt`DYZmq(g_DGx=PY2H-FG;<SndnEA4SMQoxvnlMW6JqC1kZ$(7Qg~|YL`~R
z%`E(3?nAS`0$JI~1nt80-K_8!!US!j_L*PlD&PBoybC`PjYyVy$D%*f8UNH1@=m1K
z-7abTSgb;t`;&;TuR&mBYSSUPn=}u-wS+p~jXDUO9)5HF;926)3Iyf&<?~Yx&<8F7
z<Xwbtq)`%|cZpc?`s%7HrVUvd#UQ^7Wu0+ieiJ5j&^-Oq#i(W`yZuzpiEa3^%jae9
z`dDY@9<f*x=VOqa>tJ5Qc<sXzlMv|8G^D0$B*p{Iw6ZJZ`zt)HVEbOaMuBe^rE2v<
zx1&S7De(!Jy9vlBlg|~cj8Y)>>71qdHe;|J<78}p<CvM~;ZJ~SLit<U!)%DHUyQ<W
z9hc$5j_}`^^zotHum0rp;k<cDpN3>m*#Tn1Uhckd{Hz=aX?wypA8?(Sve!JzC9%zO
z=Qj()gX<t$9!uVsSApD5wqn7~aW4tFfH3r0%fOTo&Yd2QH=2p5598TZZx8yHT0A6n
z$Si%w%*C=X2$dxDkCp>2Vpd;evfa}T_AYgHdo~O`NyBnfiFcZYRO?(>otM`o1R%d1
z9dfFWs+OuHV*2T{RTWKaZNeXuEFiL*kTtNoCQ1L&*ctsuKITn}stu85Y!7a`#G;&z
zG;{uHCEp)jwEWE<!<g8b3;CA+s;_@ZG#}TI^1WUl|G9FP>G`K(;vaXo=G2B4Pxy~5
z9d=@}S1{pcSGia@s*VF8b-prU`b(xr&;vMuCape5#nR#U$sEPrUsF#knBG4!C|VSX
z$PNMa4|d~h`(&2`{hQi)YePk+R4ilEPCqWeiP2uki|NEy64~A{T`lvBap(Jow*>dG
zIWcv_oX`&7eRH>m*aZf!9m<GiJPG!uL6QA#UtXCbUo+a5mMmxLegw7mlIJ}?56cr8
zcWTOAES8z(Z|d*Cz7GY$4vGCf7L_!kPBZs49^Vx$agc?AOA?>1r*LQ*=qQD-2V0kE
z*=UUaW;7P?2qjCp!G6%SI1?$_WE)-YxtcVl=rH9QG!T3%$`EnNd7*<gErrX8Ck!^b
zrD4KcS@Kl-jQHpW?><{){<LS4O>lG0bao%1GWrACxKGEqe+Qxgn=13OuQDEO%DUY7
zzMT4u(x19Qaz;*P>pJ&e5M9@Gz;i2cfYUG@^na@W8q*t%{G1$BBEz(4t5YAM`~ZqI
zZVHK^VbwHJs!%NReK|AD{`LdBO3bM5egj_ebsX0Owbi4*baBs7?}V@=(ih~n(WEaF
zxOOZ5Pw%~Af8LYGB*$EL{ohuMzmF8=c~H|k<kD4DSy9ZW$g^wJzl3+|mWCZQo0cJC
zlmeJSIpPA)PXuJ8>fR2D_%3jGi*nO^okavU`39`R&Du?L(3zT{I8ky!cW<#S_xKCi
zE2AYRG#&9Rni_EndMVxNGEk4C&ekJ}Dc1)LH(!^$()cv0Y7n@JiObwq00$p-a7(HU
ziu0BqY||F%>=)Uea(&d$fnCE1>fc;--%^$}yb8)Wky^0d<x17ZF)rq+h|%iAEvNMx
z03T{5{&3A7UcZHk{GbCn^R0awM*tI-*#};_X?xe;i>G~+M;by{jYQPOq*}Od$pIeO
zc=EIMJyVfIgHu5FIcDYVwlJ{=ZtVDzDy+ThG6-d{=i|?Oqt7kGNiU#~*O@t7e`VOU
zi^-YAE)keaqq91oO4cJ26b0BW;1J8RGe<eKWjym!3QgB2sPiUSv@+pwjs%DOP?*;y
zDwn8Z8?9@Eb+>%gsn%ZLMsB;&jmf%PX~<L5FpR0UvZl9d1dl7=;Z`TnQQE`wHGjE;
zo1q`;apnyfd0{+AYn^<3l(wfI4wm)z(*-T?|K&GsAv2YoGn)n)xA|$8W<W>HtjEMH
zTY+aXenpQOMdRSgvfC<8_okukzDVg2F_SaNv1!NQpY7&9DFa`{-f{1JoFadvh$}@&
z?bQF_08x)8mxV;Si=<~VEKxcZw<_pDVDyqP4U-WLMpl>V@EQGB#$#;R>J`Qha@N0e
zIuM^pW73N{y>Akr!RaiGPPnJOvpRlfBn{HzN_yuWJa?7h@_H#LGT030!qo2R&^^`d
z0T$<rpn_$V23zb+>|A03@9t8_g;8c;nEc%C^jhsxsxC~aQMH@Y#aWckp(=<UJ;Ed}
zhNkq<*;oo2u!HoTt^xs7_<=MUkFGTUviD5NGfq<tYkG?CY;lq>xJ3}|Pc}lly<iX#
z{nWa6g;0s8y^qWQNlYD-c(g-Ea@8gI(Qs%5I6ybcpsNy|TBDDi4#AB^inhPJ+GWNJ
zi6J$0s1`5yhFiJm9Z$b0DB3Hr&<{2;<hY#MY~UkfylSjXxJ<u7TImI*1!5lZrEa>e
z&!xBhx_A2dI%4nrd=pT;3liM-^pEfBKUE~>Dwd9tZMv@c9ye%TKUZeUpRrq^!NF-7
z7@6Ic_l$vGyX<Dit03tmaY<inRxkBA=J}*Lb{E@~>C^U)-WVTvFtQPkD+1h@wIA*)
zxqCf+D_xSh`3wp?|MokBHieQ+fI5AH9f9d!)L=S_7~N`~Wx#<r);W4m%2Q`1)Z2x3
z%7>4087Vwju^{yOWvOcKM=Q<$<qZaarlnoVo=c3x=9$$^Mm98tZHjX}3J>Dj6w9C2
zNZfGG;H~eiiVtbNKYlX;&{)keFlvjEvP2&vm!{ef(qiMM8?w?;)S11nBevANU&cEM
zVR`;9CB%j|zWp4d$fGL4cGk?m6u|M_h9#O&NpNDWtxVG2saYp|<lN?$+vJjLzir(&
zRq5Ts^l-Vmyg2n_<mhy#!I1Z+KMuc?1u14q-%VB)aWVeio`2JfV6>Zbv-}>algDEj
zn%$lT3i&QbDJ<3La<f4PHq6m^?!BUD@Vf??;Vn8o$Ecp*0c<2ucG`!tWrkE6&6FjW
z<j_fVm2L;u=yCW~U<bD{-WaJMVe4cvlukt>LO_Gr#2y_X3Yy&G@WIy`O21vycRYhP
z;1F&c_x%kR24E6Q%Hwznj-fH$q7k5&Ya^<Guk?<czYpwUQ(Kfo<w?wOt~&>uZ%G&j
zJ$#6z%;wZPp-!)7dMj}S=!9|s@h>F?6LnIMi=cN<wPRNN+_ac}-3JcJ;ZsPPsaCB;
zR|{@aEPkztLfWjAP<TWZRk=nbje<e@KH)8|0#D3m)!L%zQkj`EKTkcyOnbJQ^l{$k
z<$vss)ehy&Y}hz+Ci&tpyV-|2&i~m%dp&lykJc<(B0d)X6o9YYK+X%vA>Dk|?n4P0
zoIcuc))4gaV1E1^ajHBvAkm-bHl+e5KJ%hc5#MWof@us;3N|=(32SCIyt*O|ba%nQ
z+ElZ3z?mhr-mpIq4LV_IP`io))jVa)7dMLKqTU!nCuUJbsS>f0pZCf^N3>9+1#@6u
zQ~PD%wNF#m&HF_JQh1m*iKr+Xe9q0XE#c``>4W|pQ$%-{NVEj?q5yxZEzuJ$Wur6z
z-*0^IWZ&bM;;DhTjW+i)x~4!3r;agiA>=8Tx#9n+n39Us*LtMq(J^e;08JjjhF7U=
z57N6w;eVLr<0))=6n5Zbzot4mf|18%NfysLSvn=aT?ZnZY7T4+TwAS?;_`$kYytHm
z0hTv5uRp$ui*VXKBxl+n0!POTpdH_6D2zT4ogRHyu*^-1)Z8+nwXSZcVJb-blieVH
z;4_o2e$k*Hx^Jp}+U0$Memk=ZI<reUZM(oPxP}Q{sag&=vH4TAL7ayCi|FEV>4q(?
z1Aw{noex5&GI~YRcG6M|P50Ij))wSvo`@RDqDQ+eex_spY*nSd@KA`PmuFA!MR~&_
z`w;ia#zk#Q$8%SSEz<{xy@CvvonEiF{jZ`1!)E1(AH(H=g#RUYF}pi8?zm{IcRbd1
z3BNW+XG)={x}~RbiRRhSQ)|n81fR4TO)Tw<d2du#`6_}d{aMxpz(UzQn0YprIsYw7
zWN=Qrr2qM9_MGIq1f4AfR86<*ADL!4q!H~u*}V&v$!RThY++TIM)@dAN<+cY5asJ#
zdnUNebY1k@SUnWyzx<MkTW_MMaQ8W{eiY5O@{zM>h18zfnW;#gqd8VKZ?|FbwLp~H
z$KDdk(MPwLDg0zDaFPf~`Pe{#ANMNwT_v~dxaIYdcxw`_&9R%L`n&^XS8&#T2JOv}
zgkK0c_ni*#J}t_dm~wh5R79@>6FDA~f15Tl6K_x0v1adPzb*T<e=KikrT66c5ZJ}%
zqG%I(0cAY-b5X{BqHpk-{e)eN1nKMj_k?9W3%fA4JTd3NvqEMsTV3rsKX3fuBLWpW
z9+s!Z<-W_Y>p7+(`JAbx?>VFBhb3>BQEAaH(z{=FOL??Cz+W=dgC}JQ#F%2Y>A_|7
zZ)(4$H#2i?b#*>cn|3h%h%lk6;BY~-?G_P~Q;878B?ujntj!tEf4zdzd$CuQZC#WC
z74Z*y{d5|>O(h@-nhq<~Sulx3dC<Prd?e3&ECF-ln9+5EfcS6GK<_7|LY4@J|31XZ
z<j_<{d|!dS41ie6MDmFN4)5(m(+MR(upRF8TEhAia{MGfI$QurA1@Avf-HzSo7R=C
z(W}IOTycik=T)1;i2QIWgtdsq3kG5Iht4~P%tteLhi!Wri*gj2LF(@|;=eBt<d31(
zCz#XN*4F%}KQ#sa{N2aZv(2F1tLKGVPc4^WToJxxOb{e+Y#9Dli`AbxIFfrdP~@~<
zYPKScb>aE;{mE3wI~5Z%F=ycrV}RjXj8BNmEfK!=?H3(OVA#}M`$UD<=)8SfyALN?
zhR00zG3^^>N(S;^&zA3CFT%SyQEM|bvENhR-9ydlBNi!B{zDu0$@C#y5UjPSOARI_
zfbVZy*Vq~H;qG2;w2n$gLe+Z|KktWR1SW)6?ikNBF$Q}oEOIK9gY)5?m+0(PVM<Y0
zybfX~;Al&(dniLh@}OUEY#{lTCD#PGYFvxI0GynBSjMP~&vBq=mD!j<x+d@&Un7<|
zU6SVq!a!hK=%#<3&OcvOX#ChKh9^qrU;J5>{ioiVd$^m{l5Khit-ul`r8Kb$<3L@0
zN+|)Sq*$H!K>w<lasM+1Oj3FT*8(>wdnt&wP=RU_p@l&a%`D@=q?@BlVZibjM&mb{
z=nq@#?Hf0q(il0|DqX3%v+ig~-Y&iXJ`Z*6G%!&D2xq;b3zYvOnfATrMLq%tFjzu3
zz1+N}d_<<ATDIpUlr3UG!I-6!MPg#H5*riCBhbMZ1{p<A!?>QfXSHInlGO5wpVj*X
zLEuW855kSyfoL+m);Qk7zIwmB|I#g|*TDX$@$%bHi>FY{F~o_^iS7Re)*0FP2q8#N
zb3LUsDdSU}o&qfT((9WWRfdiks_+tg+gq*jgVdj)`^u-2{O(UX&j5lCon4t&(}=77
zu@?CGE<9pqv0$+;9AxsI7j`UBqCM6U$M`xtRk=i_D%#OoN9ewpQr)OB`xB))ofrsO
zKDSw211|ZvNqy!M=kd<C%v(>MSTyR8Y4L-U%9sWsmyFgk@^HS}s>2BzdhAnNhL_10
zQ!iT165o|dOvu54RSBV6z`+N7$+VVNYgx^5SGc?x9DYhDvJf!Qt#Ei(J&Cb1_XJUS
zz~ofJAGXAYB%1c-IFAX9B`GU@E=$msAxhDAG?<*4mgyC!<t<MFbem}N^b`_pUMPEN
zjHxCv?%OhaZ2`|_vG_d_GSBj-AvUNC7_?cH$&Bl_PjsMJ2K8`oeqG$;5+l=oL1%e@
zG6sa+@9ql!;h}k)!LF2GK$HvO{nr2Yv2*$?EGCn}=R#n)u<wG@!ct?`6~euTF0~2b
z=%#X-HLlQp$7Ocb8>*}frOPEDaOBQ}8&{^X-lTFz2|<E`s8vW7dO<28MbdX=h+Lp+
zjT~4K!Ok}P<jLBkxNN*~%Y}N8AfD_IPi~_Cs=@aeHm8V}nbf+I6W<T5Zox*q?HIy1
z7d<Ed8D5#h&X-=S1AQ;tXrn7AVHOYoR))cKzgATN=FvvC+$>~*;I^k{I{884N^QBY
zTSz<e(3kfQP1b<MyS^sPB2q9BR;mL?Z5jh#AzL>~+{KAK&xKh~y&-rEsb9LP;#Ywi
zLky-q_y>NH!?De@Q0-u)^acVnz{slhPu;Jw<PWUXnWq&T3;!^e`zKJio6G{pxbxIG
zIP0Gm?)RCTY~NflFx@^;L5^(CZ%4L}$%6;#ZXyJJ16_uR2-^JH(`z$D?pEpQEt5WT
zul0t=057cH1xvTi<}VrKuPfmal-?`|voauoW^~@YC8Yg6j~1E)!Ta6XHB{$s_<n9T
z*vJ3i*4}*F0!Q*|n2_&6tzn4zB3Pf0X|-&t4hLtc3-Vi)Ma_3=+-N~~{aE~Z$z6OZ
zMPpQ!$4s<{?p!7+u1g)7C{%QS4r;-lf?>pZY~b-#`i&z5c!I6paq2|_iMpP$e$YqR
zdMkyIW?TJM*`-$)&+ZMu{2>@S@)P%WY0nLxKeJgVGSe)$*8AuR{r?U{RlYonn45o5
z_qLF)ulaF<dplfRKd#sG3MWmjzjDEsts`^eLE)MJPZ6&7uU>e6JafDoD^#JqIz%M*
zi6hxp_}IUpgKN9ccuB{z)I%9>&nze(*?5DhfW<3B{y~CfTs2G2;|5QN@a5mI-&*S`
zGn|I%HfFSPKb$^av_LWL$>PeWzeu|f<~BVRJc$uOA@;d~V9^Fol81T&Y6x0bVWi%L
zwV5cfxD#$%v)ywTUTO!RZhk__83xxy2<s5;bDFB6T;|;8kz&!<ptecI+cH1Lb;^kh
zby^|)o4WGuo=|?WbZx!*NrW6(6FCr<iy$%(u)wzVPw+tHRX>kxW--T~uyk{dZ2b2l
zTNbv2>vS&8ZNu|I%Ozu>?NSv3Z8mQvW<%}HI{P>i;SZ!DRyo8+7iV?<sPvXg7`h5T
z_p$7_q1m`kKWIb6yH}mq>E;Jp!jD;<pIb4bQvrAXbxr<>cGRXC17VBqTZbDr-tLL7
z{8qG`uYoct)10_95}*q<uNs+JgxsHb=Dwil5^tl4o}pTn?v14%PcgmTk4b;OgKV2=
z2GW=>A5GiqOG}-Um3%plznz~x1HVY_#Jz;BgNt2=3d59IhknG~T7a4Op(9ey5`A6L
zJ7QL~t&IGs1>iHgyn3&3N@E~=(UmPDH;BgJBc5Wc3LofXA0u=xm1)4L{6>aq#Vj!X
z4FSqQ>71WR$uc=-^^`4Bx4vK}{)DE8-oBe_a|{*?Z6X{KAe$~`S{?h=^y^>rw&CBl
zF07YX<b+zH224UwyWajO!pJw}{>Y44(SN!nbN)_1d6tS`^iAQ`@#C?IPreWKcX_2`
z^~%l^kXggX?Vhu-KF4mGi5>uv;*(Sg2J82=GLl=xg&<S{hG&8L0#hj#fxz$`FQ==d
z-o(?|5VGN&j)3LcTL`VUMM_W&l}HXMk+}Y=m*lf(PB_-10GB6gN6iqzlS<4QxD26=
zcXqGW=p&%SP@E>VtWe53vv$$&pfHM(8C4163PnQciZoL(kwJ)0X`MI~ChMp+G(84M
zbamzn^Ql=oC%1qAHjqcPP`$S7{D>7MyGmlrbqW?28(v_iAS#;Gcdo}0hQ@~A+od|c
z>41DOM3VCDx>1q@LrGADO!{Q*a&oE1pjc1Ft!e%S6dW)sBEiXlzS+9DuHul%DjB<Z
zPbWK5NF!O1%<^QEF3NyZ*x~OO#ZE795Y3JFurc?aAWOW|S`(bNB#&+2{;zQ#*GWJ=
zZ{k(lbgWe#R{JzZB|mAuY&U8~bJ;JF#r(8)>tvs2duxd9hy%L|ooxGH>DII-wph^8
z3iZ0_VZx9)n4N3cqbsDBiS<|NV0Z#k@AE=;+ZziNr6ZI}Zq*GmE<*#@te%CHJ((N1
z^*;b!A#MvH(KTmaQm)ZDX&-|(Z^LeOw&B&))b$|mRk^AA8~1KIgke0*OMk<S8|ovs
zuTl-`bC~JB@P?>;Q0$#tXvB-{n*T%QCq4$;a?a@wnd90Y*(X`6SKY0!U1FyDf-;(g
za{UsoR@-<tT?h7;G7VIOe@Db#PJxb{%Ie@r?U411TNXfyz7CVv-iUtU<XSu~9@AqV
zo9LKh=W+!s!_qHtSg>CS(X#eVbSNx<>h0U496W`s9T|brL}TROeKo;?)HD7VM75Xa
zvS)h;`r3gQ`;}~z;*cQy+dG_lYoDysTQ)q2<%CRxCi=Wm-27hBNLb}eo5fZWn`6Co
zGdRR`wO$6@4N<BQfK5I>lK55wR7F$UtDlx)7IodNcFUGKs{kVu{XUt}>G>mCZXRVP
zf749pO*d(c=1HGT-8~Eb=OK8hk@uv$v03)3W&Eg!d58~456sKtOl4j@?NH5n#z|nu
z)+FHgcO8vrxNw<h_!VKEER5JG?thkDshH?~sc7i(2Vw6YuJ7tQ!k(aujVEE!zd2kx
z0tkqoRSolIgV|FdBN2MNRa;_p1I>-k+8=ISZ?7wNtke}56<I6cN+W}kELp#Rx%p^x
zw$Exm-B%;mhqru5+|ea!3npr9h@qozYH#3M?UbHQ(JRmvd@oL${eIn{dwNASyE{bB
zVTU>uy!}fj6TcqDJ_sX+z$xz+xjW2M;~;eS?`uT2i>S*=fk|j3dPBxOfrwC$3EP_Z
zTxA_}%$Sc1ap|O)hYN4rf6;z<f&)7PcW%Sdd%lwCK7d>Dl^s`c#T>tUq)Oa!5k?Gr
z@2h*r)x|bW;jAv?;`?a_15s*ljS6S)JWD0};Vl7*wW4=PZZ%PwZ6`}}*$6=tW8uq^
z^!_u^4^kw>E;x98GPvW}<BZM^VldpAfp~WN$eyZX)X<>dIB-;CB%u43|CNR&#uzty
zs9@Cfr;8q(iFlX0&eM#jfH_c!(X8Vq?f7`+dWaAZp@%!DFcRFZRKc+q8wHd5JK02Y
zzQsT+t2B6@(f}b?OiD%3(F<rInk1icANkXOSxx&#e_53%H2yyb7UIYV6=i$fg?uM%
zHt}Pu-^rh=TryL$u02~(B$&Lr?+rVxby>|W%tUDA%jH+AYJ#cnXWA?JfcpwYzw2dR
z8$N#O^+D^BMdL(CPYlHhN;9qpu~=YvFl!DR?3DINO$_$9D$l1t*`IEX(q#=59rHEF
zlkeC#pXppw;+U%D_vd_nAxg)qh&M94;*@L{<4HYeQatty$)uI2`BmLfaFQrH7$q#j
zj79u<N++hJvLEON*zSGR8$&+4wq8G988x`)UmN^+fr6laE(QQcfMNLj#p{mjnTF_s
z7wx?D7B!qtr7;y8{ya0GIxc;m;$j#jS7@EV)_y`M$&wQ0Z`y<|$ov-T_@C%!0jGbY
z(EE0R<+^|9ll}?s9@l_z4!LMfWrz%#B_>|6!!<B?p{<{Qb&^4f%a!k)JM5O<9+uvE
zX_NWK)hK#h)qSjcv8smA(%AT$uly%$cyp%g8(peLN;-4&9S2(0FA+pN`7z|Lq4n1j
z9ve+?Ek5T-|8f_fPIV9Xud$~}Ba$sYJvSZB0yvdus+9<aXo1E2<N2}?)pCM9Q@zyk
zcYHz+Gy*I7ZydZ!yY?P--92T<93s%UN@|@DkFiomg=D0+EIeL074zUC2;cfuhgp_Z
z<}pCvJ4_Z6)#KT!V_&NB=0nybH4K*aIS(Ebbo*<2`Yzy=mWW^4sEV2M4o|M63>7jL
zxw+8E6T1=2MYf0kaWPV{>*rh~K|KH5Kegi@zG|AKH+LvWU@nsS^U;oS?Rt5l=;usT
zMTw6tJn@Vxb7x|Fm)mr%#8<cQ2HR?FvK7}%Cg)YHXl4=8`E_!!r{_PNvHL-fn0!oi
z_xq`(U<E^vhi0fjH_cod!%f8_>;A)0fBf|JGv9oFUXza&@Ym)}z6pusT@YR+`j{2T
zhAQQGq8_P66WK)12t@&pl(j&-H{gy%(A2$po%x9k)EkV6tjw{LLNy5p*2X9-#d9_+
zT6jI3^jo6^zy}vX_mPs;r))gNwrpY;qy^u+1=wjb)xIJ@A4<lAOHMB>`E2chc!FjX
ztL%DVTml07B$0c8Jd_vhQTh}^`);seHYFrM?W|DE7K1j-d)EZZE-GRNXP<DK`2tLF
z=!+sKYX3yuRJQ$5=O4;V?jVkOKl`eGe<fc$3qnn!OcoQLybu^DGT@WqEKseS#jf(R
zD{a8~dL7yDEjdoce!x<~UOw<amb?t{kw~Ji<ayMD@EZi6g-qF+;7RfdMe~TbEF|Ad
zIUkEg=UOLvEFNO7no8>y;vY$T5%T^*X6e?MEYZ;6dp7n!u${WIy(f0IuGp?&$>ZU`
zKz70;O_kGp1r*{s(AhF=dyS`X)&t?&T>{8oD*zKDY3{-(>kgy@)Zo^zV%{DBrk!;b
zJuiOJ9?~t3uR+EWcysD{ReyJ5Zew31AN8?)kN6gAtdj2T$kv>Qea45m#gg|XCsLSm
z%(0GFBG^VUXI#H?eAE#VJ$^H1TvN6t)u#B7Lf`J|M`JBE0#0np1<bqx<Ki{ykMz{{
z?$|Jbv*tY=SM4G(xa`QzHP_70l%2fg&&I0EjhAJ-;rfD~3Kc)+StR%X`t*JLwJVd)
z=ub$me*ipugR`8iwXcB}-d?H}>odtOyeyM$WKsVyq5lkNxMa7vF=?}Befj(b!n$=y
zlr02erapQ98PD!(tCvHlXyJZw2fIwUOYiJ3mL|oqeqb_}Hf+zWJMV<-estE{ZV9Ul
z15uj-SldJ@)+g1rJVE4)zX2edLbW+cpSNjz52O{^d><Nxr0s0>C!R^ubEY@=8-wXR
z;g0#gBBM7mlJz|Of?mcwT39(gqIw}+tt>+6&0ZE&P4-cMopVwoZVa%#S1P}#vE*gi
zy~f91M(f5c)({i$qU-BISgA8{6RnE`Llmo`_vG52&&&xL){ybwiOGB*_21bt^E~=U
zCwu(cil|ws40jFrne}h*%3o_(0|8%%)x5Byz|R-rAy3GI0^8*%LkXwZrtN0nuY|p5
z>L2axqs3hStnZ1CtqIymNg>IIE^0r)t}>|AL1)hqvUK$yd(uaGfR`V5$+oW$M|d5w
zpp~hL+XFq=#HitD7BoK-&#+f6lwVE_v<Hn9@ecgc7VF2~c}TD{Eo+N=CLm6Y?`v|S
zRGr~aTG+WY!T8-hR@I+(TV9>~3O?T3=wl&}<&hGhL;EXXM^T{f=h{=M9V5o?drxJ)
zrjs6|HODkhNV^+!m@F(zgR}PRnqAL1W?`KkupPt7)D+ViHPr~&HF#}gTU3nAWGrOo
zD=^F>^JjhDk227|a5x_DtBmQtf{^@5hQ5V^8;nyqK~al+DvwELE0l>JEqIL996>+#
z{N$5`CA)sU8@}75nkug>ZDJKeQnLKbS&9A|wYqQ5)u#2=X%?{-9!Ja+H~k={fjl6;
zcIvetv|fU+@WNE^1}k+m*gV!Xh9NYvtCrGD_s&f3!|W&g(ccl077-HWAJwty<oBdy
zN)9`ibYchTBM4$RD3ReSFPx#Y7aRQRSNS#L1sGFKzL?x|%PJq^?#{KAeN74oLSe`#
zFU4OH-;sm+wip(zApEu2pG3`us4m9qzdV2wGcj@f2u{IYVm6EaJfEZ&2V#8d$glPZ
z`aOZr^f~1dgPNft)lAc|Bl=P-UQf;6D?VN&SBB&-BDijc6nB^yzbT1b#vqZqpdGST
zWO;y*6!$aE$$b^_%;U|J)7+!b!271l_B$2+_rqAYnPU35YyMs-D)5XX{&E2Fy9L|q
zWIt86{#L-#(^{|o^GhwlFU7|>z*@+I;1{}~+V>avfC}9vAUDZsLJBrjMBTGESeUy!
zO2e0zmHmSD+5j(kW~{A=(UGFy{uo6DV$hHRDB<5u=fXb{!kDCa*d55m>@Wbu%zC)$
zDEAmw%O{+gYdDj;K~ab$y0fqTI$DTw?39!!?p`v4t_mjRlV4U{WXYnR7v|?{`0cG4
z#LSx!+sbw-94Zj<(({Rr02#uUlmYCv3i{NkErI{JtJ}Z_SbvSBfK}NIzij8$0^d1Y
zR|;zJ?RsFhk`w1mYtv_z>2(%!iyVW(olEpD3B<_Hr;pvuN4Koii*-dy^rfW(g8Jhh
z&_ijYTSnh`YguaovqsXk1P@9iLA-@*4TK}lLPpP&ord#&{Sr6HFHw0ite>C!62x%?
zm&=g577zap=HjW&;y+)qhaQO=cVDc>$3?Sp^7V$>b7zPiL+nU%9=IMK#LHmBSA<k4
z<oIlnP(JhB7fFX$m5=bzF~!Erbhh<7v=A^OFaw1|)0)eu70=H9Jc{8!5-w=iwVE7c
zZ@yl8_-`^A6+R4Su3f|<qiAp+%!dyW@Cd^K8vrzl5j&@RxQxlQv&<4JocnS95Gnbt
zQh&QN7AjHa3;*|u!%c`K0f<+}E*?dwZC*AHccD0F-+v?+=mnwkuzwrRPQEEXC}U@M
zHf)<4eG3;rGQ=QXc0+L)WzqZ=G2;q>&e|vO`gHc)AC{vRx()&BXY^m2bUVPEB7$Gz
zr*>X{ueS&u7uT_hI9YXDVFiZx5!abhZJP2ZNE8X%Y_*MG$T1LGpdKa<0|R;oE6ab<
z`yapZ*u#pTf2w%T2CT-EPk9gvjG4kyh;R8@v+e@|OD=^-ZduS<e_y5_Pi1;t-h^TT
zl0=hS9uXp$b4X$@L8GWbe<|mVAxiTFElu22P*iI{w{{gSbKjDamrBA*ggmL;JD)x_
zsTNst@UHJ2uDvGvz11zKt--f@khtc{ABUCye0siC@*>tq8WP6v&*O!!yy8J(Z!T78
zrH?(2%yjL>O6gxy#Cc-4r_4X!z{~v>R?UkSER|I!CTVr1#A}VH`InTM=VKIf2ZeWV
zB}zzV1o&`_g&8QNSPGXfsb;nP?H2$3$AAhFo3BO4{ibB_lE^NsUv+02wVEnUfAL;U
z-nn}R4{AmtvXpAkD1wnu5J2TBfRqZnJA0qnzT8Np919x|22tr>C!`6289x<ndqnHv
z!=6cSjI2op0M-z~#cG&LvV^3*zP|K3`Eb!_e`uX=E0G;zO2c%eUaV4T_sb;~l2-~=
zU_rGp|8BJcz^o*H1xIIW3Gi&#m(n<02R_7Vgp|S3w|2E_0$$)w@nJyn#xC4L)z_Af
zu2B}lQB6XV{&*IKuUz-WOxrOfQ-P?w{>NNo1wq`#B0KG99riPOiatz#Hrp&Eyh=YM
z<=6M|`@?gWUo=}wE!t)DD_gqIh?KfRLmM2Z7a#y?(a$+?d;#8+1P~$fHGTWgfkyVu
z#P^_C+|r<+lrYMJaCi&3r`VsC0JD{R<4m*#>B;aUD*)Kj543u(Q0YJYWr}K{te(*3
zYCRURj`~4D+jD1I?`b}&o-P;w3SfX4Koo9pF<QwK7@&ZoXFg?_!|Q}_6_2fbw@DKR
zsHpP1zecIqhey$^>i1rKAaf!t6iM4=fL-GWouv;vcnD{iw2d%9Qvc#75IsWr91?+v
zd5t$Me_vzDL+A-}=~p@yhKNSGu@<C=)E+6pkfLgHf={RJ*`AI1f$1kZVyv>6p*?RT
zWOt?D8;{Hq&pk|b>#?32Lvdq>5-=IceI=6}wzsP@&^4TwZow&k5PY@gZm9>3LkWI^
zQ@(p6>2icIf>A>u1E$l{{z$p4^-BvCCK9IJ>2bI9z*LktQr&#aU%M4xh>vifux=@2
zZI%U#t#tX|-O?#&)D?vpcI#sC8PhYd$wkCS+ex>4B2unhqUTIP1lyN}Pw|o;ZZ}*R
z&KO14#XTxM=yJ4_z&ZM}EMW0uH3KaIi$`C0u5`Mn__?7uDnFYtHXT!X34q;039uK;
z=t-?>fl!#$gBAPlepf(=;#~2+^hG&lSNR+ZuYQ>mB1lGkj{WwD8~Yb;=kp_hxxYpH
z7nxo9)lDfK4sbLe8vZNX-9@acFZ4xkV$zE8DG8@!Fk<l{LI3Zv3k8|#8R|(Vf+zL7
z|HBmwMBw1`;4se~KBD3(z`*1lrNH?5*3}Z6K_hDVGON=8v`ic<rj_e+#=pc9Pd~j^
zJXcUL6+$!Ks#&R^Sv3@KKROHGI0DCtdl4)f*Kgdgv06)?`fFLg@uSOnni9>7M=b2F
z+^-^>2<-0^j_ki;`Q>j6Y#@VPtIo0PZ)fp(WWG~Ee^<+|j84T}{*Zom<Nlqb@_r?Q
z^e1~%IP`#k#ElXhXrYZp?pG}nCJ<LP)0T5!6i?vgYNrIe2?AujrOIYoqo{>U)Sv3Y
z+CJKg_FZ4URq0rRbHdls8|z=B>z?lOw%n9W2ji&;Yj7$q9@I){z__`Hv^^e<l;L6U
zGI0bjV#Upc-<3n;AMcTyn=t+r1Lk4uPv&Inm#(b(4OVb|4N|ID^y-W9TtqLexyaCH
zZXjUUpJ!a2G`(#CRhke|D3e1(rAIDYsq;k(+1IGOv{vtmKXCgZ=pmT_>Qp<<KgD`z
zPtLc`j4D=4ELNzVb^*vMG!`fTWk(C4Y01dumY8R2Q920e=liZ_XUdp?ykPN$Qdzd-
zuq9;|dNRGV*^fm^+pc|J^ZEiP%)&s-ZCaw;4$P%<VRElT4zhMJY7;XBVWCyh0{onj
zFFMy3;l(G$VlRnn6lhvK3VW|&K#O^XCJi!X=}wuQD^ao<lud^&JRQZvR4ax**rW!=
zJySmMWCvT}aZ>3$qhbG$NlXLE^%!dEhn$CtzuLjOw110RSexw9@5urM`IT=u@%r!1
zCeZqg5nkV!LK4J4JyIjDE2?#J;W%Ad-`1~@cC$c^zQ{Edb06ogiGB7xy~IQg`dP;5
zhkgUvun?u1<?fq(<!Om-i2JUSUUPqx9-HN{SZB1?g{V)=3||!8gA%thFh9gkRK;ql
z>Kg*I=D<w}V}1hA6cpW0+{V=4CRk4M9=d>TaCyr3bdmdY4Qo%==gw1+F7K`Vuln4P
zKkrRuCXj4o84}WtZN!+31V7JU#&$jY07FdODSV{;5;4DlF5y@Ac4`@xw{VSE){}&M
zx;y0=1C~ns#E=f{4NbSI_(xQpf(X=X+8xV$_Ik=FniziFDFH!yII|KZ?W3o<P5Q`!
zr7(Dzc|OqQ3m#jR+$uSY?K(huLDyR)+LK`Iqo6r7{RqN>5A3Gw$*NF`imO|&Yy|uJ
zUATmlyi1re2R_=AN-agg8E*tHjMuZUV~R|Y{88s+&I0Vdl`b}(TiyJp8TcoheHI`9
zyJiWkzkd^LqhF4T!XnS3#7Biz)*w${gW}vD^@w4!u-b@dY;OnwAE$k^_USnB8BUxf
z*n5yPHNC@ch&X-0#U+`Cd*kVAL`yeV_MttQ&0%4tl5_9`It**febt$b_2{%MT`;^X
zA0O>)wBW&LG#q4x7iy_^0N-JdE1{FVhS1$QKAR0L@%M?S18L5<AuGc{c4_m`uQ-*f
zb7!WXW~|lMWk1@*-=$ZF&@bRJK4c{nV@A9^0k~PmWvM`ACZv<4o)QTlR7&{m)Xddd
zR3@9viZ~(<Zar}wKfk&NtAWRVSN7`yWHLK1z;!<qf8-5)Ss!9TA&Ba&-m4-4kbDiV
zkM_t;4OI?&(7!Sklv-er-f3eDENyBIT>2V0^ikEsqLYbvws6QX86#Uw4V#OauzMjY
z+8D~p+P|Ix5Ch+}DEvjYITwclm7>u+j(UmB&onKv4ezUZzWLU8HANl|%+GafzRVlP
z2UJpUK~n_}v&Tfd>p)^1RA-cNms^q8)}MRdA)AZ9Ky(reuMhSZ^)WI{BJDEk9(mu4
zS>qi{0yp*I3m=_P!Vy1k#Ix0?ESzNV^9*dn5tGP#&km%uqwTi632zX+O7vSfzx<Di
zD<F{Ud&E|(@>KYhUo64`5E$c14-viq15{kQBIezkI|=rPrN<H4WUt7+i1@VnQ&33E
zdydF*Ld3jV^7Okx`Vl?3KCBoyeY}Q`*1xGEG~YaTiZ-Ewsg-+ge)#fEH^i+i^pt5U
zyO25``ta12u5<mwDR<zU;I@o+4>S%!QGZJYz*oXmXaw`ssdF3$PpagIJkWs&1KBME
zJL`T1N`)ky%hI8eOiR|or*6fE^HPmzi9Qc(4~((vI)5rDc6!&ekoYMasm;3Dfl}DN
z!1c<~EiH(+5|lO{ow^>1TQ<qBNLPqTeW#IF8R<G+q^3w^=#2e#u&fP#xj9$t_C!u?
zEsF+t0<DGh)>RdejF(|%)EgbY+>bMWD-gk-0Rl+OulC7}#j^U!Q;B@D;99NJzB2fh
z^q1_{S;kG2E~*!65t)NMcqq4wZj>(|oIz~gZMtI`+z7mq+Fp$P2FMX-dTOgsOPN@5
z_oO*K9jP_(i<0$_W2<cUrlrFaIEsM4+YWkurTe?oevt={^CH<_r-yT6?riuDxI6oP
z_hxV|`6cN#IT1jQ>*WqUN-LuGWq;yVbw~F3BtqB@VvrZ_JXF3DrWRJZ!L0|_L&&O3
z9JVnkIV+DOXszMu&dciLKcT(Hv?0Yem<ci+se93R8%zj}p$+}8B-vz3KLVmpB#;Q(
z*8mBk4+G^9&5VKKb!G)RsANfTXF7H>nz0U`8B?iXoZ~T=jVQ98p{pYxHN>DG5GQIb
zyTD(uf)MUqA337lHRbiPx~Q`PE~~6eItL(4`{;D_JVCB^d0E%d1xb|2M+ujk6Bp&A
zfQFBeXhOLtq$SdKMeU`UXC%)!legxW95CQ;x;19F8$+?vb*z<r4E_>3Jj21$dY9Bi
zz<<6xQWgE`5~1eIEq#!!EdlPvj8lhT20?}+E^R6~Zl96BgW(^jPE7Gmt53MS9mY7;
zi*`)R0xq`xR=r~FuTC`Pn$1YVGe>mWuYJwGV!DUKELPn(|Mjm#TaRn3C|Zm6{KdP2
z&m2R%&;NK1&A#WR6CU!6l|<kf@5eSzk0xh1Xku0L$xSK~OVNa^gAdyP*b__=mhb`8
z7?Ak1c<7l{#|3sB7VRdphI2ITUZ?mj0mFOwCNgp9N8UQuA28nFNdzh{k=l_)?4LoP
zJ;bWrZQd{qPlpl`>41Yud`4Fcg?i0&w--#3g&512%7~U_73e#s=4%H<`xI^DC_pF^
z+JeDLz~Z%sKwDQ4sOW*Zj2c`%G|N0mCp9F+dioZSnjITka*fD~**5%U#sPq}7GnEA
zcpfwR_f(iOwM11P1ERA94m6K^5B6T?`nK{Nnr6B3JSH5Ul*RQRJzMqM*f6~Gi(5UR
zi&x~jx!EM$=zco9AvM<_azVrm-+<l!WY9l<#dnWj{ub04GyitI;EOlvZFT$332HSZ
z(#4ba@y}dho(Bi6&7MT4h_Rk;>V90x9DfICM~k?!s??l(bF+wWT8m^rieNR>!#Fy4
zl7fnsNFWG1w!I`WCItE@r9uEt#Aid;F-H-y_8sRs(O~X|1*ywk8s~TYvT(u<ZpP~g
z0%ixVay_v{h2Dm+8!%C12v{>fWe7~4&h%KR+c9Qc%V#vngx3(UGOTTjPV?i9Ut@g9
zwh+WB{myc1Dq;cI43^d94>Wgr)Agh`x6|e!gpQ_b{5w&}d(*P4w=joje02`(W@3<%
zach0Oy$K&(b|gXAi@(RXvMynl3Pr9WKzGE>E~y2qS6!}+rwxjTGRLg3l&qNv<hh9{
zql#+^!}m_W=v46FpWq5}-U<CdQg2?Yak>0kC-738j8kauNK>or$BUS+%BB6wIOQHq
z*pj<lqR+ZSpI>?kC^7MNGFr3LXuWk~jtjAW2u{WjDhOnA@5%}BIGPVwxCOvmfJ0%q
z&|g!ctcmaxNhd1#vqY}UYD~ciz#6Rs@r@<EjX{l|@9KLz5!|470%H$rR393<K~via
zIQ!buCmUkI?c@48&kb;HIC<li)s!o8hhNl^Hh);=spFBIdRtt(FSZU8p4F(oRF2eT
z+I+geUW|Vc#?aF#8?E76wU=fdJQ8{r+{CC<F6jqYW?rG5Z~eJZE7oS|uCbpCw7qse
z<pk$a&>*OgQtvhk>V`D=ov@VzR_KU=o7eN=hxwO6Kb;YkjQrB?@Ve-H4_o2#<L@N?
zhlx_>Y#ym>ztI0{olr!^@X0O4>=dIY{4Gnh%Gl;sn#y7F0zUp!!^QXgQNk3#HqSJU
z`q-XHVB%G-uH0+ddTHS@kt?zTUV<}di-Lh|c^-61HGO%&3)J{4(HM)_tKcLF5WmhA
z<LAc9Q)%ru08M^q`pEYuziS7vC*~4*ktOv;L%?G#+N_LvgRjeAl3~~V;dq7!&<J*f
zF?UC^e?be2cO3+_W?4!(!^EF-&ka40_b^?)M22oS6pHNKCS}*b)u!n_3HOjzpv6L9
zq(d#l)<cm-Ywdu%--06Y(Sm)RwjH*dU+<Q-lEUTk*6DA};^6hFG@vYNMHdo+Y4ouH
z7lv%zdZ3zOp6%V{qSPj{s!adhAVUj_vcL1(2lcN^fDtp~<{&1?Ao}9hf5ZS0ihpGK
z>2U*Xu$R=DzvhN1jQbUzM(JKkcOJ98e>gtaEG5%22aAl<$)Cg}m1$o4kIRWqOpEch
zuB!1jb+F4^gI(<gutDls3XlIJgXMx^yt-C$=HX9@S28K3>iAt%C@!5B@qdfq6lTuP
zt%n+Crqd+MYTpI1_hju9ms5Gcu$xTX8yuRM%9Bj`Xs3D+u<7H69cZn+UOogmGF5LW
zfxeOo{J7tCM!|IfFNmWhW{etB@Ay`5J!mD$U630***amjJ6LE95leta%eUkc;~no|
z$|8XQ;|>4A>$>1n20e%OM|w#hBhCTnMC#V-UZid60L_=^RSnE^*E3`tJ~;EOzqj&%
zV}wLdOwfW|2exr;Fj%|mfa1k~MZ4K96kOXDvu)QRHE&c{p=$XO2M+DN-6mx_J#lur
zG$Z%%+uBp<qkASICTbGd+aElsK6dsYwwLv1zZ$zh&hrjcO%41%8b3_8?g`2<^*jIk
zUxnCj&2+#DibTk5-+s2A;g=D%JfAs*+^YP&%2f0%P84k(T5j_>f0?hh$n!POOgR<e
zjqD;mZnp`8W_r9!l#J>F6TLw}s-3U4AE-I95kR_K9X|;^W-H2<(%oIRZu?8`CeR`X
zv4{<Lvgv^9zAt-gXYAE;PX#|A)o|E8cvgqJE(j<3g>Nr+x{zWAAS@*f?bmLQpW$}w
zj~MOtE!}Jx#D?BxP*ktFzx7O!oj8@0K|v1$NLYCH^=$%eg+K>AAM@vo?&iBBzY5dk
zHIg4KJ@-V)Z3(zO4DR7k2r`-;kq)%T;wp^B{TV<-E?|9cI+bAliDiRUT{y6wXd!oi
zWZIIsH2_pNtInFoaTrg@g9OG?=-*it??xoz_rzV<_CFr@OVOq(bjr&(g@2njwJ5rm
zdP^O+%}g08m?s;c(SEgj?sv$WQud2<RNsD}o2&CT>O0;(ECo_1WsR0IeTec{Mo^aT
zLZN8d@p(9J6D~jJW=lQ3PPZPcxA)wXC@N&!+mb5J%+-AYdT{w-AJ2($&^#BA$b0p*
z{;6L!ac%|@|9zp)Xx6D!d>&R!0xLnql_Cy~pXJUSrIbnmMQ`XBa|05zU|PMMrsa{B
zqdLgN;Z=HMd%1)ufhDW6(GWzml{gSgZ&eG=Cem8p)6LBRO+{k4{sM^@e<vnM?3Mxy
zv%m0sgq;&f0uLQGr}QC)b%?Ta9a@#5Wm)_ELH22U-ttSE@Y|`yA9pS8`vU!Y9`U`u
z<^g(`589ld=CNBIu#@@*-9VxD$s2Rui5~q+1SM{a$O9$N9~z~>wTSqK7hw(@xJ;R#
zI8VL}<5V8kg)+ctlz6+3=wL<uok->~2>GT$AlFW`Tv%zX$#?0hs`M@iv2A^u0J6r^
z`mE)VJ-z6~*P_zhQ0xk2A?tk3Qye&}_K7ssvuk81cKsDuddl}88wUJ~HYNZ_Fq2Cn
zev`#KECjBi65zDWT2!dud6O83chnJE5wzzU2KWWDI4}>FLC+(Fg7RTP*T%$Z*BD7W
z==7#@j0Q;l`Icl=L|8GtS=|CdI*Hi-?>YQ#)*cFEx0)Q6vmc}Z1_^`7&_dWiCOtAd
zXHWe#u_>!*)O^&qZ`vRm-V(_M$fEx>CwQ=ssKHUOH)!so#amgk2oL{ql+*&srAzdd
z_Vp3=?3`m9AmQP_K4elLH~meY^wfTGsQnIN)MeO0vr0nm1prvwVS<o83&f$?4ZxGO
z=XOFHBypvjk2Ns$<C4N?WW*T8ubydSx{Ko7^KncLVL(8YYT!r)@nu2}<~!+7`ZR=6
z3S*iN%ak$TT#?U#lfMq5oItqz?MBb*=i>1aR0*MFE<*=!Q-%>A$mWJf+}ySx{Tiv-
zm2_gyAElBqiqr)09MG_ga+h!_pz2Jt6P2QQ1Q0VmVeJ(H$p+39UjWXRMebG7NoM0Z
za|(N>Sv(Bht@gsW>Sm<x-b_)6%I_z<Bb5JF?9iO2^!>2Q2T^AM&J%P|<?-A{h<`jl
z{Vj}x#;cybg#wXT)-s5bf9_>bpN$l>ktp6XfHU_^#aUkt*~g6o=h*0&c>>5j8tw79
z$A%x%t)}D3F)Y))$jp1X`zdt53_wCtiqPawqp>Dw>&fO^pkIiMY1*d?Z#1!}2d&}d
z2VD}@VH_M7=l%UT4}GoIjBp({*7qeSHYoHelY!G@+4&GEDcO+^`U7R#7jkO9cV83j
zvC5ZnIfE#j^$jVTyMzxMTfk4)fEy##FC>S4s*-1Rs{b;~on_no;Q)3@9AVd!N4XY|
zO4+(S+`brvn{P7+$=al8Xt|)`Fl+(pAH98k;oPavA5JgdGfEu{n4Z0emHPE)dvMU;
zajsbWiltBwk8^=s#)6;g<=F<{`TTU@j;%Nlt?0U$G*_J|fab(X(jW!$s_;>!#g#{5
z3N+0}MeY^Be;L_e<z<ak1Gl3&Lz+A@p%=Iv&((BHla&l1GRAjR!@nH)SX`nSMdebQ
zO?#id0UC!QuvKGs@w2Vml=84?FO>`#N<Y(>9=*>KY1>DAsv~0xeVp$RvS#6g-b!G5
zaGy2brT*puP(8GL&Dqx;QYPgm<Xa^MUI@;r&XwQ3lr%w>_NW7qF9jrwd>U+?)j>*C
z1Z*nix5*6?&T=T<{U@<p%a_X@4H4B9DW6u=KXf^$?=(QF`-o20?NM~nS)h&xL^n$^
zvur<stzpcdcE{an9qenjKYn;7#U`#@j+ZzOM894C-y5lcSofnaxw`tVG2c=|?HD>$
zN^N<*H<YibjO8eJmd<^P75q)9QMxh%=<AN{yKLXp1Z-@}c{bHYc1%2*_oFN}Xr()4
z<nDoWCGDwQAf(QUY?g>TeO;KSnU>&Yl@`dkvLCsfkgt8SnYcGyBhpijfs4#VfO_a7
z+Yb`gZBM~?kGM6hYZTT8ql)a_2u3n#g6vw4!uGMSQgLm;DPGxUoG3}Il3l2gEkLWY
z$Q-<Ioa}Ev2;D*CJ(9L+?@a;FM;{Gr-yqTFO>0>5)IKq(WF5%+2Z<gVL=U;AD$nag
za=jBga0_uJoyNI*y@@+rn*4n40k1pZC?&@|i>QmgG`?}2=#iAdO`_5`uA4nBNz9hE
z7<T_!kk4sYzyAYY((Iv6x*famw?nT^4t@OMR+d679QxN=(tLlx<UMD^p|7qO=iEg#
z59YAKriLzmXNs(L!=dyO$AErR^xH#hP6B&_%lKHh7VtY%>M>(^%kVacXqdz0)f9IV
z^HQbL?;jX<iyVkK58&Jm!aZuaq>`Q?fHE7)hSnSU{(%^{*LA3p0W`rhAzE^enR~-s
zsR%?vUu^=P$C*~f^v&0KZ>+WR`)6!n<p2$>Q!c=2uNAsVUJk_4mh3YOK*&*&Vsf{f
zy?h>uYwJ-L386X-LKjQR-0N1=w2G@G8aQ+%SIZM9mkAttX5+H3#NKS!rGuaogdWe!
zf1^Vs9}k!BirIHUOb#XVo<k+qFB+;m_W@(ywnDUuAzTt;M`IoWEuEg`3Kd~Uoc4np
zuE;7Mp<kc%++#fZhJu!p8u04+BfoPOy+rcbUv?;6ljp0Xxj`pBwP1tX+#hAhW~q5&
zsg^nB*Tgik>A{21Sr0KN;%Ic{-NGV?0EwuIyb|}x_MeyS%cEi~=}^_)|KJ((i5g~z
zq3X6Ic&t2^hKZuCNbcl7(}PF7^;$vj+mng*|H7FW?(o?=hJx-)Qd-?z3Q;Twgt2*j
z@MMByI06DHuJ%22rSu6KL-9LzCQw5!mLCI--Vk-iw7WpQ|BPUR%cyn+a`g`Oic97?
z3@uSp74uetQn8>XMlrL1$Wdl|&FnG&<^RZ*H+^b+bUmenijSD)k@Kb&`zbzfN#hQI
z`?5|nLQ)4qHhq#*C!y_`dNEGHnnblVAm2PG6>i4BB7m&i$l|Cf{;1?-T)M3le;;1w
zw*Lm*GUxqoWe4}@yBTe(BkLXN(r9TqW4=r3o4&uKq_8c;|4O*Ov^RVK^JW_(_vmu!
zxycna49rNsH>f23<_kEENX2Qljgxuqcf1rpq}rNt0tt{f5cZ2wzrWB+ro-CwMBJz>
zW_k!knh=*eU+`XbT8IR1m*7)K-^<SH1}N<(yQghklzpgvxF*&fy2nQx6Cq5ca_0DB
zk)av%PE*f#5r6fs+1l(n$wz04#x#MfB|0vx#FD60xBFqcg0^ko1lELxm>Gzlez)2U
z)||Tmh90HMy$6&)hlrbWRkKZx;yLk+UONCEeDIz~5ORWU4RE4_Nu&-N*A2z`=Vm(d
zTcX-EdKYHvY)=!>&iq!O_W0!9nFLVL#;7Q>@RnJ7SD~oAvB~UaNA_Qtone)^ZZl=w
zJa@8rp!RSCCfvH%MVR^R?VZ8~<N$j9wuR4GwB$X7CWk8bO*`(P3MCAKTurKtaJv&q
zzelo=H5#syt;UaL4fQY-UX2t}Qe5rm0G&bW65%ugq<cIt>__{oTyJxdsm>HP(#{E~
zv)BHpxc<*Xl#gOI5K?Tv%`o>yLD`*$zE5gBozWo$z87d0#JkUn^UVDYuU&a&3U29_
zx#XKL9?d^kyS?NK<#BELu<*U?Y$oy-EN_Wq=3;a3yvOcMF2;1NV5->FAAO{!;5Lm~
zWipYh4FJ61FvLycEL2#Sz}SbYqjY;&v^VL2C6u>!@OPD2X9j4ppSe_5u|CJr#amLB
zii5y&B{-Bij@;S(ms^uRAWmbR1|8`Om&$S+d#!vH<Zip}@vg)W$wueO4t>iQRKL^f
zdp>4t*euh17Vo8u?=M`75F|Pdm-q9QaJetvT;_g&kDDZ@vf~|-mB0jet@!)clfQWY
zbqU()=IR!}#WXJo6<PdSGM2$1%?ixL?8Z%dMddEwm-3g%D@}pMv<{fiVEpV$O9q$O
zpbV?Xv^6n_)Vz3f28JwjX^(gOKi1wm5X<)c1D4E0N?F+{Wk$*#MWR6}WS5dHJA0&}
z5-Ou)6UofpBQne0J0qLy&3l~c>7hr@^ZmW=`=9Qv`@XL8ypHpjpW}0EM-jdGa7htw
zgl^=_mo}%1;EEzT(0}BfNAHV1A~?S1_5vD&u<4j~GsDTsgA7vW>4WWo#1pNh>s%<_
zc@2Z(XZsc;&CR+{?gjuCai74{FoH+r3vNQ@_#WRVf;3Q7ESZvmtrp0Qgn;0g23UTg
zJ|h4T)nM!f#DsfPmxRl@?@>TbJFp@~^0x7qY%@2Gk!~r=uz`5!U}`Z`BrDdVbM5}3
z=R^fXP$op(4N%S#>2GAaM!NW*ADbQ!8qcA6cPu))36oE|gnO_QNaO<>O7aipNLN8E
zTa>FLUTsfb@J<^rQYifg!RH#4$5j>>cfX1P266*LbV!?AQ2JToOk17Bt&Z#sb6;fr
z-vcT9H>AAsrY!YVqeJI{X)FDio~nVkln(j0{I7RoZ@xl<6fZ3}57J?^5Fw(WiMMn|
zfIw}IL%cEalh(buhr>{Bc@E?V*X5E(1nH9#am1_FZH3hb*$;Juo!f(&0G?$Vu|ueE
zvi#u%MBr^wE5$^SRRv0TvF9m=n_yZI)V$xq1&JV0X&pDGrG*!7TLJcUWe4j=(xonU
zGj|T9E|bi4U}>4eWpjO-i^E(KOaTtpTL7+<yz5C3C2Hj(lmw`f0wQCrX)<~Y8m|m^
z4Z<O0Mw1(>a7{qJJzUQu%u#?i=O{$&yLdjonl^>G!_;n_(_pAy0#29f+Xrki(3ugC
z15yQZzu$k%A>K3@XRfqT9vxdpFJK2_7HBB?5vZj~AA0!jJ}q%+#)eb7==$zAR3Whk
zMi5DFO&B}oYkJ7(Pz*em-6YPAqH|abY=d3kU7}PHbZm`K-ahSd4d`u#!0MTexAWQH
z5*@RHP@@cCeO|CkRSHa$$|<O7_LumfV#_+_0-VDv`I?7Sl>Vok>9Fonx`qJz7QiqI
zVnzWSRt_w@zTWOLcOY|S0<A_*DREK}*>k7u^cT_LdxT&9;sW66iN;B*K08YJ>MNWm
zx9jp%rvd)CG7xe(<jmd*9E%7WTPJ(QB{}EgP;xj2N_e0(*?#UWa6zN!4pl=@ICu@>
zGDJ7Yp$<S~#Z>**_3z~>QtbIVxJwJ59&-$}vm<Mj;sELGDfJ0d+5{r<R`m#|)^dT3
zjG4*|b^uON#UaZslyNKepbctpV_yqG-59zJ%q$!}JOFXaF?SeTjhI)>eiJxw%bl<I
zRY@hNfkiNnVW$E*OanWnLa2F7`drtk^o%HSNW%<Zs-`ER$VCla2-YU<x0Bn`O-;mF
z2$-B$^y{pNP8yYp?Ga*ZwYT8;;UcsMz4ZF8kkV!2xVaRDY@OHcDC!T#&5gu!81J?t
z#<VDozxbmcG5t?1*h1>Mzm&fE70!WbM?A{iHz6}|CGoaY6ewc=nN-u3(qCo|Y=T;)
z5<#jt{Ss6|#mSmVvUF}>+qJHOmx>0@z5t6L1tQ)T{!)6^@{FZwJ9Hdb!+7xoFJ`&(
zX3uRcQUQ2{u$uiMY5?I~xziFwdldL%t{^I}uC<B0^||>pU`+xfJ6|>SFkeeHSFZnz
zkWv6RALT&g+?Up6KaKnG@;W+Z6k7gY9SGxSjjBcQp_<?Wq+hrbshzmxkt3Pa_PY5`
z;KQZ(c<}WWO0EH35IJ3g3@hE2Dn@4GSKSBrcN?CVSp~V#L@^aml%ob3D)V~Tkh2=+
z5hFAL9D}WY38PdOChA@>7hjfiPK;4btEBZmTQv!IC2FzQb#=z90l}>PLUZwfAP5Xx
zmxgoVO`lXgg-Wq6DEx&9huO;Uh^BGK@ec;=NijHy2D>`2*DQ+fe7*nJ6~?SDcf98z
zI}XDP&7W9q*G0SOHHn&Fqx_4>xl1FPjSyV%73E)~VM^ANJG|XazTY}N)@0~>M2XfF
z_^l>T#R$@YTtw9nQBpv;=sC1dfqRkP29kqZmu$O^YJtuE#y<i&r``LIwWJXLe0fWF
z=cIBNjHQZs<{&Gk)ZQC3lDk-*Z8oQlCD_@-rC*_-@*E1jidPV11U$)#a;BXdPmuN3
z6WDyIKJgz<K(g@!|9}Fh970cE`hM^zrW6T00VNyZ=3K0!3wnYOKsY0ADX}Oh<t74+
z>f<@ZFYm-Rwt;R8v*w#YP?fKb@V-)dwo$4roikgOi=LT|W0xh&GRn4OCqJ>-<EHDz
zhFXR7SJrA=tXdXOv}z9STgGY*?UlSo=K?v8^8lC7gnBoQC`5|LJLjB|+|bCD>iHY$
z-g!Qr+WNOikDFuyonKG-jZ-gXNrAKFAv{M=TreUHczU|85E$Syy*<5oUrVtuhXB3-
z{7g~+{9%m(p}hbmDfVY$N;X?}Lmoh-3fh_obN;>W0Bks_*_<xgYQ%|4<&c$^i+ozi
z08>=IIfW<G9^L$imMU}|PE^uDv)2~E)h2uk`7BjnD7pKyzo(S?r%0HT%!+aRyUpVg
z&5u^6yo13ewpCzF-h#z5?bivb!(+VWXhj#|&%K35E<>Cyz=;Yqp7H^>hAj&4Dy6FT
zIy8UwQ+3<z_@R9}^DGAGH$VGkySbr7VOAv)hTUH^$+sL&(Kr%FLB|Q8s^kOkA{qU?
za5k@njtmd=Sb!xJ42wkWMK*vsB-FU}q2xQ1?I<q!!kA`+C_2D=?f8_`#*O=kem+-5
z?^1~Gmizrw>3@GlL5HII6OKN=_3dR`fdWs)RB{cFt}%m<UY4fZ{F<Fq7|{3^D=5Ar
z<l_{WVMH1lj+(ZSljG5iG&ZdbO!q+@l~MAseyCOfb$Z{KTc#LD(wzEWOHevgvT`v(
zZnNxdeMuV!ixtU`wz=2tt=Bnup?K)<XDX~6GO&A=OEcBK<x9~?#Bfr9RU&bvCsPqL
zgGq4zIINpuuM-=I>^p;8yS7|L2g7CfIOhXSSi6aaLcU`k+xczB=2~6)aW=)&vi4DI
zH`eR#S0LKDajRBQ-s4+@$Fl^|U9U<wFdlh09?BQY*Zl2F|M;h<?au?yBM@J|8O5>9
z^62iM{Ue|L$6Q#?%b<&nc)ctpF7*a2@Z*JVvpak6f4r;j)o0HrZ1ew{wu(!6<FKXj
zhiw*^9az~c`2*q;fd{Q0iN1%}^~&Kd#@qc6m)y_&i%>Im+kP79UpLD)MY^rp^s?F3
zv2vw{kY+i=ga*T&hGUIMl(pTRh+R+rywz?d4~D;w#I}Dio<XZRqM@JstCjM{A@{U{
z9lX=!Mu~UZ5A171fNb%0rrJ~Y=b4Pk-N*ayA*X2I0c63f<{lh3`fI5h@(H<STj`6_
z_}$;}EJYoy@^>-;&U72u==<wl8B2+w)7viR*Vi(n(7L`I5nI|i1^o64P>m;s^tni}
zOpd_$Kf$B>8<yzDPY^?Nj9e6AUa{=W2K{d-|MeFjVv!-}yICi=Mcf@l$zcm0Inf_<
z5@dv1xk#T9+YAr;OP+yJZX;-S3HLYrecbl%DkGm1ZJ~us*>9N^$%nA4t0A8`iNB{}
zOgpBbk(qy={a=55&;4;L+zEAonJqn|{}3cPv5iCg=sVX|-9*=f^hHwu7r&e6N2D*L
z&pU7TDy%O46hfN+o(;RSajPY9Ual=0=$VG}MOu|e*f~+SQ##Q<9uh6gRp#f7YzMqs
zw|kOJpB1qpt8j}31ADVEf?ok!Fi`Zsq?Q~L41^V(Lf|R>?Fxh4^gpGY=D+7i-yuMj
z@$*0CNZ*4+4R)A{RS`>i!@+tV&bK9T`BQ>LBf!f-R;09^#oc&Y5&d;SETxgGTN6F`
zZBDcm0y9|kiJ!QL#ejEwk9ht~vMu}j*AH<1G+3@*`((GHjQAx!U~$AZ?7g_#v)`#}
z6FFc9N8iZqbh-#$Xle3^a?9;r*Oi|R7JY*PX32IU`nSZFCT+NQRe{7fTl`48a79?x
zkJ{88SeIotV>R{`h4SNpomBq1mx8*y*_JxiKc0+^2Moc5!uH#e>RTkk=^`qn9|^cr
zd|$K@w7bbSiS5`=cVJ=uG+LNit_s^c%U@?5t?tE!5WW=D-vXhBEAWsffBg9EL!K@O
zGf`c3=iMgR{iWKZ_Wn>!o?dmA!nV6$(`U^amSp@G-=@#d4@Dt(2D29*4rV8*>#04K
zS@SbQo1gL522Btw*ybjiy$~hRxj7L0+pqlgPb*ovR*}y#O`zi^RgT};vJLqJpyOvg
zDhBVt3s<0P74+=vG&b9E#a~ipk|&zEb_qsLn5?H1;{AG%?XP_-n(Xi@D2{{B6rZ+Z
zYr{5|G)M`>z4H{rFe?|<csuJf^RGSJ^5l;6o^cK@&(h)ca(n(Qe%t;<XN6e<oCKM<
z+!s4t(QwAi&qs%B%b}r!Jx<y5#0E5_2Y0#c+j_-M!R*5iAL+UcZkyGBm(Lb?`{(7y
zu|ii1SIq2mw9Q`pUIQsdu@09m(E~><?9eYe1IKu&w>?sHzIZ=x=&M3!M)J>_Y+5Vb
zJc0PK%i_>d?eX~vF3hb9ytfN<|GhvS>t~`2=#GyCNe#BuO`c@fwX=5<A;c$(KUM7Z
z%U1rSayIY3C#jP*um-3CxDngg`8Q*W7etZ)LI4uu_=jn&MNB{5p36ez!Ib~lOjpIf
zB$RWH<NeJK0S|_M0M5i2KAqi8D#y_4^L7rc`|Xk+|Fo2(Q(ZsuRFsG&6K`<Wwkq|n
z?}^^|wVO!|zs+auRz>mheEYuz{ylr4MP1OkemlQm^GiF3$y!~8Lwa`0`ke$(CjTGE
zl+A}gG!XT`2MUhu{`XD<I%>Fql#6wz&^Jhk-A3KS(4Sf}k8(bcPrBfQi9N->>(IPg
zKIA)Z1Fk0#q9`=8wmw=xW)~^z;On)Ty;xTu{~EAM(;c~*mg;akJ4Md@iyPwPj|W}$
zOuF^IIXjyczHy-FaM9{_P|BIGeg}k0*)0#DJN3|)()1SaHWD=WSXbD-1JoXkeAk4^
ziK0s`0%}3cL{X=c&XWx5Rpj(X#XD~B&P2`Nl}PY}{&vtdi*JT3)E+Y?V`{Xv9F4!U
z*I)VoWGEjqK*g%n#2|7b{|<O&w>6-^P-V}VxpK$@3p0vzqs3>mxrTQX5vxI9nfv@4
zgRm3OBA*3<7;1psK81XlcqMP|?hNP#>(%pW#Xn&5CQ|G3D|*yeGia>MYJ=G>&@vs$
z4P_f-fENgLcKrExbL=86EqJQ+P<Y!j`|E7p4#yxcdts{NLB+>_s3!n<Q2}tRk2K+s
zheo5l5xoqc{S|;{k^Pb%H@>nk&jc^Zc|xlVq9xpYYYZ~o&?j^dsE#C4M1dWkA4Wo)
zA`$>hq}9S#muK0Q`}iRb{gHqR66ipit==4g3eTv%0kQN201Jdez6=2a5ZhHVf5y!c
zV3Y6O(~j6}!m=Qrnj+Fw%J`hr$!K~8Wmof#)YTp%Vr~~u!Gh!v5xm9-v4C5@DU(f-
z?`aEs8;o7A4sH7E*9L6*tefnHle7u@1L&=g0`hs2dP{_nKq!=~@yrWmr|(7Rqe2`c
zhbapYVLD9BE+{?ZB+8EkeSsTTHRD0V`i|&;`#!y_?q;{XJ8-+skqF@GUI6>N0NDkT
zTJhEq==G9AV+EsWp*-V_2t&)64$Wd3PAB6Ws1YU7A0p}QlnN3$Cqc(S07{r6$gX&+
zSG4J&J<Bk`aqLT~X#Ptax0eaQ!yu!5{_eoOFAe<Vdlh=2lI;T=9Fl;_=+$MC^+rft
z#4`_U9g>h3eyfsU(xPj<vM>n<5wft#vt+7_h^+t;s!rFyQxIOai)ga^9&uWNyIC^y
z6yfbqg8tqXGvQajbaY_<;lNW@<q%*n5oY;~7$NYMze)Kqd3!Fdo`8}UHB1~SL#Je5
z)|UL8&#%6|HT;&5Yx%Xa%ye)!Z$IYes$)zZ)}nAc4~K3(kAG>}pPzyI4##sNzw>Wv
zcWOThOD_saL6~xZ($nc`{=m49c%uLl5U39649Ci71uSCBCOSK4i|CC+3Av1c@(1^Y
ztU5sX3@JD@9>~T)oo))|Qc-S*cs{BJK*Hz*kqwA_1t6M{h@&wkSKgkdn7=ID7X7ou
zdCifT&)^XCEP#rRKr!?LQS6SJ+@cqmzom<Rz9lhm;`rB7GJAf2sl1%Rz6gn(pU-(|
z6F#V8c;JTp+_%@grlCAl1UYAZS2(}kFr6Amb`guD-M{Lm9atLsL9*R0w;Y{=qj?@4
z<r(Ee0x};!sE}Jh;f|PaDM7o2K{3xh=K}{X>ZXKQihwvS?@|Y1v?)MpQI-S%Ns<{T
zu+pbqydyD5eE=Qeeu8+sYGDGw-#X5mO&tcrLVtSl19LQ56%~QV$G~*Rz%WmoLw7+v
zOo4tz-7beJ37XVwViC{TyOG%kIL?SaKQU7g7@2AoFO|G_!nSw%I`h)&%3JAQ>R_I$
zFLVZT2XvHAmq^1l4ai_%NRi#Jm*zKiCjxCq^+EPGJpJZft0DkxI34Nwdso+{gM>7~
z?I)lO_c<I{U!8Ec0PRB09zkOhGn%aToJ<l#9&r%Q3^!Z=)HP{orFATjVL^562upxF
z&(EGB3Kuot%0vZz3AOYhu%F?8V@}knq6tGY43MJxVWg^oX9|%cJS!vuCK{aG<TpTL
zSOP8t&9UY5(}r#<AY<o{P%}C+@DO3MpmIlx>ZbF3pnqa)<F#T`GHOM95ud3$-#2ow
zzn0~hR1<<eiS7|YqF=wkoTNr6NWRzAvk)C5mK?|-8Qd9<$9TbUkxwykGNl6S$I83y
z;@%&c(NO^#nARFjh~@p;(E|P{X-91S$0Bc`X}}lCS8dwqK%+SY1!zjE5PU)}gbS;H
zcP(~rkxW?O-Q|LEnX|=kV$M2CVcpGKum)ffaHr*Vk0wl<925g}`)n0av!iw_bk<x%
zp|i3NU`9MdtAGP{6{Z)IRYshZA0OaLnEK5F9OIx>$YVIz<uGwzl=Ac0o!JjM--OIc
ztvfiu8GCIAN(Nf%6@iU6Xh=-U7VL$@j&B3pJ?HqigGj<KK5WOq{X=k^RG3BSTvV)k
z22cIF+c!=EhpVSb5pXBsLH^X)&w$+!2v|w|`<LutD3}Q#_>`Byabg1A9v|<ZCvFH0
z2791I^&S9`sG;@{$CVyLRddiVz2GRK5Jscrf9cs{s+q~7?HP5v$?R%PXGb5m0_Oo8
z&{cB!9R%gn4m_p+AJ%{xt_h4GRmwswhYp^&dbkumq72_Sqic3ofFA+vRT{7POz%8;
z=Jl1%(a{&{RZx!Esn5CJ`D$k99{y3uibf!}>}n-r@9&P-=F^2m{5=~T9GfSAay|lh
zr4-6%<^{7*vV03O4CGgVMCTgnLSqp*Kmn#b0TIR)hnUGOLXrJe-`VrpR;q3^*|RWS
zj0O%)H|zQ`wCR_@r~`UvB=$lx#g<8)hdiC1#jge`2X6p2od<TdeKq<Q4X}z8fRzBz
zCtNgtAty<V<|QFsp3t*TQR(R|crj)C{(4zuGk7%Wz%kkfYTO6dRLNHtpb_mYOstUz
z<VRJ{(x*%t6lef;Lu`85-w-wu-WouPAHz_w3WKzcJ7Y^A*nKb!FL`SEd@&H^t)4cW
z8MqF%;817D<5wlwJ(RYpOjEh$Zz3BEVBF3ziY0&uu9(({tgrM2`FijVzYC}3c|9#t
zD?nSaBhhq1od2TQl42%M)vZ@OgwmxXl#LlaTwlg)dX0l?vP@m`-qYEw-FBVP)V{H+
zVjVE~>wMAB?Gvt{R#q@Ht$zT5in;n~&OfXFo_*l+tJq%qujIPL4${FQ!vhz}07+g7
zol{!-8y-$(dS(IBokXs34;<~6Ak#`&!qCf|F&O%EWf6e=dsXx8y7PCehb{vM7$G0D
z8t{E5LA;-`H#gkUIS9S0KF~BT4!{;+2k*xuOo$RC={>@<A|r6;?d&u5^LJr>&htoo
zRbmks%Epoh(!+JPQ}B=jp!rQ;?DEnyw1Wva1moH@Je<uey~acMts20n>t}jEmo845
zOyz-hRSK+dXLsr%wtXLV)nO5)R-ewD@cLf@&(gv&;5}sczCIvHG$&Wpj+aSl>K$Y#
zuo=%d2V-CTVorUl3}8Xb+ADx&u%FMrZ%@ct4d52&=LrVrJ!7;jSJ&W&St=PnZmAjT
z20gljsE0WFP{#)v`3o!&W-z(%)yjO{x;N+{)T@iioO^{Lzowd54=w=k<I=iANH1bN
z>dFG>IO4@{_`b`0u=zt6rxzkRijWC1ME>G(3g`m~l2%`+n=UQi7&Vg8uJ+1brO37c
zWMe47Rntu9G*v~83!mlp;Qj0>&9tn=V{3%ao{O=;R1P~|rP~Yh%VyN8Ld+3%v&KfP
zB&LAdEU?aoOQqWr75@NNrZRDCEQx)kL6n?7Z8#zj!Q}^EcFbm8Di(~GkHTrFIoraI
z6Yxe2403fyW&}V)wT7DFgY?_eYee#($G&y~U?wb0Qz5UT((iVB)0xc%TP2qD(sKsz
z)%^hZc!}I#Xyf!TXHS{T&yHz&jKf*WC+93*l=xO-t*p<6>4kB`D(f?gSRkQZe-6zY
zWuO=6K-rgv)^t(K*e}9%j_o1#N;x$6Y%OlG{;EtFQ1EMlTdJ7Xq<!24_>W}KwD`#t
zwU=)uf^L7V+OY(Hh;l@Q3g!P7AVrVY{C9WG%wJno49UawH!h|q$g-SZfcg#D{=`4q
zv{I}cOd10JNxL5|n!jjYLpGGR-;6$t=qu}RlAsJ<zt^jhK`eoU;gLmS(s0d`uNJxf
z*mH`@VL4WS1d6@I`}Ff_gFln>vOji5qkY<~%zinTPRgH6;#Dxfx|Y_el4G4xrPw|Q
z4UYjS9LWY+&r_@LX>O!YvlUt;W>qO_k41xrS$_-5eF&@lJ6{2pEg-O;G}<<<fr>mh
zCh*jDb(xf1We~ACe&8_j?^Y1{1`x`m)(eeq0FJ%nkL@PA9A+mlTYuFKkb3ic1oW{}
z(l?x3RWJIwkQgz3o8%)XvXd5t%obe%WfxFPGq`hP3Cy<R1l_fpVC{RB@fslH%Ci+P
zqutal?Y=a|p6E+V@9xV?yuO|W?d;bL{O#si<jn(i`giY=S7dtUCMSO)c~51iU08PK
zRY1^8LECNN04wXsIc$CNK%>Cy@e3evV>s?4mTbUbr?=k0br<NfqSqj*|0Z8we&Lhd
zLBqgC8)k|&n89INLS^cvxi5jx&isDog8vOZPV>Vp(=tc81im_+nAOq=*E{SJ<Si{q
z+67cFgr{=U%`2fzRlR6Lc_8}w07fr{Gy_kUNZG5PhI4IW&^s#Q;Qp|Z_G>Nll4u=e
z8zOcxvT$1Q>|xxJSNY{?Z`eCXuB}{H9$F~~+S0JHojAQ<&3`FRLeL%}hP!@sz*#v5
z#>n(R>`zp=Yq?WVbll;*srC*nCqKXrl?MTHz7CqsF54F!pZjF7tP&7GK8TUz-3GA&
z!0t6+P>Bg-3MiHnUsmTrM&Qf4P#JMy&}wekU^tKh?1!{NYO%)t-BPHN1^AC%daBHa
zyMKTf6FkYhbTZc$$^aC~yNZmAk4n;=XF#@6cQj09(c->>Byc8uQk>BnCGf4bnFNxO
zmjEB*Q+b{~1$5!#6CO#H!rv^+Q~;2;U+Y7ZY>8qzv$W%`LmZxxz9nd$1hgo=F<-@A
z>YsZ*AX()>LYLo+JhmU5d!Jq!k<OL_U5(ojLqOtML45TL@QKi!eS|5%?8RtXT2*uG
z{18)+twx_FQ^-1Nq2hPuN=$<=KjAO!iZ|{;X9E??GfLF>wCmRVFQjawOvGJqlb=V*
zyuF9{klf0t=%>Ts6uxc_4QOkNh}B2JURwi@I8)$6GTGA9>Gc2{<RlTWLG5L_{`5H+
zd&soBcFXm+q-z~%hSO#+T<kW~`@cT&3bSQ$^cXyeFmhB44uJaaFK3D0t;};^L*k@w
z?5sF-)xqacmi&y?+9jTj{nJ(BkOg>LX?%{_+-set$%>(I`AC+?{E4q?QpFRmfsA@D
zleQnPlUB?K&lL$sdL*hcU9owc?osvlwWT}-$)9~pcSjtP8{6i3H~C;reFTF|rq9<d
z*ze&Q!80GO{~&no8Rda#x`0m=@^)G~5@hp$N=2T6>F9mgBPt&i^|c!srYBhO<`vMS
zKGpeinG*Ns`JS%Dij7>Ycdi&*1_x>h@mKY`_z+xzKD<SF2!<<sIp%@zinDFZg(e<`
zIv9RAXjS=9V}uvnfB1G+%u4l~SY0LN<*+gF!$9Mcqd$K*`EG%ZR?Y3^xVMG3plb)8
zI*8;3-ROczVi<%q^6UOLaAn`{19Wg=G^r|ecsOGLu<Bur9`gD4MK@A}fS^meVP$y$
zKr;;u_Iw5MF-y{(<k2xCM+=$BN-ZJr2P%vXkA-8wuYSHVRK<a$>{SU^27T&0u+C2G
z=WK<*R>ngEiu|f}lOpqka+=jr=u`G5ym6!B(2W;Xh3|Qb4=bny`0zkMPTs8GEx2!6
zsWFz|BHgg^GHPD>sQ!_4R&#jI_g?w0j5c-^n2pzW8QhN_xZa6E-H_3^p^F&~`-DVB
zs>%pBvX95036{3dj5OO1bW+S$VbHFE9@n}IUF*pehr?D|3A4AxpBYl!>aB_cPci;;
zAhBd3hK$lBs_%#(3i^B~(>t~66~D7W3wS2<sl5Vtw%K5HLF+8|s--#eoo~Fm!0P~B
zV-<O&u8qz?pguiR#m0FfAH%Be`8hXL;KTC*8UwKzrRqpZ5$MB2aF~UPY#{4N1b}*&
zEi`e3Bljfm2;lzu9BQp0PJ3o2eH_c-jl><vGahz8OjimnHAm$~e&FD|iwKfT2CM1A
z!jtG^P+J1@??M|Vi&mgj#9&>j7*A}xOsnu%a$O%FuxCV!m4n3B;b=Zi!VpSM=PfwW
z2Mwn4z9|-K9ARjr66D&9vQ`)Qg|%dpd7p}}_SWg0sS+Dp@IM<2(RHA!f#$tPYk#v^
zDa6DatqU-(Pl}g0P{g&qB2ciS51f*x&A4_;>1rHbQ?J>?D5abSN_AFX4^7Af5Ur37
z&{3{iS1(TicAS(+QnZLE7TQ#6Lhn3oov~(vpn&)3rDsQGDev7+B)2NS7gD#L&~^|@
zGp*!1Hxu=3(y$U!1I-Pp`3yMset<f0Pqps@cIY%Wa?X#<{aDJLkm-d+wg;8!Fzk_M
z3KD5IDY$OCLRuj`g}Rw+Wf-gUyLy{69c4?lRgxggkX*Cjq?m3-%rM~m5eR2Hbf!bI
z15%qPIqINf`(0S;<_Cye*m)}~gNBzI4Fboz^73Y<KMvkLwGWYxLX&SsWs-Fe$Vih2
zN08A+{530u_(K5~kiSW-30-$RSTzY#ClAR@%S3bv6uf)Amc;F&_epVu15&yqOuR1#
zC-#8xIIx;?)C>*h><8ntZ&m2HWnI_VnR};JGeSk?%_|93><=7O0KUHUl>?GlId^JT
zUk)xVLY`@Eorr2UV&IYkPu=2K2i&XkAW<r6_KTXv_eB;R6!M@k36XK_jy=yWDLTer
z5E5j?;H^pfXGMr-DVe``NaTNV&itya&lnh~?aK7zbTJ~8SN4&n!w5>k;20(kcxaFd
z7W4Y-EdIwz4fU{GlI8+tvZ0~50`HDM24SP_{BDNgWypUimr#n>XUiT@3I%Uc{$N4V
zbugTu8cd>JLwik4N}#Yef4c?5KS4?9Z7G3CuFYuxwrqM|i7C<r$mTvJ)W8sDj`<ow
zom{<~eDYcRDb)HLY}q&hzLm-dP{q<mU(CJ>Epwv+aO6}y>Y}bP1zA5d-0;o{_*F&*
zU&Un(zprMN;P<8hmZKTI0%DbAN<&-eZrmV}=8q}v?i0wK6_ec60Fs@?4-8)jV2as1
zU%^4!rSR|hC-<)?0e3<dbOQ8_`Q+IH$z{wf@vP32&`u6WSS1b%gD)7jYJ!<}U2`^S
z^y?|61XnrX=3oGh3NsN)$eZU7p?Iopq{HAskst`v$!Hk@QdT1YfA0a-mDFK=CyW>q
zCpWWpu${?C+CtTD)78ucD?@cF@)hYu>qZP9-4{1DV&0f-Wg6Ydme$^*q;}%!$@SYy
zIWNiVeI&Bv{V@oHwZ7(=$EN7i6=xoC9$>PIH(hxr8y1uK)xN4ZMbB66l;`uu^D;N|
zmnbJzY=MGBX`mPS<Ms$Yn6-ysOPERfbrrt6#@M04{dhU4Vs&?PuKboK-uzEbWi0OD
zlmCMV(V3l=*5=1nEqw*;Lf<au0T07j>L8pWyf7y{HnX)6S~l)Bp#HGf#1a~t7&&(C
zLW9#8?BT2MA7W4Myaj~09<?4Ch710G4Yn)gJNo*=o{iWo5T2MANRZ9zu48KMj0c??
zrf2nq3dmRG)J8fid!OY({{2C0q(E|fWnFjd5`gD#79nMuG>o-0EQsHl2JIY8OO<t5
zvDOd{T^RiM5#+0r|IyqS7+mwQt8`bL@wL03=JVcH4`5QkeUgzIQCn!WFHMiuW_Gy0
zHBB^v-Or#l{IQ-YVZ!rLw)T^J5O=G0=K+_(GPL%M)Gc3kPQ$Xn&Eo)m?!EbRo%Zf9
z|Bhcic|_HNPp^?RI*?t(g5XT(yf;n|9*s-xI(Sy9S@Hb=OkK+TJ7-T`zMyS%u<6l#
zuHJwW;P1FS8woiC;T#!_7VDY8n+>#rmPXC*qjK*SeyIcFj^9=?I*N*8C+&rT07Y(D
ze<#yoOt&Eib$B&i024ekFP(11@)D#aQy#H>p88$~(cewgmrjdP>gKsM`D%u3XeM`4
z=GXfdbq>C^T*qoFIw7JL%pLI5d3AQ+QhN9Cf|7naeawPX!--pZ5C2U(x=%Ve=(zZc
ztR?+>*$m2uFR=X|PVM!`SGc#=$}M2SF)j#rKEP`2rJ!!tP)fiE*>y(Lv$>msOB{~Q
z0hBO?=0Vhpw=v~L(5{zNgLGi%yF?%fhA86C_9kTSYs`Z@ih?k(<8*5%PJn>z;rsUL
zR$}c^u_7H4m@8r-HP$yIMWXJoS52WP>kQ7BBFy|A%v3)CBu0mUt%7?FVE7mPG4R1!
zC;KN6t3g&wV^pBu>ncjyyJ?fW>lQq%?9ar9`(jPOz!XZj5^8iGtY>wBG)SIKCmXu`
zI~>;hKg~<n(|r{*=$sf5Tom-|cOc{u4!gV?Kj+>o<*u=;BvCoUGG$CEBB6hl+FajS
z75lC<-CpCs+cf^(v%=mql|81Y*WG-Amtgf1{@h-3+sc>e{pr2#BwC#I^6Rr-AgC+C
ze~&WXgD~&x;;ZBIfk#%LsU?a5&De%KcX5Y&@QxggO8$`;(b{#&A?rQ+^&vRU1`b<r
z@@Eq(*k6KIOJU_#J<@{jsnIu*Gl6Ayd<ye1WS*V`iltqO*J8qW!b_0Q0bQPlHNgGv
z#5cN`jQG3O?QY+GBn8Yx*K&cB@YV!$u+pJffk3eEL!DJtNPE&pJG6W;!W>+PR}_jb
zXhS71AWQbr&U4o0V=|I7l0#{iPx#z7`g9n)Hv$8zuP6|G3%RMappK<wm$o@HejLtx
zyM<K9nTNNbMYIQN(xf9^L}qsWUG*i0_-F_lv{P>nuab~!XHQQ<vke=tV(;M|aA@dK
zPF77V!V`ZJw&?$gTubTCwQ;%zr~*o+@cuW<3QGI!QBB>LXryCJFnc{A(e4CrnNSPb
z>N<q)T5As@B_6IHcIO{Hir5%quyvghDXDINM*&+d=Pdj{S#VHLR*gF$zrY#y8Gq(?
zC9)e?28i%b1r57pSIy@AnbHR1?gi8bohhXd2{2{S3ARTNkQ0!xvr>Miv#i7J&LkEz
zvw;e6KQx=1OgwqriTw(co>QOGvVRq1YXd4H=vrCVSck-qa>GWWGZe;SZNIqg|1LT|
z(vt=eijou7$7eDlx3!ibt=k95+0^%g=J#L`%8MZslt)9L_R=HR(%SrX2hB&eaGAFy
zkbgvR9QRVh^lP=CkOF+SGwT-yXXQ}*4YuaPGSv3=5&C()yR=ga5ipc|(NPC_r^7J#
zGurE@EJMjyy4=c7>XTyjoXUVme;2|BAq2rEo1N8Y3zXCIv4DTgOFDobo9uw9|2it2
zSvftxNkbIIm}M~vLTT1xa2mN#UlnAGLB@|j)#C(F(+^&x81UY!7cW#PYv+3!&0F!0
zL%5=}Q$bt3r=!z)gbg)wyWN3Sp>OKist{Nl65$Rx?G6NY<bGp&Ue!acfkotRIZW3(
zJ3z52)SW$DnG_c<`be4BBVRaxg$5=miq={;Scy915p;0pJFf{7XKEt()5@MsTB9;1
za1ASX&)rO<yaG5bw=Pf8?eZHxKKmZ-IK4;yKT6L4>A}XVBzu%Ax&AhofWp`nl;wi|
zC+WRxhxl+OOamW`zkz00abKtg)baZyb#v$kKqIeG|44mQuPMwiiJeB{TFs$27MRfA
zVZlpK9ec_w213qzEQWnB(YyTp2rI@MAHnx$CVtg|<^&e?^N>MJEte=aB7LH4!22}8
zP8wr(lWyhxOVQah^U&{T%tq^<apwzVbli0KcsGM_&OOo!<*_uKx+C9C1?1Do+YLxH
z?H!4%EyS0&IU5g^w7tt@*>(7FHyZ8*y3(jHj?YVb;;94YMNIDlwaDkS;IDaP(LSd%
zg7Gh_ruTy$#MT1NAv&8R=HqN`y~KFVyjrv^G{6+5#+$mmImbHhr5pxtAACFJXc=&e
zv)jJLWJA59<~=GYEB9kk3~eM<nT*Z`Uog%&1ubfnc_dzp`fs3_Ynrb>#MY3=T9w!7
zk}Xtkjbek!7taZJKAx9WTU4YR-yqFbiI466j0{m`#B8fRnENXH_I)b#*6UyZuA8&K
zOcZZ|+qb7y!*03Vv$weKm9aJ+n-|Tyg9p<syO_RvZkm^u!#yHuE1-I7w5qdz!I!3O
zG3>t`az;f*rr0Am3*agEf+hUa(TvaJPlq%ql)=vFS19TqMGV{JyD7O#XNUTw6L65t
zE5fI`;e}?<A|^K|g!Vv_>%7Qoe%=XGD+9$M%*)bab5izAcPw-+PC*vxaT4As-Ztxa
zBFK-OoRSe3<5q;ZGN93+*!?8QwYL1;{k@VvhZ?hDM4@gw`4Bj^J~qe7RCNML2t;Bk
zK$pp7GdBvR=Do~o^$e4@ygTUJR;tA<ipi9Cl$3%<<5%%khwY&FSPO<tyJ2;?<k>+c
zh&4}?^Aoh54GN|<yx?5|MWavn?KjO23G~(_N(b)=Bh2Xpki$(db4FQ#{`o|qO&<)G
z%;pf0$vWjYue8xF)JS=D<nXP!5!Gdi?5NX&Ig!gLc~HP0$;rK+237J)nPx-#z^ZoQ
zrz2vaFTm#pPp~lwy$^ezVgIa>JX3%`VivM-i_Q3~x2T{rn&)$(lDDwb^g?>aIbU`A
z3}iOtC(TfzYX)g#i1Lz_UIpw9zzuLbE$<BomE{3Vt(q*z$H{;@atIOcK>^N{q(~Bc
zncEFDZfFc4Wa${JpQRf_`8i;gR2JDV#jl|B2yh(r<>3T4o_D%Bx-^U>{2Vd0x26HC
zz!;O``eFeEqCA72!O|+@8}LWuy_{3r&xhtc+&TA7PXi;_QEA8B29?cI163USP3_Q{
z-GgOueD1lcDqzPxD9dh(uH4AZLSLR|G2K3@e@C@pI0+u^wtcQGL=BQHbF+xE8ImRZ
zp|=vE+EDds38~khB*&8xYA9sqR|9*eIx<}MMa*M})fIf}oVFBOx+KW8$mf7_fn#yb
zu7SzM7iuB8h4@6oDRHFaf|dz#O^RLvN=>n&$0GfEluliVNXl>Lj}UI*YlrqGzUURm
zUG9xbe?HMTVAOkpx!4m?IW)vWOXyjlfr3kbGx#EZ5YmqX>>4>T-Yz5;KMuc1T4)?#
z#Y<b-&n{MVb<mCz;!~qfrd}|S!)SyYMuhXU)-t7!*u9-UtL=cP!*uz+EYoxRwCNSd
z{G`)p0Ev@4ZL~uCxBz6q^sFE=?|4L>8Zz^GR!oC#dW%p6D@n0seLY!uOWFuKdc6|!
zW)MmHVV}Zf!;F5?=rx$otuhkPn@h+e1H=`rDIP6lwlK*r<^yir&}c6I(@`iFymWPu
zX;>i{6hG1_U?}3tHK`TVmSWPRdJkLLpr{F;5=u=eYeZxfjgd06(c>^hWI*X;-2^SQ
z8CaXtNjTKX5e3svNid4nr_EvUO8#tbFbjMPCz4vIKK)BBKK~Hr(lz4$&XhShwuSn4
zXG5Xp5K#>blyfC0!;HubzPNKIugw1W4x>WCT&P+t`!i)}wRJ;-UWz*~gdB>dR%Jvp
zg`;EGkGLhT!t@e(YE{ozg4kOfJ{;<XDIrAbK(VW#+XFsm(lE3awx*0j%kE9@FJ(Ow
za71c#&$U^!wY!<w!f`ccG3?KHHUZiH)bf7DLW2*3;1p7sUbXJH7IVEO(KXo<!h0xV
zCNEjkoF)4tD6E<tw*wDzUbEXAMrL!S?Y(e1r(~xAt{V9ERH?(zoE9^s6c26ZH%E%H
zdZQu(6Wv>Y6d<|<bpsE?<?~|3f4wiPz0=TC6e#avQg>27G@7&&T2fX<6=jgA`m!aU
zM0l}}Bw7Z@Dt4Rk_2{K9Yk&LvI&QYsxGOUt49ZDaa~xT(6;#c=HObNqHK6v|xkq>*
z_%7t4_U@Umh?^xZ^Y=J0*;z&)a-mR_1(fO8BE5_E9ub@hE6CPpspK1=8VN*UonFJm
zJcss5F^ho#4F5HDY6Iz-d<_^Lpc1?*o^zMD>9Y?MzZ3j!wla<-SO7!$Kyv2ZpM4O7
zOyE*LFciKt!-9P?#|j_?!n0N(P4_@js0~uQDcZ3>VId#<5LiL3_apB~89M&Mzu6(D
zRZb@8OaM@eFg;RJr=$`k<0(m`hQy!zD%gxn`37w3kP>UaRj5=*l$#O6RRdc?6Ekqc
zd~lDlYoy!pLEo?!LVyX4D9V(JkRpxAB_uyir<)_XgD`dPaMTb0S+Jg5kB^$EW-4--
zgL1`b`Jj)QpWpCQO$5tKV_>j`ONpPp_C9_LB13i+Z>s*Lc=axR>EwsusA*pA+;S@P
zf>X`XE4_v7&*xVI=vJ&QBnXnry@4Ji4W?*C>(ljuqn$jAtd~<q;|LHE#ZEywH@W<i
z-*Z>1QAAj~wEr0SIis$R9q0HRvJ5tSiy<(^U9mPTm35B9A@e%P$(8-U!th>Z=j32Y
z1|+UtxxXo2WU3Cin0;^RW7U;3Jv11Lb9K~$;1r^UuSkOsCtq{#oSXyCQ2?lz%dCLK
zi8JR~U6U$Wq1w#@AhITmYaynUKq_@VdQ$j_=sR_7{`2M*n;VvY!<{&uL!hG?60x-Z
zc-t~RvNO{X{tJ77fu$~zVDZct#Bps!w5;#Wrk%A12j??$$PN;BfpOkwb;y_+$-g$b
zJA*$M(hJm~3h6y@i!kjh70s%1(QB#x4CmmwIRPZL1z;wCwk$xumSYlIp`O0p!F=$T
z7@6X5U6)wNDufQ{@;2R0xG{;D3%ovG3~WJ4=BNcNkNXVERt#b;R~v)IJp}LH3EIG0
z-qRYKwqq7*CoQxGx6Wz&oLvsUFJmU<0J5T+GrGST0~b;_N$%6286@=Y0Mc5XTF4jJ
zc|ERkL<#^*f@j1Mh!l?dwdL5P#Y62^J_8(M%KkGXh=odS04DnFV)OBC<U2BxIR*&1
z#<kI0sUpD$CzUhH``O6P1qXp|k~BPgf6_?#M<~(J)A8|31vp!CI^XeDCpzwR&Duc(
zmWu}dM4q+=K0E*3!U&T!d^4`P@c^R%);fEiV$;}k>^C(StRU=72*awfQ}TfT*=w_P
zMh$x+bJ(k__`6-X!i3jJ`S0RJo_vL?47nD={~^}`;6=+Zw`$ry!&g8KoZJo=DSMIl
z`+nFd)15E8zGA9Za2>*EJ2>Ss-Lj(*mBP#VBk4;IzURZ#gm}ee7%AUfmrqi2%8tS2
zyLs@0`Hou_sgS5I(Y*zj63&yHrr$0`k8fD$d*EWpD~{b|?cHtoJ#T$AUQYm=9zu4X
z*<7GFk+)KPMOE?bRBoNL0GrQy*$OItv(l3ZrGSSGYiz)@?#RSWPKN@?p++#)IMaAJ
zr%qCjEYz2<9W4N2yM4yCEggISWr+?#^lf7;NlY47DK8fpoL-Mj>!kdkCVfvgpNx=U
zpZ<F%9W#348+DF^1<SKJICFUHBu5Xfv;xE;S}okB$Uypyu${Hh%NFY(SF6u#<nDoO
z+REw6gHY7H7wZ>?GXM6#u%oj4D&RB7w(@`)nF5m@q|Qj%*Dq$>n<}rWwTwHEcH;zy
zv%_~dva5)_zk|ejJ;T5Xs*^=4qjw^$0O&ZSLr0%><NVU}2n+<@rQp3n>{K1J8Zz{s
zHF1E5`T8Ht{Eds{H;A?8GWHp=z6TMe>v%6k?n4=<SjoQ$^;y#{z$r~|UxsAEl1CO4
zo75rO8M}H&!p$`b{h2h(uAVCxk*Op}7B=!SLM%QKoYJEDV(?93GSLJZy$+}(H!>?j
zx}YiXITz5hUx^NYeC@;79@0@!vX8}yV0=^C44=N$t>=PJEv-0YfW9%ZLj(w)<7Uk{
z9m3h<Xh5@?j&*^z!eGv!*M$HGN)AB#rl`4Z@(y6@8jXPvoBV1W<u!(xPMFoLbJ5)N
z$~(X}H%_AY5e`?ag)%xSYVIcvQe6gOk?0PidCBxC>s-^!PXZn~)r5q)du`ddb3nS@
zsn$k*#BCzTc7Vd$U(VG!I2dtupQR((vQ;9b14g-@$0mY<eMZ1rYTmt#uc9dE?W76|
zSd--O5pC0(-F`4m*F!Gt-bANUM{#spMrragQ|Dp}2Eoe(D5Kf}w*Esq6l*8eg-%{P
zndpp#DU$Uu2j@UP+rLy-`WPVhsUE<A@!#Qm8D?c@z3aK|=qy*kwE#r81>i19>@4$<
zBWPaM)V=_6G&%?v6eW^?Jk}fBu5u81W*`Gf{4fD?cf7%eJ`aZ&5tdFfY9q4p=c)kw
zMIZ?AEHzATe@DU|!7^dbnLNO#OQafh=mH2p#rV?!fEIJ#ng`faAFc26A;#RrZwGat
zy+b<{MJ-SEAJgd1(&RnOjF>8`U5LoDP{jp-mjGg>6_P|{`k7NAH|eOX{|!t4=ssv$
z0Fcvx3@BI@%01>oQxw)dJdigc4Y~lO1+k=OGf+9{cRNw(301p;i+&3br+u=zpXwb?
z0VZ899aXSJ{JRfdf&3AR2yY?2G+ktXg&z<Y(($&f+k^AggEgHnW$IT06O0jwl57qj
zz|g243V;<NI^FI9=Vz)=S3Mx)Qb|($Y(<<g4oHx#K)*kW0M=4w>j2<sQR}o{Hn`QL
zs1<M+pbRIb&XB+$l{aVxL#YL6)f1A-MUZZ8FIRYfO$t&4#7AGmHUaomWaYav!rT`d
z>(#%bZ?y>jqWA!%7VqVF{DtC+8UEo5jDEh=NOzwLvd&j^z+_x@mjuQ53IFrWzQYX9
z!3{{xdu=w+_i#f9mzJGK>WfA8;Hu+V(+*82)T9F6w+wiXxAxm0b}7>scr8EO09{gc
zA>e5?#?(~IYKl=j1nMbms_Zy+=xQhJl1-B0oBjVHkCt)(?^hE#G4t;o^mM&g*Mmvc
zk2PRmWa9By+5j`wTfiU}DB?TB6GR0>iX9vq?)eIgsw&7(upBKwjAg*_AYqkO1iTb)
zAy-fw2Y&w#Vve)td1c<Ey1#u&E&|z4$%e$ap`J_tPXSi(3qS-`o_>FJkqOP~i-PtD
z?Nn`}N7WDe<ms7u(Lf?_drKkE))|e1IUoq1eR81y4bwuTWwL6{RN4v7*H5KQ4uwv_
zpwW^mLK+n2rpo^B2$;<GTBl!fKIo6gE-fcDniw)M)X93X45DVPr|5du$EmgrOC(RP
zB8Cb?atH~?v|-&x?8loPe%lNO{wr4kaA3rb3g}6tEVIBUh4?^k+Pe~vf6|1ayu=B_
zRp}4RlP0iL2N#!4h)tZ$k!LW-6kUX2ttXy{XP#>>*UR5YXzTeBAo$IIaE}QlfSAUl
zSx<5~YW_Kn1D?>l6X&4qThj}JKJLf>83N?&bD)qwvQ3M07Pd-_o)%R8IB}aF&jv^N
zJDgeel7i&af+A&AN~p*!RJ!QhSKWP@Qv7IZ#M2H4PFUK~?(#g*#drgNO>!&0(wSwZ
zz0*k#Nskm77_!0WAQ8HA$@zC~`hC9I0<;KMBi|JKvE8Td@Qwij*8e3goE8e_sC76t
zcqUG`2!L&N$QRok5R5BD+XRlE0d*RXn+a=$gm_P|2D@j7RBD(;=O(=p__m~!0L@+U
z2a!F!Xn?waAOV>=J=T>cqQ4503a*>e0H`9N-}&CmXiMGu==mb>wP{s8TXvzOV5RdO
zVQnvbU!#LTN$;xf0dbg-T%vdU7%#y@%-SWWf#?fBFUg!L=sNj;{PxQbRV0|3pJNvC
zKbsV{3Mm4f!Z~Abx^g#C>w)ecVCL%0k#Dz4Aa|}Z@DMoEnIpdW-=K^Gm>hM{HIb}`
zGp%_3Bl&uOwYNjKqbVp#6~FSE00^8MPpwdoe5WEKZQY0VU64GwI^lS(dlpJGrUL#V
zG817?2v#)5G!MXFxrx#M0BokU>~>19`&>!H6#S6GVQt0dAqmwXfiFeZPC?>X+l6ba
zvutKVPoOy@aRX4y;+c72bO|Qoe0)EG<@B8zz||?X5nUpWUj?atRpAQ2(D2D2sV{rW
zsA!1k6r}!Dtz+2p-#{{*WYtWC3IAE1+lBR!?K6H~WH3R-5;4T!DWt9kj$Y1>Bq+K1
zK`u_Eh2OrreKl`x99!P~{ceo!$F5@~*Pl^VJK&6sz5C=9PjSip>iew}#YO2JVaQdJ
zW9=r!mL+z3eEKpAqvH`#@qBLGYn_aj!aYgxE+73Wt8_lZ@hx8R+iD-<$|E)>lHd)}
z=Clc9buoSc+@2{-*K58kn>W*UXU(CVstgzsl?*Gv=T}p>6BAi;zcBQEZ%?uAYC{t^
zW+_MYR9nFnoygP^nSNbW-(PWL6wue^O-1&whYWAn(r8NtVQ^ej7JDks>ci0prvtew
z{TxlIzH7Dj;#MSZ&#nq|JMN^^oH%L+7(nx}Wq{ZQfsls8Wdp4*Bj}Il3U%tV1?$h8
zCH1Plmj+YL?auZzP;X!VNS5nqP+M)EJr9^aHD3{%RVZHwdpTSkMJU~<XbyH=bpQ}v
z_*8OsmD5rx^_g*=BHaU@U51`IV9L$S8ycxS&?r7e98Voh7QdZ_S!1b>d}&Nd>CFvZ
zozRjlTqz7_-ambPDSFb};7RWS0F~?Iy2?%WJHO>_vtDga6_EJSs%FMs8J03%1E_Ch
z-aQ9YY3oQi)cir2Ce{p{w~ikO(!aGd)-p6dUu!t-9a=@eNq(NC4Uh#_=2>@>VTA?-
z1sQ%DfZ&VoanNfRAfI_t#J9N0>TJ+m`|r-Q$z1wV;g~6=mSdcZm{@qkbfS(JJAV5k
z*dG&<&&#wu+3z3!_3Pr`2bd`YCb}LMn%@SVTI9)Fywt%EDdVtIC!{A<6I#^DxzbC!
z-Yjo}0Opy(pSa@Fd!Cdtj039k2pI4SfDg(>!voAUOC0x1`bOxr?Ir6=O;oiWdTZVU
zr0zqPIM|jn7q2_0kW-U+!~_0maEYhW^3Y>~<azKNGOc?~Lbls(3cQg${T!{}Se)rd
z0pnmR+nAXgx)xyd?X~%0rj#o|vEFL$Bt$wdU^;vDm6>%I-Nm@ZJO?QPg6TW)d83IH
zey1XvXr3+cKq;Paj-@jUrtQdRWHHGE3hIOZQWtb;Og)P!9$;XZ{JofS7O66J!&l8U
zsVtgrj_t28=|pw0xVX`Ul2x|UW(Guluf&{GLsdOUq2uKMjfHFt3SLodR;)f5+BKnP
ze96svV9s&uHAL;5x<a+uLzZ_`clfeW*s?WaV4K9ZXT{p@o*RtX_8#myIwJrmsFx65
zllmk>uA$tH)Y<16St7zQJ+z8`oD=Bqezv&D-o|2SH7EIv&0!p|_W@+<Q@(k(@47r%
zo}9iimckj^(pXg`!!vC<mosS^w5&1(6X;@(^L~b6XAl)AGV}4t9R2b3j4y)$6?pZX
zX(=VIV=VBu6$?iUr+xTtj}0!Zf4_TZ8?pH%8#f+c19fg)GS0erlNXcia$6S>yt`ZM
zb&#i{jDxbPN~<LG%6Au$-H=|BE9}XeONoETt$n(1`YcoJr`E3*)R(T*L=jK;#-_G1
zBdRR#X$bB2>=Bx|%zl?Lfj*qT$W&*jrz`<7Ln=yP&H%N4WOh?+d<J%w8v2LbRFvlE
zR4*Q^&EI_!D*G*cA*<kiKY~zS2(|0ct5OMjsbx)n-Q&QEW_x~33R=}4iwfyA3o(5K
z*0<WYUCLSH0hFPId>aRY-#H9%TyTK^IJV}nL1Y*7AAE(jYXeJFDi5<A){<nVP5n|k
zX&*p|g3Z?+`1rY5@0wa7>Bl=qYiAbstiid}CfQHTt^<_8nQZmMwk{Atqk7qMnlCe3
zIsC|KKP`NPgLAc_AGQ8`g;^b6`nxoJRX>7<?{%T=8ZI1RJxerxhV&jF3?HbkP5Ba5
zY02#Wvg1qO^UxK6(FC#U877*k&8c%)0VNumL-83d3o+LQvh;WM*par852S<+XahQ#
z0KAL}wS9)*IUk1@1vV+zx60tTRT5`;YEr{0Dh7l*J=?Wf`>EcWZxI$?Ne*yH%{4bD
zo>mD%i$}9(bPr;F1fJTf?6KAfeBXj}I~tnotu2-oHw4h0RC?#NhMapv%PDiy(zCgx
zxe=$+n1k37b<mhE-tqL_M5&BbdwYBSfS{&WWwO)`i?)TGvV76JG80hRya_JbO=u<Z
zwnCFnG6Wl(p?#8fWmHQfn#P)1B7eT<rlbFnKS<u;1Hk(izB97r4@wP5XFo}TXk-@W
z1qTY=@VwSQ7Nmi`+H@_eu^WUF#pTSkeh-+ijZ9cbPZ)!oA)fK%JppAE!PO5QcC0ZF
z#5Gqu!8}{^X~1}|wZ+q?<Z#r-^{*7JreiEYGW&A?w9?x1g5fdINx=m;PLP<&_qHKY
zXWk-=9WGQRl64Ho8f1R*?U<W@a(^tVrJ<7XL8e}Y67uGr*II!(z3++bXToS7f-2`)
z8CJ+ESDgn@;koF~%jqjG)J#PKt;xFq*RE~}y%f=3w1;N1x>gM}Cs)Ca237!T@)R-I
zt-J4Iwbd4&d6PC5ENGg_CzOZjxnI{h1|`FHtViQ6P0aD)+<p91?`i77=-WI)8Vyk|
z`BaMm$Iun^#d3u+yY;Jw8pi<XHV<Jm5s=ks&L+HVg2HstzHn4Wuju!1+67}wy)IPH
z%uHplaoyFTsYuOgSp$(JWrrsBJBp8l$X2TWu=3IN&8@X41Iaugfz>6o)-O5xx-N(;
zmdSj31f1lP^pz!rcJA(EpXiQp-nD=-%)s#dGn!Q;tM+yq@)4s*(xa6koyJ?z0o6}>
zoEINbjvQ}aRXunuEagf+!ym)|>r|HL(mDHszuUB5$_f1;L@c>(dVDR$pwK}``;1Ps
zq{A$a6q$2Fix4PWp^-Wmm^nNg?><1t(_U>X5C^8~QNMhJNBq$J^-O-CeO<}ayQRdU
zoLQYQ-fA{!`C@ICDT^<Y!$KkL3$udTsuV4!n#LoY2wnh}HX!z&_zXAKueQOs)t6AM
zO1pMX;59%B{lW{VloDO<=v?+ZKj}hatv&9?ogh)C%i~)DApV!!*QMfOQ*J2>b%NBq
zFypj*OAE02YvX<a7Ohb_>#ag9+G?P_aa);(As4L|7toN)3Q<C=iP9H;oS<Vd9ID)J
z!t?mMbUD3k$2v*QE^@%Up44I1^7b!+@@yQ08ZW|63-2;`f9r}JHJtLPwyw2QlObp!
z_0PJKKJZRSog%PJS6SK04l@5Qpon^CF&|ghbNZbh;k+Ntf%8zqZb*0^YQq$fl6WAe
z%Z!V5zBj4PW8!T`I}5Obn|7?e!LifrKXswUMsCcGrYlRSWeFg$X7<uMxf-k*ml|mt
z2E}O4vu6&5_Kl@=89r1EbQ6u$95X%=t@@ic^6Sk7i(`g}8#G+|)BmFj-;%yl^}K4M
zGF<u0sJFMLHocTj&u2};wBj`Gsk75GHlrU@RvS`*K07J8hX>fs6+QjetZi>Dayow6
zdyYyr{V4nDfu;77#y6qe)Kpow3a|qHMJo;VMNP2FBW$Y5U-px!Rr#G{X}R;pz@-0n
zh1)WC`KbT+RUd3#j9P2$NW^&8hZ35$D>(tTtpT0S(pz8JbBK0Pa%wtgUC;`m7ndg<
zPp>A)i>yt0xKth()0g%@ZfV!bHy+cmFR4N}Mhj&!_Aiw2XeQe913~p>y;G)+{A5MX
zPWh#A^0AC91DuC@t<2Lf$Kbria#DZMr#wzO##HEAWaNA&m73e^F{3cr78loH6WXy7
zd+OaAD^}%mR0uQ3srAzQ$;3|1@8Y(d@7vOHXJb{?(wp-d43b6CV&@LMxmy!@whVgr
zUY{tVyffcxldq7s)X;Y?Z&@`F91wj@^W~0qfwL1l0iR1&7e>Vg?QJ;5zNtkK#+-oi
zen>EzIgjS0kB~_SC~~~kVzc6PRXVDyVI_ff!zG0y9y!(W&6#%PM&(I8s`*|mfAV)M
z!Xv4bSh|<=JOo%8<o=|)^S_ajlN|J1``RV(OB}ZOBOi|-{ArC}V|U)T-_!#DpYwzh
z6rEX$F1rc(<|gE&bnOd{2p${}9t?K=2m?d>S!Tm%SG6T~3O+EaxyPRK(%EdU!F6D|
zv;;=Xk5DE%lc~)fo+nL9B=|i)0SXtVSh>F0L5k;0rX?r1j_mZRuiB2^9eWJv5&7Zs
zWQ%ea-!xpcF_`1+%re%mRWBq$b-wF>42OOMlV-pJFt68&619vPjCnj-OR4LIqRX{n
zW2k!Ege)&TA@4q@!;=EJwTCB8f9Zh(9lp?I-*XatWm8GAsDVl{YJEeG=1#;F9!_va
zZ0TtmJ)uAG95^-!Fg5KFH}C1VvJVcZ(YgQivsEw`D&s2=yslMuD2?c0h?{VD0XegK
zfL9h!j+P5iRkT>N8L(7cF>v5NmR56e(#iqSCCt;U^2!EsR*k(Uy2={Ig=!4fEL!V_
z1m%^_B|n~5T=8W$L5TQR!E=*VPfS%UY@m0BWnPi0(2Uc#ZjsAI)xv>WB(23FHHmg$
z9#Wi-xbZIZpR7EmlkfRNmXi1ADD+|+XtEfeHXIMQG;3cMUnNb|W^aweNNQ*1wN8;<
zn>#D4dk5y^(;!vm0`}){krloDsi*VSOv)m{A|0QrjO|r{Gu;v<BD35gsqlkoY5gZe
zSEz_f${GV4Cmwe!X8b8`ag*QSL?tm&s!jH1^B~%Ve?9nWCw;(4J^6G1o0(9%K2?W?
zdu9CZp3r8uIU-`)wGi0M;HCcusSUo3!YKWGZr{yc+In$tHYVwym+2_-iV;&l)5$eJ
zvlAl7vuY;RW=)sO+-;)~ePwFzb`cx6<|g=feV5}VYn3*CX{#Uaw2uImmRn#0Y4?w(
zajEwM#NStMg)DF3lM$p_T~GkyJHBO!V|cJccc1L7PyOqXU_rD*t(au5e_3K5TB06?
z0sw%a?}7%|KD`A<L4Jf11E~dc^r>X--2BkVEU?lnjVY2ptrQPd>TX<ne)E@pTd6QH
zczS})3&1(-0ta2N-MEB0L!yLpr73Ui01>#%zunV}0q)798%O^Kw<1g(m=J12B$On7
zt^FT=R5PK4YSF>@>oF=j>|1|wq208KFmG)(ex&yB<_8QW+0jKJAZtqa2QegHg>7|P
z<E%aNZ%f<@OQhOq1u&C`FlYAxqm&6Dn&(gm-pfa5hBO@_+8o|jN5R<$D9>>J!<%w5
zL}z1uL7Vyy4+U>Slop*0_X0G8xljb)A1b$)F1{irWxY-enR>qq_crgIlOdL^%&0_F
z${%Fs#>_6fzT?&<5*3>tbMr62gDhPH;u@3PgFE*K5<`NW1aR5ZFtp%c#KqV9q<OzS
zNN6+6@($anB=z^(5Y>VEvVOU&bMy}j(Y^rhGBy47pCSqLyww?irU+=uc(Hl6R*!;^
zh`7HO{?i#G#xgM?+x(*ceySh8Gx!{SR_MO`mKT2>$(|I>`4fK#yG7rCG$i965%|M;
z&@pk6#%8S^x>jkvd9w#)K+Kj+Rc4=TUd5&>0BS(9{UQ5ttLj5L2+XXpVktL^p_e!W
zVlQ@k-PyPOSDj8_W0`1^T(sEyrQbItVAu8Fx+wpV&9A;X2s4*yZXkX5z#kS01U5py
z$;amJiw2*=EM%)5l09)vY|F+U-YfYcoRO{J;-8*~js++p<O_xWaNms$-n$=W17jH<
zxU}h~orK}TkUh9^zFVH-*A+MDu&_+XU2ir@{prr|=4$BZJPWnSu?;qFr>GTt^Jvv|
zonwD^rcyXOo(%~yzwY(VKW<>biy%L(a(1&K4c-GE4i=tnIK2HMk>i0Hz|nYRzD4pk
z$qc;78{D9Qk5fvVKDc?=mucXe<tOBo&j0gKHa<xNF}{6rbo0~vekJ*7q=bkSYBt|+
zPz*j?GP-R3czZF%IRod}iTq9g<DZ3toDgZ$N3zv8o0si%4!$|%=I!$*QI3K7n<C-z
z=BL^GmeBJ^T!!D+TO`N5`9&?E4~40&scs*pIyn)8Fws3yWctIA-+rN+cYE_voz&oc
zR1;{L|9;B;?YaN-V>>W+(t|*X?oN{a!>(*z=%0V(*S&LKtsyYn{r8RgAHMqQH~#+P
ze|g#ehdWurTE#fL8czQ%iT{V+_<JF{AAr2<<owLP$sGOPUgeg(ok4=Fb~M`g|HE(m
zEfW9B$^Jj=!r((#YtChP+%1p%_wD{~8+e=;&gbr;<0Fj!zx-S^0jLb?kQk!Pr~Uu!
zvx)vMYW)B3*uiA5*8gW;oFZirsTz@lNOt4_2T)-waCpsRsu<^>53G1zv~IGZz)6VY
sI$F??kS<U&-)KQcZVIG&L3iLk;{kE4Bad%X9A^LmPgg&ebxsLQ0D<l)hyVZp

diff --git a/latest/search/search_index.json b/latest/search/search_index.json
index 0dfcf54c..8e7445b6 100644
--- a/latest/search/search_index.json
+++ b/latest/search/search_index.json
@@ -1 +1 @@
-{"config":{"lang":["en"],"min_search_length":3,"prebuild_index":false,"separator":"[\\s\\-]+"},"docs":[{"location":"","text":"Welcome to RAPIDS documentation \u00b6 Warning The functionality outlined in these docs is implemented in the branch time_segments which we will merge to master soon as release 0.1 . The previous (first) release of RAPIDS along with the old docs will be labeled beta . If you landed on this page feel free to look around, just have in mind that we are polishing the last rough patches before we advertise 0.1 (Nov 16, 2020) Reproducible Analysis Pipeline for Data Streams (RAPIDS) allows you to process smartphone and wearable data to extract behavioral features (a.k.a. digital biomarkers/phenotypes). RAPIDS is open source, documented, modular, tested, and reproducible. At the moment we support smartphone data collected with AWARE and wearable data from Fitbit devices. Questions or feedback can be posted on #rapids in AWARE Framework's slack . Bugs should be reported on Github . Join our discussions on our algorithms and assumptions for feature processing . Ready to start? Go to Installation , then to Configuration , and then to Execution How does it work? \u00b6 RAPIDS is formed by R and Python scripts orchestrated by Snakemake . We suggest you read Snakemake\u2019s docs but in short: every link in the analysis chain is atomic and has files as input and output. Behavioral features are processed per sensor and per participant. What are the benefits of using RAPIDS? \u00b6 Consistent analysis . Every participant sensor dataset is analyzed in the exact same way and isolated from each other. Efficient analysis . Every analysis step is executed only once. Whenever your data or configuration changes only the affected files are updated. Parallel execution . Thanks to Snakemake, your analysis can be executed over multiple cores without changing your code. Extensible code . You can easily add your own behavioral features in R or Python and keep authorship and citations. Timezone aware . Your data is adjusted to the specified timezone (multiple timezones suport coming soon ). Flexible time segments . You can extract behavioral features on time windows of any length (e.g. 5 minutes, 3 hours, 2 days), on every day or particular days (e.g. weekends, Mondays, the 1 st of each month, etc.) or around events of interest (e.g. surveys or clinical relapses). Tested code . We are constantly adding tests to make sure our behavioral features are correct. Reproducible code . You can be sure your code will run in other computers as intended thanks to R and Python virtual environments. You can share your analysis code along your publications without any overhead. Private . All your data is processed locally. How is it organized? \u00b6 In broad terms the config.yaml , .env file , participants files , time segment files are the only ones that you will have to modify. All data is stored in data/ and all scripts are stored in src/ . For more information see RAPIDS\u2019 File Structure .","title":"Home"},{"location":"#welcome-to-rapids-documentation","text":"Warning The functionality outlined in these docs is implemented in the branch time_segments which we will merge to master soon as release 0.1 . The previous (first) release of RAPIDS along with the old docs will be labeled beta . If you landed on this page feel free to look around, just have in mind that we are polishing the last rough patches before we advertise 0.1 (Nov 16, 2020) Reproducible Analysis Pipeline for Data Streams (RAPIDS) allows you to process smartphone and wearable data to extract behavioral features (a.k.a. digital biomarkers/phenotypes). RAPIDS is open source, documented, modular, tested, and reproducible. At the moment we support smartphone data collected with AWARE and wearable data from Fitbit devices. Questions or feedback can be posted on #rapids in AWARE Framework's slack . Bugs should be reported on Github . Join our discussions on our algorithms and assumptions for feature processing . Ready to start? Go to Installation , then to Configuration , and then to Execution","title":"Welcome to RAPIDS documentation"},{"location":"#how-does-it-work","text":"RAPIDS is formed by R and Python scripts orchestrated by Snakemake . We suggest you read Snakemake\u2019s docs but in short: every link in the analysis chain is atomic and has files as input and output. Behavioral features are processed per sensor and per participant.","title":"How does it work?"},{"location":"#what-are-the-benefits-of-using-rapids","text":"Consistent analysis . Every participant sensor dataset is analyzed in the exact same way and isolated from each other. Efficient analysis . Every analysis step is executed only once. Whenever your data or configuration changes only the affected files are updated. Parallel execution . Thanks to Snakemake, your analysis can be executed over multiple cores without changing your code. Extensible code . You can easily add your own behavioral features in R or Python and keep authorship and citations. Timezone aware . Your data is adjusted to the specified timezone (multiple timezones suport coming soon ). Flexible time segments . You can extract behavioral features on time windows of any length (e.g. 5 minutes, 3 hours, 2 days), on every day or particular days (e.g. weekends, Mondays, the 1 st of each month, etc.) or around events of interest (e.g. surveys or clinical relapses). Tested code . We are constantly adding tests to make sure our behavioral features are correct. Reproducible code . You can be sure your code will run in other computers as intended thanks to R and Python virtual environments. You can share your analysis code along your publications without any overhead. Private . All your data is processed locally.","title":"What are the benefits of using RAPIDS?"},{"location":"#how-is-it-organized","text":"In broad terms the config.yaml , .env file , participants files , time segment files are the only ones that you will have to modify. All data is stored in data/ and all scripts are stored in src/ . For more information see RAPIDS\u2019 File Structure .","title":"How is it organized?"},{"location":"citation/","text":"Cite RAPIDS and providers \u00b6 RAPIDS and the community RAPIDS is a community effort and as such we want to continue recognizing the contributions from other researchers. Besides citing RAPIDS, we ask you to cite any of the authors listed below if you used those sensor providers in your analysis, thank you! RAPIDS \u00b6 If you used RAPIDS, please cite this paper . RAPIDS et al. citation Vega J, Li M, Aguillera K, Goel N, Joshi E, Durica KC, Kunta AR, Low CA RAPIDS: Reproducible Analysis Pipeline for Data Streams Collected with Mobile Devices JMIR Preprints. 18/08/2020:23246 DOI: 10.2196/preprints.23246 URL: https://preprints.jmir.org/preprint/23246 Panda (accelerometer) \u00b6 If you computed accelerometer features using the provider [PHONE_ACCLEROMETER][PANDA] cite this paper in addition to RAPIDS. Panda et al. citation Panda N, Solsky I, Huang EJ, Lipsitz S, Pradarelli JC, Delisle M, Cusack JC, Gadd MA, Lubitz CC, Mullen JT, Qadan M, Smith BL, Specht M, Stephen AE, Tanabe KK, Gawande AA, Onnela JP, Haynes AB. Using Smartphones to Capture Novel Recovery Metrics After Cancer Surgery. JAMA Surg. 2020 Feb 1;155(2):123-129. doi: 10.1001/jamasurg.2019.4702. PMID: 31657854; PMCID: PMC6820047. Stachl (applications foreground) \u00b6 If you computed applications foreground features using the app category (genre) catalogue in [PHONE_APPLICATIONS_FOREGROUND][RAPIDS] cite this paper in addition to RAPIDS. Stachl et al. citation Clemens Stachl, Quay Au, Ramona Schoedel, Samuel D. Gosling, Gabriella M. Harari, Daniel Buschek, Sarah Theres V\u00f6lkel, Tobias Schuwerk, Michelle Oldemeier, Theresa Ullmann, Heinrich Hussmann, Bernd Bischl, Markus B\u00fchner. Proceedings of the National Academy of Sciences Jul 2020, 117 (30) 17680-17687; DOI: 10.1073/pnas.1920484117 Barnett (locations) \u00b6 If you computed locations features using the provider [PHONE_LOCATIONS][BARNETT] cite this paper and this paper in addition to RAPIDS. Barnett et al. citation Ian Barnett, Jukka-Pekka Onnela, Inferring mobility measures from GPS traces with missing data, Biostatistics, Volume 21, Issue 2, April 2020, Pages e98\u2013e112, https://doi.org/10.1093/biostatistics/kxy059 Canzian et al. citation Luca Canzian and Mirco Musolesi. 2015. Trajectories of depression: unobtrusive monitoring of depressive states by means of smartphone mobility traces analysis. In Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp \u201815). Association for Computing Machinery, New York, NY, USA, 1293\u20131304. DOI: https://doi.org/10.1145/2750858.2805845 Doryab (locations) \u00b6 If you computed locations features using the provider [PHONE_LOCATIONS][DORYAB] cite this paper and this paper in addition to RAPIDS. Doryab et al. citation Doryab, A., Chikarsel, P., Liu, X., & Dey, A. K. (2019). Extraction of Behavioral Features from Smartphone and Wearable Data. ArXiv:1812.10394 [Cs, Stat]. http://arxiv.org/abs/1812.10394 Canzian et al. citation Luca Canzian and Mirco Musolesi. 2015. Trajectories of depression: unobtrusive monitoring of depressive states by means of smartphone mobility traces analysis. In Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp \u201815). Association for Computing Machinery, New York, NY, USA, 1293\u20131304. DOI: https://doi.org/10.1145/2750858.2805845","title":"Citation"},{"location":"citation/#cite-rapids-and-providers","text":"RAPIDS and the community RAPIDS is a community effort and as such we want to continue recognizing the contributions from other researchers. Besides citing RAPIDS, we ask you to cite any of the authors listed below if you used those sensor providers in your analysis, thank you!","title":"Cite RAPIDS and providers"},{"location":"citation/#rapids","text":"If you used RAPIDS, please cite this paper . RAPIDS et al. citation Vega J, Li M, Aguillera K, Goel N, Joshi E, Durica KC, Kunta AR, Low CA RAPIDS: Reproducible Analysis Pipeline for Data Streams Collected with Mobile Devices JMIR Preprints. 18/08/2020:23246 DOI: 10.2196/preprints.23246 URL: https://preprints.jmir.org/preprint/23246","title":"RAPIDS"},{"location":"citation/#panda-accelerometer","text":"If you computed accelerometer features using the provider [PHONE_ACCLEROMETER][PANDA] cite this paper in addition to RAPIDS. Panda et al. citation Panda N, Solsky I, Huang EJ, Lipsitz S, Pradarelli JC, Delisle M, Cusack JC, Gadd MA, Lubitz CC, Mullen JT, Qadan M, Smith BL, Specht M, Stephen AE, Tanabe KK, Gawande AA, Onnela JP, Haynes AB. Using Smartphones to Capture Novel Recovery Metrics After Cancer Surgery. JAMA Surg. 2020 Feb 1;155(2):123-129. doi: 10.1001/jamasurg.2019.4702. PMID: 31657854; PMCID: PMC6820047.","title":"Panda (accelerometer)"},{"location":"citation/#stachl-applications-foreground","text":"If you computed applications foreground features using the app category (genre) catalogue in [PHONE_APPLICATIONS_FOREGROUND][RAPIDS] cite this paper in addition to RAPIDS. Stachl et al. citation Clemens Stachl, Quay Au, Ramona Schoedel, Samuel D. Gosling, Gabriella M. Harari, Daniel Buschek, Sarah Theres V\u00f6lkel, Tobias Schuwerk, Michelle Oldemeier, Theresa Ullmann, Heinrich Hussmann, Bernd Bischl, Markus B\u00fchner. Proceedings of the National Academy of Sciences Jul 2020, 117 (30) 17680-17687; DOI: 10.1073/pnas.1920484117","title":"Stachl (applications foreground)"},{"location":"citation/#barnett-locations","text":"If you computed locations features using the provider [PHONE_LOCATIONS][BARNETT] cite this paper and this paper in addition to RAPIDS. Barnett et al. citation Ian Barnett, Jukka-Pekka Onnela, Inferring mobility measures from GPS traces with missing data, Biostatistics, Volume 21, Issue 2, April 2020, Pages e98\u2013e112, https://doi.org/10.1093/biostatistics/kxy059 Canzian et al. citation Luca Canzian and Mirco Musolesi. 2015. Trajectories of depression: unobtrusive monitoring of depressive states by means of smartphone mobility traces analysis. In Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp \u201815). Association for Computing Machinery, New York, NY, USA, 1293\u20131304. DOI: https://doi.org/10.1145/2750858.2805845","title":"Barnett (locations)"},{"location":"citation/#doryab-locations","text":"If you computed locations features using the provider [PHONE_LOCATIONS][DORYAB] cite this paper and this paper in addition to RAPIDS. Doryab et al. citation Doryab, A., Chikarsel, P., Liu, X., & Dey, A. K. (2019). Extraction of Behavioral Features from Smartphone and Wearable Data. ArXiv:1812.10394 [Cs, Stat]. http://arxiv.org/abs/1812.10394 Canzian et al. citation Luca Canzian and Mirco Musolesi. 2015. Trajectories of depression: unobtrusive monitoring of depressive states by means of smartphone mobility traces analysis. In Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp \u201815). Association for Computing Machinery, New York, NY, USA, 1293\u20131304. DOI: https://doi.org/10.1145/2750858.2805845","title":"Doryab (locations)"},{"location":"code_of_conduct/","text":"Contributor Covenant Code of Conduct \u00b6 Our Pledge \u00b6 We as members, contributors, and leaders pledge to make participation in our community a harassment-free experience for everyone, regardless of age, body size, visible or invisible disability, ethnicity, sex characteristics, gender identity and expression, level of experience, education, socio-economic status, nationality, personal appearance, race, religion, or sexual identity and orientation. We pledge to act and interact in ways that contribute to an open, welcoming, diverse, inclusive, and healthy community. Our Standards \u00b6 Examples of behavior that contributes to a positive environment for our community include: Demonstrating empathy and kindness toward other people Being respectful of differing opinions, viewpoints, and experiences Giving and gracefully accepting constructive feedback Accepting responsibility and apologizing to those affected by our mistakes, and learning from the experience Focusing on what is best not just for us as individuals, but for the overall community Examples of unacceptable behavior include: The use of sexualized language or imagery, and sexual attention or advances of any kind Trolling, insulting or derogatory comments, and personal or political attacks Public or private harassment Publishing others\u2019 private information, such as a physical or email address, without their explicit permission Other conduct which could reasonably be considered inappropriate in a professional setting Enforcement Responsibilities \u00b6 Community leaders are responsible for clarifying and enforcing our standards of acceptable behavior and will take appropriate and fair corrective action in response to any behavior that they deem inappropriate, threatening, offensive, or harmful. Community leaders have the right and responsibility to remove, edit, or reject comments, commits, code, wiki edits, issues, and other contributions that are not aligned to this Code of Conduct, and will communicate reasons for moderation decisions when appropriate. Scope \u00b6 This Code of Conduct applies within all community spaces, and also applies when an individual is officially representing the community in public spaces. Examples of representing our community include using an official e-mail address, posting via an official social media account, or acting as an appointed representative at an online or offline event. Enforcement \u00b6 Instances of abusive, harassing, or otherwise unacceptable behavior may be reported to the community leaders responsible for enforcement at moshi@pitt.edu . All complaints will be reviewed and investigated promptly and fairly. All community leaders are obligated to respect the privacy and security of the reporter of any incident. Enforcement Guidelines \u00b6 Community leaders will follow these Community Impact Guidelines in determining the consequences for any action they deem in violation of this Code of Conduct: 1. Correction \u00b6 Community Impact : Use of inappropriate language or other behavior deemed unprofessional or unwelcome in the community. Consequence : A private, written warning from community leaders, providing clarity around the nature of the violation and an explanation of why the behavior was inappropriate. A public apology may be requested. 2. Warning \u00b6 Community Impact : A violation through a single incident or series of actions. Consequence : A warning with consequences for continued behavior. No interaction with the people involved, including unsolicited interaction with those enforcing the Code of Conduct, for a specified period of time. This includes avoiding interactions in community spaces as well as external channels like social media. Violating these terms may lead to a temporary or permanent ban. 3. Temporary Ban \u00b6 Community Impact : A serious violation of community standards, including sustained inappropriate behavior. Consequence : A temporary ban from any sort of interaction or public communication with the community for a specified period of time. No public or private interaction with the people involved, including unsolicited interaction with those enforcing the Code of Conduct, is allowed during this period. Violating these terms may lead to a permanent ban. 4. Permanent Ban \u00b6 Community Impact : Demonstrating a pattern of violation of community standards, including sustained inappropriate behavior, harassment of an individual, or aggression toward or disparagement of classes of individuals. Consequence : A permanent ban from any sort of public interaction within the community. Attribution \u00b6 This Code of Conduct is adapted from the Contributor Covenant , version 2.0, available at https://www.contributor-covenant.org/version/2/0/code_of_conduct.html . Community Impact Guidelines were inspired by Mozilla\u2019s code of conduct enforcement ladder . For answers to common questions about this code of conduct, see the FAQ at https://www.contributor-covenant.org/faq . Translations are available at https://www.contributor-covenant.org/translations .","title":"Code of Conduct"},{"location":"code_of_conduct/#contributor-covenant-code-of-conduct","text":"","title":"Contributor Covenant Code of Conduct"},{"location":"code_of_conduct/#our-pledge","text":"We as members, contributors, and leaders pledge to make participation in our community a harassment-free experience for everyone, regardless of age, body size, visible or invisible disability, ethnicity, sex characteristics, gender identity and expression, level of experience, education, socio-economic status, nationality, personal appearance, race, religion, or sexual identity and orientation. We pledge to act and interact in ways that contribute to an open, welcoming, diverse, inclusive, and healthy community.","title":"Our Pledge"},{"location":"code_of_conduct/#our-standards","text":"Examples of behavior that contributes to a positive environment for our community include: Demonstrating empathy and kindness toward other people Being respectful of differing opinions, viewpoints, and experiences Giving and gracefully accepting constructive feedback Accepting responsibility and apologizing to those affected by our mistakes, and learning from the experience Focusing on what is best not just for us as individuals, but for the overall community Examples of unacceptable behavior include: The use of sexualized language or imagery, and sexual attention or advances of any kind Trolling, insulting or derogatory comments, and personal or political attacks Public or private harassment Publishing others\u2019 private information, such as a physical or email address, without their explicit permission Other conduct which could reasonably be considered inappropriate in a professional setting","title":"Our Standards"},{"location":"code_of_conduct/#enforcement-responsibilities","text":"Community leaders are responsible for clarifying and enforcing our standards of acceptable behavior and will take appropriate and fair corrective action in response to any behavior that they deem inappropriate, threatening, offensive, or harmful. Community leaders have the right and responsibility to remove, edit, or reject comments, commits, code, wiki edits, issues, and other contributions that are not aligned to this Code of Conduct, and will communicate reasons for moderation decisions when appropriate.","title":"Enforcement Responsibilities"},{"location":"code_of_conduct/#scope","text":"This Code of Conduct applies within all community spaces, and also applies when an individual is officially representing the community in public spaces. Examples of representing our community include using an official e-mail address, posting via an official social media account, or acting as an appointed representative at an online or offline event.","title":"Scope"},{"location":"code_of_conduct/#enforcement","text":"Instances of abusive, harassing, or otherwise unacceptable behavior may be reported to the community leaders responsible for enforcement at moshi@pitt.edu . All complaints will be reviewed and investigated promptly and fairly. All community leaders are obligated to respect the privacy and security of the reporter of any incident.","title":"Enforcement"},{"location":"code_of_conduct/#enforcement-guidelines","text":"Community leaders will follow these Community Impact Guidelines in determining the consequences for any action they deem in violation of this Code of Conduct:","title":"Enforcement Guidelines"},{"location":"code_of_conduct/#1-correction","text":"Community Impact : Use of inappropriate language or other behavior deemed unprofessional or unwelcome in the community. Consequence : A private, written warning from community leaders, providing clarity around the nature of the violation and an explanation of why the behavior was inappropriate. A public apology may be requested.","title":"1. Correction"},{"location":"code_of_conduct/#2-warning","text":"Community Impact : A violation through a single incident or series of actions. Consequence : A warning with consequences for continued behavior. No interaction with the people involved, including unsolicited interaction with those enforcing the Code of Conduct, for a specified period of time. This includes avoiding interactions in community spaces as well as external channels like social media. Violating these terms may lead to a temporary or permanent ban.","title":"2. Warning"},{"location":"code_of_conduct/#3-temporary-ban","text":"Community Impact : A serious violation of community standards, including sustained inappropriate behavior. Consequence : A temporary ban from any sort of interaction or public communication with the community for a specified period of time. No public or private interaction with the people involved, including unsolicited interaction with those enforcing the Code of Conduct, is allowed during this period. Violating these terms may lead to a permanent ban.","title":"3. Temporary Ban"},{"location":"code_of_conduct/#4-permanent-ban","text":"Community Impact : Demonstrating a pattern of violation of community standards, including sustained inappropriate behavior, harassment of an individual, or aggression toward or disparagement of classes of individuals. Consequence : A permanent ban from any sort of public interaction within the community.","title":"4. Permanent Ban"},{"location":"code_of_conduct/#attribution","text":"This Code of Conduct is adapted from the Contributor Covenant , version 2.0, available at https://www.contributor-covenant.org/version/2/0/code_of_conduct.html . Community Impact Guidelines were inspired by Mozilla\u2019s code of conduct enforcement ladder . For answers to common questions about this code of conduct, see the FAQ at https://www.contributor-covenant.org/faq . Translations are available at https://www.contributor-covenant.org/translations .","title":"Attribution"},{"location":"faq/","text":"Frequently Asked Questions \u00b6 Cannot connect to your MySQL server \u00b6 Problem **Error in .local ( drv, \\. .. ) :** **Failed to connect to database: Error: Can \\' t initialize character set unknown ( path: compiled \\_ in ) ** : Calls: dbConnect -> dbConnect -> .local -> .Call Execution halted [ Tue Mar 10 19 :40:15 2020 ] Error in rule download_dataset: jobid: 531 output: data/raw/p60/locations_raw.csv RuleException: CalledProcessError in line 20 of /home/ubuntu/rapids/rules/preprocessing.snakefile: Command 'set -euo pipefail; Rscript --vanilla /home/ubuntu/rapids/.snakemake/scripts/tmp_2jnvqs7.download_dataset.R' returned non-zero exit status 1 . File \"/home/ubuntu/rapids/rules/preprocessing.snakefile\" , line 20 , in __rule_download_dataset File \"/home/ubuntu/anaconda3/envs/moshi-env/lib/python3.7/concurrent/futures/thread.py\" , line 57 , in run Shutting down, this might take some time. Exiting because a job execution failed. Look above for error message Solution Please make sure the DATABASE_GROUP in config.yaml matches your DB credentials group in .env . Cannot start mysql in linux via brew services start mysql \u00b6 Problem Cannot start mysql in linux via brew services start mysql Solution Use mysql.server start Every time I run force the download_dataset rule all rules are executed \u00b6 Problem When running snakemake -j1 -R download_phone_data or ./rapids -j1 -R download_phone_data all the rules and files are re-computed Solution This is expected behavior. The advantage of using snakemake under the hood is that every time a file containing data is modified every rule that depends on that file will be re-executed to update their results. In this case, since download_dataset updates all the raw data, and you are forcing the rule with the flag -R every single rule that depends on those raw files will be executed. Error Table XXX doesn't exist while running the download_phone_data or download_fitbit_data rule. \u00b6 Problem Error in .local ( conn, statement, ... ) : could not run statement: Table 'db_name.table_name' doesn ' t exist Calls: colnames ... .local -> dbSendQuery -> dbSendQuery -> .local -> .Call Execution halted Solution Please make sure the sensors listed in [PHONE_VALID_SENSED_BINS][PHONE_SENSORS] and the [TABLE] of each sensor you activated in config.yaml match your database tables. How do I install RAPIDS on Ubuntu 16.04 \u00b6 Solution Install dependencies (Homebrew - if not installed): sudo apt-get install libmariadb-client-lgpl-dev libxml2-dev libssl-dev Install brew for linux and add the following line to ~/.bashrc : export PATH=$HOME/.linuxbrew/bin:$PATH source ~/.bashrc Install MySQL brew install mysql brew services start mysql Install R, pandoc and rmarkdown: brew install r brew install gcc@6 (needed due to this bug ) HOMEBREW_CC=gcc-6 brew install pandoc Install miniconda using these instructions Clone our repo: git clone https://github.com/carissalow/rapids Create a python virtual environment: cd rapids conda env create -f environment.yml -n MY_ENV_NAME conda activate MY_ENV_NAME Install R packages and virtual environment: snakemake renv_install snakemake renv_init snakemake renv_restore This step could take several minutes to complete. Please be patient and let it run until completion. mysql.h cannot be found \u00b6 Problem -------------------------- [ ERROR MESSAGE ] ---------------------------- <stdin>:1:10: fatal error: mysql.h: No such file or directory compilation terminated. ----------------------------------------------------------------------- ERROR: configuration failed for package 'RMySQL' Solution sudo apt install libmariadbclient-dev No package libcurl found \u00b6 Problem libcurl cannot be found Solution Install libcurl sudo apt install libcurl4-openssl-dev Configuration failed because openssl was not found. \u00b6 Problem openssl cannot be found Solution Install openssl sudo apt install libssl-dev Configuration failed because libxml-2.0 was not found \u00b6 Problem libxml-2.0 cannot be found Solution Install libxml-2.0 sudo apt install libxml2-dev SSL connection error when running RAPIDS \u00b6 Problem You are getting the following error message when running RAPIDS: Error: Failed to connect: SSL connection error: error:1425F102:SSL routines:ssl_choose_client_version:unsupported protocol. Solution This is a bug in Ubuntu 20.04 when trying to connect to an old MySQL server with MySQL client 8.0. You should get the same error message if you try to connect from the command line. There you can add the option --ssl-mode=DISABLED but we can't do this from the R connector. If you can't update your server, the quickest solution would be to import your database to another server or to a local environment. Alternatively, you could replace mysql-client and libmysqlclient-dev with mariadb-client and libmariadbclient-dev and reinstall renv. More info about this issue here DB_TABLES key not found \u00b6 Problem If you get the following error KeyError in line 43 of preprocessing.smk: 'PHONE_SENSORS' , it means that the indentation of the key [PHONE_SENSORS] is not matching the other child elements of PHONE_VALID_SENSED_BINS Solution You need to add or remove any leading whitespaces as needed on that line. PHONE_VALID_SENSED_BINS : COMPUTE : False # This flag is automatically ignored (set to True) if you are extracting PHONE_VALID_SENSED_DAYS or screen or Barnett's location features BIN_SIZE : &bin_size 5 # (in minutes) PHONE_SENSORS : [] Error while updating your conda environment in Ubuntu \u00b6 Problem You get the following error: CondaMultiError: CondaVerificationError: The package for tk located at /home/ubuntu/miniconda2/pkgs/tk-8.6.9-hed695b0_1003 appears to be corrupted. The path 'include/mysqlStubs.h' specified in the package manifest cannot be found. ClobberError: This transaction has incompatible packages due to a shared path. packages: conda-forge/linux-64::llvm-openmp-10.0.0-hc9558a2_0, anaconda/linux-64::intel-openmp-2019.4-243 path: 'lib/libiomp5.so' Solution Reinstall conda","title":"Frequently Asked Questions"},{"location":"faq/#frequently-asked-questions","text":"","title":"Frequently Asked Questions"},{"location":"faq/#cannot-connect-to-your-mysql-server","text":"Problem **Error in .local ( drv, \\. .. ) :** **Failed to connect to database: Error: Can \\' t initialize character set unknown ( path: compiled \\_ in ) ** : Calls: dbConnect -> dbConnect -> .local -> .Call Execution halted [ Tue Mar 10 19 :40:15 2020 ] Error in rule download_dataset: jobid: 531 output: data/raw/p60/locations_raw.csv RuleException: CalledProcessError in line 20 of /home/ubuntu/rapids/rules/preprocessing.snakefile: Command 'set -euo pipefail; Rscript --vanilla /home/ubuntu/rapids/.snakemake/scripts/tmp_2jnvqs7.download_dataset.R' returned non-zero exit status 1 . File \"/home/ubuntu/rapids/rules/preprocessing.snakefile\" , line 20 , in __rule_download_dataset File \"/home/ubuntu/anaconda3/envs/moshi-env/lib/python3.7/concurrent/futures/thread.py\" , line 57 , in run Shutting down, this might take some time. Exiting because a job execution failed. Look above for error message Solution Please make sure the DATABASE_GROUP in config.yaml matches your DB credentials group in .env .","title":"Cannot connect to your MySQL server"},{"location":"faq/#cannot-start-mysql-in-linux-via-brew-services-start-mysql","text":"Problem Cannot start mysql in linux via brew services start mysql Solution Use mysql.server start","title":"Cannot start mysql in linux via brew services start mysql"},{"location":"faq/#every-time-i-run-force-the-download_dataset-rule-all-rules-are-executed","text":"Problem When running snakemake -j1 -R download_phone_data or ./rapids -j1 -R download_phone_data all the rules and files are re-computed Solution This is expected behavior. The advantage of using snakemake under the hood is that every time a file containing data is modified every rule that depends on that file will be re-executed to update their results. In this case, since download_dataset updates all the raw data, and you are forcing the rule with the flag -R every single rule that depends on those raw files will be executed.","title":"Every time I run force the download_dataset rule all rules are executed"},{"location":"faq/#error-table-xxx-doesnt-exist-while-running-the-download_phone_data-or-download_fitbit_data-rule","text":"Problem Error in .local ( conn, statement, ... ) : could not run statement: Table 'db_name.table_name' doesn ' t exist Calls: colnames ... .local -> dbSendQuery -> dbSendQuery -> .local -> .Call Execution halted Solution Please make sure the sensors listed in [PHONE_VALID_SENSED_BINS][PHONE_SENSORS] and the [TABLE] of each sensor you activated in config.yaml match your database tables.","title":"Error Table XXX doesn't exist while running the download_phone_data or download_fitbit_data rule."},{"location":"faq/#how-do-i-install-rapids-on-ubuntu-1604","text":"Solution Install dependencies (Homebrew - if not installed): sudo apt-get install libmariadb-client-lgpl-dev libxml2-dev libssl-dev Install brew for linux and add the following line to ~/.bashrc : export PATH=$HOME/.linuxbrew/bin:$PATH source ~/.bashrc Install MySQL brew install mysql brew services start mysql Install R, pandoc and rmarkdown: brew install r brew install gcc@6 (needed due to this bug ) HOMEBREW_CC=gcc-6 brew install pandoc Install miniconda using these instructions Clone our repo: git clone https://github.com/carissalow/rapids Create a python virtual environment: cd rapids conda env create -f environment.yml -n MY_ENV_NAME conda activate MY_ENV_NAME Install R packages and virtual environment: snakemake renv_install snakemake renv_init snakemake renv_restore This step could take several minutes to complete. Please be patient and let it run until completion.","title":"How do I install RAPIDS on Ubuntu 16.04"},{"location":"faq/#mysqlh-cannot-be-found","text":"Problem -------------------------- [ ERROR MESSAGE ] ---------------------------- <stdin>:1:10: fatal error: mysql.h: No such file or directory compilation terminated. ----------------------------------------------------------------------- ERROR: configuration failed for package 'RMySQL' Solution sudo apt install libmariadbclient-dev","title":"mysql.h cannot be found"},{"location":"faq/#no-package-libcurl-found","text":"Problem libcurl cannot be found Solution Install libcurl sudo apt install libcurl4-openssl-dev","title":"No package libcurl found"},{"location":"faq/#configuration-failed-because-openssl-was-not-found","text":"Problem openssl cannot be found Solution Install openssl sudo apt install libssl-dev","title":"Configuration failed because openssl was not found."},{"location":"faq/#configuration-failed-because-libxml-20-was-not-found","text":"Problem libxml-2.0 cannot be found Solution Install libxml-2.0 sudo apt install libxml2-dev","title":"Configuration failed because libxml-2.0 was not found"},{"location":"faq/#ssl-connection-error-when-running-rapids","text":"Problem You are getting the following error message when running RAPIDS: Error: Failed to connect: SSL connection error: error:1425F102:SSL routines:ssl_choose_client_version:unsupported protocol. Solution This is a bug in Ubuntu 20.04 when trying to connect to an old MySQL server with MySQL client 8.0. You should get the same error message if you try to connect from the command line. There you can add the option --ssl-mode=DISABLED but we can't do this from the R connector. If you can't update your server, the quickest solution would be to import your database to another server or to a local environment. Alternatively, you could replace mysql-client and libmysqlclient-dev with mariadb-client and libmariadbclient-dev and reinstall renv. More info about this issue here","title":"SSL connection error when running RAPIDS"},{"location":"faq/#db_tables-key-not-found","text":"Problem If you get the following error KeyError in line 43 of preprocessing.smk: 'PHONE_SENSORS' , it means that the indentation of the key [PHONE_SENSORS] is not matching the other child elements of PHONE_VALID_SENSED_BINS Solution You need to add or remove any leading whitespaces as needed on that line. PHONE_VALID_SENSED_BINS : COMPUTE : False # This flag is automatically ignored (set to True) if you are extracting PHONE_VALID_SENSED_DAYS or screen or Barnett's location features BIN_SIZE : &bin_size 5 # (in minutes) PHONE_SENSORS : []","title":"DB_TABLES key not found"},{"location":"faq/#error-while-updating-your-conda-environment-in-ubuntu","text":"Problem You get the following error: CondaMultiError: CondaVerificationError: The package for tk located at /home/ubuntu/miniconda2/pkgs/tk-8.6.9-hed695b0_1003 appears to be corrupted. The path 'include/mysqlStubs.h' specified in the package manifest cannot be found. ClobberError: This transaction has incompatible packages due to a shared path. packages: conda-forge/linux-64::llvm-openmp-10.0.0-hc9558a2_0, anaconda/linux-64::intel-openmp-2019.4-243 path: 'lib/libiomp5.so' Solution Reinstall conda","title":"Error while updating your conda environment in Ubuntu"},{"location":"file-structure/","text":"File Structure \u00b6 Tip Read this page if you want to learn more about how RAPIDS is structured. If you want to start using it go to Installation , then to Configuration , and then to Execution All paths mentioned in this page are relative to RAPIDS\u2019 root folder. If you want to extract the behavioral features that RAPIDS offers, you will only have to create or modify the .env file , participants files , time segment files , and the config.yaml file. The config.yaml file is the heart of RAPIDS and includes parameters to manage participants, data sources, sensor data, visualizations and more. All data is saved in data/ . The data/external/ folder stores any data imported or created by the user, data/raw/ stores sensor data as imported from your database, data/interim/ has intermediate files necessary to compute behavioral features from raw data, and data/processed/ has all the final files with the behavioral features in folders per participant and sensor. All the source code is saved in src/ . The src/data/ folder stores scripts to download, clean and pre-process sensor data, src/features has scripts to extract behavioral features organized in their respective subfolders , src/models/ can host any script to create models or statistical analyses with the behavioral features you extract, and src/visualization/ has scripts to create plots of the raw and processed data. There are other important files and folders but only relevant if you are interested in extending RAPIDS (e.g. virtual env files, docs, tests, Dockerfile, the Snakefile, etc.). In the figure below, we represent the interactions between users and files. After a user modifies config.yaml and .env the Snakefile file will decide what Snakemake rules have to be executed to produce the required output files (behavioral features) and what scripts are in charge of producing such files. In addition, users can add or modifiy files in the data folder (for example to configure the participants files or the time segment files ). Interaction diagram between the user, and important files in RAPIDS","title":"File Structure"},{"location":"file-structure/#file-structure","text":"Tip Read this page if you want to learn more about how RAPIDS is structured. If you want to start using it go to Installation , then to Configuration , and then to Execution All paths mentioned in this page are relative to RAPIDS\u2019 root folder. If you want to extract the behavioral features that RAPIDS offers, you will only have to create or modify the .env file , participants files , time segment files , and the config.yaml file. The config.yaml file is the heart of RAPIDS and includes parameters to manage participants, data sources, sensor data, visualizations and more. All data is saved in data/ . The data/external/ folder stores any data imported or created by the user, data/raw/ stores sensor data as imported from your database, data/interim/ has intermediate files necessary to compute behavioral features from raw data, and data/processed/ has all the final files with the behavioral features in folders per participant and sensor. All the source code is saved in src/ . The src/data/ folder stores scripts to download, clean and pre-process sensor data, src/features has scripts to extract behavioral features organized in their respective subfolders , src/models/ can host any script to create models or statistical analyses with the behavioral features you extract, and src/visualization/ has scripts to create plots of the raw and processed data. There are other important files and folders but only relevant if you are interested in extending RAPIDS (e.g. virtual env files, docs, tests, Dockerfile, the Snakefile, etc.). In the figure below, we represent the interactions between users and files. After a user modifies config.yaml and .env the Snakefile file will decide what Snakemake rules have to be executed to produce the required output files (behavioral features) and what scripts are in charge of producing such files. In addition, users can add or modifiy files in the data folder (for example to configure the participants files or the time segment files ). Interaction diagram between the user, and important files in RAPIDS","title":"File Structure"},{"location":"team/","text":"RAPIDS Team \u00b6 If you are interested in contributing feel free to submit a pull request or contact us. Core Team \u00b6 Julio Vega (Designer and Lead Developer) \u00b6 About Julio Vega is a postdoctoral associate at the Mobile Sensing + Health Institute. He is interested in personalized methodologies to monitor chronic conditions that affect daily human behavior using mobile and wearable data. vegaju at upmc . edu Personal Website Meng Li \u00b6 About Meng Li received her Master of Science degree in Information Science from the University of Pittsburgh. She is interested in applying machine learning algorithms to the medical field. lim11 at upmc . edu Linkedin Profile Github Profile Abhineeth Reddy Kunta \u00b6 About Abhineeth Reddy Kunta is a Senior Software Engineer with the Mobile Sensing + Health Institute. He is experienced in software development and specializes in building solutions using machine learning. Abhineeth likes exploring ways to leverage technology in advancing medicine and education. Previously he worked as a Computer Programmer at Georgia Department of Public Health. He has a master\u2019s degree in Computer Science from George Mason University. Kwesi Aguillera \u00b6 About Kwesi Aguillera is currently in his first year at the University of Pittsburgh pursuing a Master of Sciences in Information Science specializing in Big Data Analytics. He received his Bachelor of Science degree in Computer Science and Management from the University of the West Indies. Kwesi considers himself a full stack developer and looks forward to applying this knowledge to big data analysis. Linkedin Profile Echhit Joshi \u00b6 About Echhit Joshi is a Masters student at the School of Computing and Information at University of Pittsburgh. His areas of interest are Machine/Deep Learning, Data Mining, and Analytics. Linkedin Profile Nicolas Leo \u00b6 About Nicolas is a rising senior studying computer science at the University of Pittsburgh. His academic interests include databases, machine learning, and application development. After completing his undergraduate degree, he plans to attend graduate school for a MS in Computer Science with a focus on Intelligent Systems. Nikunj Goel \u00b6 About Nik is a graduate student at the University of Pittsburgh pursuing Master of Science in Information Science. He earned his Bachelor of Technology degree in Information Technology from India. He is a Data Enthusiasts and passionate about finding the meaning out of raw data. In a long term, his goal is to create a breakthrough in Data Science and Deep Learning. Linkedin Profile Community Contributors \u00b6 Agam Kumar \u00b6 About Agam is a junior at Carnegie Mellon University studying Statistics and Machine Learning and pursuing an additional major in Computer Science. He is a member of the Data Science team in the Health and Human Performance Lab at CMU and has keen interests in software development and data science. His research interests include ML applications in medicine. Linkedin Profile Github Profile Yasaman S. Sefidgar \u00b6 About Linkedin Profile Advisors \u00b6 Afsaneh Doryab \u00b6 About Personal Website Carissa Low \u00b6 About Profile","title":"Team"},{"location":"team/#rapids-team","text":"If you are interested in contributing feel free to submit a pull request or contact us.","title":"RAPIDS Team"},{"location":"team/#core-team","text":"","title":"Core Team"},{"location":"team/#julio-vega-designer-and-lead-developer","text":"About Julio Vega is a postdoctoral associate at the Mobile Sensing + Health Institute. He is interested in personalized methodologies to monitor chronic conditions that affect daily human behavior using mobile and wearable data. vegaju at upmc . edu Personal Website","title":"Julio Vega (Designer and Lead Developer)"},{"location":"team/#meng-li","text":"About Meng Li received her Master of Science degree in Information Science from the University of Pittsburgh. She is interested in applying machine learning algorithms to the medical field. lim11 at upmc . edu Linkedin Profile Github Profile","title":"Meng Li"},{"location":"team/#abhineeth-reddy-kunta","text":"About Abhineeth Reddy Kunta is a Senior Software Engineer with the Mobile Sensing + Health Institute. He is experienced in software development and specializes in building solutions using machine learning. Abhineeth likes exploring ways to leverage technology in advancing medicine and education. Previously he worked as a Computer Programmer at Georgia Department of Public Health. He has a master\u2019s degree in Computer Science from George Mason University.","title":"Abhineeth Reddy Kunta"},{"location":"team/#kwesi-aguillera","text":"About Kwesi Aguillera is currently in his first year at the University of Pittsburgh pursuing a Master of Sciences in Information Science specializing in Big Data Analytics. He received his Bachelor of Science degree in Computer Science and Management from the University of the West Indies. Kwesi considers himself a full stack developer and looks forward to applying this knowledge to big data analysis. Linkedin Profile","title":"Kwesi Aguillera"},{"location":"team/#echhit-joshi","text":"About Echhit Joshi is a Masters student at the School of Computing and Information at University of Pittsburgh. His areas of interest are Machine/Deep Learning, Data Mining, and Analytics. Linkedin Profile","title":"Echhit Joshi"},{"location":"team/#nicolas-leo","text":"About Nicolas is a rising senior studying computer science at the University of Pittsburgh. His academic interests include databases, machine learning, and application development. After completing his undergraduate degree, he plans to attend graduate school for a MS in Computer Science with a focus on Intelligent Systems.","title":"Nicolas Leo"},{"location":"team/#nikunj-goel","text":"About Nik is a graduate student at the University of Pittsburgh pursuing Master of Science in Information Science. He earned his Bachelor of Technology degree in Information Technology from India. He is a Data Enthusiasts and passionate about finding the meaning out of raw data. In a long term, his goal is to create a breakthrough in Data Science and Deep Learning. Linkedin Profile","title":"Nikunj Goel"},{"location":"team/#community-contributors","text":"","title":"Community Contributors"},{"location":"team/#agam-kumar","text":"About Agam is a junior at Carnegie Mellon University studying Statistics and Machine Learning and pursuing an additional major in Computer Science. He is a member of the Data Science team in the Health and Human Performance Lab at CMU and has keen interests in software development and data science. His research interests include ML applications in medicine. Linkedin Profile Github Profile","title":"Agam Kumar"},{"location":"team/#yasaman-s-sefidgar","text":"About Linkedin Profile","title":"Yasaman S. Sefidgar"},{"location":"team/#advisors","text":"","title":"Advisors"},{"location":"team/#afsaneh-doryab","text":"About Personal Website","title":"Afsaneh Doryab"},{"location":"team/#carissa-low","text":"About Profile","title":"Carissa Low"},{"location":"developers/documentation/","text":"Documentation \u00b6 We use mkdocs with the material theme to write these docs. Whenever you make any changes, just push them back to the repo and the documentation will be deployed automatically. Set up development environment \u00b6 Make sure your conda environment is active pip install mkdocs pip install mkdocs-material Preview \u00b6 Run the following command in RAPIDS root folder and go to http://127.0.0.1:8000 : mkdocs serve File Structure \u00b6 The documentation config file is /mkdocs.yml , if you are adding new .md files to the docs modify the nav attribute at the bottom of that file. You can use the hierarchy there to find all the files that appear in the documentation. Reference \u00b6 Check this page to get familiar with the different visual elements we can use in the docs (admonitions, code blocks, tables, etc.) You can also refer to /docs/setup/installation.md and /docs/setup/configuration.md to see practical examples of these elements. Hint Any links to internal pages should be relative to the current page. For example, any link from this page (documentation) which is inside ./developers should begin with ../ to go one folder level up like: [ mylink ]( ../setup/installation.md ) Extras \u00b6 You can insert emojis using this syntax :[SOURCE]-[ICON_NAME] from the following sources: https://materialdesignicons.com/ https://fontawesome.com/icons/tasks?style=solid https://primer.style/octicons/ You can use this page to create markdown tables more easily","title":"Documentation"},{"location":"developers/documentation/#documentation","text":"We use mkdocs with the material theme to write these docs. Whenever you make any changes, just push them back to the repo and the documentation will be deployed automatically.","title":"Documentation"},{"location":"developers/documentation/#set-up-development-environment","text":"Make sure your conda environment is active pip install mkdocs pip install mkdocs-material","title":"Set up development environment"},{"location":"developers/documentation/#preview","text":"Run the following command in RAPIDS root folder and go to http://127.0.0.1:8000 : mkdocs serve","title":"Preview"},{"location":"developers/documentation/#file-structure","text":"The documentation config file is /mkdocs.yml , if you are adding new .md files to the docs modify the nav attribute at the bottom of that file. You can use the hierarchy there to find all the files that appear in the documentation.","title":"File Structure"},{"location":"developers/documentation/#reference","text":"Check this page to get familiar with the different visual elements we can use in the docs (admonitions, code blocks, tables, etc.) You can also refer to /docs/setup/installation.md and /docs/setup/configuration.md to see practical examples of these elements. Hint Any links to internal pages should be relative to the current page. For example, any link from this page (documentation) which is inside ./developers should begin with ../ to go one folder level up like: [ mylink ]( ../setup/installation.md )","title":"Reference"},{"location":"developers/documentation/#extras","text":"You can insert emojis using this syntax :[SOURCE]-[ICON_NAME] from the following sources: https://materialdesignicons.com/ https://fontawesome.com/icons/tasks?style=solid https://primer.style/octicons/ You can use this page to create markdown tables more easily","title":"Extras"},{"location":"developers/remote-support/","text":"Remote Support \u00b6 We use the Live Share extension of Visual Studio Code to debug bugs when sharing data or database credentials is not possible. Install Visual Studio Code Open you RAPIDS root folder in a new VSCode window Open a new Terminal Terminal > New terminal Install the Live Share extension pack Press Ctrl + P or Cmd + P and run this command: >live share: start collaboration session 6. Follow the instructions and share the session link you receive","title":"Remote Support"},{"location":"developers/remote-support/#remote-support","text":"We use the Live Share extension of Visual Studio Code to debug bugs when sharing data or database credentials is not possible. Install Visual Studio Code Open you RAPIDS root folder in a new VSCode window Open a new Terminal Terminal > New terminal Install the Live Share extension pack Press Ctrl + P or Cmd + P and run this command: >live share: start collaboration session 6. Follow the instructions and share the session link you receive","title":"Remote Support"},{"location":"developers/test-cases/","text":"Test Cases \u00b6 Along with the continued development and the addition of new sensors and features to the RAPIDS pipeline, tests for the currently available sensors and features are being implemented. Since this is a Work In Progress this page will be updated with the list of sensors and features for which testing is available. For each of the sensors listed a description of the data used for testing (test cases) are outline. Currently for all intent and testing purposes the tests/data/raw/test01/ contains all the test data files for testing android data formats and tests/data/raw/test02/ contains all the test data files for testing iOS data formats. It follows that the expected (verified output) are contained in the tests/data/processed/test01/ and tests/data/processed/test02/ for Android and iOS respectively. tests/data/raw/test03/ and tests/data/raw/test04/ contain data files for testing empty raw data files for android and iOS respectively. The following is a list of the sensors that testing is currently available. Messages (SMS) \u00b6 The raw message data file contains data for 2 separate days. The data for the first day contains records 5 records for every epoch . The second day's data contains 6 records for each of only 2 epoch (currently morning and evening ) The raw message data contains records for both message_types (i.e. recieved and sent ) in both days in all epochs. The number records with each message_types per epoch is randomly distributed There is at least one records with each message_types per epoch. There is one raw message data file each, as described above, for testing both iOS and Android data. There is also an additional empty data file for both android and iOS for testing empty data files Calls \u00b6 Due to the difference in the format of the raw call data for iOS and Android the following is the expected results the calls_with_datetime_unified.csv . This would give a better idea of the use cases being tested since the calls_with_datetime_unified.csv would make both the iOS and Android data comparable. The call data would contain data for 2 days. The data for the first day contains 6 records for every epoch . The second day's data contains 6 records for each of only 2 epoch (currently morning and evening ) The call data contains records for all call_types (i.e. incoming , outgoing and missed ) in both days in all epochs. The number records with each of the call_types per epoch is randomly distributed. There is at least one records with each call_types per epoch. There is one call data file each, as described above, for testing both iOS and Android data. There is also an additional empty data file for both android and iOS for testing empty data files Screen \u00b6 Due to the difference in the format of the raw screen data for iOS and Android the following is the expected results the screen_deltas.csv . This would give a better idea of the use cases being tested since the screen_eltas.csv would make both the iOS and Android data comparable These files are used to calculate the features for the screen sensor The screen delta data file contains data for 1 day. The screen delta data contains 1 record to represent an unlock episode that falls within an epoch for every epoch . The screen delta data contains 1 record to represent an unlock episode that falls across the boundary of 2 epochs. Namely the unlock episode starts in one epoch and ends in the next, thus there is a record for unlock episodes that fall across night to morning , morning to afternoon and finally afternoon to night The testing is done for unlock episode_type. There is one screen data file each for testing both iOS and Android data formats. There is also an additional empty data file for both android and iOS for testing empty data files Battery \u00b6 Due to the difference in the format of the raw battery data for iOS and Android as well as versions of iOS the following is the expected results the battery_deltas.csv . This would give a better idea of the use cases being tested since the battery_deltas.csv would make both the iOS and Android data comparable. These files are used to calculate the features for the battery sensor. The battery delta data file contains data for 1 day. The battery delta data contains 1 record each for a charging and discharging episode that falls within an epoch for every epoch . Thus, for the daily epoch there would be multiple charging and discharging episodes Since either a charging episode or a discharging episode and not both can occur across epochs, in order to test episodes that occur across epochs alternating episodes of charging and discharging episodes that fall across night to morning , morning to afternoon and finally afternoon to night are present in the battery delta data. This starts with a discharging episode that begins in night and end in morning . There is one battery data file each, for testing both iOS and Android data formats. There is also an additional empty data file for both android and iOS for testing empty data files Bluetooth \u00b6 The raw Bluetooth data file contains data for 1 day. The raw Bluetooth data contains at least 2 records for each epoch . Each epoch has a record with a timestamp for the beginning boundary for that epoch and a record with a timestamp for the ending boundary for that epoch . (e.g. For the morning epoch there is a record with a timestamp for 6:00AM and another record with a timestamp for 11:59:59AM . These are to test edge cases) An option of 5 Bluetooth devices are randomly distributed throughout the data records. There is one raw Bluetooth data file each, for testing both iOS and Android data formats. There is also an additional empty data file for both android and iOS for testing empty data files. WIFI \u00b6 There are 2 data files ( wifi_raw.csv and sensor_wifi_raw.csv ) for each fake participant for each phone platform. The raw WIFI data files contain data for 1 day. The sensor_wifi_raw.csv data contains at least 2 records for each epoch . Each epoch has a record with a timestamp for the beginning boundary for that epoch and a record with a timestamp for the ending boundary for that epoch . (e.g. For the morning epoch there is a record with a timestamp for 6:00AM and another record with a timestamp for 11:59:59AM . These are to test edge cases) The wifi_raw.csv data contains 3 records with random timestamps for each epoch to represent visible broadcasting WIFI network. This file is empty for the iOS phone testing data. An option of 10 access point devices is randomly distributed throughout the data records. 5 each for sensor_wifi_raw.csv and wifi_raw.csv . There data files for testing both iOS and Android data formats. There are also additional empty data files for both android and iOS for testing empty data files. Light \u00b6 The raw light data file contains data for 1 day. The raw light data contains 3 or 4 rows of data for each epoch except night . The single row of data for night is for testing features for single values inputs. (Example testing the standard deviation of one input value) Since light is only available for Android there is only one file that contains data for Android. All other files (i.e. for iPhone) are empty data files. Application Foreground \u00b6 The raw application foreground data file contains data for 1 day. The raw application foreground data contains 7 - 9 rows of data for each epoch . The records for each epoch contains apps that are randomly selected from a list of apps that are from the MULTIPLE_CATEGORIES and SINGLE_CATEGORIES (See testing_config.yaml ). There are also records in each epoch that have apps randomly selected from a list of apps that are from the EXCLUDED_CATEGORIES and EXCLUDED_APPS . This is to test that these apps are actually being excluded from the calculations of features. There are also records to test SINGLE_APPS calculations. Since application foreground is only available for Android there is only one file that contains data for Android. All other files (i.e. for iPhone) are empty data files. Activity Recognition \u00b6 The raw Activity Recognition data file contains data for 1 day. The raw Activity Recognition data each epoch period contains rows that records 2 - 5 different activity_types . The is such that durations of activities can be tested. Additionally, there are records that mimic the duration of an activity over the time boundary of neighboring epochs. (For example, there a set of records that mimic the participant in_vehicle from afternoon into evening ) There is one file each with raw Activity Recognition data for testing both iOS and Android data formats. (plugin_google_activity_recognition_raw.csv for android and plugin_ios_activity_recognition_raw.csv for iOS) There is also an additional empty data file for both android and iOS for testing empty data files. Conversation \u00b6 The raw conversation data file contains data for 2 day. The raw conversation data contains records with a sample of both datatypes (i.e. voice/noise = 0 , and conversation = 2 ) as well as rows with for samples of each of the inference values (i.e. silence = 0 , noise = 1 , voice = 2 , and unknown = 3 ) for each epoch . The different datatype and inference records are randomly distributed throughout the epoch . Additionally there are 2 - 5 records for conversations ( datatype = 2, and inference = -1) in each epoch and for each epoch except night, there is a conversation record that has a double_convo_start timestamp that is from the previous epoch . This is to test the calculations of features across epochs . There is a raw conversation data file for both android and iOS platforms ( plugin_studentlife_audio_android_raw.csv and plugin_studentlife_audio_raw.csv respectively). Finally, there are also additional empty data files for both android and iOS for testing empty data files","title":"Test cases"},{"location":"developers/test-cases/#test-cases","text":"Along with the continued development and the addition of new sensors and features to the RAPIDS pipeline, tests for the currently available sensors and features are being implemented. Since this is a Work In Progress this page will be updated with the list of sensors and features for which testing is available. For each of the sensors listed a description of the data used for testing (test cases) are outline. Currently for all intent and testing purposes the tests/data/raw/test01/ contains all the test data files for testing android data formats and tests/data/raw/test02/ contains all the test data files for testing iOS data formats. It follows that the expected (verified output) are contained in the tests/data/processed/test01/ and tests/data/processed/test02/ for Android and iOS respectively. tests/data/raw/test03/ and tests/data/raw/test04/ contain data files for testing empty raw data files for android and iOS respectively. The following is a list of the sensors that testing is currently available.","title":"Test Cases"},{"location":"developers/test-cases/#messages-sms","text":"The raw message data file contains data for 2 separate days. The data for the first day contains records 5 records for every epoch . The second day's data contains 6 records for each of only 2 epoch (currently morning and evening ) The raw message data contains records for both message_types (i.e. recieved and sent ) in both days in all epochs. The number records with each message_types per epoch is randomly distributed There is at least one records with each message_types per epoch. There is one raw message data file each, as described above, for testing both iOS and Android data. There is also an additional empty data file for both android and iOS for testing empty data files","title":"Messages (SMS)"},{"location":"developers/test-cases/#calls","text":"Due to the difference in the format of the raw call data for iOS and Android the following is the expected results the calls_with_datetime_unified.csv . This would give a better idea of the use cases being tested since the calls_with_datetime_unified.csv would make both the iOS and Android data comparable. The call data would contain data for 2 days. The data for the first day contains 6 records for every epoch . The second day's data contains 6 records for each of only 2 epoch (currently morning and evening ) The call data contains records for all call_types (i.e. incoming , outgoing and missed ) in both days in all epochs. The number records with each of the call_types per epoch is randomly distributed. There is at least one records with each call_types per epoch. There is one call data file each, as described above, for testing both iOS and Android data. There is also an additional empty data file for both android and iOS for testing empty data files","title":"Calls"},{"location":"developers/test-cases/#screen","text":"Due to the difference in the format of the raw screen data for iOS and Android the following is the expected results the screen_deltas.csv . This would give a better idea of the use cases being tested since the screen_eltas.csv would make both the iOS and Android data comparable These files are used to calculate the features for the screen sensor The screen delta data file contains data for 1 day. The screen delta data contains 1 record to represent an unlock episode that falls within an epoch for every epoch . The screen delta data contains 1 record to represent an unlock episode that falls across the boundary of 2 epochs. Namely the unlock episode starts in one epoch and ends in the next, thus there is a record for unlock episodes that fall across night to morning , morning to afternoon and finally afternoon to night The testing is done for unlock episode_type. There is one screen data file each for testing both iOS and Android data formats. There is also an additional empty data file for both android and iOS for testing empty data files","title":"Screen"},{"location":"developers/test-cases/#battery","text":"Due to the difference in the format of the raw battery data for iOS and Android as well as versions of iOS the following is the expected results the battery_deltas.csv . This would give a better idea of the use cases being tested since the battery_deltas.csv would make both the iOS and Android data comparable. These files are used to calculate the features for the battery sensor. The battery delta data file contains data for 1 day. The battery delta data contains 1 record each for a charging and discharging episode that falls within an epoch for every epoch . Thus, for the daily epoch there would be multiple charging and discharging episodes Since either a charging episode or a discharging episode and not both can occur across epochs, in order to test episodes that occur across epochs alternating episodes of charging and discharging episodes that fall across night to morning , morning to afternoon and finally afternoon to night are present in the battery delta data. This starts with a discharging episode that begins in night and end in morning . There is one battery data file each, for testing both iOS and Android data formats. There is also an additional empty data file for both android and iOS for testing empty data files","title":"Battery"},{"location":"developers/test-cases/#bluetooth","text":"The raw Bluetooth data file contains data for 1 day. The raw Bluetooth data contains at least 2 records for each epoch . Each epoch has a record with a timestamp for the beginning boundary for that epoch and a record with a timestamp for the ending boundary for that epoch . (e.g. For the morning epoch there is a record with a timestamp for 6:00AM and another record with a timestamp for 11:59:59AM . These are to test edge cases) An option of 5 Bluetooth devices are randomly distributed throughout the data records. There is one raw Bluetooth data file each, for testing both iOS and Android data formats. There is also an additional empty data file for both android and iOS for testing empty data files.","title":"Bluetooth"},{"location":"developers/test-cases/#wifi","text":"There are 2 data files ( wifi_raw.csv and sensor_wifi_raw.csv ) for each fake participant for each phone platform. The raw WIFI data files contain data for 1 day. The sensor_wifi_raw.csv data contains at least 2 records for each epoch . Each epoch has a record with a timestamp for the beginning boundary for that epoch and a record with a timestamp for the ending boundary for that epoch . (e.g. For the morning epoch there is a record with a timestamp for 6:00AM and another record with a timestamp for 11:59:59AM . These are to test edge cases) The wifi_raw.csv data contains 3 records with random timestamps for each epoch to represent visible broadcasting WIFI network. This file is empty for the iOS phone testing data. An option of 10 access point devices is randomly distributed throughout the data records. 5 each for sensor_wifi_raw.csv and wifi_raw.csv . There data files for testing both iOS and Android data formats. There are also additional empty data files for both android and iOS for testing empty data files.","title":"WIFI"},{"location":"developers/test-cases/#light","text":"The raw light data file contains data for 1 day. The raw light data contains 3 or 4 rows of data for each epoch except night . The single row of data for night is for testing features for single values inputs. (Example testing the standard deviation of one input value) Since light is only available for Android there is only one file that contains data for Android. All other files (i.e. for iPhone) are empty data files.","title":"Light"},{"location":"developers/test-cases/#application-foreground","text":"The raw application foreground data file contains data for 1 day. The raw application foreground data contains 7 - 9 rows of data for each epoch . The records for each epoch contains apps that are randomly selected from a list of apps that are from the MULTIPLE_CATEGORIES and SINGLE_CATEGORIES (See testing_config.yaml ). There are also records in each epoch that have apps randomly selected from a list of apps that are from the EXCLUDED_CATEGORIES and EXCLUDED_APPS . This is to test that these apps are actually being excluded from the calculations of features. There are also records to test SINGLE_APPS calculations. Since application foreground is only available for Android there is only one file that contains data for Android. All other files (i.e. for iPhone) are empty data files.","title":"Application Foreground"},{"location":"developers/test-cases/#activity-recognition","text":"The raw Activity Recognition data file contains data for 1 day. The raw Activity Recognition data each epoch period contains rows that records 2 - 5 different activity_types . The is such that durations of activities can be tested. Additionally, there are records that mimic the duration of an activity over the time boundary of neighboring epochs. (For example, there a set of records that mimic the participant in_vehicle from afternoon into evening ) There is one file each with raw Activity Recognition data for testing both iOS and Android data formats. (plugin_google_activity_recognition_raw.csv for android and plugin_ios_activity_recognition_raw.csv for iOS) There is also an additional empty data file for both android and iOS for testing empty data files.","title":"Activity Recognition"},{"location":"developers/test-cases/#conversation","text":"The raw conversation data file contains data for 2 day. The raw conversation data contains records with a sample of both datatypes (i.e. voice/noise = 0 , and conversation = 2 ) as well as rows with for samples of each of the inference values (i.e. silence = 0 , noise = 1 , voice = 2 , and unknown = 3 ) for each epoch . The different datatype and inference records are randomly distributed throughout the epoch . Additionally there are 2 - 5 records for conversations ( datatype = 2, and inference = -1) in each epoch and for each epoch except night, there is a conversation record that has a double_convo_start timestamp that is from the previous epoch . This is to test the calculations of features across epochs . There is a raw conversation data file for both android and iOS platforms ( plugin_studentlife_audio_android_raw.csv and plugin_studentlife_audio_raw.csv respectively). Finally, there are also additional empty data files for both android and iOS for testing empty data files","title":"Conversation"},{"location":"developers/testing/","text":"Testing \u00b6 The following is a simple guide to testing RAPIDS. All files necessary for testing are stored in the /tests directory Steps for Testing \u00b6 To begin testing RAPIDS place the fake raw input data csv files in tests/data/raw/ . The fake participant files should be placed in tests/data/external/ . The expected output files of RAPIDS after processing the input data should be placed in tests/data/processesd/ . The Snakemake rule(s) that are to be tested must be placed in the tests/Snakemake file. The current tests/Snakemake is a good example of how to define them. (At the time of writing this documentation the snakefile contains rules messages (SMS), calls and screen) Edit the tests/settings/config.yaml . Add and/or remove the rules to be run for testing from the forcerun list. Edit the tests/settings/testing_config.yaml with the necessary configuration settings for running the rules to be tested. Add any additional testscripts in tests/scripts . Uncomment or comment off lines in the testing shell script tests/scripts/run_tests.sh . Run the testing shell script. tests/scripts/run_tests.sh The following is a snippet of the output you should see after running your test. test_sensors_files_exist ( test_sensor_features.TestSensorFeatures ) ... ok test_sensors_features_calculations ( test_sensor_features.TestSensorFeatures ) ... FAIL ====================================================================== FAIL: test_sensors_features_calculations ( test_sensor_features.TestSensorFeatures ) ---------------------------------------------------------------------- The results above show that the first test test_sensors_files_exist passed while test_sensors_features_calculations failed. In addition you should get the traceback of the failure (not shown here). For more information on how to implement test scripts and use unittest please see Unittest Documentation Testing of the RAPIDS sensors and features is a work-in-progress. Please see test-cases for a list of sensors and features that have testing currently available. Currently the repository is set up to test a number of sensors out of the box by simply running the tests/scripts/run_tests.sh command once the RAPIDS python environment is active.","title":"Testing"},{"location":"developers/testing/#testing","text":"The following is a simple guide to testing RAPIDS. All files necessary for testing are stored in the /tests directory","title":"Testing"},{"location":"developers/testing/#steps-for-testing","text":"To begin testing RAPIDS place the fake raw input data csv files in tests/data/raw/ . The fake participant files should be placed in tests/data/external/ . The expected output files of RAPIDS after processing the input data should be placed in tests/data/processesd/ . The Snakemake rule(s) that are to be tested must be placed in the tests/Snakemake file. The current tests/Snakemake is a good example of how to define them. (At the time of writing this documentation the snakefile contains rules messages (SMS), calls and screen) Edit the tests/settings/config.yaml . Add and/or remove the rules to be run for testing from the forcerun list. Edit the tests/settings/testing_config.yaml with the necessary configuration settings for running the rules to be tested. Add any additional testscripts in tests/scripts . Uncomment or comment off lines in the testing shell script tests/scripts/run_tests.sh . Run the testing shell script. tests/scripts/run_tests.sh The following is a snippet of the output you should see after running your test. test_sensors_files_exist ( test_sensor_features.TestSensorFeatures ) ... ok test_sensors_features_calculations ( test_sensor_features.TestSensorFeatures ) ... FAIL ====================================================================== FAIL: test_sensors_features_calculations ( test_sensor_features.TestSensorFeatures ) ---------------------------------------------------------------------- The results above show that the first test test_sensors_files_exist passed while test_sensors_features_calculations failed. In addition you should get the traceback of the failure (not shown here). For more information on how to implement test scripts and use unittest please see Unittest Documentation Testing of the RAPIDS sensors and features is a work-in-progress. Please see test-cases for a list of sensors and features that have testing currently available. Currently the repository is set up to test a number of sensors out of the box by simply running the tests/scripts/run_tests.sh command once the RAPIDS python environment is active.","title":"Steps for Testing"},{"location":"developers/virtual-environments/","text":"Virtual Environments \u00b6 Add new packages \u00b6 Try to install any new package using conda . If a package is not available in one of conda \u2018s channels you can install it with pip but make sure your virtual environment is active. Update your conda environment.yaml \u00b6 After installing a new package you can use the following command in your terminal to update your environment.yaml before publishing your pipeline. Note that we ignore the package version for libfortran to keep compatibility with Linux: conda env export --no-builds | sed 's/^.*libgfortran.*$/ - libgfortran/' > environment.yml Update and prune your conda environment from a environment.yaml file \u00b6 Execute the following command in your terminal, see these docs for more information conda env update --prefix ./env --file environment.yml --prune","title":"Virtual Environments"},{"location":"developers/virtual-environments/#virtual-environments","text":"","title":"Virtual Environments"},{"location":"developers/virtual-environments/#add-new-packages","text":"Try to install any new package using conda . If a package is not available in one of conda \u2018s channels you can install it with pip but make sure your virtual environment is active.","title":"Add new packages"},{"location":"developers/virtual-environments/#update-your-conda-environmentyaml","text":"After installing a new package you can use the following command in your terminal to update your environment.yaml before publishing your pipeline. Note that we ignore the package version for libfortran to keep compatibility with Linux: conda env export --no-builds | sed 's/^.*libgfortran.*$/ - libgfortran/' > environment.yml","title":"Update your conda environment.yaml"},{"location":"developers/virtual-environments/#update-and-prune-your-conda-environment-from-a-environmentyaml-file","text":"Execute the following command in your terminal, see these docs for more information conda env update --prefix ./env --file environment.yml --prune","title":"Update and prune your conda environment from a environment.yaml file"},{"location":"features/add-new-features/","text":"Add New Features \u00b6 Hint We recommend reading the Behavioral Features Introduction before reading this page Hint You won\u2019t have to deal with time zones, dates, times, data cleaning or preprocessing. The data that RAPIDS pipes to your feature extraction code is ready to process. New Features for Existing Sensors \u00b6 You can add new features to any existing sensors (see list below) by adding a new provider in three steps: Modify the config.yaml file Create a provider folder, script and function Implement your features extraction code As a tutorial, we will add a new provider for PHONE_ACCELEROMETER called VEGA that extracts feature1 , feature2 , feature3 in Python and that it requires a parameter from the user called MY_PARAMETER . Existing Sensors An existing sensor is any of the phone or Fitbit sensors with a configuration entry in config.yaml : Phone Accelerometer Phone Activity Recognition Phone Applications Foreground Phone Battery Phone Bluetooth Phone Calls Phone Conversation Phone Data Yield Phone Light Phone Locations Phone Messages Phone Screen Phone WiFI Connected Phone WiFI Visible Fitbit Heart Rate Summary Fitbit Heart Rate Intraday Fitbit Sleep Summary Fitbit Steps Summary Fitbit Steps Intraday Modify the config.yaml file \u00b6 In this step you need to add your provider configuration section under the relevant sensor in config.yaml . See our example for our tutorial\u2019s VEGA provider for PHONE_ACCELEROMETER : Example configuration for a new accelerometer provider VEGA PHONE_ACCELEROMETER : TABLE : accelerometer PROVIDERS : RAPIDS : COMPUTE : False ... PANDA : COMPUTE : False ... VEGA : COMPUTE : False FEATURES : [ \"feature1\" , \"feature2\" , \"feature3\" ] MY_PARAMTER : a_string SRC_FOLDER : \"vega\" SRC_LANGUAGE : \"python\" Key Description [COMPUTE] Flag to activate/deactivate your provider [FEATURES] List of features your provider supports. Your provider code should only return the features on this list [MY_PARAMTER] An arbitrary parameter that our example provider VEGA needs. This can be a boolean, integer, float, string or an array of any of such types. [SRC_LANGUAGE] The programming language of your provider script, it can be python or r , in our example python [SRC_FOLDER] The name of your provider in lower case, in our example vega (this will be the name of your folder in the next step) Create a provider folder, script and function \u00b6 In this step you need to add a folder, script and function for your provider. Create your provider folder under src/feature/DEVICE_SENSOR/YOUR_PROVIDER , in our example src/feature/phone_accelerometer/vega (same as [SRC_FOLDER] in the step above). Create your provider script inside your provider folder, it can be a Python file called main.py or an R file called main.R . Add your provider function in your provider script. The name of such function should be [providername]_features , in our example vega_features Python function def [ providername ] _features ( sensor_data_files , time_segment , provider , filter_data_by_segment , * args , ** kwargs ): R function [ providername ] _ features <- function ( sensor_data , time_segment , provider ) Implement your feature extraction code \u00b6 The provider function that you created in the step above will receive the following parameters: Parameter Description sensor_data_files Path to the CSV file containing the data of a single participant. This data has been cleaned and preprocessed. Your function will be automatically called for each participant in your study (in the [PIDS] array in config.yaml ) time_segment The label of the time segment that should be processed. provider The parameters you configured for your provider in config.yaml will be available in this variable as a dictionary in Python or a list in R. In our example this dictionary contains {MY_PARAMETER:\"a_string\"} filter_data_by_segment Python only. A function that you will use to filter your data. In R this function is already available in the environment. *args Python only. Not used for now **kwargs Python only. Not used for now The code to extract your behavioral features should be implemented in your provider function and in general terms it will have three stages: 1. Read a participant\u2019s data by loading the CSV data stored in the file pointed by sensor_data_files acc_data = pd . read_csv ( sensor_data_files [ \"sensor_data\" ]) Note that phone\u2019s battery, screen, and activity recognition data is given as episodes instead of event rows (for example, start and end timestamps of the periods the phone screen was on) 2. Filter your data to process only those rows that belong to time_segment This step is only one line of code, but to undersand why we need it, keep reading. acc_data = filter_data_by_segment ( acc_data , time_segment ) You should use the filter_data_by_segment() function to process and group those rows that belong to each of the time segments RAPIDS could be configured with . Let\u2019s understand the filter_data_by_segment() function with an example. A RAPIDS user can extract features on any arbitrary time segment . A time segment is a period of time that has a label and one or more instances. For example, the user (or you) could have requested features on a daily, weekly, and week-end basis for p01 . The labels are arbritrary and the instances depend on the days a participant was monitored for: the daily segment could be named my_days and if p01 was monitored for 14 days, it would have 14 instances the weekly segment could be named my_weeks and if p01 was monitored for 14 days, it would have 2 instances. the weekend segment could be named my_weekends and if p01 was monitored for 14 days, it would have 2 instances. For this example, RAPIDS will call your provider function three times for p01 , once where time_segment is my_days , once where time_segment is my_weeks and once where time_segment is my_weekends . In this example not every row in p01 \u2018s data needs to take part in the feature computation for either segment and the rows need to be grouped differently. Thus filter_data_by_segment() comes in handy, it will return a data frame that contains the rows that were logged during a time segment plus an extra column called local_segment . This new column will have as many unique values as time segment instances exist (14, 2, and 2 for our p01 \u2018s my_days , my_weeks , and my_weekends examples). After filtering, you should group the data frame by this column and compute any desired features , for example: acc_features [ \"maxmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . max () The reason RAPIDS does not filter the participant\u2019s data set for you is because your code might need to compute something based on a participant\u2019s complete dataset before computing their features. For example, you might want to identify the number that called a participant the most throughout the study before computing a feature with the number of calls the participant received from this number. 3. Return a data frame with your features After filtering, grouping your data, and computing your features, your provider function should return a data frame that has: One row per time segment instance (e.g. 14 our p01 \u2018s my_days example) The local_segment column added by filter_data_by_segment() One column per feature. By convention the name of your features should only contain letters or numbers ( feature1 ). RAPIDS will automatically add the right sensor and provider prefix ( phone_accelerometr_vega_ ) PHONE_ACCELEROMETER Provider Example For your reference, this a short example of our own provider ( RAPIDS ) for PHONE_ACCELEROMETER that computes five acceleration features def rapids_features ( sensor_data_files , time_segment , provider , filter_data_by_segment , * args , ** kwargs ): acc_data = pd . read_csv ( sensor_data_files [ \"sensor_data\" ]) requested_features = provider [ \"FEATURES\" ] # name of the features this function can compute base_features_names = [ \"maxmagnitude\" , \"minmagnitude\" , \"avgmagnitude\" , \"medianmagnitude\" , \"stdmagnitude\" ] # the subset of requested features this function can compute features_to_compute = list ( set ( requested_features ) & set ( base_features_names )) acc_features = pd . DataFrame ( columns = [ \"local_segment\" ] + features_to_compute ) if not acc_data . empty : acc_data = filter_data_by_segment ( acc_data , time_segment ) if not acc_data . empty : acc_features = pd . DataFrame () # get magnitude related features: magnitude = sqrt(x^2+y^2+z^2) magnitude = acc_data . apply ( lambda row : np . sqrt ( row [ \"double_values_0\" ] ** 2 + row [ \"double_values_1\" ] ** 2 + row [ \"double_values_2\" ] ** 2 ), axis = 1 ) acc_data = acc_data . assign ( magnitude = magnitude . values ) if \"maxmagnitude\" in features_to_compute : acc_features [ \"maxmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . max () if \"minmagnitude\" in features_to_compute : acc_features [ \"minmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . min () if \"avgmagnitude\" in features_to_compute : acc_features [ \"avgmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . mean () if \"medianmagnitude\" in features_to_compute : acc_features [ \"medianmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . median () if \"stdmagnitude\" in features_to_compute : acc_features [ \"stdmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . std () acc_features = acc_features . reset_index () return acc_features New Features for Non-Existing Sensors \u00b6 If you want to add features for a device or a sensor that we do not support at the moment (those that do not appear in the \"Existing Sensors\" list above), contact us or request it on Slack and we can add the necessary code so you can follow the instructions above.","title":"Add New Features"},{"location":"features/add-new-features/#add-new-features","text":"Hint We recommend reading the Behavioral Features Introduction before reading this page Hint You won\u2019t have to deal with time zones, dates, times, data cleaning or preprocessing. The data that RAPIDS pipes to your feature extraction code is ready to process.","title":"Add New Features"},{"location":"features/add-new-features/#new-features-for-existing-sensors","text":"You can add new features to any existing sensors (see list below) by adding a new provider in three steps: Modify the config.yaml file Create a provider folder, script and function Implement your features extraction code As a tutorial, we will add a new provider for PHONE_ACCELEROMETER called VEGA that extracts feature1 , feature2 , feature3 in Python and that it requires a parameter from the user called MY_PARAMETER . Existing Sensors An existing sensor is any of the phone or Fitbit sensors with a configuration entry in config.yaml : Phone Accelerometer Phone Activity Recognition Phone Applications Foreground Phone Battery Phone Bluetooth Phone Calls Phone Conversation Phone Data Yield Phone Light Phone Locations Phone Messages Phone Screen Phone WiFI Connected Phone WiFI Visible Fitbit Heart Rate Summary Fitbit Heart Rate Intraday Fitbit Sleep Summary Fitbit Steps Summary Fitbit Steps Intraday","title":"New Features for Existing Sensors"},{"location":"features/add-new-features/#modify-the-configyaml-file","text":"In this step you need to add your provider configuration section under the relevant sensor in config.yaml . See our example for our tutorial\u2019s VEGA provider for PHONE_ACCELEROMETER : Example configuration for a new accelerometer provider VEGA PHONE_ACCELEROMETER : TABLE : accelerometer PROVIDERS : RAPIDS : COMPUTE : False ... PANDA : COMPUTE : False ... VEGA : COMPUTE : False FEATURES : [ \"feature1\" , \"feature2\" , \"feature3\" ] MY_PARAMTER : a_string SRC_FOLDER : \"vega\" SRC_LANGUAGE : \"python\" Key Description [COMPUTE] Flag to activate/deactivate your provider [FEATURES] List of features your provider supports. Your provider code should only return the features on this list [MY_PARAMTER] An arbitrary parameter that our example provider VEGA needs. This can be a boolean, integer, float, string or an array of any of such types. [SRC_LANGUAGE] The programming language of your provider script, it can be python or r , in our example python [SRC_FOLDER] The name of your provider in lower case, in our example vega (this will be the name of your folder in the next step)","title":"Modify the config.yaml file"},{"location":"features/add-new-features/#create-a-provider-folder-script-and-function","text":"In this step you need to add a folder, script and function for your provider. Create your provider folder under src/feature/DEVICE_SENSOR/YOUR_PROVIDER , in our example src/feature/phone_accelerometer/vega (same as [SRC_FOLDER] in the step above). Create your provider script inside your provider folder, it can be a Python file called main.py or an R file called main.R . Add your provider function in your provider script. The name of such function should be [providername]_features , in our example vega_features Python function def [ providername ] _features ( sensor_data_files , time_segment , provider , filter_data_by_segment , * args , ** kwargs ): R function [ providername ] _ features <- function ( sensor_data , time_segment , provider )","title":"Create a provider folder, script and function"},{"location":"features/add-new-features/#implement-your-feature-extraction-code","text":"The provider function that you created in the step above will receive the following parameters: Parameter Description sensor_data_files Path to the CSV file containing the data of a single participant. This data has been cleaned and preprocessed. Your function will be automatically called for each participant in your study (in the [PIDS] array in config.yaml ) time_segment The label of the time segment that should be processed. provider The parameters you configured for your provider in config.yaml will be available in this variable as a dictionary in Python or a list in R. In our example this dictionary contains {MY_PARAMETER:\"a_string\"} filter_data_by_segment Python only. A function that you will use to filter your data. In R this function is already available in the environment. *args Python only. Not used for now **kwargs Python only. Not used for now The code to extract your behavioral features should be implemented in your provider function and in general terms it will have three stages: 1. Read a participant\u2019s data by loading the CSV data stored in the file pointed by sensor_data_files acc_data = pd . read_csv ( sensor_data_files [ \"sensor_data\" ]) Note that phone\u2019s battery, screen, and activity recognition data is given as episodes instead of event rows (for example, start and end timestamps of the periods the phone screen was on) 2. Filter your data to process only those rows that belong to time_segment This step is only one line of code, but to undersand why we need it, keep reading. acc_data = filter_data_by_segment ( acc_data , time_segment ) You should use the filter_data_by_segment() function to process and group those rows that belong to each of the time segments RAPIDS could be configured with . Let\u2019s understand the filter_data_by_segment() function with an example. A RAPIDS user can extract features on any arbitrary time segment . A time segment is a period of time that has a label and one or more instances. For example, the user (or you) could have requested features on a daily, weekly, and week-end basis for p01 . The labels are arbritrary and the instances depend on the days a participant was monitored for: the daily segment could be named my_days and if p01 was monitored for 14 days, it would have 14 instances the weekly segment could be named my_weeks and if p01 was monitored for 14 days, it would have 2 instances. the weekend segment could be named my_weekends and if p01 was monitored for 14 days, it would have 2 instances. For this example, RAPIDS will call your provider function three times for p01 , once where time_segment is my_days , once where time_segment is my_weeks and once where time_segment is my_weekends . In this example not every row in p01 \u2018s data needs to take part in the feature computation for either segment and the rows need to be grouped differently. Thus filter_data_by_segment() comes in handy, it will return a data frame that contains the rows that were logged during a time segment plus an extra column called local_segment . This new column will have as many unique values as time segment instances exist (14, 2, and 2 for our p01 \u2018s my_days , my_weeks , and my_weekends examples). After filtering, you should group the data frame by this column and compute any desired features , for example: acc_features [ \"maxmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . max () The reason RAPIDS does not filter the participant\u2019s data set for you is because your code might need to compute something based on a participant\u2019s complete dataset before computing their features. For example, you might want to identify the number that called a participant the most throughout the study before computing a feature with the number of calls the participant received from this number. 3. Return a data frame with your features After filtering, grouping your data, and computing your features, your provider function should return a data frame that has: One row per time segment instance (e.g. 14 our p01 \u2018s my_days example) The local_segment column added by filter_data_by_segment() One column per feature. By convention the name of your features should only contain letters or numbers ( feature1 ). RAPIDS will automatically add the right sensor and provider prefix ( phone_accelerometr_vega_ ) PHONE_ACCELEROMETER Provider Example For your reference, this a short example of our own provider ( RAPIDS ) for PHONE_ACCELEROMETER that computes five acceleration features def rapids_features ( sensor_data_files , time_segment , provider , filter_data_by_segment , * args , ** kwargs ): acc_data = pd . read_csv ( sensor_data_files [ \"sensor_data\" ]) requested_features = provider [ \"FEATURES\" ] # name of the features this function can compute base_features_names = [ \"maxmagnitude\" , \"minmagnitude\" , \"avgmagnitude\" , \"medianmagnitude\" , \"stdmagnitude\" ] # the subset of requested features this function can compute features_to_compute = list ( set ( requested_features ) & set ( base_features_names )) acc_features = pd . DataFrame ( columns = [ \"local_segment\" ] + features_to_compute ) if not acc_data . empty : acc_data = filter_data_by_segment ( acc_data , time_segment ) if not acc_data . empty : acc_features = pd . DataFrame () # get magnitude related features: magnitude = sqrt(x^2+y^2+z^2) magnitude = acc_data . apply ( lambda row : np . sqrt ( row [ \"double_values_0\" ] ** 2 + row [ \"double_values_1\" ] ** 2 + row [ \"double_values_2\" ] ** 2 ), axis = 1 ) acc_data = acc_data . assign ( magnitude = magnitude . values ) if \"maxmagnitude\" in features_to_compute : acc_features [ \"maxmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . max () if \"minmagnitude\" in features_to_compute : acc_features [ \"minmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . min () if \"avgmagnitude\" in features_to_compute : acc_features [ \"avgmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . mean () if \"medianmagnitude\" in features_to_compute : acc_features [ \"medianmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . median () if \"stdmagnitude\" in features_to_compute : acc_features [ \"stdmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . std () acc_features = acc_features . reset_index () return acc_features","title":"Implement your feature extraction code"},{"location":"features/add-new-features/#new-features-for-non-existing-sensors","text":"If you want to add features for a device or a sensor that we do not support at the moment (those that do not appear in the \"Existing Sensors\" list above), contact us or request it on Slack and we can add the necessary code so you can follow the instructions above.","title":"New Features for Non-Existing Sensors"},{"location":"features/feature-introduction/","text":"Behavioral Features Introduction \u00b6 Every phone or Fitbit sensor has a corresponding config section in config.yaml , these sections follow a similar structure and we\u2019ll use PHONE_ACCELEROMETER as an example to explain this structure. Hint We recommend reading this page if you are using RAPIDS for the first time All computed sensor features are stored under /data/processed/features on files per sensor, per participant and per study (all participants). Every time you change any sensor parameters, provider parameters or provider features, all the necessary files will be updated as soon as you execute RAPIDS. Config section example for PHONE_ACCELEROMETER # 1) Config section PHONE_ACCELEROMETER : # 2) Parameters for PHONE_ACCELEROMETER TABLE : accelerometer # 3) Providers for PHONE_ACCELEROMETER PROVIDERS : # 4) RAPIDS provider RAPIDS : # 4.1) Parameters of RAPIDS provider of PHONE_ACCELEROMETER COMPUTE : False # 4.2) Features of RAPIDS provider of PHONE_ACCELEROMETER FEATURES : [ \"maxmagnitude\" , \"minmagnitude\" , \"avgmagnitude\" , \"medianmagnitude\" , \"stdmagnitude\" ] SRC_FOLDER : \"rapids\" # inside src/features/phone_accelerometer SRC_LANGUAGE : \"python\" # 5) PANDA provider PANDA : # 5.1) Parameters of PANDA provider of PHONE_ACCELEROMETER COMPUTE : False VALID_SENSED_MINUTES : False # 5.2) Features of PANDA provider of PHONE_ACCELEROMETER FEATURES : exertional_activity_episode : [ \"sumduration\" , \"maxduration\" , \"minduration\" , \"avgduration\" , \"medianduration\" , \"stdduration\" ] nonexertional_activity_episode : [ \"sumduration\" , \"maxduration\" , \"minduration\" , \"avgduration\" , \"medianduration\" , \"stdduration\" ] SRC_FOLDER : \"panda\" # inside src/features/phone_accelerometer SRC_LANGUAGE : \"python\" Sensor Parameters \u00b6 Each sensor configuration section has a \u201cparameters\u201d subsection (see #2 in the example). These are parameters that affect different aspects of how the raw data is downloaded, and processed. The TABLE parameter exists for every sensor, but some sensors will have extra parameters like [PHONE_LOCATIONS] . We explain these parameters in a table at the top of each sensor documentation page. Sensor Providers \u00b6 Each sensor configuration section can have zero, one or more behavioral feature providers (see #3 in the example). A provider is a script created by the core RAPIDS team or other researchers that extracts behavioral features for that sensor. In this example, accelerometer has two providers: RAPIDS (see #4 ) and PANDA (see #5 ). Provider Parameters \u00b6 Each provider has parameters that affect the computation of the behavioral features it offers (see #4.1 or #5.1 in the example). These parameters will include at least a [COMPUTE] flag that you switch to True to extract a provider\u2019s behavioral features. We explain every provider\u2019s parameter in a table under the Parameters description heading on each provider documentation page. Provider Features \u00b6 Each provider offers a set of behavioral features (see #4.2 or #5.2 in the example). For some providers these features are grouped in an array (like those for RAPIDS provider in #4.2 ) but for others they are grouped in a collection of arrays depending on the meaning and purpose of those features (like those for PANDAS provider in #5.2 ). In either case, you can delete the features you are not interested in and they will not be included in the sensor\u2019s output feature file. We explain each behavioral feature in a table under the Features description heading on each provider documentation page.","title":"Introduction"},{"location":"features/feature-introduction/#behavioral-features-introduction","text":"Every phone or Fitbit sensor has a corresponding config section in config.yaml , these sections follow a similar structure and we\u2019ll use PHONE_ACCELEROMETER as an example to explain this structure. Hint We recommend reading this page if you are using RAPIDS for the first time All computed sensor features are stored under /data/processed/features on files per sensor, per participant and per study (all participants). Every time you change any sensor parameters, provider parameters or provider features, all the necessary files will be updated as soon as you execute RAPIDS. Config section example for PHONE_ACCELEROMETER # 1) Config section PHONE_ACCELEROMETER : # 2) Parameters for PHONE_ACCELEROMETER TABLE : accelerometer # 3) Providers for PHONE_ACCELEROMETER PROVIDERS : # 4) RAPIDS provider RAPIDS : # 4.1) Parameters of RAPIDS provider of PHONE_ACCELEROMETER COMPUTE : False # 4.2) Features of RAPIDS provider of PHONE_ACCELEROMETER FEATURES : [ \"maxmagnitude\" , \"minmagnitude\" , \"avgmagnitude\" , \"medianmagnitude\" , \"stdmagnitude\" ] SRC_FOLDER : \"rapids\" # inside src/features/phone_accelerometer SRC_LANGUAGE : \"python\" # 5) PANDA provider PANDA : # 5.1) Parameters of PANDA provider of PHONE_ACCELEROMETER COMPUTE : False VALID_SENSED_MINUTES : False # 5.2) Features of PANDA provider of PHONE_ACCELEROMETER FEATURES : exertional_activity_episode : [ \"sumduration\" , \"maxduration\" , \"minduration\" , \"avgduration\" , \"medianduration\" , \"stdduration\" ] nonexertional_activity_episode : [ \"sumduration\" , \"maxduration\" , \"minduration\" , \"avgduration\" , \"medianduration\" , \"stdduration\" ] SRC_FOLDER : \"panda\" # inside src/features/phone_accelerometer SRC_LANGUAGE : \"python\"","title":"Behavioral Features Introduction"},{"location":"features/feature-introduction/#sensor-parameters","text":"Each sensor configuration section has a \u201cparameters\u201d subsection (see #2 in the example). These are parameters that affect different aspects of how the raw data is downloaded, and processed. The TABLE parameter exists for every sensor, but some sensors will have extra parameters like [PHONE_LOCATIONS] . We explain these parameters in a table at the top of each sensor documentation page.","title":"Sensor Parameters"},{"location":"features/feature-introduction/#sensor-providers","text":"Each sensor configuration section can have zero, one or more behavioral feature providers (see #3 in the example). A provider is a script created by the core RAPIDS team or other researchers that extracts behavioral features for that sensor. In this example, accelerometer has two providers: RAPIDS (see #4 ) and PANDA (see #5 ).","title":"Sensor Providers"},{"location":"features/feature-introduction/#provider-parameters","text":"Each provider has parameters that affect the computation of the behavioral features it offers (see #4.1 or #5.1 in the example). These parameters will include at least a [COMPUTE] flag that you switch to True to extract a provider\u2019s behavioral features. We explain every provider\u2019s parameter in a table under the Parameters description heading on each provider documentation page.","title":"Provider Parameters"},{"location":"features/feature-introduction/#provider-features","text":"Each provider offers a set of behavioral features (see #4.2 or #5.2 in the example). For some providers these features are grouped in an array (like those for RAPIDS provider in #4.2 ) but for others they are grouped in a collection of arrays depending on the meaning and purpose of those features (like those for PANDAS provider in #5.2 ). In either case, you can delete the features you are not interested in and they will not be included in the sensor\u2019s output feature file. We explain each behavioral feature in a table under the Features description heading on each provider documentation page.","title":"Provider Features"},{"location":"features/fitbit-heartrate-intraday/","text":"Fitbit Heart Rate Intraday \u00b6 Sensor parameters description for [FITBIT_HEARTRATE_INTRADAY] : Key Description [TABLE] Database table name or file path where the heart rate intraday data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-07\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:1200.6102,\u201dmax\u201d:88,\u201dmin\u201d:31,\u201dminutes\u201d:1058,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:760.3020,\u201dmax\u201d:120,\u201dmin\u201d:86,\u201dminutes\u201d:366,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:15.2048,\u201dmax\u201d:146,\u201dmin\u201d:120,\u201dminutes\u201d:2,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:221,\u201dmin\u201d:148,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:72}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:68},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:67},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:67},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-08\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:1100.1120,\u201dmax\u201d:89,\u201dmin\u201d:30,\u201dminutes\u201d:921,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:660.0012,\u201dmax\u201d:118,\u201dmin\u201d:82,\u201dminutes\u201d:361,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:23.7088,\u201dmax\u201d:142,\u201dmin\u201d:108,\u201dminutes\u201d:3,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:221,\u201dmin\u201d:148,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:70}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:77},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:75},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:73},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-09\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:750.3615,\u201dmax\u201d:77,\u201dmin\u201d:30,\u201dminutes\u201d:851,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:734.1516,\u201dmax\u201d:107,\u201dmin\u201d:77,\u201dminutes\u201d:550,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:131.8579,\u201dmax\u201d:130,\u201dmin\u201d:107,\u201dminutes\u201d:29,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:220,\u201dmin\u201d:130,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:69}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:90},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:89},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:88},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} PLAIN_TEXT device_id local_date_time heartrate heartrate_zone a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:00:00 68 outofrange a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:01:00 67 outofrange a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:02:00 67 outofrange RAPIDS provider \u00b6 Available time segments Available for all time segments File Sequence - data/raw/ { pid } /fitbit_heartrate_intraday_raw.csv - data/raw/ { pid } /fitbit_heartrate_intraday_parsed.csv - data/raw/ { pid } /fitbit_heartrate_intraday_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_heartrate_intraday_features/fitbit_heartrate_intraday_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_heartrate_intraday.csv Parameters description for [FITBIT_HEARTRATE_INTRADAY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_HEARTRATE_INTRADAY features from the RAPIDS provider [FEATURES] Features to be computed from heart rate intraday data, see table below Features description for [FITBIT_HEARTRATE_INTRADAY][PROVIDERS][RAPIDS] : Feature Units Description maxhr beats/mins The maximum heart rate during a time segment. minhr beats/mins The minimum heart rate during a time segment. avghr beats/mins The average heart rate during a time segment. medianhr beats/mins The median of heart rate during a time segment. modehr beats/mins The mode of heart rate during a time segment. stdhr beats/mins The standard deviation of heart rate during a time segment. diffmaxmodehr beats/mins The difference between the maximum and mode heart rate during a time segment. diffminmodehr beats/mins The difference between the mode and minimum heart rate during a time segment. entropyhr nats Shannon\u2019s entropy measurement based on heart rate during a time segment. minutesonZONE minutes Number of minutes the user\u2019s heart rate fell within each heartrate_zone during a time segment. Assumptions/Observations There are four heart rate zones (ZONE): outofrange , fatburn , cardio , and peak . Please refer to Fitbit documentation for more information about the way they are computed.","title":"Fitbit Heart Rate Intraday"},{"location":"features/fitbit-heartrate-intraday/#fitbit-heart-rate-intraday","text":"Sensor parameters description for [FITBIT_HEARTRATE_INTRADAY] : Key Description [TABLE] Database table name or file path where the heart rate intraday data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-07\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:1200.6102,\u201dmax\u201d:88,\u201dmin\u201d:31,\u201dminutes\u201d:1058,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:760.3020,\u201dmax\u201d:120,\u201dmin\u201d:86,\u201dminutes\u201d:366,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:15.2048,\u201dmax\u201d:146,\u201dmin\u201d:120,\u201dminutes\u201d:2,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:221,\u201dmin\u201d:148,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:72}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:68},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:67},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:67},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-08\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:1100.1120,\u201dmax\u201d:89,\u201dmin\u201d:30,\u201dminutes\u201d:921,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:660.0012,\u201dmax\u201d:118,\u201dmin\u201d:82,\u201dminutes\u201d:361,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:23.7088,\u201dmax\u201d:142,\u201dmin\u201d:108,\u201dminutes\u201d:3,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:221,\u201dmin\u201d:148,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:70}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:77},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:75},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:73},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-09\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:750.3615,\u201dmax\u201d:77,\u201dmin\u201d:30,\u201dminutes\u201d:851,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:734.1516,\u201dmax\u201d:107,\u201dmin\u201d:77,\u201dminutes\u201d:550,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:131.8579,\u201dmax\u201d:130,\u201dmin\u201d:107,\u201dminutes\u201d:29,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:220,\u201dmin\u201d:130,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:69}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:90},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:89},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:88},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} PLAIN_TEXT device_id local_date_time heartrate heartrate_zone a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:00:00 68 outofrange a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:01:00 67 outofrange a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:02:00 67 outofrange","title":"Fitbit Heart Rate Intraday"},{"location":"features/fitbit-heartrate-intraday/#rapids-provider","text":"Available time segments Available for all time segments File Sequence - data/raw/ { pid } /fitbit_heartrate_intraday_raw.csv - data/raw/ { pid } /fitbit_heartrate_intraday_parsed.csv - data/raw/ { pid } /fitbit_heartrate_intraday_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_heartrate_intraday_features/fitbit_heartrate_intraday_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_heartrate_intraday.csv Parameters description for [FITBIT_HEARTRATE_INTRADAY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_HEARTRATE_INTRADAY features from the RAPIDS provider [FEATURES] Features to be computed from heart rate intraday data, see table below Features description for [FITBIT_HEARTRATE_INTRADAY][PROVIDERS][RAPIDS] : Feature Units Description maxhr beats/mins The maximum heart rate during a time segment. minhr beats/mins The minimum heart rate during a time segment. avghr beats/mins The average heart rate during a time segment. medianhr beats/mins The median of heart rate during a time segment. modehr beats/mins The mode of heart rate during a time segment. stdhr beats/mins The standard deviation of heart rate during a time segment. diffmaxmodehr beats/mins The difference between the maximum and mode heart rate during a time segment. diffminmodehr beats/mins The difference between the mode and minimum heart rate during a time segment. entropyhr nats Shannon\u2019s entropy measurement based on heart rate during a time segment. minutesonZONE minutes Number of minutes the user\u2019s heart rate fell within each heartrate_zone during a time segment. Assumptions/Observations There are four heart rate zones (ZONE): outofrange , fatburn , cardio , and peak . Please refer to Fitbit documentation for more information about the way they are computed.","title":"RAPIDS provider"},{"location":"features/fitbit-heartrate-summary/","text":"Fitbit Heart Rate Summary \u00b6 Sensor parameters description for [FITBIT_HEARTRATE_SUMMARY] : Key Description [TABLE] Database table name or file path where the heart rate summary data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-07\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:1200.6102,\u201dmax\u201d:88,\u201dmin\u201d:31,\u201dminutes\u201d:1058,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:760.3020,\u201dmax\u201d:120,\u201dmin\u201d:86,\u201dminutes\u201d:366,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:15.2048,\u201dmax\u201d:146,\u201dmin\u201d:120,\u201dminutes\u201d:2,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:221,\u201dmin\u201d:148,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:72}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:68},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:67},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:67},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-08\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:1100.1120,\u201dmax\u201d:89,\u201dmin\u201d:30,\u201dminutes\u201d:921,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:660.0012,\u201dmax\u201d:118,\u201dmin\u201d:82,\u201dminutes\u201d:361,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:23.7088,\u201dmax\u201d:142,\u201dmin\u201d:108,\u201dminutes\u201d:3,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:221,\u201dmin\u201d:148,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:70}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:77},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:75},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:73},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-09\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:750.3615,\u201dmax\u201d:77,\u201dmin\u201d:30,\u201dminutes\u201d:851,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:734.1516,\u201dmax\u201d:107,\u201dmin\u201d:77,\u201dminutes\u201d:550,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:131.8579,\u201dmax\u201d:130,\u201dmin\u201d:107,\u201dminutes\u201d:29,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:220,\u201dmin\u201d:130,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:69}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:90},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:89},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:88},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} PLAIN_TEXT device_id local_date_time heartrate_daily_restinghr heartrate_daily_caloriesoutofrange heartrate_daily_caloriesfatburn heartrate_daily_caloriescardio heartrate_daily_caloriespeak a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 72 1200.6102 760.3020 15.2048 0 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-08 70 1100.1120 660.0012 23.7088 0 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-09 69 750.3615 734.1516 131.8579 0 RAPIDS provider \u00b6 Available time segments Only available for segments that span 1 or more complete days (e.g. Jan 1 st 00:00 to Jan 3 rd 23:59) File Sequence - data/raw/ { pid } /fitbit_heartrate_summary_raw.csv - data/raw/ { pid } /fitbit_heartrate_summary_parsed.csv - data/raw/ { pid } /fitbit_heartrate_summary_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_heartrate_summary_features/fitbit_heartrate_summary_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_heartrate_summary.csv Parameters description for [FITBIT_HEARTRATE_SUMMARY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_HEARTRATE_SUMMARY features from the RAPIDS provider [FEATURES] Features to be computed from heart rate summary data, see table below Features description for [FITBIT_HEARTRATE_SUMMARY][PROVIDERS][RAPIDS] : Feature Units Description maxrestinghr beats/mins The maximum daily resting heart rate during a time segment. minrestinghr beats/mins The minimum daily resting heart rate during a time segment. avgrestinghr beats/mins The average daily resting heart rate during a time segment. medianrestinghr beats/mins The median of daily resting heart rate during a time segment. moderestinghr beats/mins The mode of daily resting heart rate during a time segment. stdrestinghr beats/mins The standard deviation of daily resting heart rate during a time segment. diffmaxmoderestinghr beats/mins The difference between the maximum and mode daily resting heart rate during a time segment. diffminmoderestinghr beats/mins The difference between the mode and minimum daily resting heart rate during a time segment. entropyrestinghr nats Shannon\u2019s entropy measurement based on daily resting heart rate during a time segment. sumcaloriesZONE cals The total daily calories burned within heartrate_zone during a time segment. maxcaloriesZONE cals The maximum daily calories burned within heartrate_zone during a time segment. mincaloriesZONE cals The minimum daily calories burned within heartrate_zone during a time segment. avgcaloriesZONE cals The average daily calories burned within heartrate_zone during a time segment. mediancaloriesZONE cals The median of daily calories burned within heartrate_zone during a time segment. stdcaloriesZONE cals The standard deviation of daily calories burned within heartrate_zone during a time segment. entropycaloriesZONE nats Shannon\u2019s entropy measurement based on daily calories burned within heartrate_zone during a time segment. Assumptions/Observations There are four heart rate zones (ZONE): outofrange , fatburn , cardio , and peak . Please refer to Fitbit documentation for more information about the way they are computed. Calories\u2019 accuracy depends on the users\u2019 Fitbit profile (weight, height, etc.).","title":"Fitbit Heart Rate Summary"},{"location":"features/fitbit-heartrate-summary/#fitbit-heart-rate-summary","text":"Sensor parameters description for [FITBIT_HEARTRATE_SUMMARY] : Key Description [TABLE] Database table name or file path where the heart rate summary data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-07\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:1200.6102,\u201dmax\u201d:88,\u201dmin\u201d:31,\u201dminutes\u201d:1058,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:760.3020,\u201dmax\u201d:120,\u201dmin\u201d:86,\u201dminutes\u201d:366,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:15.2048,\u201dmax\u201d:146,\u201dmin\u201d:120,\u201dminutes\u201d:2,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:221,\u201dmin\u201d:148,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:72}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:68},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:67},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:67},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-08\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:1100.1120,\u201dmax\u201d:89,\u201dmin\u201d:30,\u201dminutes\u201d:921,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:660.0012,\u201dmax\u201d:118,\u201dmin\u201d:82,\u201dminutes\u201d:361,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:23.7088,\u201dmax\u201d:142,\u201dmin\u201d:108,\u201dminutes\u201d:3,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:221,\u201dmin\u201d:148,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:70}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:77},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:75},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:73},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-09\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:750.3615,\u201dmax\u201d:77,\u201dmin\u201d:30,\u201dminutes\u201d:851,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:734.1516,\u201dmax\u201d:107,\u201dmin\u201d:77,\u201dminutes\u201d:550,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:131.8579,\u201dmax\u201d:130,\u201dmin\u201d:107,\u201dminutes\u201d:29,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:220,\u201dmin\u201d:130,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:69}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:90},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:89},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:88},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} PLAIN_TEXT device_id local_date_time heartrate_daily_restinghr heartrate_daily_caloriesoutofrange heartrate_daily_caloriesfatburn heartrate_daily_caloriescardio heartrate_daily_caloriespeak a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 72 1200.6102 760.3020 15.2048 0 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-08 70 1100.1120 660.0012 23.7088 0 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-09 69 750.3615 734.1516 131.8579 0","title":"Fitbit Heart Rate Summary"},{"location":"features/fitbit-heartrate-summary/#rapids-provider","text":"Available time segments Only available for segments that span 1 or more complete days (e.g. Jan 1 st 00:00 to Jan 3 rd 23:59) File Sequence - data/raw/ { pid } /fitbit_heartrate_summary_raw.csv - data/raw/ { pid } /fitbit_heartrate_summary_parsed.csv - data/raw/ { pid } /fitbit_heartrate_summary_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_heartrate_summary_features/fitbit_heartrate_summary_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_heartrate_summary.csv Parameters description for [FITBIT_HEARTRATE_SUMMARY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_HEARTRATE_SUMMARY features from the RAPIDS provider [FEATURES] Features to be computed from heart rate summary data, see table below Features description for [FITBIT_HEARTRATE_SUMMARY][PROVIDERS][RAPIDS] : Feature Units Description maxrestinghr beats/mins The maximum daily resting heart rate during a time segment. minrestinghr beats/mins The minimum daily resting heart rate during a time segment. avgrestinghr beats/mins The average daily resting heart rate during a time segment. medianrestinghr beats/mins The median of daily resting heart rate during a time segment. moderestinghr beats/mins The mode of daily resting heart rate during a time segment. stdrestinghr beats/mins The standard deviation of daily resting heart rate during a time segment. diffmaxmoderestinghr beats/mins The difference between the maximum and mode daily resting heart rate during a time segment. diffminmoderestinghr beats/mins The difference between the mode and minimum daily resting heart rate during a time segment. entropyrestinghr nats Shannon\u2019s entropy measurement based on daily resting heart rate during a time segment. sumcaloriesZONE cals The total daily calories burned within heartrate_zone during a time segment. maxcaloriesZONE cals The maximum daily calories burned within heartrate_zone during a time segment. mincaloriesZONE cals The minimum daily calories burned within heartrate_zone during a time segment. avgcaloriesZONE cals The average daily calories burned within heartrate_zone during a time segment. mediancaloriesZONE cals The median of daily calories burned within heartrate_zone during a time segment. stdcaloriesZONE cals The standard deviation of daily calories burned within heartrate_zone during a time segment. entropycaloriesZONE nats Shannon\u2019s entropy measurement based on daily calories burned within heartrate_zone during a time segment. Assumptions/Observations There are four heart rate zones (ZONE): outofrange , fatburn , cardio , and peak . Please refer to Fitbit documentation for more information about the way they are computed. Calories\u2019 accuracy depends on the users\u2019 Fitbit profile (weight, height, etc.).","title":"RAPIDS provider"},{"location":"features/fitbit-sleep-summary/","text":"Fitbit Sleep Summary \u00b6 Sensor parameters description for [FITBIT_SLEEP_SUMMARY] : Key Description [TABLE] Database table name or file path where the sleep summary data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data with Fitbit\u2019s sleep API Version 1 JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d: [{\u201cawakeCount\u201d: 2, \u201cawakeDuration\u201d: 3, \u201cawakeningsCount\u201d: 10, \u201cdateOfSleep\u201d: \u201c2020-10-07\u201d, \u201cduration\u201d: 8100000, \u201cefficiency\u201d: 91, \u201cendTime\u201d: \u201c2020-10-07T18:10:00.000\u201d, \u201cisMainSleep\u201d: true, \u201clogId\u201d: 14147921940, \u201cminuteData\u201d: [{\u201cdateTime\u201d: \u201c15:55:00\u201d, \u201cvalue\u201d: \u201c3\u201d}, {\u201cdateTime\u201d: \u201c15:56:00\u201d, \u201cvalue\u201d: \u201c3\u201d}, {\u201cdateTime\u201d: \u201c15:57:00\u201d, \u201cvalue\u201d: \u201c2\u201d},\u2026], \u201cminutesAfterWakeup\u201d: 0, \u201cminutesAsleep\u201d: 123, \u201cminutesAwake\u201d: 12, \u201cminutesToFallAsleep\u201d: 0, \u201crestlessCount\u201d: 8, \u201crestlessDuration\u201d: 9, \u201cstartTime\u201d: \u201c2020-10-07T15:55:00.000\u201d, \u201ctimeInBed\u201d: 135}, {\u201cawakeCount\u201d: 0, \u201cawakeDuration\u201d: 0, \u201cawakeningsCount\u201d: 1, \u201cdateOfSleep\u201d: \u201c2020-10-07\u201d, \u201cduration\u201d: 3780000, \u201cefficiency\u201d: 100, \u201cendTime\u201d: \u201c2020-10-07T10:52:30.000\u201d, \u201cisMainSleep\u201d: false, \u201clogId\u201d: 14144903977, \u201cminuteData\u201d: [{\u201cdateTime\u201d: \u201c09:49:00\u201d, \u201cvalue\u201d: \u201c1\u201d}, {\u201cdateTime\u201d: \u201c09:50:00\u201d, \u201cvalue\u201d: \u201c1\u201d}, {\u201cdateTime\u201d: \u201c09:51:00\u201d, \u201cvalue\u201d: \u201c1\u201d},\u2026], \u201cminutesAfterWakeup\u201d: 1, \u201cminutesAsleep\u201d: 62, \u201cminutesAwake\u201d: 0, \u201cminutesToFallAsleep\u201d: 0, \u201crestlessCount\u201d: 1, \u201crestlessDuration\u201d: 1, \u201cstartTime\u201d: \u201c2020-10-07T09:49:00.000\u201d, \u201ctimeInBed\u201d: 63}], \u201csummary\u201d: {\u201ctotalMinutesAsleep\u201d: 185, \u201ctotalSleepRecords\u201d: 2, \u201ctotalTimeInBed\u201d: 198}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d: [{\u201cawakeCount\u201d: 3, \u201cawakeDuration\u201d: 21, \u201cawakeningsCount\u201d: 16, \u201cdateOfSleep\u201d: \u201c2020-10-08\u201d, \u201cduration\u201d: 19260000, \u201cefficiency\u201d: 89, \u201cendTime\u201d: \u201c2020-10-08T06:01:30.000\u201d, \u201cisMainSleep\u201d: true, \u201clogId\u201d: 14150613895, \u201cminuteData\u201d: [{\u201cdateTime\u201d: \u201c00:40:00\u201d, \u201cvalue\u201d: \u201c3\u201d}, {\u201cdateTime\u201d: \u201c00:41:00\u201d, \u201cvalue\u201d: \u201c3\u201d}, {\u201cdateTime\u201d: \u201c00:42:00\u201d, \u201cvalue\u201d: \u201c3\u201d},\u2026], \u201cminutesAfterWakeup\u201d: 0, \u201cminutesAsleep\u201d: 275, \u201cminutesAwake\u201d: 33, \u201cminutesToFallAsleep\u201d: 0, \u201crestlessCount\u201d: 13, \u201crestlessDuration\u201d: 25, \u201cstartTime\u201d: \u201c2020-10-08T00:40:00.000\u201d, \u201ctimeInBed\u201d: 321}], \u201csummary\u201d: {\u201ctotalMinutesAsleep\u201d: 275, \u201ctotalSleepRecords\u201d: 1, \u201ctotalTimeInBed\u201d: 321}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d: [{\u201cawakeCount\u201d: 1, \u201cawakeDuration\u201d: 3, \u201cawakeningsCount\u201d: 8, \u201cdateOfSleep\u201d: \u201c2020-10-09\u201d, \u201cduration\u201d: 19320000, \u201cefficiency\u201d: 96, \u201cendTime\u201d: \u201c2020-10-09T05:57:30.000\u201d, \u201cisMainSleep\u201d: true, \u201clogId\u201d: 14161136803, \u201cminuteData\u201d: [{\u201cdateTime\u201d: \u201c00:35:30\u201d, \u201cvalue\u201d: \u201c2\u201d}, {\u201cdateTime\u201d: \u201c00:36:30\u201d, \u201cvalue\u201d: \u201c1\u201d}, {\u201cdateTime\u201d: \u201c00:37:30\u201d, \u201cvalue\u201d: \u201c1\u201d},\u2026], \u201cminutesAfterWakeup\u201d: 0, \u201cminutesAsleep\u201d: 309, \u201cminutesAwake\u201d: 13, \u201cminutesToFallAsleep\u201d: 0, \u201crestlessCount\u201d: 7, \u201crestlessDuration\u201d: 10, \u201cstartTime\u201d: \u201c2020-10-09T00:35:30.000\u201d, \u201ctimeInBed\u201d: 322}], \u201csummary\u201d: {\u201ctotalMinutesAsleep\u201d: 309, \u201ctotalSleepRecords\u201d: 1, \u201ctotalTimeInBed\u201d: 322}} PLAIN_TEXT device_id local_start_date_time local_end_date_time efficiency minutes_after_wakeup minutes_asleep minutes_awake minutes_to_fall_asleep minutes_in_bed is_main_sleep type count_awake duration_awake count_awakenings count_restless duration_restless a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 15:55:00 2020-10-07 18:10:00 91 0 123 12 0 135 1 classic 2 3 10 8 9 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 09:49:00 2020-10-07 10:52:30 100 1 62 0 0 63 0 classic 0 0 1 1 1 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-08 00:40:00 2020-10-08 06:01:30 89 0 275 33 0 321 1 classic 3 21 16 13 25 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-09 00:35:30 2020-10-09 05:57:30 96 0 309 13 0 322 1 classic 1 3 8 7 10 Example of the structure of source data with Fitbit\u2019s sleep API Version 1.2 JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d:[{\u201cdateOfSleep\u201d:\u201d2020-10-10\u201d,\u201dduration\u201d:3600000,\u201defficiency\u201d:92,\u201dendTime\u201d:\u201d2020-10-10T16:37:00.000\u201d,\u201dinfoCode\u201d:2,\u201disMainSleep\u201d:false,\u201dlevels\u201d:{\u201cdata\u201d:[{\u201cdateTime\u201d:\u201d2020-10-10T15:36:30.000\u201d,\u201dlevel\u201d:\u201drestless\u201d,\u201dseconds\u201d:60},{\u201cdateTime\u201d:\u201d2020-10-10T15:37:30.000\u201d,\u201dlevel\u201d:\u201dasleep\u201d,\u201dseconds\u201d:660},{\u201cdateTime\u201d:\u201d2020-10-10T15:48:30.000\u201d,\u201dlevel\u201d:\u201drestless\u201d,\u201dseconds\u201d:60},\u2026], \u201csummary\u201d:{\u201casleep\u201d:{\u201ccount\u201d:0,\u201dminutes\u201d:56},\u201dawake\u201d:{\u201ccount\u201d:0,\u201dminutes\u201d:0},\u201drestless\u201d:{\u201ccount\u201d:3,\u201dminutes\u201d:4}}},\u201dlogId\u201d:26315914306,\u201dminutesAfterWakeup\u201d:0,\u201dminutesAsleep\u201d:55,\u201dminutesAwake\u201d:5,\u201dminutesToFallAsleep\u201d:0,\u201dstartTime\u201d:\u201d2020-10-10T15:36:30.000\u201d,\u201dtimeInBed\u201d:60,\u201dtype\u201d:\u201dclassic\u201d},{\u201cdateOfSleep\u201d:\u201d2020-10-10\u201d,\u201dduration\u201d:22980000,\u201defficiency\u201d:88,\u201dendTime\u201d:\u201d2020-10-10T08:10:00.000\u201d,\u201dinfoCode\u201d:0,\u201disMainSleep\u201d:true,\u201dlevels\u201d:{\u201cdata\u201d:[{\u201cdateTime\u201d:\u201d2020-10-10T01:46:30.000\u201d,\u201dlevel\u201d:\u201dlight\u201d,\u201dseconds\u201d:420},{\u201cdateTime\u201d:\u201d2020-10-10T01:53:30.000\u201d,\u201dlevel\u201d:\u201ddeep\u201d,\u201dseconds\u201d:1230},{\u201cdateTime\u201d:\u201d2020-10-10T02:14:00.000\u201d,\u201dlevel\u201d:\u201dlight\u201d,\u201dseconds\u201d:360},\u2026], \u201csummary\u201d:{\u201cdeep\u201d:{\u201ccount\u201d:3,\u201dminutes\u201d:92,\u201dthirtyDayAvgMinutes\u201d:0},\u201dlight\u201d:{\u201ccount\u201d:29,\u201dminutes\u201d:193,\u201dthirtyDayAvgMinutes\u201d:0},\u201drem\u201d:{\u201ccount\u201d:4,\u201dminutes\u201d:33,\u201dthirtyDayAvgMinutes\u201d:0},\u201dwake\u201d:{\u201ccount\u201d:28,\u201dminutes\u201d:65,\u201dthirtyDayAvgMinutes\u201d:0}}},\u201dlogId\u201d:26311786557,\u201dminutesAfterWakeup\u201d:0,\u201dminutesAsleep\u201d:318,\u201dminutesAwake\u201d:65,\u201dminutesToFallAsleep\u201d:0,\u201dstartTime\u201d:\u201d2020-10-10T01:46:30.000\u201d,\u201dtimeInBed\u201d:383,\u201dtype\u201d:\u201dstages\u201d}],\u201dsummary\u201d:{\u201cstages\u201d:{\u201cdeep\u201d:92,\u201dlight\u201d:193,\u201drem\u201d:33,\u201dwake\u201d:65},\u201dtotalMinutesAsleep\u201d:373,\u201dtotalSleepRecords\u201d:2,\u201dtotalTimeInBed\u201d:443}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d:[{\u201cdateOfSleep\u201d:\u201d2020-10-11\u201d,\u201dduration\u201d:41640000,\u201defficiency\u201d:89,\u201dendTime\u201d:\u201d2020-10-11T11:47:00.000\u201d,\u201dinfoCode\u201d:0,\u201disMainSleep\u201d:true,\u201dlevels\u201d:{\u201cdata\u201d:[{\u201cdateTime\u201d:\u201d2020-10-11T00:12:30.000\u201d,\u201dlevel\u201d:\u201dwake\u201d,\u201dseconds\u201d:450},{\u201cdateTime\u201d:\u201d2020-10-11T00:20:00.000\u201d,\u201dlevel\u201d:\u201dlight\u201d,\u201dseconds\u201d:870},{\u201cdateTime\u201d:\u201d2020-10-11T00:34:30.000\u201d,\u201dlevel\u201d:\u201dwake\u201d,\u201dseconds\u201d:780},\u2026], \u201csummary\u201d:{\u201cdeep\u201d:{\u201ccount\u201d:4,\u201dminutes\u201d:52,\u201dthirtyDayAvgMinutes\u201d:62},\u201dlight\u201d:{\u201ccount\u201d:32,\u201dminutes\u201d:442,\u201dthirtyDayAvgMinutes\u201d:364},\u201drem\u201d:{\u201ccount\u201d:6,\u201dminutes\u201d:68,\u201dthirtyDayAvgMinutes\u201d:58},\u201dwake\u201d:{\u201ccount\u201d:29,\u201dminutes\u201d:132,\u201dthirtyDayAvgMinutes\u201d:94}}},\u201dlogId\u201d:26589710670,\u201dminutesAfterWakeup\u201d:1,\u201dminutesAsleep\u201d:562,\u201dminutesAwake\u201d:132,\u201dminutesToFallAsleep\u201d:0,\u201dstartTime\u201d:\u201d2020-10-11T00:12:30.000\u201d,\u201dtimeInBed\u201d:694,\u201dtype\u201d:\u201dstages\u201d}],\u201dsummary\u201d:{\u201cstages\u201d:{\u201cdeep\u201d:52,\u201dlight\u201d:442,\u201drem\u201d:68,\u201dwake\u201d:132},\u201dtotalMinutesAsleep\u201d:562,\u201dtotalSleepRecords\u201d:1,\u201dtotalTimeInBed\u201d:694}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d:[{\u201cdateOfSleep\u201d:\u201d2020-10-12\u201d,\u201dduration\u201d:28980000,\u201defficiency\u201d:93,\u201dendTime\u201d:\u201d2020-10-12T09:34:30.000\u201d,\u201dinfoCode\u201d:0,\u201disMainSleep\u201d:true,\u201dlevels\u201d:{\u201cdata\u201d:[{\u201cdateTime\u201d:\u201d2020-10-12T01:31:00.000\u201d,\u201dlevel\u201d:\u201dwake\u201d,\u201dseconds\u201d:600},{\u201cdateTime\u201d:\u201d2020-10-12T01:41:00.000\u201d,\u201dlevel\u201d:\u201dlight\u201d,\u201dseconds\u201d:60},{\u201cdateTime\u201d:\u201d2020-10-12T01:42:00.000\u201d,\u201dlevel\u201d:\u201ddeep\u201d,\u201dseconds\u201d:2340},\u2026], \u201csummary\u201d:{\u201cdeep\u201d:{\u201ccount\u201d:4,\u201dminutes\u201d:63,\u201dthirtyDayAvgMinutes\u201d:59},\u201dlight\u201d:{\u201ccount\u201d:27,\u201dminutes\u201d:257,\u201dthirtyDayAvgMinutes\u201d:364},\u201drem\u201d:{\u201ccount\u201d:5,\u201dminutes\u201d:94,\u201dthirtyDayAvgMinutes\u201d:58},\u201dwake\u201d:{\u201ccount\u201d:24,\u201dminutes\u201d:69,\u201dthirtyDayAvgMinutes\u201d:95}}},\u201dlogId\u201d:26589710673,\u201dminutesAfterWakeup\u201d:0,\u201dminutesAsleep\u201d:415,\u201dminutesAwake\u201d:68,\u201dminutesToFallAsleep\u201d:0,\u201dstartTime\u201d:\u201d2020-10-12T01:31:00.000\u201d,\u201dtimeInBed\u201d:483,\u201dtype\u201d:\u201dstages\u201d}],\u201dsummary\u201d:{\u201cstages\u201d:{\u201cdeep\u201d:63,\u201dlight\u201d:257,\u201drem\u201d:94,\u201dwake\u201d:69},\u201dtotalMinutesAsleep\u201d:415,\u201dtotalSleepRecords\u201d:1,\u201dtotalTimeInBed\u201d:483}} PLAIN_TEXT device_id local_start_date_time local_end_date_time efficiency minutes_after_wakeup minutes_asleep minutes_awake minutes_to_fall_asleep minutes_in_bed is_main_sleep type a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-10 15:36:30 2020-10-10 16:37:00 92 0 55 5 0 60 0 classic a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-10 01:46:30 2020-10-10 08:10:00 88 0 318 65 0 383 1 stages a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-11 00:12:30 2020-10-11 11:47:00 89 1 562 132 0 694 1 stages a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-12 01:31:00 2020-10-12 09:34:30 93 0 415 68 0 483 1 stages RAPIDS provider \u00b6 Available time segments Only available for segments that span 1 or more complete days (e.g. Jan 1 st 00:00 to Jan 3 rd 23:59) File Sequence - data/raw/ { pid } /fitbit_sleep_summary_raw.csv - data/raw/ { pid } /fitbit_sleep_summary_parsed.csv - data/raw/ { pid } /fitbit_sleep_summary_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_sleep_summary_features/fitbit_sleep_summary_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_sleep_summary.csv Parameters description for [FITBIT_SLEEP_SUMMARY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_SLEEP_SUMMARY features from the RAPIDS provider [SLEEP_TYPES] Types of sleep to be included in the feature extraction computation. Fitbit provides 3 types of sleep: main , nap , all . [FEATURES] Features to be computed from sleep summary data, see table below Features description for [FITBIT_SLEEP_SUMMARY][PROVIDERS][RAPIDS] : Feature Units Description countepisodeTYPE episodes Number of sleep episodes for a certain sleep type during a time segment. avgefficiencyTYPE scores Average sleep efficiency for a certain sleep type during a time segment. sumdurationafterwakeupTYPE minutes Total duration the user stayed in bed after waking up for a certain sleep type during a time segment. sumdurationasleepTYPE minutes Total sleep duration for a certain sleep type during a time segment. sumdurationawakeTYPE minutes Total duration the user stayed awake but still in bed for a certain sleep type during a time segment. sumdurationtofallasleepTYPE minutes Total duration the user spent to fall asleep for a certain sleep type during a time segment. sumdurationinbedTYPE minutes Total duration the user stayed in bed (sumdurationtofallasleep + sumdurationawake + sumdurationasleep + sumdurationafterwakeup) for a certain sleep type during a time segment. avgdurationafterwakeupTYPE minutes Average duration the user stayed in bed after waking up for a certain sleep type during a time segment. avgdurationasleepTYPE minutes Average sleep duration for a certain sleep type during a time segment. avgdurationawakeTYPE minutes Average duration the user stayed awake but still in bed for a certain sleep type during a time segment. avgdurationtofallasleepTYPE minutes Average duration the user spent to fall asleep for a certain sleep type during a time segment. avgdurationinbedTYPE minutes Average duration the user stayed in bed (sumdurationtofallasleep + sumdurationawake + sumdurationasleep + sumdurationafterwakeup) for a certain sleep type during a time segment. Assumptions/Observations There are three sleep types (TYPE): main , nap , all . The all type contains both main sleep and naps. There are two versions of Fitbit\u2019s sleep API ( version 1 and version 1.2 ), and each provides raw sleep data in a different format: Count & duration summaries . v1 contains count_awake , duration_awake , count_awakenings , count_restless , and duration_restless fields for every sleep record but v1.2 does not. API columns . Features are computed based on the values provided by Fitbit\u2019s API: efficiency , minutes_after_wakeup , minutes_asleep , minutes_awake , minutes_to_fall_asleep , minutes_in_bed , is_main_sleep and type .","title":"Fitbit Sleep Summary"},{"location":"features/fitbit-sleep-summary/#fitbit-sleep-summary","text":"Sensor parameters description for [FITBIT_SLEEP_SUMMARY] : Key Description [TABLE] Database table name or file path where the sleep summary data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data with Fitbit\u2019s sleep API Version 1 JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d: [{\u201cawakeCount\u201d: 2, \u201cawakeDuration\u201d: 3, \u201cawakeningsCount\u201d: 10, \u201cdateOfSleep\u201d: \u201c2020-10-07\u201d, \u201cduration\u201d: 8100000, \u201cefficiency\u201d: 91, \u201cendTime\u201d: \u201c2020-10-07T18:10:00.000\u201d, \u201cisMainSleep\u201d: true, \u201clogId\u201d: 14147921940, \u201cminuteData\u201d: [{\u201cdateTime\u201d: \u201c15:55:00\u201d, \u201cvalue\u201d: \u201c3\u201d}, {\u201cdateTime\u201d: \u201c15:56:00\u201d, \u201cvalue\u201d: \u201c3\u201d}, {\u201cdateTime\u201d: \u201c15:57:00\u201d, \u201cvalue\u201d: \u201c2\u201d},\u2026], \u201cminutesAfterWakeup\u201d: 0, \u201cminutesAsleep\u201d: 123, \u201cminutesAwake\u201d: 12, \u201cminutesToFallAsleep\u201d: 0, \u201crestlessCount\u201d: 8, \u201crestlessDuration\u201d: 9, \u201cstartTime\u201d: \u201c2020-10-07T15:55:00.000\u201d, \u201ctimeInBed\u201d: 135}, {\u201cawakeCount\u201d: 0, \u201cawakeDuration\u201d: 0, \u201cawakeningsCount\u201d: 1, \u201cdateOfSleep\u201d: \u201c2020-10-07\u201d, \u201cduration\u201d: 3780000, \u201cefficiency\u201d: 100, \u201cendTime\u201d: \u201c2020-10-07T10:52:30.000\u201d, \u201cisMainSleep\u201d: false, \u201clogId\u201d: 14144903977, \u201cminuteData\u201d: [{\u201cdateTime\u201d: \u201c09:49:00\u201d, \u201cvalue\u201d: \u201c1\u201d}, {\u201cdateTime\u201d: \u201c09:50:00\u201d, \u201cvalue\u201d: \u201c1\u201d}, {\u201cdateTime\u201d: \u201c09:51:00\u201d, \u201cvalue\u201d: \u201c1\u201d},\u2026], \u201cminutesAfterWakeup\u201d: 1, \u201cminutesAsleep\u201d: 62, \u201cminutesAwake\u201d: 0, \u201cminutesToFallAsleep\u201d: 0, \u201crestlessCount\u201d: 1, \u201crestlessDuration\u201d: 1, \u201cstartTime\u201d: \u201c2020-10-07T09:49:00.000\u201d, \u201ctimeInBed\u201d: 63}], \u201csummary\u201d: {\u201ctotalMinutesAsleep\u201d: 185, \u201ctotalSleepRecords\u201d: 2, \u201ctotalTimeInBed\u201d: 198}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d: [{\u201cawakeCount\u201d: 3, \u201cawakeDuration\u201d: 21, \u201cawakeningsCount\u201d: 16, \u201cdateOfSleep\u201d: \u201c2020-10-08\u201d, \u201cduration\u201d: 19260000, \u201cefficiency\u201d: 89, \u201cendTime\u201d: \u201c2020-10-08T06:01:30.000\u201d, \u201cisMainSleep\u201d: true, \u201clogId\u201d: 14150613895, \u201cminuteData\u201d: [{\u201cdateTime\u201d: \u201c00:40:00\u201d, \u201cvalue\u201d: \u201c3\u201d}, {\u201cdateTime\u201d: \u201c00:41:00\u201d, \u201cvalue\u201d: \u201c3\u201d}, {\u201cdateTime\u201d: \u201c00:42:00\u201d, \u201cvalue\u201d: \u201c3\u201d},\u2026], \u201cminutesAfterWakeup\u201d: 0, \u201cminutesAsleep\u201d: 275, \u201cminutesAwake\u201d: 33, \u201cminutesToFallAsleep\u201d: 0, \u201crestlessCount\u201d: 13, \u201crestlessDuration\u201d: 25, \u201cstartTime\u201d: \u201c2020-10-08T00:40:00.000\u201d, \u201ctimeInBed\u201d: 321}], \u201csummary\u201d: {\u201ctotalMinutesAsleep\u201d: 275, \u201ctotalSleepRecords\u201d: 1, \u201ctotalTimeInBed\u201d: 321}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d: [{\u201cawakeCount\u201d: 1, \u201cawakeDuration\u201d: 3, \u201cawakeningsCount\u201d: 8, \u201cdateOfSleep\u201d: \u201c2020-10-09\u201d, \u201cduration\u201d: 19320000, \u201cefficiency\u201d: 96, \u201cendTime\u201d: \u201c2020-10-09T05:57:30.000\u201d, \u201cisMainSleep\u201d: true, \u201clogId\u201d: 14161136803, \u201cminuteData\u201d: [{\u201cdateTime\u201d: \u201c00:35:30\u201d, \u201cvalue\u201d: \u201c2\u201d}, {\u201cdateTime\u201d: \u201c00:36:30\u201d, \u201cvalue\u201d: \u201c1\u201d}, {\u201cdateTime\u201d: \u201c00:37:30\u201d, \u201cvalue\u201d: \u201c1\u201d},\u2026], \u201cminutesAfterWakeup\u201d: 0, \u201cminutesAsleep\u201d: 309, \u201cminutesAwake\u201d: 13, \u201cminutesToFallAsleep\u201d: 0, \u201crestlessCount\u201d: 7, \u201crestlessDuration\u201d: 10, \u201cstartTime\u201d: \u201c2020-10-09T00:35:30.000\u201d, \u201ctimeInBed\u201d: 322}], \u201csummary\u201d: {\u201ctotalMinutesAsleep\u201d: 309, \u201ctotalSleepRecords\u201d: 1, \u201ctotalTimeInBed\u201d: 322}} PLAIN_TEXT device_id local_start_date_time local_end_date_time efficiency minutes_after_wakeup minutes_asleep minutes_awake minutes_to_fall_asleep minutes_in_bed is_main_sleep type count_awake duration_awake count_awakenings count_restless duration_restless a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 15:55:00 2020-10-07 18:10:00 91 0 123 12 0 135 1 classic 2 3 10 8 9 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 09:49:00 2020-10-07 10:52:30 100 1 62 0 0 63 0 classic 0 0 1 1 1 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-08 00:40:00 2020-10-08 06:01:30 89 0 275 33 0 321 1 classic 3 21 16 13 25 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-09 00:35:30 2020-10-09 05:57:30 96 0 309 13 0 322 1 classic 1 3 8 7 10 Example of the structure of source data with Fitbit\u2019s sleep API Version 1.2 JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d:[{\u201cdateOfSleep\u201d:\u201d2020-10-10\u201d,\u201dduration\u201d:3600000,\u201defficiency\u201d:92,\u201dendTime\u201d:\u201d2020-10-10T16:37:00.000\u201d,\u201dinfoCode\u201d:2,\u201disMainSleep\u201d:false,\u201dlevels\u201d:{\u201cdata\u201d:[{\u201cdateTime\u201d:\u201d2020-10-10T15:36:30.000\u201d,\u201dlevel\u201d:\u201drestless\u201d,\u201dseconds\u201d:60},{\u201cdateTime\u201d:\u201d2020-10-10T15:37:30.000\u201d,\u201dlevel\u201d:\u201dasleep\u201d,\u201dseconds\u201d:660},{\u201cdateTime\u201d:\u201d2020-10-10T15:48:30.000\u201d,\u201dlevel\u201d:\u201drestless\u201d,\u201dseconds\u201d:60},\u2026], \u201csummary\u201d:{\u201casleep\u201d:{\u201ccount\u201d:0,\u201dminutes\u201d:56},\u201dawake\u201d:{\u201ccount\u201d:0,\u201dminutes\u201d:0},\u201drestless\u201d:{\u201ccount\u201d:3,\u201dminutes\u201d:4}}},\u201dlogId\u201d:26315914306,\u201dminutesAfterWakeup\u201d:0,\u201dminutesAsleep\u201d:55,\u201dminutesAwake\u201d:5,\u201dminutesToFallAsleep\u201d:0,\u201dstartTime\u201d:\u201d2020-10-10T15:36:30.000\u201d,\u201dtimeInBed\u201d:60,\u201dtype\u201d:\u201dclassic\u201d},{\u201cdateOfSleep\u201d:\u201d2020-10-10\u201d,\u201dduration\u201d:22980000,\u201defficiency\u201d:88,\u201dendTime\u201d:\u201d2020-10-10T08:10:00.000\u201d,\u201dinfoCode\u201d:0,\u201disMainSleep\u201d:true,\u201dlevels\u201d:{\u201cdata\u201d:[{\u201cdateTime\u201d:\u201d2020-10-10T01:46:30.000\u201d,\u201dlevel\u201d:\u201dlight\u201d,\u201dseconds\u201d:420},{\u201cdateTime\u201d:\u201d2020-10-10T01:53:30.000\u201d,\u201dlevel\u201d:\u201ddeep\u201d,\u201dseconds\u201d:1230},{\u201cdateTime\u201d:\u201d2020-10-10T02:14:00.000\u201d,\u201dlevel\u201d:\u201dlight\u201d,\u201dseconds\u201d:360},\u2026], \u201csummary\u201d:{\u201cdeep\u201d:{\u201ccount\u201d:3,\u201dminutes\u201d:92,\u201dthirtyDayAvgMinutes\u201d:0},\u201dlight\u201d:{\u201ccount\u201d:29,\u201dminutes\u201d:193,\u201dthirtyDayAvgMinutes\u201d:0},\u201drem\u201d:{\u201ccount\u201d:4,\u201dminutes\u201d:33,\u201dthirtyDayAvgMinutes\u201d:0},\u201dwake\u201d:{\u201ccount\u201d:28,\u201dminutes\u201d:65,\u201dthirtyDayAvgMinutes\u201d:0}}},\u201dlogId\u201d:26311786557,\u201dminutesAfterWakeup\u201d:0,\u201dminutesAsleep\u201d:318,\u201dminutesAwake\u201d:65,\u201dminutesToFallAsleep\u201d:0,\u201dstartTime\u201d:\u201d2020-10-10T01:46:30.000\u201d,\u201dtimeInBed\u201d:383,\u201dtype\u201d:\u201dstages\u201d}],\u201dsummary\u201d:{\u201cstages\u201d:{\u201cdeep\u201d:92,\u201dlight\u201d:193,\u201drem\u201d:33,\u201dwake\u201d:65},\u201dtotalMinutesAsleep\u201d:373,\u201dtotalSleepRecords\u201d:2,\u201dtotalTimeInBed\u201d:443}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d:[{\u201cdateOfSleep\u201d:\u201d2020-10-11\u201d,\u201dduration\u201d:41640000,\u201defficiency\u201d:89,\u201dendTime\u201d:\u201d2020-10-11T11:47:00.000\u201d,\u201dinfoCode\u201d:0,\u201disMainSleep\u201d:true,\u201dlevels\u201d:{\u201cdata\u201d:[{\u201cdateTime\u201d:\u201d2020-10-11T00:12:30.000\u201d,\u201dlevel\u201d:\u201dwake\u201d,\u201dseconds\u201d:450},{\u201cdateTime\u201d:\u201d2020-10-11T00:20:00.000\u201d,\u201dlevel\u201d:\u201dlight\u201d,\u201dseconds\u201d:870},{\u201cdateTime\u201d:\u201d2020-10-11T00:34:30.000\u201d,\u201dlevel\u201d:\u201dwake\u201d,\u201dseconds\u201d:780},\u2026], \u201csummary\u201d:{\u201cdeep\u201d:{\u201ccount\u201d:4,\u201dminutes\u201d:52,\u201dthirtyDayAvgMinutes\u201d:62},\u201dlight\u201d:{\u201ccount\u201d:32,\u201dminutes\u201d:442,\u201dthirtyDayAvgMinutes\u201d:364},\u201drem\u201d:{\u201ccount\u201d:6,\u201dminutes\u201d:68,\u201dthirtyDayAvgMinutes\u201d:58},\u201dwake\u201d:{\u201ccount\u201d:29,\u201dminutes\u201d:132,\u201dthirtyDayAvgMinutes\u201d:94}}},\u201dlogId\u201d:26589710670,\u201dminutesAfterWakeup\u201d:1,\u201dminutesAsleep\u201d:562,\u201dminutesAwake\u201d:132,\u201dminutesToFallAsleep\u201d:0,\u201dstartTime\u201d:\u201d2020-10-11T00:12:30.000\u201d,\u201dtimeInBed\u201d:694,\u201dtype\u201d:\u201dstages\u201d}],\u201dsummary\u201d:{\u201cstages\u201d:{\u201cdeep\u201d:52,\u201dlight\u201d:442,\u201drem\u201d:68,\u201dwake\u201d:132},\u201dtotalMinutesAsleep\u201d:562,\u201dtotalSleepRecords\u201d:1,\u201dtotalTimeInBed\u201d:694}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d:[{\u201cdateOfSleep\u201d:\u201d2020-10-12\u201d,\u201dduration\u201d:28980000,\u201defficiency\u201d:93,\u201dendTime\u201d:\u201d2020-10-12T09:34:30.000\u201d,\u201dinfoCode\u201d:0,\u201disMainSleep\u201d:true,\u201dlevels\u201d:{\u201cdata\u201d:[{\u201cdateTime\u201d:\u201d2020-10-12T01:31:00.000\u201d,\u201dlevel\u201d:\u201dwake\u201d,\u201dseconds\u201d:600},{\u201cdateTime\u201d:\u201d2020-10-12T01:41:00.000\u201d,\u201dlevel\u201d:\u201dlight\u201d,\u201dseconds\u201d:60},{\u201cdateTime\u201d:\u201d2020-10-12T01:42:00.000\u201d,\u201dlevel\u201d:\u201ddeep\u201d,\u201dseconds\u201d:2340},\u2026], \u201csummary\u201d:{\u201cdeep\u201d:{\u201ccount\u201d:4,\u201dminutes\u201d:63,\u201dthirtyDayAvgMinutes\u201d:59},\u201dlight\u201d:{\u201ccount\u201d:27,\u201dminutes\u201d:257,\u201dthirtyDayAvgMinutes\u201d:364},\u201drem\u201d:{\u201ccount\u201d:5,\u201dminutes\u201d:94,\u201dthirtyDayAvgMinutes\u201d:58},\u201dwake\u201d:{\u201ccount\u201d:24,\u201dminutes\u201d:69,\u201dthirtyDayAvgMinutes\u201d:95}}},\u201dlogId\u201d:26589710673,\u201dminutesAfterWakeup\u201d:0,\u201dminutesAsleep\u201d:415,\u201dminutesAwake\u201d:68,\u201dminutesToFallAsleep\u201d:0,\u201dstartTime\u201d:\u201d2020-10-12T01:31:00.000\u201d,\u201dtimeInBed\u201d:483,\u201dtype\u201d:\u201dstages\u201d}],\u201dsummary\u201d:{\u201cstages\u201d:{\u201cdeep\u201d:63,\u201dlight\u201d:257,\u201drem\u201d:94,\u201dwake\u201d:69},\u201dtotalMinutesAsleep\u201d:415,\u201dtotalSleepRecords\u201d:1,\u201dtotalTimeInBed\u201d:483}} PLAIN_TEXT device_id local_start_date_time local_end_date_time efficiency minutes_after_wakeup minutes_asleep minutes_awake minutes_to_fall_asleep minutes_in_bed is_main_sleep type a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-10 15:36:30 2020-10-10 16:37:00 92 0 55 5 0 60 0 classic a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-10 01:46:30 2020-10-10 08:10:00 88 0 318 65 0 383 1 stages a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-11 00:12:30 2020-10-11 11:47:00 89 1 562 132 0 694 1 stages a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-12 01:31:00 2020-10-12 09:34:30 93 0 415 68 0 483 1 stages","title":"Fitbit Sleep Summary"},{"location":"features/fitbit-sleep-summary/#rapids-provider","text":"Available time segments Only available for segments that span 1 or more complete days (e.g. Jan 1 st 00:00 to Jan 3 rd 23:59) File Sequence - data/raw/ { pid } /fitbit_sleep_summary_raw.csv - data/raw/ { pid } /fitbit_sleep_summary_parsed.csv - data/raw/ { pid } /fitbit_sleep_summary_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_sleep_summary_features/fitbit_sleep_summary_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_sleep_summary.csv Parameters description for [FITBIT_SLEEP_SUMMARY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_SLEEP_SUMMARY features from the RAPIDS provider [SLEEP_TYPES] Types of sleep to be included in the feature extraction computation. Fitbit provides 3 types of sleep: main , nap , all . [FEATURES] Features to be computed from sleep summary data, see table below Features description for [FITBIT_SLEEP_SUMMARY][PROVIDERS][RAPIDS] : Feature Units Description countepisodeTYPE episodes Number of sleep episodes for a certain sleep type during a time segment. avgefficiencyTYPE scores Average sleep efficiency for a certain sleep type during a time segment. sumdurationafterwakeupTYPE minutes Total duration the user stayed in bed after waking up for a certain sleep type during a time segment. sumdurationasleepTYPE minutes Total sleep duration for a certain sleep type during a time segment. sumdurationawakeTYPE minutes Total duration the user stayed awake but still in bed for a certain sleep type during a time segment. sumdurationtofallasleepTYPE minutes Total duration the user spent to fall asleep for a certain sleep type during a time segment. sumdurationinbedTYPE minutes Total duration the user stayed in bed (sumdurationtofallasleep + sumdurationawake + sumdurationasleep + sumdurationafterwakeup) for a certain sleep type during a time segment. avgdurationafterwakeupTYPE minutes Average duration the user stayed in bed after waking up for a certain sleep type during a time segment. avgdurationasleepTYPE minutes Average sleep duration for a certain sleep type during a time segment. avgdurationawakeTYPE minutes Average duration the user stayed awake but still in bed for a certain sleep type during a time segment. avgdurationtofallasleepTYPE minutes Average duration the user spent to fall asleep for a certain sleep type during a time segment. avgdurationinbedTYPE minutes Average duration the user stayed in bed (sumdurationtofallasleep + sumdurationawake + sumdurationasleep + sumdurationafterwakeup) for a certain sleep type during a time segment. Assumptions/Observations There are three sleep types (TYPE): main , nap , all . The all type contains both main sleep and naps. There are two versions of Fitbit\u2019s sleep API ( version 1 and version 1.2 ), and each provides raw sleep data in a different format: Count & duration summaries . v1 contains count_awake , duration_awake , count_awakenings , count_restless , and duration_restless fields for every sleep record but v1.2 does not. API columns . Features are computed based on the values provided by Fitbit\u2019s API: efficiency , minutes_after_wakeup , minutes_asleep , minutes_awake , minutes_to_fall_asleep , minutes_in_bed , is_main_sleep and type .","title":"RAPIDS provider"},{"location":"features/fitbit-steps-intraday/","text":"Fitbit Steps Intraday \u00b6 Sensor parameters description for [FITBIT_STEPS_INTRADAY] : Key Description [TABLE] Database table name or file path where the steps intraday data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-07\u201d,\u201dvalue\u201d:\u201d1775\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:5},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:3},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:0},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-08\u201d,\u201dvalue\u201d:\u201d3201\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:14},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:11},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:10},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-09\u201d,\u201dvalue\u201d:\u201d998\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:0},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:0},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:0},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} PLAIN_TEXT device_id local_date_time steps a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:00:00 5 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:01:00 3 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:02:00 0 RAPIDS provider \u00b6 Available time segments Available for all time segments File Sequence - data/raw/ { pid } /fitbit_steps_intraday_raw.csv - data/raw/ { pid } /fitbit_steps_intraday_parsed.csv - data/raw/ { pid } /fitbit_steps_intraday_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_steps_intraday_features/fitbit_steps_intraday_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_steps_intraday.csv Parameters description for [FITBIT_STEPS_INTRADAY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_STEPS_INTRADAY features from the RAPIDS provider [FEATURES] Features to be computed from steps intraday data, see table below [THRESHOLD_ACTIVE_BOUT] Every minute with Fitbit steps data wil be labelled as sedentary if its step count is below this threshold, otherwise, active . [INCLUDE_ZERO_STEP_ROWS] Whether or not to include time segments with a 0 step count during the whole day. Features description for [FITBIT_STEPS_INTRADAY][PROVIDERS][RAPIDS] : Feature Units Description sumsteps steps The total step count during a time segment. maxsteps steps The maximum step count during a time segment. minsteps steps The minimum step count during a time segment. avgsteps steps The average step count during a time segment. stdsteps steps The standard deviation of step count during a time segment. countepisodesedentarybout bouts Number of sedentary bouts during a time segment. sumdurationsedentarybout minutes Total duration of all sedentary bouts during a time segment. maxdurationsedentarybout minutes The maximum duration of any sedentary bout during a time segment. mindurationsedentarybout minutes The minimum duration of any sedentary bout during a time segment. avgdurationsedentarybout minutes The average duration of sedentary bouts during a time segment. stddurationsedentarybout minutes The standard deviation of the duration of sedentary bouts during a time segment. countepisodeactivebout bouts Number of active bouts during a time segment. sumdurationactivebout minutes Total duration of all active bouts during a time segment. maxdurationactivebout minutes The maximum duration of any active bout during a time segment. mindurationactivebout minutes The minimum duration of any active bout during a time segment. avgdurationactivebout minutes The average duration of active bouts during a time segment. stddurationactivebout minutes The standard deviation of the duration of active bouts during a time segment. Assumptions/Observations Active and sedentary bouts . If the step count per minute is smaller than THRESHOLD_ACTIVE_BOUT (default value is 10), that minute is labelled as sedentary, otherwise, is labelled as active. Active and sedentary bouts are periods of consecutive minutes labelled as active or sedentary .","title":"Fitbit Steps Intraday"},{"location":"features/fitbit-steps-intraday/#fitbit-steps-intraday","text":"Sensor parameters description for [FITBIT_STEPS_INTRADAY] : Key Description [TABLE] Database table name or file path where the steps intraday data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-07\u201d,\u201dvalue\u201d:\u201d1775\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:5},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:3},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:0},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-08\u201d,\u201dvalue\u201d:\u201d3201\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:14},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:11},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:10},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-09\u201d,\u201dvalue\u201d:\u201d998\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:0},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:0},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:0},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} PLAIN_TEXT device_id local_date_time steps a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:00:00 5 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:01:00 3 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:02:00 0","title":"Fitbit Steps Intraday"},{"location":"features/fitbit-steps-intraday/#rapids-provider","text":"Available time segments Available for all time segments File Sequence - data/raw/ { pid } /fitbit_steps_intraday_raw.csv - data/raw/ { pid } /fitbit_steps_intraday_parsed.csv - data/raw/ { pid } /fitbit_steps_intraday_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_steps_intraday_features/fitbit_steps_intraday_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_steps_intraday.csv Parameters description for [FITBIT_STEPS_INTRADAY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_STEPS_INTRADAY features from the RAPIDS provider [FEATURES] Features to be computed from steps intraday data, see table below [THRESHOLD_ACTIVE_BOUT] Every minute with Fitbit steps data wil be labelled as sedentary if its step count is below this threshold, otherwise, active . [INCLUDE_ZERO_STEP_ROWS] Whether or not to include time segments with a 0 step count during the whole day. Features description for [FITBIT_STEPS_INTRADAY][PROVIDERS][RAPIDS] : Feature Units Description sumsteps steps The total step count during a time segment. maxsteps steps The maximum step count during a time segment. minsteps steps The minimum step count during a time segment. avgsteps steps The average step count during a time segment. stdsteps steps The standard deviation of step count during a time segment. countepisodesedentarybout bouts Number of sedentary bouts during a time segment. sumdurationsedentarybout minutes Total duration of all sedentary bouts during a time segment. maxdurationsedentarybout minutes The maximum duration of any sedentary bout during a time segment. mindurationsedentarybout minutes The minimum duration of any sedentary bout during a time segment. avgdurationsedentarybout minutes The average duration of sedentary bouts during a time segment. stddurationsedentarybout minutes The standard deviation of the duration of sedentary bouts during a time segment. countepisodeactivebout bouts Number of active bouts during a time segment. sumdurationactivebout minutes Total duration of all active bouts during a time segment. maxdurationactivebout minutes The maximum duration of any active bout during a time segment. mindurationactivebout minutes The minimum duration of any active bout during a time segment. avgdurationactivebout minutes The average duration of active bouts during a time segment. stddurationactivebout minutes The standard deviation of the duration of active bouts during a time segment. Assumptions/Observations Active and sedentary bouts . If the step count per minute is smaller than THRESHOLD_ACTIVE_BOUT (default value is 10), that minute is labelled as sedentary, otherwise, is labelled as active. Active and sedentary bouts are periods of consecutive minutes labelled as active or sedentary .","title":"RAPIDS provider"},{"location":"features/fitbit-steps-summary/","text":"Fitbit Steps Summary \u00b6 Sensor parameters description for [FITBIT_STEPS_SUMMARY] : Key Description [TABLE] Database table name or file path where the steps summary data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-07\u201d,\u201dvalue\u201d:\u201d1775\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:5},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:3},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:0},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-08\u201d,\u201dvalue\u201d:\u201d3201\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:14},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:11},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:10},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-09\u201d,\u201dvalue\u201d:\u201d998\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:0},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:0},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:0},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} PLAIN_TEXT device_id local_date_time steps a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 1775 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-08 3201 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-09 998 RAPIDS provider \u00b6 Available time segments Only available for segments that span 1 or more complete days (e.g. Jan 1 st 00:00 to Jan 3 rd 23:59) File Sequence - data/raw/ { pid } /fitbit_steps_summary_raw.csv - data/raw/ { pid } /fitbit_steps_summary_parsed.csv - data/raw/ { pid } /fitbit_steps_summary_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_steps_summary_features/fitbit_steps_summary_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_steps_summary.csv Parameters description for [FITBIT_STEPS_SUMMARY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_STEPS_SUMMARY features from the RAPIDS provider [FEATURES] Features to be computed from steps summary data, see table below Features description for [FITBIT_STEPS_SUMMARY][PROVIDERS][RAPIDS] : Feature Units Description maxsumsteps steps The maximum daily step count during a time segment. minsumsteps steps The minimum daily step count during a time segment. avgsumsteps steps The average daily step count during a time segment. mediansumsteps steps The median of daily step count during a time segment. stdsumsteps steps The standard deviation of daily step count during a time segment. Assumptions/Observations NA","title":"Fitbit Steps Summary"},{"location":"features/fitbit-steps-summary/#fitbit-steps-summary","text":"Sensor parameters description for [FITBIT_STEPS_SUMMARY] : Key Description [TABLE] Database table name or file path where the steps summary data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-07\u201d,\u201dvalue\u201d:\u201d1775\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:5},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:3},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:0},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-08\u201d,\u201dvalue\u201d:\u201d3201\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:14},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:11},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:10},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-09\u201d,\u201dvalue\u201d:\u201d998\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:0},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:0},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:0},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} PLAIN_TEXT device_id local_date_time steps a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 1775 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-08 3201 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-09 998","title":"Fitbit Steps Summary"},{"location":"features/fitbit-steps-summary/#rapids-provider","text":"Available time segments Only available for segments that span 1 or more complete days (e.g. Jan 1 st 00:00 to Jan 3 rd 23:59) File Sequence - data/raw/ { pid } /fitbit_steps_summary_raw.csv - data/raw/ { pid } /fitbit_steps_summary_parsed.csv - data/raw/ { pid } /fitbit_steps_summary_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_steps_summary_features/fitbit_steps_summary_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_steps_summary.csv Parameters description for [FITBIT_STEPS_SUMMARY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_STEPS_SUMMARY features from the RAPIDS provider [FEATURES] Features to be computed from steps summary data, see table below Features description for [FITBIT_STEPS_SUMMARY][PROVIDERS][RAPIDS] : Feature Units Description maxsumsteps steps The maximum daily step count during a time segment. minsumsteps steps The minimum daily step count during a time segment. avgsumsteps steps The average daily step count during a time segment. mediansumsteps steps The median of daily step count during a time segment. stdsumsteps steps The standard deviation of daily step count during a time segment. Assumptions/Observations NA","title":"RAPIDS provider"},{"location":"features/phone-accelerometer/","text":"Phone Accelerometer \u00b6 Sensor parameters description for [PHONE_ACCELEROMETER] : Key Description [TABLE] Database table where the accelerometer data is stored RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_accelerometer_raw.csv - data/raw/ { pid } /phone_accelerometer_with_datetime.csv - data/interim/ { pid } /phone_accelerometer_features/phone_accelerometer_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_accelerometer.csv Parameters description for [PHONE_ACCELEROMETER][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_ACCELEROMETER features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_ACCELEROMETER][PROVIDERS][RAPIDS] : Feature Units Description maxmagnitude m/s 2 The maximum magnitude of acceleration ( \\(\\|acceleration\\| = \\sqrt{x^2 + y^2 + z^2}\\) ). minmagnitude m/s 2 The minimum magnitude of acceleration. avgmagnitude m/s 2 The average magnitude of acceleration. medianmagnitude m/s 2 The median magnitude of acceleration. stdmagnitude m/s 2 The standard deviation of acceleration. Assumptions/Observations Analyzing accelerometer data is a memory intensive task. If RAPIDS crashes is likely because the accelerometer dataset for a participant is to big to fit in memory. We are considering different alternatives to overcome this problem. PANDA provider \u00b6 These features are based on the work by Panda et al . Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_accelerometer_raw.csv - data/raw/ { pid } /phone_accelerometer_with_datetime.csv - data/interim/ { pid } /phone_accelerometer_features/phone_accelerometer_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_accelerometer.csv Parameters description for [PHONE_ACCELEROMETER][PROVIDERS][PANDA] : Key Description [COMPUTE] Set to True to extract PHONE_ACCELEROMETER features from the PANDA provider [FEATURES] Features to be computed for exertional and non-exertional activity episodes, see table below Features description for [PHONE_ACCELEROMETER][PROVIDERS][PANDA] : Feature Units Description sumduration minutes Total duration of all exertional or non-exertional activity episodes. maxduration minutes Longest duration of any exertional or non-exertional activity episode. minduration minutes Shortest duration of any exertional or non-exertional activity episode. avgduration minutes Average duration of any exertional or non-exertional activity episode. medianduration minutes Median duration of any exertional or non-exertional activity episode. stdduration minutes Standard deviation of the duration of all exertional or non-exertional activity episodes. Assumptions/Observations Analyzing accelerometer data is a memory intensive task. If RAPIDS crashes is likely because the accelerometer dataset for a participant is to big to fit in memory. We are considering different alternatives to overcome this problem. See Panda et al for a definition of exertional and non-exertional activity episodes","title":"Phone Accelerometer"},{"location":"features/phone-accelerometer/#phone-accelerometer","text":"Sensor parameters description for [PHONE_ACCELEROMETER] : Key Description [TABLE] Database table where the accelerometer data is stored","title":"Phone Accelerometer"},{"location":"features/phone-accelerometer/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_accelerometer_raw.csv - data/raw/ { pid } /phone_accelerometer_with_datetime.csv - data/interim/ { pid } /phone_accelerometer_features/phone_accelerometer_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_accelerometer.csv Parameters description for [PHONE_ACCELEROMETER][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_ACCELEROMETER features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_ACCELEROMETER][PROVIDERS][RAPIDS] : Feature Units Description maxmagnitude m/s 2 The maximum magnitude of acceleration ( \\(\\|acceleration\\| = \\sqrt{x^2 + y^2 + z^2}\\) ). minmagnitude m/s 2 The minimum magnitude of acceleration. avgmagnitude m/s 2 The average magnitude of acceleration. medianmagnitude m/s 2 The median magnitude of acceleration. stdmagnitude m/s 2 The standard deviation of acceleration. Assumptions/Observations Analyzing accelerometer data is a memory intensive task. If RAPIDS crashes is likely because the accelerometer dataset for a participant is to big to fit in memory. We are considering different alternatives to overcome this problem.","title":"RAPIDS provider"},{"location":"features/phone-accelerometer/#panda-provider","text":"These features are based on the work by Panda et al . Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_accelerometer_raw.csv - data/raw/ { pid } /phone_accelerometer_with_datetime.csv - data/interim/ { pid } /phone_accelerometer_features/phone_accelerometer_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_accelerometer.csv Parameters description for [PHONE_ACCELEROMETER][PROVIDERS][PANDA] : Key Description [COMPUTE] Set to True to extract PHONE_ACCELEROMETER features from the PANDA provider [FEATURES] Features to be computed for exertional and non-exertional activity episodes, see table below Features description for [PHONE_ACCELEROMETER][PROVIDERS][PANDA] : Feature Units Description sumduration minutes Total duration of all exertional or non-exertional activity episodes. maxduration minutes Longest duration of any exertional or non-exertional activity episode. minduration minutes Shortest duration of any exertional or non-exertional activity episode. avgduration minutes Average duration of any exertional or non-exertional activity episode. medianduration minutes Median duration of any exertional or non-exertional activity episode. stdduration minutes Standard deviation of the duration of all exertional or non-exertional activity episodes. Assumptions/Observations Analyzing accelerometer data is a memory intensive task. If RAPIDS crashes is likely because the accelerometer dataset for a participant is to big to fit in memory. We are considering different alternatives to overcome this problem. See Panda et al for a definition of exertional and non-exertional activity episodes","title":"PANDA provider"},{"location":"features/phone-activity-recognition/","text":"Phone Activity Recognition \u00b6 Sensor parameters description for [PHONE_ACTIVITY_RECOGNITION] : Key Description [TABLE][ANDROID] Database table where the activity data from Android devices is stored (the AWARE client saves this data on different tables for Android and iOS) [TABLE][IOS] Database table where the activity data from iOS devices is stored (the AWARE client saves this data on different tables for Android and iOS) [EPISODE_THRESHOLD_BETWEEN_ROWS] Difference in minutes between any two rows for them to be considered part of the same activity episode RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_activity_recognition_raw.csv - data/raw/ { pid } /phone_activity_recognition_with_datetime.csv - data/raw/ { pid } /phone_activity_recognition_with_datetime_unified.csv - data/interim/ { pid } /phone_activity_recognition_episodes.csv - data/interim/ { pid } /phone_activity_recognition_episodes_resampled.csv - data/interim/ { pid } /phone_activity_recognition_episodes_resampled_with_datetime.csv - data/interim/ { pid } /phone_activity_recognition_features/phone_activity_recognition_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_activity_recognition.csv Parameters description for [PHONE_ACTIVITY_RECOGNITION][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_ACTIVITY_RECOGNITION features from the RAPIDS provider [FEATURES] Features to be computed, see table below [ACTIVITY_CLASSES][STATIONARY] An array of the activity labels to be considered in the STATIONARY category choose any of still , tilting [ACTIVITY_CLASSES][MOBILE] An array of the activity labels to be considered in the MOBILE category choose any of on_foot , walking , running , on_bicycle [ACTIVITY_CLASSES][VEHICLE] An array of the activity labels to be considered in the VEHICLE category choose any of in_vehicule Features description for [PHONE_ACTIVITY_RECOGNITION][PROVIDERS][RAPIDS] : Feature Units Description count rows Number of episodes. mostcommonactivity activity type The most common activity type (e.g. still , on_foot , etc.). If there is a tie, the first one is chosen. countuniqueactivities activity type Number of unique activities. durationstationary minutes The total duration of [ACTIVITY_CLASSES][STATIONARY] episodes durationmobile minutes The total duration of [ACTIVITY_CLASSES][MOBILE] episodes of on foot, running, and on bicycle activities durationvehicle minutes The total duration of [ACTIVITY_CLASSES][VEHICLE] episodes of on vehicle activity Assumptions/Observations iOS Activity Recognition names and types are unified with Android labels: iOS Activity Name Android Activity Name Android Activity Type walking walking 7 running running 8 cycling on_bicycle 1 automotive in_vehicle 0 stationary still 3 unknown unknown 4 In AWARE, Activity Recognition data for Android and iOS are stored in two different database tables, RAPIDS automatically infers what platform each participant belongs to based on their participant file .","title":"Phone Activity Recognition"},{"location":"features/phone-activity-recognition/#phone-activity-recognition","text":"Sensor parameters description for [PHONE_ACTIVITY_RECOGNITION] : Key Description [TABLE][ANDROID] Database table where the activity data from Android devices is stored (the AWARE client saves this data on different tables for Android and iOS) [TABLE][IOS] Database table where the activity data from iOS devices is stored (the AWARE client saves this data on different tables for Android and iOS) [EPISODE_THRESHOLD_BETWEEN_ROWS] Difference in minutes between any two rows for them to be considered part of the same activity episode","title":"Phone Activity Recognition"},{"location":"features/phone-activity-recognition/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_activity_recognition_raw.csv - data/raw/ { pid } /phone_activity_recognition_with_datetime.csv - data/raw/ { pid } /phone_activity_recognition_with_datetime_unified.csv - data/interim/ { pid } /phone_activity_recognition_episodes.csv - data/interim/ { pid } /phone_activity_recognition_episodes_resampled.csv - data/interim/ { pid } /phone_activity_recognition_episodes_resampled_with_datetime.csv - data/interim/ { pid } /phone_activity_recognition_features/phone_activity_recognition_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_activity_recognition.csv Parameters description for [PHONE_ACTIVITY_RECOGNITION][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_ACTIVITY_RECOGNITION features from the RAPIDS provider [FEATURES] Features to be computed, see table below [ACTIVITY_CLASSES][STATIONARY] An array of the activity labels to be considered in the STATIONARY category choose any of still , tilting [ACTIVITY_CLASSES][MOBILE] An array of the activity labels to be considered in the MOBILE category choose any of on_foot , walking , running , on_bicycle [ACTIVITY_CLASSES][VEHICLE] An array of the activity labels to be considered in the VEHICLE category choose any of in_vehicule Features description for [PHONE_ACTIVITY_RECOGNITION][PROVIDERS][RAPIDS] : Feature Units Description count rows Number of episodes. mostcommonactivity activity type The most common activity type (e.g. still , on_foot , etc.). If there is a tie, the first one is chosen. countuniqueactivities activity type Number of unique activities. durationstationary minutes The total duration of [ACTIVITY_CLASSES][STATIONARY] episodes durationmobile minutes The total duration of [ACTIVITY_CLASSES][MOBILE] episodes of on foot, running, and on bicycle activities durationvehicle minutes The total duration of [ACTIVITY_CLASSES][VEHICLE] episodes of on vehicle activity Assumptions/Observations iOS Activity Recognition names and types are unified with Android labels: iOS Activity Name Android Activity Name Android Activity Type walking walking 7 running running 8 cycling on_bicycle 1 automotive in_vehicle 0 stationary still 3 unknown unknown 4 In AWARE, Activity Recognition data for Android and iOS are stored in two different database tables, RAPIDS automatically infers what platform each participant belongs to based on their participant file .","title":"RAPIDS provider"},{"location":"features/phone-applications-foreground/","text":"Phone Applications Foreground \u00b6 Sensor parameters description for [PHONE_APPLICATIONS_FOREGROUND] (these parameters are used by the only provider available at the moment, RAPIDS): Key Description [TABLE] Database table where the applications foreground data is stored [APPLICATION_CATEGORIES][CATALOGUE_SOURCE] FILE or GOOGLE . If FILE , app categories (genres) are read from [CATALOGUE_FILE] . If [GOOGLE] , app categories (genres) are scrapped from the Play Store [APPLICATION_CATEGORIES][CATALOGUE_FILE] CSV file with a package_name and genre column. By default we provide the catalogue created by Stachl et al in data/external/stachl_application_genre_catalogue.csv [APPLICATION_CATEGORIES][UPDATE_CATALOGUE_FILE] if [CATALOGUE_SOURCE] is equal to FILE , this flag signals whether or not to update [CATALOGUE_FILE] , if [CATALOGUE_SOURCE] is equal to GOOGLE all scraped genres will be saved to [CATALOGUE_FILE] [APPLICATION_CATEGORIES][SCRAPE_MISSING_CATEGORIES] This flag signals whether or not to scrape categories (genres) missing from the [CATALOGUE_FILE] . If [CATALOGUE_SOURCE] is equal to GOOGLE , all genres are scraped anyway (this flag is ignored) RAPIDS provider \u00b6 The app category (genre) catalogue used in these features was originally created by Stachl et al . Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_applications_foreground_raw.csv - data/raw/ { pid } /phone_applications_foreground_with_datetime.csv - data/raw/ { pid } /phone_applications_foreground_with_datetime_with_categories.csv - data/interim/ { pid } /phone_applications_foreground_features/phone_applications_foreground_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_applications_foreground.csv Parameters description for [PHONE_APPLICATIONS_FOREGROUND][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_APPLICATIONS_FOREGROUND features from the RAPIDS provider [FEATURES] Features to be computed, see table below [SINGLE_CATEGORIES] An array of app categories to be included in the feature extraction computation. The special keyword all represents a category with all the apps from each participant. By default we use the category catalogue pointed by [APPLICATION_CATEGORIES][CATALOGUE_FILE] (see the Sensor parameters description table above) [MULTIPLE_CATEGORIES] An array of collections representing meta-categories (a group of categories). They key of each element is the name of the meta-category and the value is an array of member app categories. By default we use the category catalogue pointed by [APPLICATION_CATEGORIES][CATALOGUE_FILE] (see the Sensor parameters description table above) [SINGLE_APPS] An array of apps to be included in the feature extraction computation. Use their package name (e.g. com.google.android.youtube ) or the reserved keyword top1global (the most used app by a participant over the whole monitoring study) [EXCLUDED_CATEGORIES] An array of app categories to be excluded from the feature extraction computation. By default we use the category catalogue pointed by [APPLICATION_CATEGORIES][CATALOGUE_FILE] (see the Sensor parameters description table above) [EXCLUDED_APPS] An array of apps to be excluded from the feature extraction computation. Use their package name, for example: com.google.android.youtube Features description for [PHONE_APPLICATIONS_FOREGROUND][PROVIDERS][RAPIDS] : Feature Units Description count apps Number of times a single app or apps within a category were used (i.e. they were brought to the foreground either by tapping their icon or switching to it from another app) timeoffirstuse minutes The time in minutes between 12:00am (midnight) and the first use of a single app or apps within a category during a time_segment timeoflastuse minutes The time in minutes between 12:00am (midnight) and the last use of a single app or apps within a category during a time_segment frequencyentropy nats The entropy of the used apps within a category during a time_segment (each app is seen as a unique event, the more apps were used, the higher the entropy). This is especially relevant when computed over all apps. Entropy cannot be obtained for a single app Assumptions/Observations Features can be computed by app, by apps grouped under a single category (genre) and by multiple categories grouped together (meta-categories). For example, we can get features for Facebook (single app), for Social Network apps (a category including Facebook and other social media apps) or for Social (a meta-category formed by Social Network and Social Media Tools categories). Apps installed by default like YouTube are considered systems apps on some phones. We do an exact match to exclude apps where \u201cgenre\u201d == EXCLUDED_CATEGORIES or \u201cpackage_name\u201d == EXCLUDED_APPS . We provide three ways of classifying and app within a category (genre): a) by automatically scraping its official category from the Google Play Store, b) by using the catalogue created by Stachl et al. which we provide in RAPIDS ( data/external/stachl_application_genre_catalogue.csv ), or c) by manually creating a personalized catalogue. You can choose a, b or c by modifying [APPLICATION_GENRES] keys and values (see the Sensor parameters description table above).","title":"Phone Applications Foreground"},{"location":"features/phone-applications-foreground/#phone-applications-foreground","text":"Sensor parameters description for [PHONE_APPLICATIONS_FOREGROUND] (these parameters are used by the only provider available at the moment, RAPIDS): Key Description [TABLE] Database table where the applications foreground data is stored [APPLICATION_CATEGORIES][CATALOGUE_SOURCE] FILE or GOOGLE . If FILE , app categories (genres) are read from [CATALOGUE_FILE] . If [GOOGLE] , app categories (genres) are scrapped from the Play Store [APPLICATION_CATEGORIES][CATALOGUE_FILE] CSV file with a package_name and genre column. By default we provide the catalogue created by Stachl et al in data/external/stachl_application_genre_catalogue.csv [APPLICATION_CATEGORIES][UPDATE_CATALOGUE_FILE] if [CATALOGUE_SOURCE] is equal to FILE , this flag signals whether or not to update [CATALOGUE_FILE] , if [CATALOGUE_SOURCE] is equal to GOOGLE all scraped genres will be saved to [CATALOGUE_FILE] [APPLICATION_CATEGORIES][SCRAPE_MISSING_CATEGORIES] This flag signals whether or not to scrape categories (genres) missing from the [CATALOGUE_FILE] . If [CATALOGUE_SOURCE] is equal to GOOGLE , all genres are scraped anyway (this flag is ignored)","title":"Phone Applications Foreground"},{"location":"features/phone-applications-foreground/#rapids-provider","text":"The app category (genre) catalogue used in these features was originally created by Stachl et al . Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_applications_foreground_raw.csv - data/raw/ { pid } /phone_applications_foreground_with_datetime.csv - data/raw/ { pid } /phone_applications_foreground_with_datetime_with_categories.csv - data/interim/ { pid } /phone_applications_foreground_features/phone_applications_foreground_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_applications_foreground.csv Parameters description for [PHONE_APPLICATIONS_FOREGROUND][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_APPLICATIONS_FOREGROUND features from the RAPIDS provider [FEATURES] Features to be computed, see table below [SINGLE_CATEGORIES] An array of app categories to be included in the feature extraction computation. The special keyword all represents a category with all the apps from each participant. By default we use the category catalogue pointed by [APPLICATION_CATEGORIES][CATALOGUE_FILE] (see the Sensor parameters description table above) [MULTIPLE_CATEGORIES] An array of collections representing meta-categories (a group of categories). They key of each element is the name of the meta-category and the value is an array of member app categories. By default we use the category catalogue pointed by [APPLICATION_CATEGORIES][CATALOGUE_FILE] (see the Sensor parameters description table above) [SINGLE_APPS] An array of apps to be included in the feature extraction computation. Use their package name (e.g. com.google.android.youtube ) or the reserved keyword top1global (the most used app by a participant over the whole monitoring study) [EXCLUDED_CATEGORIES] An array of app categories to be excluded from the feature extraction computation. By default we use the category catalogue pointed by [APPLICATION_CATEGORIES][CATALOGUE_FILE] (see the Sensor parameters description table above) [EXCLUDED_APPS] An array of apps to be excluded from the feature extraction computation. Use their package name, for example: com.google.android.youtube Features description for [PHONE_APPLICATIONS_FOREGROUND][PROVIDERS][RAPIDS] : Feature Units Description count apps Number of times a single app or apps within a category were used (i.e. they were brought to the foreground either by tapping their icon or switching to it from another app) timeoffirstuse minutes The time in minutes between 12:00am (midnight) and the first use of a single app or apps within a category during a time_segment timeoflastuse minutes The time in minutes between 12:00am (midnight) and the last use of a single app or apps within a category during a time_segment frequencyentropy nats The entropy of the used apps within a category during a time_segment (each app is seen as a unique event, the more apps were used, the higher the entropy). This is especially relevant when computed over all apps. Entropy cannot be obtained for a single app Assumptions/Observations Features can be computed by app, by apps grouped under a single category (genre) and by multiple categories grouped together (meta-categories). For example, we can get features for Facebook (single app), for Social Network apps (a category including Facebook and other social media apps) or for Social (a meta-category formed by Social Network and Social Media Tools categories). Apps installed by default like YouTube are considered systems apps on some phones. We do an exact match to exclude apps where \u201cgenre\u201d == EXCLUDED_CATEGORIES or \u201cpackage_name\u201d == EXCLUDED_APPS . We provide three ways of classifying and app within a category (genre): a) by automatically scraping its official category from the Google Play Store, b) by using the catalogue created by Stachl et al. which we provide in RAPIDS ( data/external/stachl_application_genre_catalogue.csv ), or c) by manually creating a personalized catalogue. You can choose a, b or c by modifying [APPLICATION_GENRES] keys and values (see the Sensor parameters description table above).","title":"RAPIDS provider"},{"location":"features/phone-battery/","text":"Phone Battery \u00b6 Sensor parameters description for [PHONE_BATTERY] : Key Description [TABLE] Database table where the battery data is stored [EPISODE_THRESHOLD_BETWEEN_ROWS] Difference in minutes between any two rows for them to be considered part of the same battery charge or discharge episode RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_battery_raw.csv - data/interim/ { pid } /phone_battery_episodes.csv - data/interim/ { pid } /phone_battery_episodes_resampled.csv - data/interim/ { pid } /phone_battery_episodes_resampled_with_datetime.csv - data/interim/ { pid } /phone_battery_features/phone_battery_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_battery.csv Parameters description for [PHONE_BATTERY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_BATTERY features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_BATTERY][PROVIDERS][RAPIDS] : Feature Units Description countdischarge episodes Number of discharging episodes. sumdurationdischarge minutes The total duration of all discharging episodes. countcharge episodes Number of battery charging episodes. sumdurationcharge minutes The total duration of all charging episodes. avgconsumptionrate episodes/minutes The average of all episodes\u2019 consumption rates. An episode\u2019s consumption rate is defined as the ratio between its battery delta and duration maxconsumptionrate episodes/minutes The highest of all episodes\u2019 consumption rates. An episode\u2019s consumption rate is defined as the ratio between its battery delta and duration Assumptions/Observations We convert battery data collected with iOS client v1 (autodetected because battery status 4 do not exist) to match Android battery format: we swap status 3 for 5 and 1 for 3 We group battery data into discharge or charge episodes considering any contiguous rows with consecutive reductions or increases of the battery level if they are logged within [EPISODE_THRESHOLD_BETWEEN_ROWS] minutes from each other.","title":"Phone Battery"},{"location":"features/phone-battery/#phone-battery","text":"Sensor parameters description for [PHONE_BATTERY] : Key Description [TABLE] Database table where the battery data is stored [EPISODE_THRESHOLD_BETWEEN_ROWS] Difference in minutes between any two rows for them to be considered part of the same battery charge or discharge episode","title":"Phone Battery"},{"location":"features/phone-battery/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_battery_raw.csv - data/interim/ { pid } /phone_battery_episodes.csv - data/interim/ { pid } /phone_battery_episodes_resampled.csv - data/interim/ { pid } /phone_battery_episodes_resampled_with_datetime.csv - data/interim/ { pid } /phone_battery_features/phone_battery_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_battery.csv Parameters description for [PHONE_BATTERY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_BATTERY features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_BATTERY][PROVIDERS][RAPIDS] : Feature Units Description countdischarge episodes Number of discharging episodes. sumdurationdischarge minutes The total duration of all discharging episodes. countcharge episodes Number of battery charging episodes. sumdurationcharge minutes The total duration of all charging episodes. avgconsumptionrate episodes/minutes The average of all episodes\u2019 consumption rates. An episode\u2019s consumption rate is defined as the ratio between its battery delta and duration maxconsumptionrate episodes/minutes The highest of all episodes\u2019 consumption rates. An episode\u2019s consumption rate is defined as the ratio between its battery delta and duration Assumptions/Observations We convert battery data collected with iOS client v1 (autodetected because battery status 4 do not exist) to match Android battery format: we swap status 3 for 5 and 1 for 3 We group battery data into discharge or charge episodes considering any contiguous rows with consecutive reductions or increases of the battery level if they are logged within [EPISODE_THRESHOLD_BETWEEN_ROWS] minutes from each other.","title":"RAPIDS provider"},{"location":"features/phone-bluetooth/","text":"Phone Bluetooth \u00b6 Sensor parameters description for [PHONE_BLUETOOTH] : Key Description [TABLE] Database table where the bluetooth data is stored RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_bluetooth_raw.csv - data/raw/ { pid } /phone_bluetooth_with_datetime.csv - data/interim/ { pid } /phone_bluetooth_features/phone_bluetooth_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_bluetooth.csv \" Parameters description for [PHONE_BLUETOOTH][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_BLUETOOTH features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_BLUETOOTH][PROVIDERS][RAPIDS] : Feature Units Description countscans devices Number of scanned devices during a time_segment , a device can be detected multiple times over time and these appearances are counted separately uniquedevices devices Number of unique devices during a time_segment as identified by their hardware ( bt_address ) address countscansmostuniquedevice scans Number of scans of the most scanned device during a time_segment across the whole monitoring period Assumptions/Observations NA","title":"Phone Bluetooth"},{"location":"features/phone-bluetooth/#phone-bluetooth","text":"Sensor parameters description for [PHONE_BLUETOOTH] : Key Description [TABLE] Database table where the bluetooth data is stored","title":"Phone Bluetooth"},{"location":"features/phone-bluetooth/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_bluetooth_raw.csv - data/raw/ { pid } /phone_bluetooth_with_datetime.csv - data/interim/ { pid } /phone_bluetooth_features/phone_bluetooth_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_bluetooth.csv \" Parameters description for [PHONE_BLUETOOTH][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_BLUETOOTH features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_BLUETOOTH][PROVIDERS][RAPIDS] : Feature Units Description countscans devices Number of scanned devices during a time_segment , a device can be detected multiple times over time and these appearances are counted separately uniquedevices devices Number of unique devices during a time_segment as identified by their hardware ( bt_address ) address countscansmostuniquedevice scans Number of scans of the most scanned device during a time_segment across the whole monitoring period Assumptions/Observations NA","title":"RAPIDS provider"},{"location":"features/phone-calls/","text":"Phone Calls \u00b6 Sensor parameters description for [PHONE_CALLS] : Key Description [TABLE] Database table where the calls data is stored RAPIDS Provider \u00b6 Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_calls_raw.csv - data/raw/ { pid } /phone_calls_with_datetime.csv - data/raw/ { pid } /phone_calls_with_datetime_unified.csv - data/interim/ { pid } /phone_calls_features/phone_calls_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_calls.csv Parameters description for [PHONE_CALLS][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_CALLS features from the RAPIDS provider [CALL_TYPES] The particular call_type that will be analyzed. The options for this parameter are incoming, outgoing or missed. [FEATURES] Features to be computed for outgoing , incoming , and missed calls. Note that the same features are available for both incoming and outgoing calls, while missed calls has its own set of features. See the tables below. Features description for [PHONE_CALLS][PROVIDERS][RAPIDS] incoming and outgoing calls: Feature Units Description count calls Number of calls of a particular call_type occurred during a particular time_segment . distinctcontacts contacts Number of distinct contacts that are associated with a particular call_type for a particular time_segment meanduration seconds The mean duration of all calls of a particular call_type during a particular time_segment . sumduration seconds The sum of the duration of all calls of a particular call_type during a particular time_segment . minduration seconds The duration of the shortest call of a particular call_type during a particular time_segment . maxduration seconds The duration of the longest call of a particular call_type during a particular time_segment . stdduration seconds The standard deviation of the duration of all the calls of a particular call_type during a particular time_segment . modeduration seconds The mode of the duration of all the calls of a particular call_type during a particular time_segment . entropyduration nats The estimate of the Shannon entropy for the the duration of all the calls of a particular call_type during a particular time_segment . timefirstcall minutes The time in minutes between 12:00am (midnight) and the first call of call_type . timelastcall minutes The time in minutes between 12:00am (midnight) and the last call of call_type . countmostfrequentcontact calls The number of calls of a particular call_type during a particular time_segment of the most frequent contact throughout the monitored period. Features description for [PHONE_CALLS][PROVIDERS][RAPIDS] missed calls: Feature Units Description count calls Number of missed calls that occurred during a particular time_segment . distinctcontacts contacts Number of distinct contacts that are associated with missed calls for a particular time_segment timefirstcall minutes The time in hours from 12:00am (Midnight) that the first missed call occurred. timelastcall minutes The time in hours from 12:00am (Midnight) that the last missed call occurred. countmostfrequentcontact calls The number of missed calls during a particular time_segment of the most frequent contact throughout the monitored period. Assumptions/Observations Traces for iOS calls are unique even for the same contact calling a participant more than once which renders countmostfrequentcontact meaningless and distinctcontacts equal to the total number of traces. [CALL_TYPES] and [FEATURES] keys in config.yaml need to match. For example, [CALL_TYPES] outgoing matches the [FEATURES] key outgoing iOS calls data is transformed to match Android calls data format. See our algorithm","title":"Phone Calls"},{"location":"features/phone-calls/#phone-calls","text":"Sensor parameters description for [PHONE_CALLS] : Key Description [TABLE] Database table where the calls data is stored","title":"Phone Calls"},{"location":"features/phone-calls/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_calls_raw.csv - data/raw/ { pid } /phone_calls_with_datetime.csv - data/raw/ { pid } /phone_calls_with_datetime_unified.csv - data/interim/ { pid } /phone_calls_features/phone_calls_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_calls.csv Parameters description for [PHONE_CALLS][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_CALLS features from the RAPIDS provider [CALL_TYPES] The particular call_type that will be analyzed. The options for this parameter are incoming, outgoing or missed. [FEATURES] Features to be computed for outgoing , incoming , and missed calls. Note that the same features are available for both incoming and outgoing calls, while missed calls has its own set of features. See the tables below. Features description for [PHONE_CALLS][PROVIDERS][RAPIDS] incoming and outgoing calls: Feature Units Description count calls Number of calls of a particular call_type occurred during a particular time_segment . distinctcontacts contacts Number of distinct contacts that are associated with a particular call_type for a particular time_segment meanduration seconds The mean duration of all calls of a particular call_type during a particular time_segment . sumduration seconds The sum of the duration of all calls of a particular call_type during a particular time_segment . minduration seconds The duration of the shortest call of a particular call_type during a particular time_segment . maxduration seconds The duration of the longest call of a particular call_type during a particular time_segment . stdduration seconds The standard deviation of the duration of all the calls of a particular call_type during a particular time_segment . modeduration seconds The mode of the duration of all the calls of a particular call_type during a particular time_segment . entropyduration nats The estimate of the Shannon entropy for the the duration of all the calls of a particular call_type during a particular time_segment . timefirstcall minutes The time in minutes between 12:00am (midnight) and the first call of call_type . timelastcall minutes The time in minutes between 12:00am (midnight) and the last call of call_type . countmostfrequentcontact calls The number of calls of a particular call_type during a particular time_segment of the most frequent contact throughout the monitored period. Features description for [PHONE_CALLS][PROVIDERS][RAPIDS] missed calls: Feature Units Description count calls Number of missed calls that occurred during a particular time_segment . distinctcontacts contacts Number of distinct contacts that are associated with missed calls for a particular time_segment timefirstcall minutes The time in hours from 12:00am (Midnight) that the first missed call occurred. timelastcall minutes The time in hours from 12:00am (Midnight) that the last missed call occurred. countmostfrequentcontact calls The number of missed calls during a particular time_segment of the most frequent contact throughout the monitored period. Assumptions/Observations Traces for iOS calls are unique even for the same contact calling a participant more than once which renders countmostfrequentcontact meaningless and distinctcontacts equal to the total number of traces. [CALL_TYPES] and [FEATURES] keys in config.yaml need to match. For example, [CALL_TYPES] outgoing matches the [FEATURES] key outgoing iOS calls data is transformed to match Android calls data format. See our algorithm","title":"RAPIDS Provider"},{"location":"features/phone-conversation/","text":"Phone Conversation \u00b6 Sensor parameters description for [PHONE_CONVERSATION] : Key Description [TABLE][ANDROID] Database table where the conversation data from Android devices is stored (the AWARE client saves this data on different tables for Android and iOS) [TABLE][IOS] Database table where the conversation data from iOS devices is stored (the AWARE client saves this data on different tables for Android and iOS) RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_conversation_raw.csv - data/raw/ { pid } /phone_conversation_with_datetime.csv - data/raw/ { pid } /phone_conversation_with_datetime_unified.csv - data/interim/ { pid } /phone_conversation_features/phone_conversation_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_conversation.csv Parameters description for [PHONE_CONVERSATION][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_CONVERSATION features from the RAPIDS provider [FEATURES] Features to be computed, see table below [RECORDING_MINUTES] Minutes the plugin was recording audio (default 1 min) [PAUSED_MINUTES] Minutes the plugin was NOT recording audio (default 3 min) Features description for [PHONE_CONVERSATION][PROVIDERS][RAPIDS] : Feature Units Description minutessilence minutes Minutes labeled as silence minutesnoise minutes Minutes labeled as noise minutesvoice minutes Minutes labeled as voice minutesunknown minutes Minutes labeled as unknown sumconversationduration minutes Total duration of all conversations maxconversationduration minutes Longest duration of all conversations minconversationduration minutes Shortest duration of all conversations avgconversationduration minutes Average duration of all conversations sdconversationduration minutes Standard Deviation of the duration of all conversations timefirstconversation minutes Minutes since midnight when the first conversation for a time segment was detected timelastconversation minutes Minutes since midnight when the last conversation for a time segment was detected noisesumenergy L2-norm Sum of all energy values when inference is noise noiseavgenergy L2-norm Average of all energy values when inference is noise noisesdenergy L2-norm Standard Deviation of all energy values when inference is noise noiseminenergy L2-norm Minimum of all energy values when inference is noise noisemaxenergy L2-norm Maximum of all energy values when inference is noise voicesumenergy L2-norm Sum of all energy values when inference is voice voiceavgenergy L2-norm Average of all energy values when inference is voice voicesdenergy L2-norm Standard Deviation of all energy values when inference is voice voiceminenergy L2-norm Minimum of all energy values when inference is voice voicemaxenergy L2-norm Maximum of all energy values when inference is voice silencesensedfraction - Ratio between minutessilence and the sum of (minutessilence, minutesnoise, minutesvoice, minutesunknown) noisesensedfraction - Ratio between minutesnoise and the sum of (minutessilence, minutesnoise, minutesvoice, minutesunknown) voicesensedfraction - Ratio between minutesvoice and the sum of (minutessilence, minutesnoise, minutesvoice, minutesunknown) unknownsensedfraction - Ratio between minutesunknown and the sum of (minutessilence, minutesnoise, minutesvoice, minutesunknown) silenceexpectedfraction - Ration between minutessilence and the number of minutes that in theory should have been sensed based on the record and pause cycle of the plugin (1440 / recordingMinutes+pausedMinutes) noiseexpectedfraction - Ration between minutesnoise and the number of minutes that in theory should have been sensed based on the record and pause cycle of the plugin (1440 / recordingMinutes+pausedMinutes) voiceexpectedfraction - Ration between minutesvoice and the number of minutes that in theory should have been sensed based on the record and pause cycle of the plugin (1440 / recordingMinutes+pausedMinutes) unknownexpectedfraction - Ration between minutesunknown and the number of minutes that in theory should have been sensed based on the record and pause cycle of the plugin (1440 / recordingMinutes+pausedMinutes) Assumptions/Observations The timestamp of conversation rows in iOS is in seconds so we convert it to milliseconds to match Android\u2019s format","title":"Phone Conversation"},{"location":"features/phone-conversation/#phone-conversation","text":"Sensor parameters description for [PHONE_CONVERSATION] : Key Description [TABLE][ANDROID] Database table where the conversation data from Android devices is stored (the AWARE client saves this data on different tables for Android and iOS) [TABLE][IOS] Database table where the conversation data from iOS devices is stored (the AWARE client saves this data on different tables for Android and iOS)","title":"Phone Conversation"},{"location":"features/phone-conversation/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_conversation_raw.csv - data/raw/ { pid } /phone_conversation_with_datetime.csv - data/raw/ { pid } /phone_conversation_with_datetime_unified.csv - data/interim/ { pid } /phone_conversation_features/phone_conversation_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_conversation.csv Parameters description for [PHONE_CONVERSATION][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_CONVERSATION features from the RAPIDS provider [FEATURES] Features to be computed, see table below [RECORDING_MINUTES] Minutes the plugin was recording audio (default 1 min) [PAUSED_MINUTES] Minutes the plugin was NOT recording audio (default 3 min) Features description for [PHONE_CONVERSATION][PROVIDERS][RAPIDS] : Feature Units Description minutessilence minutes Minutes labeled as silence minutesnoise minutes Minutes labeled as noise minutesvoice minutes Minutes labeled as voice minutesunknown minutes Minutes labeled as unknown sumconversationduration minutes Total duration of all conversations maxconversationduration minutes Longest duration of all conversations minconversationduration minutes Shortest duration of all conversations avgconversationduration minutes Average duration of all conversations sdconversationduration minutes Standard Deviation of the duration of all conversations timefirstconversation minutes Minutes since midnight when the first conversation for a time segment was detected timelastconversation minutes Minutes since midnight when the last conversation for a time segment was detected noisesumenergy L2-norm Sum of all energy values when inference is noise noiseavgenergy L2-norm Average of all energy values when inference is noise noisesdenergy L2-norm Standard Deviation of all energy values when inference is noise noiseminenergy L2-norm Minimum of all energy values when inference is noise noisemaxenergy L2-norm Maximum of all energy values when inference is noise voicesumenergy L2-norm Sum of all energy values when inference is voice voiceavgenergy L2-norm Average of all energy values when inference is voice voicesdenergy L2-norm Standard Deviation of all energy values when inference is voice voiceminenergy L2-norm Minimum of all energy values when inference is voice voicemaxenergy L2-norm Maximum of all energy values when inference is voice silencesensedfraction - Ratio between minutessilence and the sum of (minutessilence, minutesnoise, minutesvoice, minutesunknown) noisesensedfraction - Ratio between minutesnoise and the sum of (minutessilence, minutesnoise, minutesvoice, minutesunknown) voicesensedfraction - Ratio between minutesvoice and the sum of (minutessilence, minutesnoise, minutesvoice, minutesunknown) unknownsensedfraction - Ratio between minutesunknown and the sum of (minutessilence, minutesnoise, minutesvoice, minutesunknown) silenceexpectedfraction - Ration between minutessilence and the number of minutes that in theory should have been sensed based on the record and pause cycle of the plugin (1440 / recordingMinutes+pausedMinutes) noiseexpectedfraction - Ration between minutesnoise and the number of minutes that in theory should have been sensed based on the record and pause cycle of the plugin (1440 / recordingMinutes+pausedMinutes) voiceexpectedfraction - Ration between minutesvoice and the number of minutes that in theory should have been sensed based on the record and pause cycle of the plugin (1440 / recordingMinutes+pausedMinutes) unknownexpectedfraction - Ration between minutesunknown and the number of minutes that in theory should have been sensed based on the record and pause cycle of the plugin (1440 / recordingMinutes+pausedMinutes) Assumptions/Observations The timestamp of conversation rows in iOS is in seconds so we convert it to milliseconds to match Android\u2019s format","title":"RAPIDS provider"},{"location":"features/phone-data-yield/","text":"Phone Data Yield \u00b6 This is a combinatorial sensor which means that we use the data from multiple sensors to extract data yield features. Data yield features can be used to remove rows ( time segments ) that do not contain enough data. You should decide what is your \u201cenough\u201d threshold depending on the type of sensors you collected (frequency vs event based, e.g. acceleroemter vs calls), the length of your study, and the rates of missing data that your analysis could handle. Why is data yield important? Imagine that you want to extract PHONE_CALL features on daily segments ( 00:00 to 23:59 ). Let\u2019s say that on day 1 the phone logged 10 calls and 23 hours of data from other sensors and on day 2 the phone logged 10 calls and only 2 hours of data from other sensors. It\u2019s more likely that other calls were placed on the 22 hours of data that you didn\u2019t log on day 2 than on the 1 hour of data you didn\u2019t log on day 1, and so including day 2 in your analysis could bias your results. Sensor parameters description for [PHONE_DATA_YIELD] : Key Description [SENSORS] One or more phone sensor config keys (e.g. PHONE_MESSAGE ). The more keys you include the more accurately RAPIDS can approximate the time an smartphone was sensing data. The supported phone sensors you can include in this list are outlined below ( do NOT include Fitbit sensors ). Supported phone sensors for [PHONE_DATA_YIELD][SENSORS] PHONE_ACCELEROMETER PHONE_ACTIVITY_RECOGNITION PHONE_APPLICATIONS_FOREGROUND PHONE_BATTERY PHONE_BLUETOOTH PHONE_CALLS PHONE_CONVERSATION PHONE_MESSAGES PHONE_LIGHT PHONE_LOCATIONS PHONE_SCREEN PHONE_WIFI_VISIBLE PHONE_WIFI_CONNECTED RAPIDS provider \u00b6 Before explaining the data yield features, let\u2019s define the following relevant concepts: A valid minute is any 60 second window when any phone sensor logged at least 1 row of data A valid hour is any 60 minute window with at least X valid minutes. The X or threshold is given by [MINUTE_RATIO_THRESHOLD_FOR_VALID_YIELDED_HOURS] The timestamps of all sensors are concatenated and then grouped per time segment. Minute and hour windows are created from the beginning of each time segment instance and these windows are marked as valid based on the definitions above. The duration of each time segment is taken into account to compute the features described below. Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } / { sensor } _raw.csv # one for every [PHONE_DATA_YIELD][SENSORS] - data/interim/ { pid } /phone_yielded_timestamps.csv - data/interim/ { pid } /phone_yielded_timestamps_with_datetime.csv - data/interim/ { pid } /phone_data_yield_features/phone_data_yield_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_data_yield.csv Parameters description for [PHONE_DATA_YIELD][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_DATA_YIELD features from the RAPIDS provider [FEATURES] Features to be computed, see table below [MINUTE_RATIO_THRESHOLD_FOR_VALID_YIELDED_HOURS] The proportion [0.0 ,1.0] of valid minutes in a 60-minute window necessary to flag that window as valid. Features description for [PHONE_DATA_YIELD][PROVIDERS][RAPIDS] : Feature Units Description ratiovalidyieldedminutes rows The ratio between the number of valid minutes and the duration in minutes of a time segment. ratiovalidyieldedhours lux The ratio between the number of valid hours and the duration in hours of a time segment. If the time segment is shorter than 1 hour this feature will always be 1. Assumptions/Observations We recommend using ratiovalidyieldedminutes on time segments that are shorter than two or three hours and ratiovalidyieldedhours for longer segments. This is because relying on yielded minutes only can be misleading when a big chunk of those missing minutes are clustered together. For example, let\u2019s assume we are working with a 24-hour time segment that is missing 12 hours of data. Two extreme cases can occur: the 12 missing hours are from the beginning of the segment or 30 minutes could be missing from every hour (24 * 30 minutes = 12 hours). ratiovalidyieldedminutes would be 0.5 for both a and b (hinting the missing circumstances are similar). However, ratiovalidyieldedhours would be 0.5 for a and 1.0 for b if [MINUTE_RATIO_THRESHOLD_FOR_VALID_YIELDED_HOURS] is between [0.0 and 0.49] (hinting that the missing circumstances might be more favorable for b . In other words, sensed data for b is more evenly spread compared to a .","title":"Phone Data Yield"},{"location":"features/phone-data-yield/#phone-data-yield","text":"This is a combinatorial sensor which means that we use the data from multiple sensors to extract data yield features. Data yield features can be used to remove rows ( time segments ) that do not contain enough data. You should decide what is your \u201cenough\u201d threshold depending on the type of sensors you collected (frequency vs event based, e.g. acceleroemter vs calls), the length of your study, and the rates of missing data that your analysis could handle. Why is data yield important? Imagine that you want to extract PHONE_CALL features on daily segments ( 00:00 to 23:59 ). Let\u2019s say that on day 1 the phone logged 10 calls and 23 hours of data from other sensors and on day 2 the phone logged 10 calls and only 2 hours of data from other sensors. It\u2019s more likely that other calls were placed on the 22 hours of data that you didn\u2019t log on day 2 than on the 1 hour of data you didn\u2019t log on day 1, and so including day 2 in your analysis could bias your results. Sensor parameters description for [PHONE_DATA_YIELD] : Key Description [SENSORS] One or more phone sensor config keys (e.g. PHONE_MESSAGE ). The more keys you include the more accurately RAPIDS can approximate the time an smartphone was sensing data. The supported phone sensors you can include in this list are outlined below ( do NOT include Fitbit sensors ). Supported phone sensors for [PHONE_DATA_YIELD][SENSORS] PHONE_ACCELEROMETER PHONE_ACTIVITY_RECOGNITION PHONE_APPLICATIONS_FOREGROUND PHONE_BATTERY PHONE_BLUETOOTH PHONE_CALLS PHONE_CONVERSATION PHONE_MESSAGES PHONE_LIGHT PHONE_LOCATIONS PHONE_SCREEN PHONE_WIFI_VISIBLE PHONE_WIFI_CONNECTED","title":"Phone Data Yield"},{"location":"features/phone-data-yield/#rapids-provider","text":"Before explaining the data yield features, let\u2019s define the following relevant concepts: A valid minute is any 60 second window when any phone sensor logged at least 1 row of data A valid hour is any 60 minute window with at least X valid minutes. The X or threshold is given by [MINUTE_RATIO_THRESHOLD_FOR_VALID_YIELDED_HOURS] The timestamps of all sensors are concatenated and then grouped per time segment. Minute and hour windows are created from the beginning of each time segment instance and these windows are marked as valid based on the definitions above. The duration of each time segment is taken into account to compute the features described below. Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } / { sensor } _raw.csv # one for every [PHONE_DATA_YIELD][SENSORS] - data/interim/ { pid } /phone_yielded_timestamps.csv - data/interim/ { pid } /phone_yielded_timestamps_with_datetime.csv - data/interim/ { pid } /phone_data_yield_features/phone_data_yield_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_data_yield.csv Parameters description for [PHONE_DATA_YIELD][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_DATA_YIELD features from the RAPIDS provider [FEATURES] Features to be computed, see table below [MINUTE_RATIO_THRESHOLD_FOR_VALID_YIELDED_HOURS] The proportion [0.0 ,1.0] of valid minutes in a 60-minute window necessary to flag that window as valid. Features description for [PHONE_DATA_YIELD][PROVIDERS][RAPIDS] : Feature Units Description ratiovalidyieldedminutes rows The ratio between the number of valid minutes and the duration in minutes of a time segment. ratiovalidyieldedhours lux The ratio between the number of valid hours and the duration in hours of a time segment. If the time segment is shorter than 1 hour this feature will always be 1. Assumptions/Observations We recommend using ratiovalidyieldedminutes on time segments that are shorter than two or three hours and ratiovalidyieldedhours for longer segments. This is because relying on yielded minutes only can be misleading when a big chunk of those missing minutes are clustered together. For example, let\u2019s assume we are working with a 24-hour time segment that is missing 12 hours of data. Two extreme cases can occur: the 12 missing hours are from the beginning of the segment or 30 minutes could be missing from every hour (24 * 30 minutes = 12 hours). ratiovalidyieldedminutes would be 0.5 for both a and b (hinting the missing circumstances are similar). However, ratiovalidyieldedhours would be 0.5 for a and 1.0 for b if [MINUTE_RATIO_THRESHOLD_FOR_VALID_YIELDED_HOURS] is between [0.0 and 0.49] (hinting that the missing circumstances might be more favorable for b . In other words, sensed data for b is more evenly spread compared to a .","title":"RAPIDS provider"},{"location":"features/phone-light/","text":"Phone Light \u00b6 Sensor parameters description for [PHONE_LIGHT] : Key Description [TABLE] Database table where the light data is stored RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_light_raw.csv - data/raw/ { pid } /phone_light_with_datetime.csv - data/interim/ { pid } /phone_light_features/phone_light_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_light.csv Parameters description for [PHONE_LIGHT][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_LIGHT features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_LIGHT][PROVIDERS][RAPIDS] : Feature Units Description count rows Number light sensor rows recorded. maxlux lux The maximum ambient luminance. minlux lux The minimum ambient luminance. avglux lux The average ambient luminance. medianlux lux The median ambient luminance. stdlux lux The standard deviation of ambient luminance. Assumptions/Observations NA","title":"Phone Light"},{"location":"features/phone-light/#phone-light","text":"Sensor parameters description for [PHONE_LIGHT] : Key Description [TABLE] Database table where the light data is stored","title":"Phone Light"},{"location":"features/phone-light/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_light_raw.csv - data/raw/ { pid } /phone_light_with_datetime.csv - data/interim/ { pid } /phone_light_features/phone_light_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_light.csv Parameters description for [PHONE_LIGHT][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_LIGHT features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_LIGHT][PROVIDERS][RAPIDS] : Feature Units Description count rows Number light sensor rows recorded. maxlux lux The maximum ambient luminance. minlux lux The minimum ambient luminance. avglux lux The average ambient luminance. medianlux lux The median ambient luminance. stdlux lux The standard deviation of ambient luminance. Assumptions/Observations NA","title":"RAPIDS provider"},{"location":"features/phone-locations/","text":"Phone Locations \u00b6 Sensor parameters description for [PHONE_LOCATIONS] : Key Description [TABLE] Database table where the location data is stored [LOCATIONS_TO_USE] Type of location data to use, one of ALL , GPS or FUSED_RESAMPLED . This filter is based on the provider column of the AWARE locations table, ALL includes every row, GPS only includes rows where provider is gps, and FUSED_RESAMPLED only includes rows where provider is fused after being resampled. [FUSED_RESAMPLED_CONSECUTIVE_THRESHOLD] if FUSED_RESAMPLED is used, the original fused data has to be resampled, a location row will be resampled to the next valid timestamp (see the Assumptions/Observations below) only if the time difference between them is less or equal than this threshold (in minutes). [FUSED_RESAMPLED_TIME_SINCE_VALID_LOCATION] if FUSED_RESAMPLED is used, the original fused data has to be resampled, a location row will be resampled at most for this long (in minutes) Assumptions/Observations Types of location data to use AWARE Android and iOS clients can collect location coordinates through the phone's GPS, the network cellular towers around the phone or Google's fused location API. If you want to use only the GPS provider set [LOCATIONS_TO_USE] to GPS , if you want to use all providers (not recommended due to the difference in accuracy) set [LOCATIONS_TO_USE] to ALL , if your AWARE client was configured to use fused location only or want to focus only on this provider, set [LOCATIONS_TO_USE] to RESAMPLE_FUSED . RESAMPLE_FUSED takes the original fused location coordinates and replicates each pair forward in time as long as the phone was sensing data as indicated by the joined timestamps of [PHONE_DATA_YIELD][SENSORS] , this is done because Google's API only logs a new location coordinate pair when it is sufficiently different in time or space from the previous one. There are two parameters associated with resampling fused location. FUSED_RESAMPLED_CONSECUTIVE_THRESHOLD (in minutes, default 30) controls the maximum gap between any two coordinate pairs to replicate the last known pair (for example, participant A's phone did not collect data between 10.30am and 10:50am and between 11:05am and 11:40am, the last known coordinate pair will be replicated during the first period but not the second, in other words, we assume that we cannot longer guarantee the participant stayed at the last known location if the phone did not sense data for more than 30 minutes). FUSED_RESAMPLED_TIME_SINCE_VALID_LOCATION (in minutes, default 720 or 12 hours) stops the last known fused location from being replicated longer that this threshold even if the phone was sensing data continuously (for example, participant A went home at 9pm and their phone was sensing data without gaps until 11am the next morning, the last known location will only be replicated until 9am). If you have suggestions to modify or improve this resampling, let us know. BARNETT provider \u00b6 These features are based on the original open-source implementation by Barnett et al and some features created by Canzian et al . Available time segments and platforms Available only for segments that start at 00:00:00 and end at 23:59:59 of the same day (daily segments) Available for Android and iOS File Sequence - data/raw/ { pid } /phone_locations_raw.csv - data/interim/ { pid } /phone_locations_processed.csv - data/interim/ { pid } /phone_locations_processed_with_datetime.csv - data/interim/ { pid } /phone_locations_features/phone_locations_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_locations.csv Parameters description for [PHONE_LOCATIONS][PROVIDERS][BARNETT] : Key Description [COMPUTE] Set to True to extract PHONE_LOCATIONS features from the BARNETT provider [FEATURES] Features to be computed, see table below [ACCURACY_LIMIT] An integer in meters, any location rows with an accuracy higher than this will be dropped. This number means there\u2019s a 68% probability the true location is within this radius [TIMEZONE] Timezone where the location data was collected. By default points to the one defined in the Configuration [MINUTES_DATA_USED] Set to True to include an extra column in the final location feature file containing the number of minutes used to compute the features on each time segment. Use this for quality control purposes, the more data minutes exist for a period, the more reliable its features should be. For fused location, a single minute can contain more than one coordinate pair if the participant is moving fast enough. Features description for [PHONE_LOCATIONS][PROVIDERS][BARNETT] adapted from Beiwe Summary Statistics : Feature Units Description hometime minutes Time at home. Time spent at home in minutes. Home is the most visited significant location between 8 pm and 8 am including any pauses within a 200-meter radius. disttravelled meters Total distance travelled over a day (flights). rog meters The Radius of Gyration (rog) is a measure in meters of the area covered by a person over a day. A centroid is calculated for all the places (pauses) visited during a day and a weighted distance between all the places and that centroid is computed. The weights are proportional to the time spent in each place. maxdiam meters The maximum diameter is the largest distance between any two pauses. maxhomedist meters The maximum distance from home in meters. siglocsvisited locations The number of significant locations visited during the day. Significant locations are computed using k-means clustering over pauses found in the whole monitoring period. The number of clusters is found iterating k from 1 to 200 stopping until the centroids of two significant locations are within 400 meters of one another. avgflightlen meters Mean length of all flights. stdflightlen meters Standard deviation of the length of all flights. avgflightdur seconds Mean duration of all flights. stdflightdur seconds The standard deviation of the duration of all flights. probpause - The fraction of a day spent in a pause (as opposed to a flight) siglocentropy nats Shannon\u2019s entropy measurement based on the proportion of time spent at each significant location visited during a day. circdnrtn - A continuous metric quantifying a person\u2019s circadian routine that can take any value between 0 and 1, where 0 represents a daily routine completely different from any other sensed days and 1 a routine the same as every other sensed day. wkenddayrtn - Same as circdnrtn but computed separately for weekends and weekdays. Assumptions/Observations Barnett's et al features These features are based on a Pause-Flight model. A pause is defined as a mobiity trace (location pings) within a certain duration and distance (by default 300 seconds and 60 meters). A flight is any mobility trace between two pauses. Data is resampled and imputed before the features are computed. See Barnett et al for more information. In RAPIDS we only expose two parameters for these features (timezone and accuracy limit). You can change other parameters in src/features/phone_locations/barnett/library/MobilityFeatures.R . Significant Locations Significant locations are determined using K-means clustering on pauses longer than 10 minutes. The number of clusters (K) is increased until no two clusters are within 400 meters from each other. After this, pauses within a certain range of a cluster (200 meters by default) will count as a visit to that significant location. This description was adapted from the Supplementary Materials of Barnett et al . The Circadian Calculation For a detailed description of how this is calculated, see Canzian et al . DORYAB provider \u00b6 These features are based on the original implementation by Doryab et al. . Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_locations_raw.csv - data/interim/ { pid } /phone_locations_processed.csv - data/interim/ { pid } /phone_locations_processed_with_datetime.csv - data/interim/ { pid } /phone_locations_features/phone_locations_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_locations.csv Parameters description for [PHONE_LOCATIONS][PROVIDERS][BARNETT] : Key Description [COMPUTE] Set to True to extract PHONE_LOCATIONS features from the BARNETT provider [FEATURES] Features to be computed, see table below [DBSCAN_EPS] The maximum distance in meters between two samples for one to be considered as in the neighborhood of the other. This is not a maximum bound on the distances of points within a cluster. This is the most important DBSCAN parameter to choose appropriately for your data set and distance function. [DBSCAN_MINSAMPLES] The number of samples (or total weight) in a neighborhood for a point to be considered as a core point of a cluster. This includes the point itself. [THRESHOLD_STATIC] It is the threshold value in km/hr which labels a row as Static or Moving. [MAXIMUM_GAP_ALLOWED] The maximum gap (in seconds) allowed between any two consecutive rows for them to be considered part of the same displacement. If this threshold is too high, it can throw speed and distance calculations off for periods when the the phone was not sensing. [MINUTES_DATA_USED] Set to True to include an extra column in the final location feature file containing the number of minutes used to compute the features on each time segment. Use this for quality control purposes, the more data minutes exist for a period, the more reliable its features should be. For fused location, a single minute can contain more than one coordinate pair if the participant is moving fast enough. [SAMPLING_FREQUENCY] Expected time difference between any two location rows in minutes. If set to 0 , the sampling frequency will be inferred automatically as the median of all the differences between any two consecutive row timestamps (recommended if you are using FUSED_RESAMPLED data). This parameter impacts all the time calculations. Features description for [PHONE_LOCATIONS][PROVIDERS][BARNETT] : Feature Units Description locationvariance \\(meters^2\\) The sum of the variances of the latitude and longitude columns. loglocationvariance - Log of the sum of the variances of the latitude and longitude columns. totaldistance meters Total distance travelled in a time segment using the haversine formula. averagespeed km/hr Average speed in a time segment considering only the instances labeled as Moving. varspeed km/hr Speed variance in a time segment considering only the instances labeled as Moving. circadianmovement - \"It encodes the extent to which a person\u2019s location patterns follow a 24-hour circadian cycle.\" Doryab et al. . numberofsignificantplaces places Number of significant locations visited. It is calculated using the DBSCAN clustering algorithm which takes in EPS and MIN_SAMPLES as parameters to identify clusters. Each cluster is a significant place. numberlocationtransitions transitions Number of movements between any two clusters in a time segment. radiusgyration meters Quantifies the area covered by a participant timeattop1location minutes Time spent at the most significant location. timeattop2location minutes Time spent at the 2 nd most significant location. timeattop3location minutes Time spent at the 3 rd most significant location. movingtostaticratio - Ratio between the number of rows labeled Moving versus Static outlierstimepercent - Ratio between the number of rows that belong to non-significant clusters divided by the total number of rows in a time segment. maxlengthstayatclusters minutes Maximum time spent in a cluster (significant location). minlengthstayatclusters minutes Minimum time spent in a cluster (significant location). meanlengthstayatclusters minutes Average time spent in a cluster (significant location). stdlengthstayatclusters minutes Standard deviation of time spent in a cluster (significant location). locationentropy nats Shannon Entropy computed over the row count of each cluster (significant location), it will be higher the more rows belong to a cluster (i.e. the more time a participant spent at a significant location). normalizedlocationentropy nats Shannon Entropy computed over the row count of each cluster (significant location) divided by the number of clusters, it will be higher the more rows belong to a cluster (i.e. the more time a participant spent at a significant location). Assumptions/Observations Significant Locations Identified Significant locations are determined using DBSCAN clustering on locations that a patient visit over the course of the period of data collection. The Circadian Calculation For a detailed description of how this is calculated, see Canzian et al .","title":"Phone Locations"},{"location":"features/phone-locations/#phone-locations","text":"Sensor parameters description for [PHONE_LOCATIONS] : Key Description [TABLE] Database table where the location data is stored [LOCATIONS_TO_USE] Type of location data to use, one of ALL , GPS or FUSED_RESAMPLED . This filter is based on the provider column of the AWARE locations table, ALL includes every row, GPS only includes rows where provider is gps, and FUSED_RESAMPLED only includes rows where provider is fused after being resampled. [FUSED_RESAMPLED_CONSECUTIVE_THRESHOLD] if FUSED_RESAMPLED is used, the original fused data has to be resampled, a location row will be resampled to the next valid timestamp (see the Assumptions/Observations below) only if the time difference between them is less or equal than this threshold (in minutes). [FUSED_RESAMPLED_TIME_SINCE_VALID_LOCATION] if FUSED_RESAMPLED is used, the original fused data has to be resampled, a location row will be resampled at most for this long (in minutes) Assumptions/Observations Types of location data to use AWARE Android and iOS clients can collect location coordinates through the phone's GPS, the network cellular towers around the phone or Google's fused location API. If you want to use only the GPS provider set [LOCATIONS_TO_USE] to GPS , if you want to use all providers (not recommended due to the difference in accuracy) set [LOCATIONS_TO_USE] to ALL , if your AWARE client was configured to use fused location only or want to focus only on this provider, set [LOCATIONS_TO_USE] to RESAMPLE_FUSED . RESAMPLE_FUSED takes the original fused location coordinates and replicates each pair forward in time as long as the phone was sensing data as indicated by the joined timestamps of [PHONE_DATA_YIELD][SENSORS] , this is done because Google's API only logs a new location coordinate pair when it is sufficiently different in time or space from the previous one. There are two parameters associated with resampling fused location. FUSED_RESAMPLED_CONSECUTIVE_THRESHOLD (in minutes, default 30) controls the maximum gap between any two coordinate pairs to replicate the last known pair (for example, participant A's phone did not collect data between 10.30am and 10:50am and between 11:05am and 11:40am, the last known coordinate pair will be replicated during the first period but not the second, in other words, we assume that we cannot longer guarantee the participant stayed at the last known location if the phone did not sense data for more than 30 minutes). FUSED_RESAMPLED_TIME_SINCE_VALID_LOCATION (in minutes, default 720 or 12 hours) stops the last known fused location from being replicated longer that this threshold even if the phone was sensing data continuously (for example, participant A went home at 9pm and their phone was sensing data without gaps until 11am the next morning, the last known location will only be replicated until 9am). If you have suggestions to modify or improve this resampling, let us know.","title":"Phone Locations"},{"location":"features/phone-locations/#barnett-provider","text":"These features are based on the original open-source implementation by Barnett et al and some features created by Canzian et al . Available time segments and platforms Available only for segments that start at 00:00:00 and end at 23:59:59 of the same day (daily segments) Available for Android and iOS File Sequence - data/raw/ { pid } /phone_locations_raw.csv - data/interim/ { pid } /phone_locations_processed.csv - data/interim/ { pid } /phone_locations_processed_with_datetime.csv - data/interim/ { pid } /phone_locations_features/phone_locations_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_locations.csv Parameters description for [PHONE_LOCATIONS][PROVIDERS][BARNETT] : Key Description [COMPUTE] Set to True to extract PHONE_LOCATIONS features from the BARNETT provider [FEATURES] Features to be computed, see table below [ACCURACY_LIMIT] An integer in meters, any location rows with an accuracy higher than this will be dropped. This number means there\u2019s a 68% probability the true location is within this radius [TIMEZONE] Timezone where the location data was collected. By default points to the one defined in the Configuration [MINUTES_DATA_USED] Set to True to include an extra column in the final location feature file containing the number of minutes used to compute the features on each time segment. Use this for quality control purposes, the more data minutes exist for a period, the more reliable its features should be. For fused location, a single minute can contain more than one coordinate pair if the participant is moving fast enough. Features description for [PHONE_LOCATIONS][PROVIDERS][BARNETT] adapted from Beiwe Summary Statistics : Feature Units Description hometime minutes Time at home. Time spent at home in minutes. Home is the most visited significant location between 8 pm and 8 am including any pauses within a 200-meter radius. disttravelled meters Total distance travelled over a day (flights). rog meters The Radius of Gyration (rog) is a measure in meters of the area covered by a person over a day. A centroid is calculated for all the places (pauses) visited during a day and a weighted distance between all the places and that centroid is computed. The weights are proportional to the time spent in each place. maxdiam meters The maximum diameter is the largest distance between any two pauses. maxhomedist meters The maximum distance from home in meters. siglocsvisited locations The number of significant locations visited during the day. Significant locations are computed using k-means clustering over pauses found in the whole monitoring period. The number of clusters is found iterating k from 1 to 200 stopping until the centroids of two significant locations are within 400 meters of one another. avgflightlen meters Mean length of all flights. stdflightlen meters Standard deviation of the length of all flights. avgflightdur seconds Mean duration of all flights. stdflightdur seconds The standard deviation of the duration of all flights. probpause - The fraction of a day spent in a pause (as opposed to a flight) siglocentropy nats Shannon\u2019s entropy measurement based on the proportion of time spent at each significant location visited during a day. circdnrtn - A continuous metric quantifying a person\u2019s circadian routine that can take any value between 0 and 1, where 0 represents a daily routine completely different from any other sensed days and 1 a routine the same as every other sensed day. wkenddayrtn - Same as circdnrtn but computed separately for weekends and weekdays. Assumptions/Observations Barnett's et al features These features are based on a Pause-Flight model. A pause is defined as a mobiity trace (location pings) within a certain duration and distance (by default 300 seconds and 60 meters). A flight is any mobility trace between two pauses. Data is resampled and imputed before the features are computed. See Barnett et al for more information. In RAPIDS we only expose two parameters for these features (timezone and accuracy limit). You can change other parameters in src/features/phone_locations/barnett/library/MobilityFeatures.R . Significant Locations Significant locations are determined using K-means clustering on pauses longer than 10 minutes. The number of clusters (K) is increased until no two clusters are within 400 meters from each other. After this, pauses within a certain range of a cluster (200 meters by default) will count as a visit to that significant location. This description was adapted from the Supplementary Materials of Barnett et al . The Circadian Calculation For a detailed description of how this is calculated, see Canzian et al .","title":"BARNETT provider"},{"location":"features/phone-locations/#doryab-provider","text":"These features are based on the original implementation by Doryab et al. . Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_locations_raw.csv - data/interim/ { pid } /phone_locations_processed.csv - data/interim/ { pid } /phone_locations_processed_with_datetime.csv - data/interim/ { pid } /phone_locations_features/phone_locations_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_locations.csv Parameters description for [PHONE_LOCATIONS][PROVIDERS][BARNETT] : Key Description [COMPUTE] Set to True to extract PHONE_LOCATIONS features from the BARNETT provider [FEATURES] Features to be computed, see table below [DBSCAN_EPS] The maximum distance in meters between two samples for one to be considered as in the neighborhood of the other. This is not a maximum bound on the distances of points within a cluster. This is the most important DBSCAN parameter to choose appropriately for your data set and distance function. [DBSCAN_MINSAMPLES] The number of samples (or total weight) in a neighborhood for a point to be considered as a core point of a cluster. This includes the point itself. [THRESHOLD_STATIC] It is the threshold value in km/hr which labels a row as Static or Moving. [MAXIMUM_GAP_ALLOWED] The maximum gap (in seconds) allowed between any two consecutive rows for them to be considered part of the same displacement. If this threshold is too high, it can throw speed and distance calculations off for periods when the the phone was not sensing. [MINUTES_DATA_USED] Set to True to include an extra column in the final location feature file containing the number of minutes used to compute the features on each time segment. Use this for quality control purposes, the more data minutes exist for a period, the more reliable its features should be. For fused location, a single minute can contain more than one coordinate pair if the participant is moving fast enough. [SAMPLING_FREQUENCY] Expected time difference between any two location rows in minutes. If set to 0 , the sampling frequency will be inferred automatically as the median of all the differences between any two consecutive row timestamps (recommended if you are using FUSED_RESAMPLED data). This parameter impacts all the time calculations. Features description for [PHONE_LOCATIONS][PROVIDERS][BARNETT] : Feature Units Description locationvariance \\(meters^2\\) The sum of the variances of the latitude and longitude columns. loglocationvariance - Log of the sum of the variances of the latitude and longitude columns. totaldistance meters Total distance travelled in a time segment using the haversine formula. averagespeed km/hr Average speed in a time segment considering only the instances labeled as Moving. varspeed km/hr Speed variance in a time segment considering only the instances labeled as Moving. circadianmovement - \"It encodes the extent to which a person\u2019s location patterns follow a 24-hour circadian cycle.\" Doryab et al. . numberofsignificantplaces places Number of significant locations visited. It is calculated using the DBSCAN clustering algorithm which takes in EPS and MIN_SAMPLES as parameters to identify clusters. Each cluster is a significant place. numberlocationtransitions transitions Number of movements between any two clusters in a time segment. radiusgyration meters Quantifies the area covered by a participant timeattop1location minutes Time spent at the most significant location. timeattop2location minutes Time spent at the 2 nd most significant location. timeattop3location minutes Time spent at the 3 rd most significant location. movingtostaticratio - Ratio between the number of rows labeled Moving versus Static outlierstimepercent - Ratio between the number of rows that belong to non-significant clusters divided by the total number of rows in a time segment. maxlengthstayatclusters minutes Maximum time spent in a cluster (significant location). minlengthstayatclusters minutes Minimum time spent in a cluster (significant location). meanlengthstayatclusters minutes Average time spent in a cluster (significant location). stdlengthstayatclusters minutes Standard deviation of time spent in a cluster (significant location). locationentropy nats Shannon Entropy computed over the row count of each cluster (significant location), it will be higher the more rows belong to a cluster (i.e. the more time a participant spent at a significant location). normalizedlocationentropy nats Shannon Entropy computed over the row count of each cluster (significant location) divided by the number of clusters, it will be higher the more rows belong to a cluster (i.e. the more time a participant spent at a significant location). Assumptions/Observations Significant Locations Identified Significant locations are determined using DBSCAN clustering on locations that a patient visit over the course of the period of data collection. The Circadian Calculation For a detailed description of how this is calculated, see Canzian et al .","title":"DORYAB provider"},{"location":"features/phone-messages/","text":"Phone Messages \u00b6 Sensor parameters description for [PHONE_MESSAGES] : Key Description [TABLE] Database table where the messages data is stored RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_messages_raw.csv - data/raw/ { pid } /phone_messages_with_datetime.csv - data/interim/ { pid } /phone_messages_features/phone_messages_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_messages.csv Parameters description for [PHONE_MESSAGES][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_MESSAGES features from the RAPIDS provider [MESSAGES_TYPES] The messages_type that will be analyzed. The options for this parameter are received or sent . [FEATURES] Features to be computed, see table below for [MESSAGES_TYPES] received and sent Features description for [PHONE_MESSAGES][PROVIDERS][RAPIDS] : Feature Units Description count messages Number of messages of type messages_type that occurred during a particular time_segment . distinctcontacts contacts Number of distinct contacts that are associated with a particular messages_type during a particular time_segment . timefirstmessages minutes Number of minutes between 12:00am (midnight) and the first message of a particular messages_type during a particular time_segment . timelastmessages minutes Number of minutes between 12:00am (midnight) and the last message of a particular messages_type during a particular time_segment . countmostfrequentcontact messages Number of messages from the contact with the most messages of messages_type during a time_segment throughout the whole dataset of each participant. Assumptions/Observations [MESSAGES_TYPES] and [FEATURES] keys in config.yaml need to match. For example, [MESSAGES_TYPES] sent matches the [FEATURES] key sent","title":"Phone Messages"},{"location":"features/phone-messages/#phone-messages","text":"Sensor parameters description for [PHONE_MESSAGES] : Key Description [TABLE] Database table where the messages data is stored","title":"Phone Messages"},{"location":"features/phone-messages/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_messages_raw.csv - data/raw/ { pid } /phone_messages_with_datetime.csv - data/interim/ { pid } /phone_messages_features/phone_messages_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_messages.csv Parameters description for [PHONE_MESSAGES][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_MESSAGES features from the RAPIDS provider [MESSAGES_TYPES] The messages_type that will be analyzed. The options for this parameter are received or sent . [FEATURES] Features to be computed, see table below for [MESSAGES_TYPES] received and sent Features description for [PHONE_MESSAGES][PROVIDERS][RAPIDS] : Feature Units Description count messages Number of messages of type messages_type that occurred during a particular time_segment . distinctcontacts contacts Number of distinct contacts that are associated with a particular messages_type during a particular time_segment . timefirstmessages minutes Number of minutes between 12:00am (midnight) and the first message of a particular messages_type during a particular time_segment . timelastmessages minutes Number of minutes between 12:00am (midnight) and the last message of a particular messages_type during a particular time_segment . countmostfrequentcontact messages Number of messages from the contact with the most messages of messages_type during a time_segment throughout the whole dataset of each participant. Assumptions/Observations [MESSAGES_TYPES] and [FEATURES] keys in config.yaml need to match. For example, [MESSAGES_TYPES] sent matches the [FEATURES] key sent","title":"RAPIDS provider"},{"location":"features/phone-screen/","text":"Phone Screen \u00b6 Sensor parameters description for [PHONE_SCREEN] : Key Description [TABLE] Database table where the screen data is stored RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_screen_raw.csv - data/raw/ { pid } /phone_screen_with_datetime.csv - data/raw/ { pid } /phone_screen_with_datetime_unified.csv - data/interim/ { pid } /phone_screen_episodes.csv - data/interim/ { pid } /phone_screen_episodes_resampled.csv - data/interim/ { pid } /phone_screen_episodes_resampled_with_datetime.csv - data/interim/ { pid } /phone_screen_features/phone_screen_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_screen.csv Parameters description for [PHONE_SCREEN][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_SCREEN features from the RAPIDS provider [FEATURES] Features to be computed, see table below [REFERENCE_HOUR_FIRST_USE] The reference point from which firstuseafter is to be computed, default is midnight [IGNORE_EPISODES_SHORTER_THAN] Ignore episodes that are shorter than this threshold (minutes). Set to 0 to disable this filter. [IGNORE_EPISODES_LONGER_THAN] Ignore episodes that are longer than this threshold (minutes). Set to 0 to disable this filter. [EPISODE_TYPES] Currently we only support unlock episodes (from when the phone is unlocked until the screen is off) Features description for [PHONE_SCREEN][PROVIDERS][RAPIDS] : Feature Units Description sumduration minutes Total duration of all unlock episodes. maxduration minutes Longest duration of any unlock episode. minduration minutes Shortest duration of any unlock episode. avgduration minutes Average duration of all unlock episodes. stdduration minutes Standard deviation duration of all unlock episodes. countepisode episodes Number of all unlock episodes |firstuseafter |minutes |Minutes until the first unlock episode. Assumptions/Observations In Android, lock events can happen right after an off event, after a few seconds of an off event, or never happen depending on the phone's settings, therefore, an unlock episode is defined as the time between an unlock and a off event. In iOS, on and off events do not exist, so an unlock episode is defined as the time between an unlock and a lock event. Events in iOS are recorded reliably albeit some duplicated lock events within milliseconds from each other, so we only keep consecutive unlock/lock pairs. In Android you cand find multiple consecutive unlock or lock events, so we only keep consecutive unlock/off pairs. In our experiments these cases are less than 10% of the screen events collected and this happens because ACTION_SCREEN_OFF and ACTION_SCREEN_ON are sent when the device becomes non-interactive which may have nothing to do with the screen turning off . In addition to unlock/off episodes, in Android it is possible to measure the time spent on the lock screen before an unlock event as well as the total screen time (i.e. ON to OFF ) but these are not implemented at the moment. We transform iOS screen events to match Android\u2019s format, we replace lock episodes with off episodes (2 with 0) in iOS. However, as mentioned above this is still computing unlock to lock episodes.","title":"Phone Screen"},{"location":"features/phone-screen/#phone-screen","text":"Sensor parameters description for [PHONE_SCREEN] : Key Description [TABLE] Database table where the screen data is stored","title":"Phone Screen"},{"location":"features/phone-screen/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_screen_raw.csv - data/raw/ { pid } /phone_screen_with_datetime.csv - data/raw/ { pid } /phone_screen_with_datetime_unified.csv - data/interim/ { pid } /phone_screen_episodes.csv - data/interim/ { pid } /phone_screen_episodes_resampled.csv - data/interim/ { pid } /phone_screen_episodes_resampled_with_datetime.csv - data/interim/ { pid } /phone_screen_features/phone_screen_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_screen.csv Parameters description for [PHONE_SCREEN][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_SCREEN features from the RAPIDS provider [FEATURES] Features to be computed, see table below [REFERENCE_HOUR_FIRST_USE] The reference point from which firstuseafter is to be computed, default is midnight [IGNORE_EPISODES_SHORTER_THAN] Ignore episodes that are shorter than this threshold (minutes). Set to 0 to disable this filter. [IGNORE_EPISODES_LONGER_THAN] Ignore episodes that are longer than this threshold (minutes). Set to 0 to disable this filter. [EPISODE_TYPES] Currently we only support unlock episodes (from when the phone is unlocked until the screen is off) Features description for [PHONE_SCREEN][PROVIDERS][RAPIDS] : Feature Units Description sumduration minutes Total duration of all unlock episodes. maxduration minutes Longest duration of any unlock episode. minduration minutes Shortest duration of any unlock episode. avgduration minutes Average duration of all unlock episodes. stdduration minutes Standard deviation duration of all unlock episodes. countepisode episodes Number of all unlock episodes |firstuseafter |minutes |Minutes until the first unlock episode. Assumptions/Observations In Android, lock events can happen right after an off event, after a few seconds of an off event, or never happen depending on the phone's settings, therefore, an unlock episode is defined as the time between an unlock and a off event. In iOS, on and off events do not exist, so an unlock episode is defined as the time between an unlock and a lock event. Events in iOS are recorded reliably albeit some duplicated lock events within milliseconds from each other, so we only keep consecutive unlock/lock pairs. In Android you cand find multiple consecutive unlock or lock events, so we only keep consecutive unlock/off pairs. In our experiments these cases are less than 10% of the screen events collected and this happens because ACTION_SCREEN_OFF and ACTION_SCREEN_ON are sent when the device becomes non-interactive which may have nothing to do with the screen turning off . In addition to unlock/off episodes, in Android it is possible to measure the time spent on the lock screen before an unlock event as well as the total screen time (i.e. ON to OFF ) but these are not implemented at the moment. We transform iOS screen events to match Android\u2019s format, we replace lock episodes with off episodes (2 with 0) in iOS. However, as mentioned above this is still computing unlock to lock episodes.","title":"RAPIDS provider"},{"location":"features/phone-wifi-connected/","text":"Phone WiFi Connected \u00b6 Sensor parameters description for [PHONE_WIFI_CONNECTED] : Key Description [TABLE] Database table where the wifi (connected) data is stored RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_wifi_connected_raw.csv - data/raw/ { pid } /phone_wifi_connected_with_datetime.csv - data/interim/ { pid } /phone_wifi_connected_features/phone_wifi_connected_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_wifi_connected.csv Parameters description for [PHONE_WIFI_CONNECTED][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_WIFI_CONNECTED features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_WIFI_CONNECTED][PROVIDERS][RAPIDS] : Feature Units Description countscans devices Number of scanned WiFi access points connected during a time_segment, an access point can be detected multiple times over time and these appearances are counted separately uniquedevices devices Number of unique access point during a time_segment as identified by their hardware address countscansmostuniquedevice scans Number of scans of the most scanned access point during a time_segment across the whole monitoring period Assumptions/Observations A connected WiFI access point is one that a phone was connected to. By default AWARE stores this data in the sensor_wifi table.","title":"Phone WiFI Connected"},{"location":"features/phone-wifi-connected/#phone-wifi-connected","text":"Sensor parameters description for [PHONE_WIFI_CONNECTED] : Key Description [TABLE] Database table where the wifi (connected) data is stored","title":"Phone WiFi Connected"},{"location":"features/phone-wifi-connected/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_wifi_connected_raw.csv - data/raw/ { pid } /phone_wifi_connected_with_datetime.csv - data/interim/ { pid } /phone_wifi_connected_features/phone_wifi_connected_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_wifi_connected.csv Parameters description for [PHONE_WIFI_CONNECTED][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_WIFI_CONNECTED features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_WIFI_CONNECTED][PROVIDERS][RAPIDS] : Feature Units Description countscans devices Number of scanned WiFi access points connected during a time_segment, an access point can be detected multiple times over time and these appearances are counted separately uniquedevices devices Number of unique access point during a time_segment as identified by their hardware address countscansmostuniquedevice scans Number of scans of the most scanned access point during a time_segment across the whole monitoring period Assumptions/Observations A connected WiFI access point is one that a phone was connected to. By default AWARE stores this data in the sensor_wifi table.","title":"RAPIDS provider"},{"location":"features/phone-wifi-visible/","text":"Phone WiFi Visible \u00b6 Sensor parameters description for [PHONE_WIFI_VISIBLE] : Key Description [TABLE] Database table where the wifi (visible) data is stored RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_wifi_visible_raw.csv - data/raw/ { pid } /phone_wifi_visible_with_datetime.csv - data/interim/ { pid } /phone_wifi_visible_features/phone_wifi_visible_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_wifi_visible.csv Parameters description for [PHONE_WIFI_VISIBLE][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_WIFI_VISIBLE features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_WIFI_VISIBLE][PROVIDERS][RAPIDS] : Feature Units Description countscans devices Number of scanned WiFi access points visible during a time_segment, an access point can be detected multiple times over time and these appearances are counted separately uniquedevices devices Number of unique access point during a time_segment as identified by their hardware address countscansmostuniquedevice scans Number of scans of the most scanned access point during a time_segment across the whole monitoring period Assumptions/Observations A visible WiFI access point is one that a phone sensed around itself but that it was not connected to. Due to API restrictions, this sensor is not available on iOS. By default AWARE stores this data in the wifi table.","title":"Phone WiFI Visible"},{"location":"features/phone-wifi-visible/#phone-wifi-visible","text":"Sensor parameters description for [PHONE_WIFI_VISIBLE] : Key Description [TABLE] Database table where the wifi (visible) data is stored","title":"Phone WiFi Visible"},{"location":"features/phone-wifi-visible/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_wifi_visible_raw.csv - data/raw/ { pid } /phone_wifi_visible_with_datetime.csv - data/interim/ { pid } /phone_wifi_visible_features/phone_wifi_visible_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_wifi_visible.csv Parameters description for [PHONE_WIFI_VISIBLE][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_WIFI_VISIBLE features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_WIFI_VISIBLE][PROVIDERS][RAPIDS] : Feature Units Description countscans devices Number of scanned WiFi access points visible during a time_segment, an access point can be detected multiple times over time and these appearances are counted separately uniquedevices devices Number of unique access point during a time_segment as identified by their hardware address countscansmostuniquedevice scans Number of scans of the most scanned access point during a time_segment across the whole monitoring period Assumptions/Observations A visible WiFI access point is one that a phone sensed around itself but that it was not connected to. Due to API restrictions, this sensor is not available on iOS. By default AWARE stores this data in the wifi table.","title":"RAPIDS provider"},{"location":"setup/configuration/","text":"Configuration \u00b6 You need to follow these steps to configure your RAPIDS deployment before you can extract behavioral features Add your database credentials Choose the timezone of your study Create your participants files Select what time segments you want to extract features on Modify your device data source configuration Select what sensors and features you want to process When you are done with this configuration, go to executing RAPIDS . Hint Every time you see config[\"KEY\"] or [KEY] in these docs we are referring to the corresponding key in the config.yaml file. Database credentials \u00b6 Create an empty file called .env in your RAPIDS root directory Add the following lines and replace your database-specific credentials (user, password, host, and database): [ MY_GROUP ] user=MY_USER password=MY_PASSWORD host=MY_HOST port=3306 database=MY_DATABASE Warning The label MY_GROUP is arbitrary but it has to match the following config.yaml key: DATABASE_GROUP : &database_group MY_GROUP Note You can ignore this step if you are only processing Fitbit data in CSV files. Timezone of your study \u00b6 Single timezone \u00b6 If your study only happened in a single time zone, select the appropriate code form this list and change the following config key. Double check your timezone code pick, for example US Eastern Time is America/New_York not EST TIMEZONE : &timezone America/New_York Multiple timezones \u00b6 Support coming soon. Participant files \u00b6 Participant files link together multiple devices (smartphones and wearables) to specific participants and identify them throughout RAPIDS. You can create these files manually or automatically . Participant files are stored in data/external/participant_files/pxx.yaml and follow a unified structure . Note The list PIDS in config.yaml needs to have the participant file names of the people you want to process. For example, if you created p01.yaml , p02.yaml and p03.yaml files in /data/external/participant_files/ , then PIDS should be: PIDS : [ p01 , p02 , p03 ] Tip Attribute values of the [PHONE] and [FITBIT] sections in every participant file are optional which allows you to analyze data from participants that only carried smartphones, only Fitbit devices, or both. Optional: Migrating participants files with the old format If you were using the pre-release version of RAPIDS with participant files in plain text (as opposed to yaml), you can run the following command and your old files will be converted into yaml files stored in data/external/participant_files/ python tools/update_format_participant_files.py Structure of participants files \u00b6 Example of the structure of a participant file In this example, the participant used an android phone, an ios phone, and a fitbit device throughout the study between Apr 23 rd 2020 and Oct 28 th 2020 PHONE : DEVICE_IDS : [ a748ee1a-1d0b-4ae9-9074-279a2b6ba524 , dsadas-2324-fgsf-sdwr-gdfgs4rfsdf43 ] PLATFORMS : [ android , ios ] LABEL : test01 START_DATE : 2020-04-23 END_DATE : 2020-10-28 FITBIT : DEVICE_IDS : [ fitbit1 ] LABEL : test01 START_DATE : 2020-04-23 END_DATE : 2020-10-28 For [PHONE] Key Description [DEVICE_IDS] An array of the strings that uniquely identify each smartphone, you can have more than one for when participants changed phones in the middle of the study, in this case, data from all their devices will be joined and relabeled with the last 1 on this list. [PLATFORMS] An array that specifies the OS of each smartphone in [DEVICE_IDS] , use a combination of android or ios (we support participants that changed platforms in the middle of your study!). If you have an aware_device table in your database you can set [PLATFORMS]: [multiple] and RAPIDS will infer them automatically. [LABEL] A string that is used in reports and visualizations. [START_DATE] A string with format YYY-MM-DD . Only data collected after this date will be included in the analysis [END_DATE] A string with format YYY-MM-DD . Only data collected before this date will be included in the analysis For [FITBIT] Key Description [DEVICE_IDS] An array of the strings that uniquely identify each Fitbit, you can have more than one in case the participant changed devices in the middle of the study, in this case, data from all devices will be joined and relabeled with the last device_id on this list. [LABEL] A string that is used in reports and visualizations. [START_DATE] A string with format YYY-MM-DD . Only data collected after this date will be included in the analysis [END_DATE] A string with format YYY-MM-DD . Only data collected before this date will be included in the analysis Automatic creation of participant files \u00b6 You have two options a) use the aware_device table in your database or b) use a CSV file. In either case, in your config.yaml , set [PHONE_SECTION][ADD] or [FITBIT_SECTION][ADD] to TRUE depending on what devices you used in your study. Set [DEVICE_ID_COLUMN] to the name of the column that uniquely identifies each device and include any device ids you want to ignore in [IGNORED_DEVICE_IDS] . aware_device table Set the following keys in your config.yaml CREATE_PARTICIPANT_FILES : SOURCE : TYPE : AWARE_DEVICE_TABLE DATABASE_GROUP : *database_group CSV_FILE_PATH : \"\" TIMEZONE : *timezone PHONE_SECTION : ADD : TRUE # or FALSE DEVICE_ID_COLUMN : device_id # column name IGNORED_DEVICE_IDS : [] FITBIT_SECTION : ADD : TRUE # or FALSE DEVICE_ID_COLUMN : fitbit_id # column name IGNORED_DEVICE_IDS : [] Then run snakemake -j1 create_participants_files CSV file Set the following keys in your config.yaml . CREATE_PARTICIPANT_FILES : SOURCE : TYPE : CSV_FILE DATABASE_GROUP : \"\" CSV_FILE_PATH : \"your_path/to_your.csv\" TIMEZONE : *timezone PHONE_SECTION : ADD : TRUE # or FALSE DEVICE_ID_COLUMN : device_id # column name IGNORED_DEVICE_IDS : [] FITBIT_SECTION : ADD : TRUE # or FALSE DEVICE_ID_COLUMN : fitbit_id # column name IGNORED_DEVICE_IDS : [] Your CSV file ( [SOURCE][CSV_FILE_PATH] ) should have the following columns but you can omit any values you don\u2019t have on each column: Column Description phone device id The name of this column has to match [PHONE_SECTION][DEVICE_ID_COLUMN] . Separate multiple ids with ; fitbit device id The name of this column has to match [FITBIT_SECTION][DEVICE_ID_COLUMN] . Separate multiple ids with ; pid Unique identifiers with the format pXXX (your participant files will be named with this string platform Use android , ios or multiple as explained above, separate values with ; label A human readable string that is used in reports and visualizations. start_date A string with format YYY-MM-DD . end_date A string with format YYY-MM-DD . Example device_id,pid,label,platform,start_date,end_date,fitbit_id a748ee1a-1d0b-4ae9-9074-279a2b6ba524;dsadas-2324-fgsf-sdwr-gdfgs4rfsdf43,p01,julio,android;ios,2020-01-01,2021-01-01,fitbit1 4c4cf7a1-0340-44bc-be0f-d5053bf7390c,p02,meng,ios,2021-01-01,2022-01-01,fitbit2 Then run snakemake -j1 create_participants_files Time Segments \u00b6 Time segments (or epochs) are the time windows on which you want to extract behavioral features. For example, you might want to process data on every day, every morning, or only during weekends. RAPIDS offers three categories of time segments that are flexible enough to cover most use cases: frequency (short time windows every day), periodic (arbitrary time windows on any day), and event (arbitrary time windows around events of interest). See also our examples . Frequency Segments These segments are computed on every day and all have the same duration (for example 30 minutes). Set the following keys in your config.yaml TIME_SEGMENTS : &time_segments TYPE : FREQUENCY FILE : \"data/external/your_frequency_segments.csv\" INCLUDE_PAST_PERIODIC_SEGMENTS : FALSE The file pointed by [TIME_SEGMENTS][FILE] should have the following format and can only have 1 row. Column Description label A string that is used as a prefix in the name of your time segments length An integer representing the duration of your time segments in minutes Example label,length thirtyminutes,30 This configuration will compute 48 time segments for every day when any data from any participant was sensed. For example: start_time,length,label 00:00,30,thirtyminutes0000 00:30,30,thirtyminutes0001 01:00,30,thirtyminutes0002 01:30,30,thirtyminutes0003 ... Periodic Segments These segments can be computed every day, or on specific days of the week, month, quarter, and year. Their minimum duration is 1 minute but they can be as long as you want. Set the following keys in your config.yaml . TIME_SEGMENTS : &time_segments TYPE : PERIODIC FILE : \"data/external/your_periodic_segments.csv\" INCLUDE_PAST_PERIODIC_SEGMENTS : FALSE # or TRUE If [INCLUDE_PAST_PERIODIC_SEGMENTS] is set to TRUE , RAPIDS will consider instances of your segments back enough in the past as to include the first row of data of each participant. For example, if the first row of data from a participant happened on Saturday March 7 th 2020 and the requested segment duration is 7 days starting on every Sunday, the first segment to be considered would start on Sunday March 1 st if [INCLUDE_PAST_PERIODIC_SEGMENTS] is TRUE or on Sunday March 8 th if FALSE . The file pointed by [TIME_SEGMENTS][FILE] should have the following format and can have multiple rows. Column Description label A string that is used as a prefix in the name of your time segments. It has to be unique between rows start_time A string with format HH:MM:SS representing the starting time of this segment on any day length A string representing the length of this segment.It can have one or more of the following strings XXD XXH XXM XXS to represent days, hours, minutes and seconds. For example 7D 23H 59M 59S repeats_on One of the follow options every_day , wday , qday , mday , and yday . The last four represent a week, quarter, month and year day repeats_value An integer complementing repeats_on . If you set repeats_on to every_day set this to 0 , otherwise 1-7 represent a wday starting from Mondays, 1-31 represent a mday , 1-91 represent a qday , and 1-366 represent a yday Example label,start_time,length,repeats_on,repeats_value daily,00:00:00,23H 59M 59S,every_day,0 morning,06:00:00,5H 59M 59S,every_day,0 afternoon,12:00:00,5H 59M 59S,every_day,0 evening,18:00:00,5H 59M 59S,every_day,0 night,00:00:00,5H 59M 59S,every_day,0 This configuration will create five segments instances ( daily , morning , afternoon , evening , night ) on any given day ( every_day set to 0). The daily segment will start at midnight and will last 23:59:59 , the other four segments will start at 6am, 12pm, 6pm, and 12am respectively and last for 05:59:59 . Event segments These segments can be computed before or after an event of interest (defined as any UNIX timestamp). Their minimum duration is 1 minute but they can be as long as you want. The start of each segment can be shifted backwards or forwards from the specified timestamp. Set the following keys in your config.yaml . TIME_SEGMENTS : &time_segments TYPE : EVENT FILE : \"data/external/your_event_segments.csv\" INCLUDE_PAST_PERIODIC_SEGMENTS : FALSE # or TRUE The file pointed by [TIME_SEGMENTS][FILE] should have the following format and can have multiple rows. Column Description label A string that is used as a prefix in the name of your time segments. If labels are unique, every segment is independent; if two or more segments have the same label, their data will be grouped when computing auxiliary data for features like the most frequent contact for calls (the most frequent contact will be computed across all these segments). There cannot be two overlaping event segments with the same label (RAPIDS will throw an error) event_timestamp A UNIX timestamp that represents the moment an event of interest happened (clinical relapse, survey, readmission, etc.). The corresponding time segment will be computed around this moment using length , shift , and shift_direction length A string representing the length of this segment. It can have one or more of the following keys XXD XXH XXM XXS to represent a number of days, hours, minutes, and seconds. For example 7D 23H 59M 59S shift A string representing the time shift from event_timestamp . It can have one or more of the following keys XXD XXH XXM XXS to represent a number of days, hours, minutes and seconds. For example 7D 23H 59M 59S . Use this value to change the start of a segment with respect to its event_timestamp . For example, set this variable to 1H to create a segment that starts 1 hour from an event of interest ( shift_direction determines if it\u2019s before or after). shift_direction An integer representing whether the shift is before ( -1 ) or after ( 1 ) an event_timestamp device_id The device id (smartphone or fitbit) to whom this segment belongs to. You have to create a line in this event segment file for each event of a participant that you want to analyse. If you have participants with multiple device ids you can choose any of them Example label,event_timestamp,length,shift,shift_direction,device_id stress1,1587661220000,1H,5M,1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 stress2,1587747620000,4H,4H,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 stress3,1587906020000,3H,5M,1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 stress4,1584291600000,7H,4H,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 stress5,1588172420000,9H,5M,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 mood,1587661220000,1H,0,0,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 mood,1587747620000,1D,0,0,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 mood,1587906020000,7D,0,0,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 This example will create eight segments for a single participant ( a748ee1a... ), five independent stressX segments with various lengths (1,4,3,7, and 9 hours). Segments stress1 , stress3 , and stress5 are shifted forwards by 5 minutes and stress2 and stress4 are shifted backwards by 4 hours (that is, if the stress4 event happened on March 15 th at 1pm EST ( 1584291600000 ), the time segment will start on that day at 9am and end at 4pm). The three mood segments are 1 hour, 1 day and 7 days long and have no shift. In addition, these mood segments are grouped together, meaning that although RAPIDS will compute features on each one of them, some necessary information to compute a few of such features will be extracted from all three segments, for example the phone contact that called a participant the most or the location clusters visited by a participant. Segment Examples \u00b6 5-minutes Use the following Frequency segment file to create 288 (12 * 60 * 24) 5-minute segments starting from midnight of every day in your study label,length fiveminutes,5 Daily Use the following Periodic segment file to create daily segments starting from midnight of every day in your study label,start_time,length,repeats_on,repeats_value daily,00:00:00,23H 59M 59S,every_day,0 Morning Use the following Periodic segment file to create morning segments starting at 06:00:00 and ending at 11:59:59 of every day in your study label,start_time,length,repeats_on,repeats_value morning,06:00:00,5H 59M 59S,every_day,0 Overnight Use the following Periodic segment file to create overnight segments starting at 20:00:00 and ending at 07:59:59 (next day) of every day in your study label,start_time,length,repeats_on,repeats_value morning,20:00:00,11H 59M 59S,every_day,0 Weekly Use the following Periodic segment file to create non-overlapping weekly segments starting at midnight of every Monday in your study label,start_time,length,repeats_on,repeats_value weekly,00:00:00,6D 23H 59M 59S,wday,1 Use the following Periodic segment file to create overlapping weekly segments starting at midnight of every day in your study label,start_time,length,repeats_on,repeats_value weekly,00:00:00,6D 23H 59M 59S,every_day,0 Week-ends Use the following Periodic segment file to create week-end segments starting at midnight of every Saturday in your study label,start_time,length,repeats_on,repeats_value weekend,00:00:00,1D 23H 59M 59S,wday,6 Around surveys Use the following Event segment file to create two 2-hour segments that start 1 hour before surveys answered by 3 participants label,event_timestamp,length,shift,shift_direction,device_id survey1,1587661220000,2H,1H,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 survey2,1587747620000,2H,1H,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 survey1,1587906020000,2H,1H,-1,rqtertsd-43ff-34fr-3eeg-efe4fergregr survey2,1584291600000,2H,1H,-1,rqtertsd-43ff-34fr-3eeg-efe4fergregr survey1,1588172420000,2H,1H,-1,klj34oi2-8frk-2343-21kk-324ljklewlr3 survey2,1584291600000,2H,1H,-1,klj34oi2-8frk-2343-21kk-324ljklewlr3 Device Data Source Configuration \u00b6 You might need to modify the following config keys in your config.yaml depending on what devices your participants used and where you are storing your data. You can ignore [PHONE_DATA_CONFIGURATION] or [FITBIT_DATA_CONFIGURATION] if you are not working with either devices. Phone The relevant config.yaml section looks like this by default: PHONE_DATA_CONFIGURATION : SOURCE : TYPE : DATABASE DATABASE_GROUP : *database_group DEVICE_ID_COLUMN : device_id # column name TIMEZONE : TYPE : SINGLE # SINGLE (MULTIPLE support coming soon) VALUE : *timezone Parameters for [PHONE_DATA_CONFIGURATION] Key Description [SOURCE] [TYPE] Only DATABASE is supported (phone data will be pulled from a database) [SOURCE] [DATABASE_GROUP] *database_group points to the value defined before in Database credentials [SOURCE] [DEVICE_ID_COLUMN] A column that contains strings that uniquely identify smartphones. For data collected with AWARE this is usually device_id [TIMEZONE] [TYPE] Only SINGLE is supported for now [TIMEZONE] [VALUE] *timezone points to the value defined before in Timezone of your study Fitbit The relevant config.yaml section looks like this by default: FITBIT_DATA_CONFIGURATION : SOURCE : TYPE : DATABASE # DATABASE or FILES (set each [FITBIT_SENSOR][TABLE] attribute with a table name or a file path accordingly) COLUMN_FORMAT : JSON # JSON or PLAIN_TEXT DATABASE_GROUP : *database_group DEVICE_ID_COLUMN : device_id # column name TIMEZONE : TYPE : SINGLE # Fitbit devices don't support time zones so we read this data in the timezone indicated by VALUE VALUE : *timezone Parameters for For [FITBIT_DATA_CONFIGURATION] Key Description [SOURCE] [TYPE] DATABASE or FILES (set each [FITBIT_SENSOR] [TABLE] attribute accordingly with a table name or a file path) [SOURCE] [COLUMN_FORMAT] JSON or PLAIN_TEXT . Column format of the source data. If you pulled your data directly from the Fitbit API the column containing the sensor data will be in JSON format [SOURCE] [DATABASE_GROUP] *database_group points to the value defined before in Database credentials . Only used if [TYPE] is DATABASE . [SOURCE] [DEVICE_ID_COLUMN] A column that contains strings that uniquely identify Fitbit devices. [TIMEZONE] [TYPE] Only SINGLE is supported (Fitbit devices always store data in local time). [TIMEZONE] [VALUE] *timezone points to the value defined before in Timezone of your study Sensor and Features to Process \u00b6 Finally, you need to modify the config.yaml section of the sensors you want to extract behavioral features from. All sensors follow the same naming nomenclature ( DEVICE_SENSOR ) and parameter structure which we explain in the Behavioral Features Introduction . Done Head over to Execution to learn how to execute RAPIDS.","title":"Configuration"},{"location":"setup/configuration/#configuration","text":"You need to follow these steps to configure your RAPIDS deployment before you can extract behavioral features Add your database credentials Choose the timezone of your study Create your participants files Select what time segments you want to extract features on Modify your device data source configuration Select what sensors and features you want to process When you are done with this configuration, go to executing RAPIDS . Hint Every time you see config[\"KEY\"] or [KEY] in these docs we are referring to the corresponding key in the config.yaml file.","title":"Configuration"},{"location":"setup/configuration/#database-credentials","text":"Create an empty file called .env in your RAPIDS root directory Add the following lines and replace your database-specific credentials (user, password, host, and database): [ MY_GROUP ] user=MY_USER password=MY_PASSWORD host=MY_HOST port=3306 database=MY_DATABASE Warning The label MY_GROUP is arbitrary but it has to match the following config.yaml key: DATABASE_GROUP : &database_group MY_GROUP Note You can ignore this step if you are only processing Fitbit data in CSV files.","title":"Database credentials"},{"location":"setup/configuration/#timezone-of-your-study","text":"","title":"Timezone of your study"},{"location":"setup/configuration/#single-timezone","text":"If your study only happened in a single time zone, select the appropriate code form this list and change the following config key. Double check your timezone code pick, for example US Eastern Time is America/New_York not EST TIMEZONE : &timezone America/New_York","title":"Single timezone"},{"location":"setup/configuration/#multiple-timezones","text":"Support coming soon.","title":"Multiple timezones"},{"location":"setup/configuration/#participant-files","text":"Participant files link together multiple devices (smartphones and wearables) to specific participants and identify them throughout RAPIDS. You can create these files manually or automatically . Participant files are stored in data/external/participant_files/pxx.yaml and follow a unified structure . Note The list PIDS in config.yaml needs to have the participant file names of the people you want to process. For example, if you created p01.yaml , p02.yaml and p03.yaml files in /data/external/participant_files/ , then PIDS should be: PIDS : [ p01 , p02 , p03 ] Tip Attribute values of the [PHONE] and [FITBIT] sections in every participant file are optional which allows you to analyze data from participants that only carried smartphones, only Fitbit devices, or both. Optional: Migrating participants files with the old format If you were using the pre-release version of RAPIDS with participant files in plain text (as opposed to yaml), you can run the following command and your old files will be converted into yaml files stored in data/external/participant_files/ python tools/update_format_participant_files.py","title":"Participant files"},{"location":"setup/configuration/#structure-of-participants-files","text":"Example of the structure of a participant file In this example, the participant used an android phone, an ios phone, and a fitbit device throughout the study between Apr 23 rd 2020 and Oct 28 th 2020 PHONE : DEVICE_IDS : [ a748ee1a-1d0b-4ae9-9074-279a2b6ba524 , dsadas-2324-fgsf-sdwr-gdfgs4rfsdf43 ] PLATFORMS : [ android , ios ] LABEL : test01 START_DATE : 2020-04-23 END_DATE : 2020-10-28 FITBIT : DEVICE_IDS : [ fitbit1 ] LABEL : test01 START_DATE : 2020-04-23 END_DATE : 2020-10-28 For [PHONE] Key Description [DEVICE_IDS] An array of the strings that uniquely identify each smartphone, you can have more than one for when participants changed phones in the middle of the study, in this case, data from all their devices will be joined and relabeled with the last 1 on this list. [PLATFORMS] An array that specifies the OS of each smartphone in [DEVICE_IDS] , use a combination of android or ios (we support participants that changed platforms in the middle of your study!). If you have an aware_device table in your database you can set [PLATFORMS]: [multiple] and RAPIDS will infer them automatically. [LABEL] A string that is used in reports and visualizations. [START_DATE] A string with format YYY-MM-DD . Only data collected after this date will be included in the analysis [END_DATE] A string with format YYY-MM-DD . Only data collected before this date will be included in the analysis For [FITBIT] Key Description [DEVICE_IDS] An array of the strings that uniquely identify each Fitbit, you can have more than one in case the participant changed devices in the middle of the study, in this case, data from all devices will be joined and relabeled with the last device_id on this list. [LABEL] A string that is used in reports and visualizations. [START_DATE] A string with format YYY-MM-DD . Only data collected after this date will be included in the analysis [END_DATE] A string with format YYY-MM-DD . Only data collected before this date will be included in the analysis","title":"Structure of participants files"},{"location":"setup/configuration/#automatic-creation-of-participant-files","text":"You have two options a) use the aware_device table in your database or b) use a CSV file. In either case, in your config.yaml , set [PHONE_SECTION][ADD] or [FITBIT_SECTION][ADD] to TRUE depending on what devices you used in your study. Set [DEVICE_ID_COLUMN] to the name of the column that uniquely identifies each device and include any device ids you want to ignore in [IGNORED_DEVICE_IDS] . aware_device table Set the following keys in your config.yaml CREATE_PARTICIPANT_FILES : SOURCE : TYPE : AWARE_DEVICE_TABLE DATABASE_GROUP : *database_group CSV_FILE_PATH : \"\" TIMEZONE : *timezone PHONE_SECTION : ADD : TRUE # or FALSE DEVICE_ID_COLUMN : device_id # column name IGNORED_DEVICE_IDS : [] FITBIT_SECTION : ADD : TRUE # or FALSE DEVICE_ID_COLUMN : fitbit_id # column name IGNORED_DEVICE_IDS : [] Then run snakemake -j1 create_participants_files CSV file Set the following keys in your config.yaml . CREATE_PARTICIPANT_FILES : SOURCE : TYPE : CSV_FILE DATABASE_GROUP : \"\" CSV_FILE_PATH : \"your_path/to_your.csv\" TIMEZONE : *timezone PHONE_SECTION : ADD : TRUE # or FALSE DEVICE_ID_COLUMN : device_id # column name IGNORED_DEVICE_IDS : [] FITBIT_SECTION : ADD : TRUE # or FALSE DEVICE_ID_COLUMN : fitbit_id # column name IGNORED_DEVICE_IDS : [] Your CSV file ( [SOURCE][CSV_FILE_PATH] ) should have the following columns but you can omit any values you don\u2019t have on each column: Column Description phone device id The name of this column has to match [PHONE_SECTION][DEVICE_ID_COLUMN] . Separate multiple ids with ; fitbit device id The name of this column has to match [FITBIT_SECTION][DEVICE_ID_COLUMN] . Separate multiple ids with ; pid Unique identifiers with the format pXXX (your participant files will be named with this string platform Use android , ios or multiple as explained above, separate values with ; label A human readable string that is used in reports and visualizations. start_date A string with format YYY-MM-DD . end_date A string with format YYY-MM-DD . Example device_id,pid,label,platform,start_date,end_date,fitbit_id a748ee1a-1d0b-4ae9-9074-279a2b6ba524;dsadas-2324-fgsf-sdwr-gdfgs4rfsdf43,p01,julio,android;ios,2020-01-01,2021-01-01,fitbit1 4c4cf7a1-0340-44bc-be0f-d5053bf7390c,p02,meng,ios,2021-01-01,2022-01-01,fitbit2 Then run snakemake -j1 create_participants_files","title":"Automatic creation of participant files"},{"location":"setup/configuration/#time-segments","text":"Time segments (or epochs) are the time windows on which you want to extract behavioral features. For example, you might want to process data on every day, every morning, or only during weekends. RAPIDS offers three categories of time segments that are flexible enough to cover most use cases: frequency (short time windows every day), periodic (arbitrary time windows on any day), and event (arbitrary time windows around events of interest). See also our examples . Frequency Segments These segments are computed on every day and all have the same duration (for example 30 minutes). Set the following keys in your config.yaml TIME_SEGMENTS : &time_segments TYPE : FREQUENCY FILE : \"data/external/your_frequency_segments.csv\" INCLUDE_PAST_PERIODIC_SEGMENTS : FALSE The file pointed by [TIME_SEGMENTS][FILE] should have the following format and can only have 1 row. Column Description label A string that is used as a prefix in the name of your time segments length An integer representing the duration of your time segments in minutes Example label,length thirtyminutes,30 This configuration will compute 48 time segments for every day when any data from any participant was sensed. For example: start_time,length,label 00:00,30,thirtyminutes0000 00:30,30,thirtyminutes0001 01:00,30,thirtyminutes0002 01:30,30,thirtyminutes0003 ... Periodic Segments These segments can be computed every day, or on specific days of the week, month, quarter, and year. Their minimum duration is 1 minute but they can be as long as you want. Set the following keys in your config.yaml . TIME_SEGMENTS : &time_segments TYPE : PERIODIC FILE : \"data/external/your_periodic_segments.csv\" INCLUDE_PAST_PERIODIC_SEGMENTS : FALSE # or TRUE If [INCLUDE_PAST_PERIODIC_SEGMENTS] is set to TRUE , RAPIDS will consider instances of your segments back enough in the past as to include the first row of data of each participant. For example, if the first row of data from a participant happened on Saturday March 7 th 2020 and the requested segment duration is 7 days starting on every Sunday, the first segment to be considered would start on Sunday March 1 st if [INCLUDE_PAST_PERIODIC_SEGMENTS] is TRUE or on Sunday March 8 th if FALSE . The file pointed by [TIME_SEGMENTS][FILE] should have the following format and can have multiple rows. Column Description label A string that is used as a prefix in the name of your time segments. It has to be unique between rows start_time A string with format HH:MM:SS representing the starting time of this segment on any day length A string representing the length of this segment.It can have one or more of the following strings XXD XXH XXM XXS to represent days, hours, minutes and seconds. For example 7D 23H 59M 59S repeats_on One of the follow options every_day , wday , qday , mday , and yday . The last four represent a week, quarter, month and year day repeats_value An integer complementing repeats_on . If you set repeats_on to every_day set this to 0 , otherwise 1-7 represent a wday starting from Mondays, 1-31 represent a mday , 1-91 represent a qday , and 1-366 represent a yday Example label,start_time,length,repeats_on,repeats_value daily,00:00:00,23H 59M 59S,every_day,0 morning,06:00:00,5H 59M 59S,every_day,0 afternoon,12:00:00,5H 59M 59S,every_day,0 evening,18:00:00,5H 59M 59S,every_day,0 night,00:00:00,5H 59M 59S,every_day,0 This configuration will create five segments instances ( daily , morning , afternoon , evening , night ) on any given day ( every_day set to 0). The daily segment will start at midnight and will last 23:59:59 , the other four segments will start at 6am, 12pm, 6pm, and 12am respectively and last for 05:59:59 . Event segments These segments can be computed before or after an event of interest (defined as any UNIX timestamp). Their minimum duration is 1 minute but they can be as long as you want. The start of each segment can be shifted backwards or forwards from the specified timestamp. Set the following keys in your config.yaml . TIME_SEGMENTS : &time_segments TYPE : EVENT FILE : \"data/external/your_event_segments.csv\" INCLUDE_PAST_PERIODIC_SEGMENTS : FALSE # or TRUE The file pointed by [TIME_SEGMENTS][FILE] should have the following format and can have multiple rows. Column Description label A string that is used as a prefix in the name of your time segments. If labels are unique, every segment is independent; if two or more segments have the same label, their data will be grouped when computing auxiliary data for features like the most frequent contact for calls (the most frequent contact will be computed across all these segments). There cannot be two overlaping event segments with the same label (RAPIDS will throw an error) event_timestamp A UNIX timestamp that represents the moment an event of interest happened (clinical relapse, survey, readmission, etc.). The corresponding time segment will be computed around this moment using length , shift , and shift_direction length A string representing the length of this segment. It can have one or more of the following keys XXD XXH XXM XXS to represent a number of days, hours, minutes, and seconds. For example 7D 23H 59M 59S shift A string representing the time shift from event_timestamp . It can have one or more of the following keys XXD XXH XXM XXS to represent a number of days, hours, minutes and seconds. For example 7D 23H 59M 59S . Use this value to change the start of a segment with respect to its event_timestamp . For example, set this variable to 1H to create a segment that starts 1 hour from an event of interest ( shift_direction determines if it\u2019s before or after). shift_direction An integer representing whether the shift is before ( -1 ) or after ( 1 ) an event_timestamp device_id The device id (smartphone or fitbit) to whom this segment belongs to. You have to create a line in this event segment file for each event of a participant that you want to analyse. If you have participants with multiple device ids you can choose any of them Example label,event_timestamp,length,shift,shift_direction,device_id stress1,1587661220000,1H,5M,1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 stress2,1587747620000,4H,4H,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 stress3,1587906020000,3H,5M,1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 stress4,1584291600000,7H,4H,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 stress5,1588172420000,9H,5M,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 mood,1587661220000,1H,0,0,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 mood,1587747620000,1D,0,0,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 mood,1587906020000,7D,0,0,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 This example will create eight segments for a single participant ( a748ee1a... ), five independent stressX segments with various lengths (1,4,3,7, and 9 hours). Segments stress1 , stress3 , and stress5 are shifted forwards by 5 minutes and stress2 and stress4 are shifted backwards by 4 hours (that is, if the stress4 event happened on March 15 th at 1pm EST ( 1584291600000 ), the time segment will start on that day at 9am and end at 4pm). The three mood segments are 1 hour, 1 day and 7 days long and have no shift. In addition, these mood segments are grouped together, meaning that although RAPIDS will compute features on each one of them, some necessary information to compute a few of such features will be extracted from all three segments, for example the phone contact that called a participant the most or the location clusters visited by a participant.","title":"Time Segments"},{"location":"setup/configuration/#segment-examples","text":"5-minutes Use the following Frequency segment file to create 288 (12 * 60 * 24) 5-minute segments starting from midnight of every day in your study label,length fiveminutes,5 Daily Use the following Periodic segment file to create daily segments starting from midnight of every day in your study label,start_time,length,repeats_on,repeats_value daily,00:00:00,23H 59M 59S,every_day,0 Morning Use the following Periodic segment file to create morning segments starting at 06:00:00 and ending at 11:59:59 of every day in your study label,start_time,length,repeats_on,repeats_value morning,06:00:00,5H 59M 59S,every_day,0 Overnight Use the following Periodic segment file to create overnight segments starting at 20:00:00 and ending at 07:59:59 (next day) of every day in your study label,start_time,length,repeats_on,repeats_value morning,20:00:00,11H 59M 59S,every_day,0 Weekly Use the following Periodic segment file to create non-overlapping weekly segments starting at midnight of every Monday in your study label,start_time,length,repeats_on,repeats_value weekly,00:00:00,6D 23H 59M 59S,wday,1 Use the following Periodic segment file to create overlapping weekly segments starting at midnight of every day in your study label,start_time,length,repeats_on,repeats_value weekly,00:00:00,6D 23H 59M 59S,every_day,0 Week-ends Use the following Periodic segment file to create week-end segments starting at midnight of every Saturday in your study label,start_time,length,repeats_on,repeats_value weekend,00:00:00,1D 23H 59M 59S,wday,6 Around surveys Use the following Event segment file to create two 2-hour segments that start 1 hour before surveys answered by 3 participants label,event_timestamp,length,shift,shift_direction,device_id survey1,1587661220000,2H,1H,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 survey2,1587747620000,2H,1H,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 survey1,1587906020000,2H,1H,-1,rqtertsd-43ff-34fr-3eeg-efe4fergregr survey2,1584291600000,2H,1H,-1,rqtertsd-43ff-34fr-3eeg-efe4fergregr survey1,1588172420000,2H,1H,-1,klj34oi2-8frk-2343-21kk-324ljklewlr3 survey2,1584291600000,2H,1H,-1,klj34oi2-8frk-2343-21kk-324ljklewlr3","title":"Segment Examples"},{"location":"setup/configuration/#device-data-source-configuration","text":"You might need to modify the following config keys in your config.yaml depending on what devices your participants used and where you are storing your data. You can ignore [PHONE_DATA_CONFIGURATION] or [FITBIT_DATA_CONFIGURATION] if you are not working with either devices. Phone The relevant config.yaml section looks like this by default: PHONE_DATA_CONFIGURATION : SOURCE : TYPE : DATABASE DATABASE_GROUP : *database_group DEVICE_ID_COLUMN : device_id # column name TIMEZONE : TYPE : SINGLE # SINGLE (MULTIPLE support coming soon) VALUE : *timezone Parameters for [PHONE_DATA_CONFIGURATION] Key Description [SOURCE] [TYPE] Only DATABASE is supported (phone data will be pulled from a database) [SOURCE] [DATABASE_GROUP] *database_group points to the value defined before in Database credentials [SOURCE] [DEVICE_ID_COLUMN] A column that contains strings that uniquely identify smartphones. For data collected with AWARE this is usually device_id [TIMEZONE] [TYPE] Only SINGLE is supported for now [TIMEZONE] [VALUE] *timezone points to the value defined before in Timezone of your study Fitbit The relevant config.yaml section looks like this by default: FITBIT_DATA_CONFIGURATION : SOURCE : TYPE : DATABASE # DATABASE or FILES (set each [FITBIT_SENSOR][TABLE] attribute with a table name or a file path accordingly) COLUMN_FORMAT : JSON # JSON or PLAIN_TEXT DATABASE_GROUP : *database_group DEVICE_ID_COLUMN : device_id # column name TIMEZONE : TYPE : SINGLE # Fitbit devices don't support time zones so we read this data in the timezone indicated by VALUE VALUE : *timezone Parameters for For [FITBIT_DATA_CONFIGURATION] Key Description [SOURCE] [TYPE] DATABASE or FILES (set each [FITBIT_SENSOR] [TABLE] attribute accordingly with a table name or a file path) [SOURCE] [COLUMN_FORMAT] JSON or PLAIN_TEXT . Column format of the source data. If you pulled your data directly from the Fitbit API the column containing the sensor data will be in JSON format [SOURCE] [DATABASE_GROUP] *database_group points to the value defined before in Database credentials . Only used if [TYPE] is DATABASE . [SOURCE] [DEVICE_ID_COLUMN] A column that contains strings that uniquely identify Fitbit devices. [TIMEZONE] [TYPE] Only SINGLE is supported (Fitbit devices always store data in local time). [TIMEZONE] [VALUE] *timezone points to the value defined before in Timezone of your study","title":"Device Data Source Configuration"},{"location":"setup/configuration/#sensor-and-features-to-process","text":"Finally, you need to modify the config.yaml section of the sensors you want to extract behavioral features from. All sensors follow the same naming nomenclature ( DEVICE_SENSOR ) and parameter structure which we explain in the Behavioral Features Introduction . Done Head over to Execution to learn how to execute RAPIDS.","title":"Sensor and Features to Process"},{"location":"setup/execution/","text":"Execution \u00b6 After you have installed and configured RAPIDS, use the following command to execute it. ./rapids -j1 Ready to extract behavioral features If you are ready to extract features head over to the Behavioral Features Introduction Info The script ./rapids is a wrapper around Snakemake so you can pass any parameters that Snakemake accepts (e.g. -j1 ). Updating RAPIDS output after modifying config.yaml Any changes to the config.yaml file will be applied automatically and only the relevant files will be updated. This means that after modifying the features list for PHONE_MESSAGE for example, RAPIDS will update the output file with the correct features. Multi-core You can run RAPIDS over multiple cores by modifying the -j argument (e.g. use -j8 to use 8 cores). However , take into account that this means multiple sensor datasets for different participants will be load in memory at the same time. If RAPIDS crashes because it ran out of memory reduce the number of cores and try again. As reference, we have run RAPIDS over 12 cores and 32 Gb of RAM without problems for a study with 200 participants with 14 days of low-frequency smartphone data (no accelerometer, gyroscope, or magnetometer). Forcing a complete rerun If you want to update your data from your database or rerun the whole pipeline from scratch run one or both of the following commands depending on the devices you are using: ./rapids -j1 -R download_phone_data ./rapids -j1 -R download_fitbit_data Deleting RAPIDS output If you want to delete all the output files RAPIDS produces you can execute the following command: ./rapids -j1 --delete-all-output","title":"Execution"},{"location":"setup/execution/#execution","text":"After you have installed and configured RAPIDS, use the following command to execute it. ./rapids -j1 Ready to extract behavioral features If you are ready to extract features head over to the Behavioral Features Introduction Info The script ./rapids is a wrapper around Snakemake so you can pass any parameters that Snakemake accepts (e.g. -j1 ). Updating RAPIDS output after modifying config.yaml Any changes to the config.yaml file will be applied automatically and only the relevant files will be updated. This means that after modifying the features list for PHONE_MESSAGE for example, RAPIDS will update the output file with the correct features. Multi-core You can run RAPIDS over multiple cores by modifying the -j argument (e.g. use -j8 to use 8 cores). However , take into account that this means multiple sensor datasets for different participants will be load in memory at the same time. If RAPIDS crashes because it ran out of memory reduce the number of cores and try again. As reference, we have run RAPIDS over 12 cores and 32 Gb of RAM without problems for a study with 200 participants with 14 days of low-frequency smartphone data (no accelerometer, gyroscope, or magnetometer). Forcing a complete rerun If you want to update your data from your database or rerun the whole pipeline from scratch run one or both of the following commands depending on the devices you are using: ./rapids -j1 -R download_phone_data ./rapids -j1 -R download_fitbit_data Deleting RAPIDS output If you want to delete all the output files RAPIDS produces you can execute the following command: ./rapids -j1 --delete-all-output","title":"Execution"},{"location":"setup/installation/","text":"Installation \u00b6 You can install RAPIDS using Docker (the fastest), or native instructions for MacOS and Ubuntu Docker Install Docker Pull our RAPIDS container docker pull agamk/rapids:latest ` Run RAPIDS' container (after this step is done you should see a prompt in the main RAPIDS folder with its python environment active) docker run -it agamk/rapids:latest Pull the latest version of RAPIDS git pull Make RAPIDS script executable chmod +x rapids Check that RAPIDS is working ./rapids -j1 Optional . You can edit RAPIDS files with vim but we recommend using Visual Studio Code and its Remote Containers extension How to configure Remote Containers extension Make sure RAPIDS container is running Install the Remote - Containers extension Go to the Remote Explorer panel on the left hand sidebar On the top right dropdown menu choose Containers Double click on the agamk/rapids container in the CONTAINERS tree A new VS Code session should open on RAPIDS main folder insidethe container. MacOS We tested these instructions in Catalina Install brew Install MySQL brew install mysql brew services start mysql Install R 4.0, pandoc and rmarkdown. If you have other instances of R, we recommend uninstalling them brew install r brew install pandoc Rscript --vanilla -e 'install.packages(\"rmarkdown\", repos=\"http://cran.us.r-project.org\")' Install miniconda (restart your terminal afterwards) brew cask install miniconda conda init zsh # (or conda init bash) Clone our repo git clone https://github.com/carissalow/rapids Create a python virtual environment cd rapids conda env create -f environment.yml -n rapids conda activate rapids Install R packages and virtual environment: snakemake -j1 renv_install snakemake -j1 renv_restore Note This step could take several minutes to complete, especially if you have less than 3Gb of RAM or packages need to be compiled from source. Please be patient and let it run until completion. Make RAPIDS script executable chmod +x rapids Check that RAPIDS is working ./rapids -j1 Ubuntu We tested on Ubuntu 18.04 & 20.04 Install dependencies sudo apt install libcurl4-openssl-dev sudo apt install libssl-dev sudo apt install libxml2-dev Install MySQL sudo apt install libmysqlclient-dev sudo apt install mysql-server Add key for R\u2019s repository. sudo apt-key adv --keyserver keyserver.ubuntu.com --recv-keys E298A3A825C0D65DFD57CBB651716619E084DAB9 Add R\u2019s repository For 18.04 sudo add-apt-repository 'deb https://cloud.r-project.org/bin/linux/ubuntu bionic-cran40/' For 20.04 sudo add-apt-repository 'deb https://cloud.r-project.org/bin/linux/ubuntu focal-cran40/' Install R 4.0. If you have other instances of R, we recommend uninstalling them sudo apt update sudo apt install r-base Install Pandoc and rmarkdown sudo apt install pandoc Rscript --vanilla -e 'install.packages(\"rmarkdown\", repos=\"http://cran.us.r-project.org\")' Install git sudo apt install git Install miniconda Restart your current shell Clone our repo: git clone https://github.com/carissalow/rapids Create a python virtual environment: cd rapids conda env create -f environment.yml -n MY_ENV_NAME conda activate MY_ENV_NAME Install R packages and virtual environment: snakemake -j1 renv_install snakemake -j1 renv_restore Note This step could take several minutes to complete, especially if you have less than 3Gb of RAM or packages need to be compiled from source. Please be patient and let it run until completion. Make RAPIDS script executable chmod +x rapids Check that RAPIDS is working ./rapids -j1","title":"Installation"},{"location":"setup/installation/#installation","text":"You can install RAPIDS using Docker (the fastest), or native instructions for MacOS and Ubuntu Docker Install Docker Pull our RAPIDS container docker pull agamk/rapids:latest ` Run RAPIDS' container (after this step is done you should see a prompt in the main RAPIDS folder with its python environment active) docker run -it agamk/rapids:latest Pull the latest version of RAPIDS git pull Make RAPIDS script executable chmod +x rapids Check that RAPIDS is working ./rapids -j1 Optional . You can edit RAPIDS files with vim but we recommend using Visual Studio Code and its Remote Containers extension How to configure Remote Containers extension Make sure RAPIDS container is running Install the Remote - Containers extension Go to the Remote Explorer panel on the left hand sidebar On the top right dropdown menu choose Containers Double click on the agamk/rapids container in the CONTAINERS tree A new VS Code session should open on RAPIDS main folder insidethe container. MacOS We tested these instructions in Catalina Install brew Install MySQL brew install mysql brew services start mysql Install R 4.0, pandoc and rmarkdown. If you have other instances of R, we recommend uninstalling them brew install r brew install pandoc Rscript --vanilla -e 'install.packages(\"rmarkdown\", repos=\"http://cran.us.r-project.org\")' Install miniconda (restart your terminal afterwards) brew cask install miniconda conda init zsh # (or conda init bash) Clone our repo git clone https://github.com/carissalow/rapids Create a python virtual environment cd rapids conda env create -f environment.yml -n rapids conda activate rapids Install R packages and virtual environment: snakemake -j1 renv_install snakemake -j1 renv_restore Note This step could take several minutes to complete, especially if you have less than 3Gb of RAM or packages need to be compiled from source. Please be patient and let it run until completion. Make RAPIDS script executable chmod +x rapids Check that RAPIDS is working ./rapids -j1 Ubuntu We tested on Ubuntu 18.04 & 20.04 Install dependencies sudo apt install libcurl4-openssl-dev sudo apt install libssl-dev sudo apt install libxml2-dev Install MySQL sudo apt install libmysqlclient-dev sudo apt install mysql-server Add key for R\u2019s repository. sudo apt-key adv --keyserver keyserver.ubuntu.com --recv-keys E298A3A825C0D65DFD57CBB651716619E084DAB9 Add R\u2019s repository For 18.04 sudo add-apt-repository 'deb https://cloud.r-project.org/bin/linux/ubuntu bionic-cran40/' For 20.04 sudo add-apt-repository 'deb https://cloud.r-project.org/bin/linux/ubuntu focal-cran40/' Install R 4.0. If you have other instances of R, we recommend uninstalling them sudo apt update sudo apt install r-base Install Pandoc and rmarkdown sudo apt install pandoc Rscript --vanilla -e 'install.packages(\"rmarkdown\", repos=\"http://cran.us.r-project.org\")' Install git sudo apt install git Install miniconda Restart your current shell Clone our repo: git clone https://github.com/carissalow/rapids Create a python virtual environment: cd rapids conda env create -f environment.yml -n MY_ENV_NAME conda activate MY_ENV_NAME Install R packages and virtual environment: snakemake -j1 renv_install snakemake -j1 renv_restore Note This step could take several minutes to complete, especially if you have less than 3Gb of RAM or packages need to be compiled from source. Please be patient and let it run until completion. Make RAPIDS script executable chmod +x rapids Check that RAPIDS is working ./rapids -j1","title":"Installation"},{"location":"workflow-examples/minimal/","text":"Minimal Working Example \u00b6 This is a quick guide for creating and running a simple pipeline to extract missing, outgoing, and incoming call features for daily and night epochs of one participant monitored on the US East coast. Install RAPIDS and make sure your conda environment is active (see Installation ) Make the changes listed below for the corresponding Configuration step (we provide an example of what the relevant sections in your config.yml will look like after you are done) Things to change on each configuration step 1. Setup your database connection credentials in .env . We assume your credentials group is called MY_GROUP . 2. America/New_York should be the default timezone 3. Create a participant file p01.yaml based on one of your participants and add p01 to [PIDS] in config.yaml . The following would be the content of your p01.yaml participant file: PHONE : DEVICE_IDS : [ aaaaaaaa-1111-bbbb-2222-cccccccccccc ] # your participant's AWARE device id PLATFORMS : [ android ] # or ios LABEL : MyTestP01 # any string START_DATE : 2020-01-01 # this can also be empty END_DATE : 2021-01-01 # this can also be empty 4. [TIME_SEGMENTS][TYPE] should be the default PERIODIC . Change [TIME_SEGMENTS][FILE] with the path of a file containing the following lines: label,start_time,length,repeats_on,repeats_value daily,00:00:00,23H 59M 59S,every_day,0 night,00:00:00,5H 59M 59S,every_day,0 5. If you collected data with AWARE you won\u2019t need to modify the attributes of [DEVICE_DATA][PHONE] 6. Set [PHONE_CALLS][PROVIDERS][RAPIDS][COMPUTE] to True Example of the config.yaml sections after the changes outlined above PIDS: [p01] TIMEZONE: &timezone America/New_York DATABASE_GROUP: &database_group MY_GROUP # ... other irrelevant sections TIME_SEGMENTS: &time_segments TYPE: PERIODIC FILE: \"data/external/timesegments_periodic.csv\" # make sure the three lines specified above are in the file INCLUDE_PAST_PERIODIC_SEGMENTS: FALSE # No need to change this if you collected AWARE data on a database and your credentials are grouped under `MY_GROUP` in `.env` DEVICE_DATA: PHONE: SOURCE: TYPE: DATABASE DATABASE_GROUP: *database_group DEVICE_ID_COLUMN: device_id # column name TIMEZONE: TYPE: SINGLE # SINGLE or MULTIPLE VALUE: *timezone ############## PHONE ########################################################### ################################################################################ # ... other irrelevant sections # Communication call features config, TYPES and FEATURES keys need to match PHONE_CALLS: TABLE: calls # change if your calls table has a different name PROVIDERS: RAPIDS: COMPUTE: True # set this to True! CALL_TYPES: ... Run RAPIDS ./rapids -j1 The call features for daily and morning time segments will be in /data/processed/features/p01/phone_calls.csv","title":"Minimal"},{"location":"workflow-examples/minimal/#minimal-working-example","text":"This is a quick guide for creating and running a simple pipeline to extract missing, outgoing, and incoming call features for daily and night epochs of one participant monitored on the US East coast. Install RAPIDS and make sure your conda environment is active (see Installation ) Make the changes listed below for the corresponding Configuration step (we provide an example of what the relevant sections in your config.yml will look like after you are done) Things to change on each configuration step 1. Setup your database connection credentials in .env . We assume your credentials group is called MY_GROUP . 2. America/New_York should be the default timezone 3. Create a participant file p01.yaml based on one of your participants and add p01 to [PIDS] in config.yaml . The following would be the content of your p01.yaml participant file: PHONE : DEVICE_IDS : [ aaaaaaaa-1111-bbbb-2222-cccccccccccc ] # your participant's AWARE device id PLATFORMS : [ android ] # or ios LABEL : MyTestP01 # any string START_DATE : 2020-01-01 # this can also be empty END_DATE : 2021-01-01 # this can also be empty 4. [TIME_SEGMENTS][TYPE] should be the default PERIODIC . Change [TIME_SEGMENTS][FILE] with the path of a file containing the following lines: label,start_time,length,repeats_on,repeats_value daily,00:00:00,23H 59M 59S,every_day,0 night,00:00:00,5H 59M 59S,every_day,0 5. If you collected data with AWARE you won\u2019t need to modify the attributes of [DEVICE_DATA][PHONE] 6. Set [PHONE_CALLS][PROVIDERS][RAPIDS][COMPUTE] to True Example of the config.yaml sections after the changes outlined above PIDS: [p01] TIMEZONE: &timezone America/New_York DATABASE_GROUP: &database_group MY_GROUP # ... other irrelevant sections TIME_SEGMENTS: &time_segments TYPE: PERIODIC FILE: \"data/external/timesegments_periodic.csv\" # make sure the three lines specified above are in the file INCLUDE_PAST_PERIODIC_SEGMENTS: FALSE # No need to change this if you collected AWARE data on a database and your credentials are grouped under `MY_GROUP` in `.env` DEVICE_DATA: PHONE: SOURCE: TYPE: DATABASE DATABASE_GROUP: *database_group DEVICE_ID_COLUMN: device_id # column name TIMEZONE: TYPE: SINGLE # SINGLE or MULTIPLE VALUE: *timezone ############## PHONE ########################################################### ################################################################################ # ... other irrelevant sections # Communication call features config, TYPES and FEATURES keys need to match PHONE_CALLS: TABLE: calls # change if your calls table has a different name PROVIDERS: RAPIDS: COMPUTE: True # set this to True! CALL_TYPES: ... Run RAPIDS ./rapids -j1 The call features for daily and morning time segments will be in /data/processed/features/p01/phone_calls.csv","title":"Minimal Working Example"}]}
\ No newline at end of file
+{"config":{"lang":["en"],"min_search_length":3,"prebuild_index":false,"separator":"[\\s\\-]+"},"docs":[{"location":"","text":"Welcome to RAPIDS documentation \u00b6 Warning The functionality outlined in these docs is implemented in the branch time_segments which we will merge to master soon as release 0.1 . The previous (first) release of RAPIDS along with the old docs will be labeled beta . If you landed on this page feel free to look around, just have in mind that we are polishing the last rough patches before we advertise 0.1 (Nov 16, 2020) Reproducible Analysis Pipeline for Data Streams (RAPIDS) allows you to process smartphone and wearable data to extract behavioral features (a.k.a. digital biomarkers/phenotypes). RAPIDS is open source, documented, modular, tested, and reproducible. At the moment we support smartphone data collected with AWARE and wearable data from Fitbit devices. Questions or feedback can be posted on #rapids in AWARE Framework's slack . Bugs should be reported on Github . Join our discussions on our algorithms and assumptions for feature processing . Ready to start? Go to Installation , then to Configuration , and then to Execution How does it work? \u00b6 RAPIDS is formed by R and Python scripts orchestrated by Snakemake . We suggest you read Snakemake\u2019s docs but in short: every link in the analysis chain is atomic and has files as input and output. Behavioral features are processed per sensor and per participant. What are the benefits of using RAPIDS? \u00b6 Consistent analysis . Every participant sensor dataset is analyzed in the exact same way and isolated from each other. Efficient analysis . Every analysis step is executed only once. Whenever your data or configuration changes only the affected files are updated. Parallel execution . Thanks to Snakemake, your analysis can be executed over multiple cores without changing your code. Extensible code . You can easily add your own behavioral features in R or Python and keep authorship and citations. Timezone aware . Your data is adjusted to the specified timezone (multiple timezones suport coming soon ). Flexible time segments . You can extract behavioral features on time windows of any length (e.g. 5 minutes, 3 hours, 2 days), on every day or particular days (e.g. weekends, Mondays, the 1 st of each month, etc.) or around events of interest (e.g. surveys or clinical relapses). Tested code . We are constantly adding tests to make sure our behavioral features are correct. Reproducible code . You can be sure your code will run in other computers as intended thanks to R and Python virtual environments. You can share your analysis code along your publications without any overhead. Private . All your data is processed locally. How is it organized? \u00b6 In broad terms the config.yaml , .env file , participants files , time segment files are the only ones that you will have to modify. All data is stored in data/ and all scripts are stored in src/ . For more information see RAPIDS\u2019 File Structure .","title":"Home"},{"location":"#welcome-to-rapids-documentation","text":"Warning The functionality outlined in these docs is implemented in the branch time_segments which we will merge to master soon as release 0.1 . The previous (first) release of RAPIDS along with the old docs will be labeled beta . If you landed on this page feel free to look around, just have in mind that we are polishing the last rough patches before we advertise 0.1 (Nov 16, 2020) Reproducible Analysis Pipeline for Data Streams (RAPIDS) allows you to process smartphone and wearable data to extract behavioral features (a.k.a. digital biomarkers/phenotypes). RAPIDS is open source, documented, modular, tested, and reproducible. At the moment we support smartphone data collected with AWARE and wearable data from Fitbit devices. Questions or feedback can be posted on #rapids in AWARE Framework's slack . Bugs should be reported on Github . Join our discussions on our algorithms and assumptions for feature processing . Ready to start? Go to Installation , then to Configuration , and then to Execution","title":"Welcome to RAPIDS documentation"},{"location":"#how-does-it-work","text":"RAPIDS is formed by R and Python scripts orchestrated by Snakemake . We suggest you read Snakemake\u2019s docs but in short: every link in the analysis chain is atomic and has files as input and output. Behavioral features are processed per sensor and per participant.","title":"How does it work?"},{"location":"#what-are-the-benefits-of-using-rapids","text":"Consistent analysis . Every participant sensor dataset is analyzed in the exact same way and isolated from each other. Efficient analysis . Every analysis step is executed only once. Whenever your data or configuration changes only the affected files are updated. Parallel execution . Thanks to Snakemake, your analysis can be executed over multiple cores without changing your code. Extensible code . You can easily add your own behavioral features in R or Python and keep authorship and citations. Timezone aware . Your data is adjusted to the specified timezone (multiple timezones suport coming soon ). Flexible time segments . You can extract behavioral features on time windows of any length (e.g. 5 minutes, 3 hours, 2 days), on every day or particular days (e.g. weekends, Mondays, the 1 st of each month, etc.) or around events of interest (e.g. surveys or clinical relapses). Tested code . We are constantly adding tests to make sure our behavioral features are correct. Reproducible code . You can be sure your code will run in other computers as intended thanks to R and Python virtual environments. You can share your analysis code along your publications without any overhead. Private . All your data is processed locally.","title":"What are the benefits of using RAPIDS?"},{"location":"#how-is-it-organized","text":"In broad terms the config.yaml , .env file , participants files , time segment files are the only ones that you will have to modify. All data is stored in data/ and all scripts are stored in src/ . For more information see RAPIDS\u2019 File Structure .","title":"How is it organized?"},{"location":"citation/","text":"Cite RAPIDS and providers \u00b6 RAPIDS and the community RAPIDS is a community effort and as such we want to continue recognizing the contributions from other researchers. Besides citing RAPIDS, we ask you to cite any of the authors listed below if you used those sensor providers in your analysis, thank you! RAPIDS \u00b6 If you used RAPIDS, please cite this paper . RAPIDS et al. citation Vega J, Li M, Aguillera K, Goel N, Joshi E, Durica KC, Kunta AR, Low CA RAPIDS: Reproducible Analysis Pipeline for Data Streams Collected with Mobile Devices JMIR Preprints. 18/08/2020:23246 DOI: 10.2196/preprints.23246 URL: https://preprints.jmir.org/preprint/23246 Panda (accelerometer) \u00b6 If you computed accelerometer features using the provider [PHONE_ACCLEROMETER][PANDA] cite this paper in addition to RAPIDS. Panda et al. citation Panda N, Solsky I, Huang EJ, Lipsitz S, Pradarelli JC, Delisle M, Cusack JC, Gadd MA, Lubitz CC, Mullen JT, Qadan M, Smith BL, Specht M, Stephen AE, Tanabe KK, Gawande AA, Onnela JP, Haynes AB. Using Smartphones to Capture Novel Recovery Metrics After Cancer Surgery. JAMA Surg. 2020 Feb 1;155(2):123-129. doi: 10.1001/jamasurg.2019.4702. PMID: 31657854; PMCID: PMC6820047. Stachl (applications foreground) \u00b6 If you computed applications foreground features using the app category (genre) catalogue in [PHONE_APPLICATIONS_FOREGROUND][RAPIDS] cite this paper in addition to RAPIDS. Stachl et al. citation Clemens Stachl, Quay Au, Ramona Schoedel, Samuel D. Gosling, Gabriella M. Harari, Daniel Buschek, Sarah Theres V\u00f6lkel, Tobias Schuwerk, Michelle Oldemeier, Theresa Ullmann, Heinrich Hussmann, Bernd Bischl, Markus B\u00fchner. Proceedings of the National Academy of Sciences Jul 2020, 117 (30) 17680-17687; DOI: 10.1073/pnas.1920484117 Barnett (locations) \u00b6 If you computed locations features using the provider [PHONE_LOCATIONS][BARNETT] cite this paper and this paper in addition to RAPIDS. Barnett et al. citation Ian Barnett, Jukka-Pekka Onnela, Inferring mobility measures from GPS traces with missing data, Biostatistics, Volume 21, Issue 2, April 2020, Pages e98\u2013e112, https://doi.org/10.1093/biostatistics/kxy059 Canzian et al. citation Luca Canzian and Mirco Musolesi. 2015. Trajectories of depression: unobtrusive monitoring of depressive states by means of smartphone mobility traces analysis. In Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp \u201815). Association for Computing Machinery, New York, NY, USA, 1293\u20131304. DOI: https://doi.org/10.1145/2750858.2805845 Doryab (locations) \u00b6 If you computed locations features using the provider [PHONE_LOCATIONS][DORYAB] cite this paper and this paper in addition to RAPIDS. Doryab et al. citation Doryab, A., Chikarsel, P., Liu, X., & Dey, A. K. (2019). Extraction of Behavioral Features from Smartphone and Wearable Data. ArXiv:1812.10394 [Cs, Stat]. http://arxiv.org/abs/1812.10394 Canzian et al. citation Luca Canzian and Mirco Musolesi. 2015. Trajectories of depression: unobtrusive monitoring of depressive states by means of smartphone mobility traces analysis. In Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp \u201815). Association for Computing Machinery, New York, NY, USA, 1293\u20131304. DOI: https://doi.org/10.1145/2750858.2805845","title":"Citation"},{"location":"citation/#cite-rapids-and-providers","text":"RAPIDS and the community RAPIDS is a community effort and as such we want to continue recognizing the contributions from other researchers. Besides citing RAPIDS, we ask you to cite any of the authors listed below if you used those sensor providers in your analysis, thank you!","title":"Cite RAPIDS and providers"},{"location":"citation/#rapids","text":"If you used RAPIDS, please cite this paper . RAPIDS et al. citation Vega J, Li M, Aguillera K, Goel N, Joshi E, Durica KC, Kunta AR, Low CA RAPIDS: Reproducible Analysis Pipeline for Data Streams Collected with Mobile Devices JMIR Preprints. 18/08/2020:23246 DOI: 10.2196/preprints.23246 URL: https://preprints.jmir.org/preprint/23246","title":"RAPIDS"},{"location":"citation/#panda-accelerometer","text":"If you computed accelerometer features using the provider [PHONE_ACCLEROMETER][PANDA] cite this paper in addition to RAPIDS. Panda et al. citation Panda N, Solsky I, Huang EJ, Lipsitz S, Pradarelli JC, Delisle M, Cusack JC, Gadd MA, Lubitz CC, Mullen JT, Qadan M, Smith BL, Specht M, Stephen AE, Tanabe KK, Gawande AA, Onnela JP, Haynes AB. Using Smartphones to Capture Novel Recovery Metrics After Cancer Surgery. JAMA Surg. 2020 Feb 1;155(2):123-129. doi: 10.1001/jamasurg.2019.4702. PMID: 31657854; PMCID: PMC6820047.","title":"Panda (accelerometer)"},{"location":"citation/#stachl-applications-foreground","text":"If you computed applications foreground features using the app category (genre) catalogue in [PHONE_APPLICATIONS_FOREGROUND][RAPIDS] cite this paper in addition to RAPIDS. Stachl et al. citation Clemens Stachl, Quay Au, Ramona Schoedel, Samuel D. Gosling, Gabriella M. Harari, Daniel Buschek, Sarah Theres V\u00f6lkel, Tobias Schuwerk, Michelle Oldemeier, Theresa Ullmann, Heinrich Hussmann, Bernd Bischl, Markus B\u00fchner. Proceedings of the National Academy of Sciences Jul 2020, 117 (30) 17680-17687; DOI: 10.1073/pnas.1920484117","title":"Stachl (applications foreground)"},{"location":"citation/#barnett-locations","text":"If you computed locations features using the provider [PHONE_LOCATIONS][BARNETT] cite this paper and this paper in addition to RAPIDS. Barnett et al. citation Ian Barnett, Jukka-Pekka Onnela, Inferring mobility measures from GPS traces with missing data, Biostatistics, Volume 21, Issue 2, April 2020, Pages e98\u2013e112, https://doi.org/10.1093/biostatistics/kxy059 Canzian et al. citation Luca Canzian and Mirco Musolesi. 2015. Trajectories of depression: unobtrusive monitoring of depressive states by means of smartphone mobility traces analysis. In Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp \u201815). Association for Computing Machinery, New York, NY, USA, 1293\u20131304. DOI: https://doi.org/10.1145/2750858.2805845","title":"Barnett (locations)"},{"location":"citation/#doryab-locations","text":"If you computed locations features using the provider [PHONE_LOCATIONS][DORYAB] cite this paper and this paper in addition to RAPIDS. Doryab et al. citation Doryab, A., Chikarsel, P., Liu, X., & Dey, A. K. (2019). Extraction of Behavioral Features from Smartphone and Wearable Data. ArXiv:1812.10394 [Cs, Stat]. http://arxiv.org/abs/1812.10394 Canzian et al. citation Luca Canzian and Mirco Musolesi. 2015. Trajectories of depression: unobtrusive monitoring of depressive states by means of smartphone mobility traces analysis. In Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp \u201815). Association for Computing Machinery, New York, NY, USA, 1293\u20131304. DOI: https://doi.org/10.1145/2750858.2805845","title":"Doryab (locations)"},{"location":"code_of_conduct/","text":"Contributor Covenant Code of Conduct \u00b6 Our Pledge \u00b6 We as members, contributors, and leaders pledge to make participation in our community a harassment-free experience for everyone, regardless of age, body size, visible or invisible disability, ethnicity, sex characteristics, gender identity and expression, level of experience, education, socio-economic status, nationality, personal appearance, race, religion, or sexual identity and orientation. We pledge to act and interact in ways that contribute to an open, welcoming, diverse, inclusive, and healthy community. Our Standards \u00b6 Examples of behavior that contributes to a positive environment for our community include: Demonstrating empathy and kindness toward other people Being respectful of differing opinions, viewpoints, and experiences Giving and gracefully accepting constructive feedback Accepting responsibility and apologizing to those affected by our mistakes, and learning from the experience Focusing on what is best not just for us as individuals, but for the overall community Examples of unacceptable behavior include: The use of sexualized language or imagery, and sexual attention or advances of any kind Trolling, insulting or derogatory comments, and personal or political attacks Public or private harassment Publishing others\u2019 private information, such as a physical or email address, without their explicit permission Other conduct which could reasonably be considered inappropriate in a professional setting Enforcement Responsibilities \u00b6 Community leaders are responsible for clarifying and enforcing our standards of acceptable behavior and will take appropriate and fair corrective action in response to any behavior that they deem inappropriate, threatening, offensive, or harmful. Community leaders have the right and responsibility to remove, edit, or reject comments, commits, code, wiki edits, issues, and other contributions that are not aligned to this Code of Conduct, and will communicate reasons for moderation decisions when appropriate. Scope \u00b6 This Code of Conduct applies within all community spaces, and also applies when an individual is officially representing the community in public spaces. Examples of representing our community include using an official e-mail address, posting via an official social media account, or acting as an appointed representative at an online or offline event. Enforcement \u00b6 Instances of abusive, harassing, or otherwise unacceptable behavior may be reported to the community leaders responsible for enforcement at moshi@pitt.edu . All complaints will be reviewed and investigated promptly and fairly. All community leaders are obligated to respect the privacy and security of the reporter of any incident. Enforcement Guidelines \u00b6 Community leaders will follow these Community Impact Guidelines in determining the consequences for any action they deem in violation of this Code of Conduct: 1. Correction \u00b6 Community Impact : Use of inappropriate language or other behavior deemed unprofessional or unwelcome in the community. Consequence : A private, written warning from community leaders, providing clarity around the nature of the violation and an explanation of why the behavior was inappropriate. A public apology may be requested. 2. Warning \u00b6 Community Impact : A violation through a single incident or series of actions. Consequence : A warning with consequences for continued behavior. No interaction with the people involved, including unsolicited interaction with those enforcing the Code of Conduct, for a specified period of time. This includes avoiding interactions in community spaces as well as external channels like social media. Violating these terms may lead to a temporary or permanent ban. 3. Temporary Ban \u00b6 Community Impact : A serious violation of community standards, including sustained inappropriate behavior. Consequence : A temporary ban from any sort of interaction or public communication with the community for a specified period of time. No public or private interaction with the people involved, including unsolicited interaction with those enforcing the Code of Conduct, is allowed during this period. Violating these terms may lead to a permanent ban. 4. Permanent Ban \u00b6 Community Impact : Demonstrating a pattern of violation of community standards, including sustained inappropriate behavior, harassment of an individual, or aggression toward or disparagement of classes of individuals. Consequence : A permanent ban from any sort of public interaction within the community. Attribution \u00b6 This Code of Conduct is adapted from the Contributor Covenant , version 2.0, available at https://www.contributor-covenant.org/version/2/0/code_of_conduct.html . Community Impact Guidelines were inspired by Mozilla\u2019s code of conduct enforcement ladder . For answers to common questions about this code of conduct, see the FAQ at https://www.contributor-covenant.org/faq . Translations are available at https://www.contributor-covenant.org/translations .","title":"Code of Conduct"},{"location":"code_of_conduct/#contributor-covenant-code-of-conduct","text":"","title":"Contributor Covenant Code of Conduct"},{"location":"code_of_conduct/#our-pledge","text":"We as members, contributors, and leaders pledge to make participation in our community a harassment-free experience for everyone, regardless of age, body size, visible or invisible disability, ethnicity, sex characteristics, gender identity and expression, level of experience, education, socio-economic status, nationality, personal appearance, race, religion, or sexual identity and orientation. We pledge to act and interact in ways that contribute to an open, welcoming, diverse, inclusive, and healthy community.","title":"Our Pledge"},{"location":"code_of_conduct/#our-standards","text":"Examples of behavior that contributes to a positive environment for our community include: Demonstrating empathy and kindness toward other people Being respectful of differing opinions, viewpoints, and experiences Giving and gracefully accepting constructive feedback Accepting responsibility and apologizing to those affected by our mistakes, and learning from the experience Focusing on what is best not just for us as individuals, but for the overall community Examples of unacceptable behavior include: The use of sexualized language or imagery, and sexual attention or advances of any kind Trolling, insulting or derogatory comments, and personal or political attacks Public or private harassment Publishing others\u2019 private information, such as a physical or email address, without their explicit permission Other conduct which could reasonably be considered inappropriate in a professional setting","title":"Our Standards"},{"location":"code_of_conduct/#enforcement-responsibilities","text":"Community leaders are responsible for clarifying and enforcing our standards of acceptable behavior and will take appropriate and fair corrective action in response to any behavior that they deem inappropriate, threatening, offensive, or harmful. Community leaders have the right and responsibility to remove, edit, or reject comments, commits, code, wiki edits, issues, and other contributions that are not aligned to this Code of Conduct, and will communicate reasons for moderation decisions when appropriate.","title":"Enforcement Responsibilities"},{"location":"code_of_conduct/#scope","text":"This Code of Conduct applies within all community spaces, and also applies when an individual is officially representing the community in public spaces. Examples of representing our community include using an official e-mail address, posting via an official social media account, or acting as an appointed representative at an online or offline event.","title":"Scope"},{"location":"code_of_conduct/#enforcement","text":"Instances of abusive, harassing, or otherwise unacceptable behavior may be reported to the community leaders responsible for enforcement at moshi@pitt.edu . All complaints will be reviewed and investigated promptly and fairly. All community leaders are obligated to respect the privacy and security of the reporter of any incident.","title":"Enforcement"},{"location":"code_of_conduct/#enforcement-guidelines","text":"Community leaders will follow these Community Impact Guidelines in determining the consequences for any action they deem in violation of this Code of Conduct:","title":"Enforcement Guidelines"},{"location":"code_of_conduct/#1-correction","text":"Community Impact : Use of inappropriate language or other behavior deemed unprofessional or unwelcome in the community. Consequence : A private, written warning from community leaders, providing clarity around the nature of the violation and an explanation of why the behavior was inappropriate. A public apology may be requested.","title":"1. Correction"},{"location":"code_of_conduct/#2-warning","text":"Community Impact : A violation through a single incident or series of actions. Consequence : A warning with consequences for continued behavior. No interaction with the people involved, including unsolicited interaction with those enforcing the Code of Conduct, for a specified period of time. This includes avoiding interactions in community spaces as well as external channels like social media. Violating these terms may lead to a temporary or permanent ban.","title":"2. Warning"},{"location":"code_of_conduct/#3-temporary-ban","text":"Community Impact : A serious violation of community standards, including sustained inappropriate behavior. Consequence : A temporary ban from any sort of interaction or public communication with the community for a specified period of time. No public or private interaction with the people involved, including unsolicited interaction with those enforcing the Code of Conduct, is allowed during this period. Violating these terms may lead to a permanent ban.","title":"3. Temporary Ban"},{"location":"code_of_conduct/#4-permanent-ban","text":"Community Impact : Demonstrating a pattern of violation of community standards, including sustained inappropriate behavior, harassment of an individual, or aggression toward or disparagement of classes of individuals. Consequence : A permanent ban from any sort of public interaction within the community.","title":"4. Permanent Ban"},{"location":"code_of_conduct/#attribution","text":"This Code of Conduct is adapted from the Contributor Covenant , version 2.0, available at https://www.contributor-covenant.org/version/2/0/code_of_conduct.html . Community Impact Guidelines were inspired by Mozilla\u2019s code of conduct enforcement ladder . For answers to common questions about this code of conduct, see the FAQ at https://www.contributor-covenant.org/faq . Translations are available at https://www.contributor-covenant.org/translations .","title":"Attribution"},{"location":"faq/","text":"Frequently Asked Questions \u00b6 Cannot connect to your MySQL server \u00b6 Problem **Error in .local ( drv, \\. .. ) :** **Failed to connect to database: Error: Can \\' t initialize character set unknown ( path: compiled \\_ in ) ** : Calls: dbConnect -> dbConnect -> .local -> .Call Execution halted [ Tue Mar 10 19 :40:15 2020 ] Error in rule download_dataset: jobid: 531 output: data/raw/p60/locations_raw.csv RuleException: CalledProcessError in line 20 of /home/ubuntu/rapids/rules/preprocessing.snakefile: Command 'set -euo pipefail; Rscript --vanilla /home/ubuntu/rapids/.snakemake/scripts/tmp_2jnvqs7.download_dataset.R' returned non-zero exit status 1 . File \"/home/ubuntu/rapids/rules/preprocessing.snakefile\" , line 20 , in __rule_download_dataset File \"/home/ubuntu/anaconda3/envs/moshi-env/lib/python3.7/concurrent/futures/thread.py\" , line 57 , in run Shutting down, this might take some time. Exiting because a job execution failed. Look above for error message Solution Please make sure the DATABASE_GROUP in config.yaml matches your DB credentials group in .env . Cannot start mysql in linux via brew services start mysql \u00b6 Problem Cannot start mysql in linux via brew services start mysql Solution Use mysql.server start Every time I run force the download_dataset rule all rules are executed \u00b6 Problem When running snakemake -j1 -R download_phone_data or ./rapids -j1 -R download_phone_data all the rules and files are re-computed Solution This is expected behavior. The advantage of using snakemake under the hood is that every time a file containing data is modified every rule that depends on that file will be re-executed to update their results. In this case, since download_dataset updates all the raw data, and you are forcing the rule with the flag -R every single rule that depends on those raw files will be executed. Error Table XXX doesn't exist while running the download_phone_data or download_fitbit_data rule. \u00b6 Problem Error in .local ( conn, statement, ... ) : could not run statement: Table 'db_name.table_name' doesn ' t exist Calls: colnames ... .local -> dbSendQuery -> dbSendQuery -> .local -> .Call Execution halted Solution Please make sure the sensors listed in [PHONE_VALID_SENSED_BINS][PHONE_SENSORS] and the [TABLE] of each sensor you activated in config.yaml match your database tables. How do I install RAPIDS on Ubuntu 16.04 \u00b6 Solution Install dependencies (Homebrew - if not installed): sudo apt-get install libmariadb-client-lgpl-dev libxml2-dev libssl-dev Install brew for linux and add the following line to ~/.bashrc : export PATH=$HOME/.linuxbrew/bin:$PATH source ~/.bashrc Install MySQL brew install mysql brew services start mysql Install R, pandoc and rmarkdown: brew install r brew install gcc@6 (needed due to this bug ) HOMEBREW_CC=gcc-6 brew install pandoc Install miniconda using these instructions Clone our repo: git clone https://github.com/carissalow/rapids Create a python virtual environment: cd rapids conda env create -f environment.yml -n MY_ENV_NAME conda activate MY_ENV_NAME Install R packages and virtual environment: snakemake renv_install snakemake renv_init snakemake renv_restore This step could take several minutes to complete. Please be patient and let it run until completion. mysql.h cannot be found \u00b6 Problem -------------------------- [ ERROR MESSAGE ] ---------------------------- <stdin>:1:10: fatal error: mysql.h: No such file or directory compilation terminated. ----------------------------------------------------------------------- ERROR: configuration failed for package 'RMySQL' Solution sudo apt install libmariadbclient-dev No package libcurl found \u00b6 Problem libcurl cannot be found Solution Install libcurl sudo apt install libcurl4-openssl-dev Configuration failed because openssl was not found. \u00b6 Problem openssl cannot be found Solution Install openssl sudo apt install libssl-dev Configuration failed because libxml-2.0 was not found \u00b6 Problem libxml-2.0 cannot be found Solution Install libxml-2.0 sudo apt install libxml2-dev SSL connection error when running RAPIDS \u00b6 Problem You are getting the following error message when running RAPIDS: Error: Failed to connect: SSL connection error: error:1425F102:SSL routines:ssl_choose_client_version:unsupported protocol. Solution This is a bug in Ubuntu 20.04 when trying to connect to an old MySQL server with MySQL client 8.0. You should get the same error message if you try to connect from the command line. There you can add the option --ssl-mode=DISABLED but we can't do this from the R connector. If you can't update your server, the quickest solution would be to import your database to another server or to a local environment. Alternatively, you could replace mysql-client and libmysqlclient-dev with mariadb-client and libmariadbclient-dev and reinstall renv. More info about this issue here DB_TABLES key not found \u00b6 Problem If you get the following error KeyError in line 43 of preprocessing.smk: 'PHONE_SENSORS' , it means that the indentation of the key [PHONE_SENSORS] is not matching the other child elements of PHONE_VALID_SENSED_BINS Solution You need to add or remove any leading whitespaces as needed on that line. PHONE_VALID_SENSED_BINS : COMPUTE : False # This flag is automatically ignored (set to True) if you are extracting PHONE_VALID_SENSED_DAYS or screen or Barnett's location features BIN_SIZE : &bin_size 5 # (in minutes) PHONE_SENSORS : [] Error while updating your conda environment in Ubuntu \u00b6 Problem You get the following error: CondaMultiError: CondaVerificationError: The package for tk located at /home/ubuntu/miniconda2/pkgs/tk-8.6.9-hed695b0_1003 appears to be corrupted. The path 'include/mysqlStubs.h' specified in the package manifest cannot be found. ClobberError: This transaction has incompatible packages due to a shared path. packages: conda-forge/linux-64::llvm-openmp-10.0.0-hc9558a2_0, anaconda/linux-64::intel-openmp-2019.4-243 path: 'lib/libiomp5.so' Solution Reinstall conda","title":"Frequently Asked Questions"},{"location":"faq/#frequently-asked-questions","text":"","title":"Frequently Asked Questions"},{"location":"faq/#cannot-connect-to-your-mysql-server","text":"Problem **Error in .local ( drv, \\. .. ) :** **Failed to connect to database: Error: Can \\' t initialize character set unknown ( path: compiled \\_ in ) ** : Calls: dbConnect -> dbConnect -> .local -> .Call Execution halted [ Tue Mar 10 19 :40:15 2020 ] Error in rule download_dataset: jobid: 531 output: data/raw/p60/locations_raw.csv RuleException: CalledProcessError in line 20 of /home/ubuntu/rapids/rules/preprocessing.snakefile: Command 'set -euo pipefail; Rscript --vanilla /home/ubuntu/rapids/.snakemake/scripts/tmp_2jnvqs7.download_dataset.R' returned non-zero exit status 1 . File \"/home/ubuntu/rapids/rules/preprocessing.snakefile\" , line 20 , in __rule_download_dataset File \"/home/ubuntu/anaconda3/envs/moshi-env/lib/python3.7/concurrent/futures/thread.py\" , line 57 , in run Shutting down, this might take some time. Exiting because a job execution failed. Look above for error message Solution Please make sure the DATABASE_GROUP in config.yaml matches your DB credentials group in .env .","title":"Cannot connect to your MySQL server"},{"location":"faq/#cannot-start-mysql-in-linux-via-brew-services-start-mysql","text":"Problem Cannot start mysql in linux via brew services start mysql Solution Use mysql.server start","title":"Cannot start mysql in linux via brew services start mysql"},{"location":"faq/#every-time-i-run-force-the-download_dataset-rule-all-rules-are-executed","text":"Problem When running snakemake -j1 -R download_phone_data or ./rapids -j1 -R download_phone_data all the rules and files are re-computed Solution This is expected behavior. The advantage of using snakemake under the hood is that every time a file containing data is modified every rule that depends on that file will be re-executed to update their results. In this case, since download_dataset updates all the raw data, and you are forcing the rule with the flag -R every single rule that depends on those raw files will be executed.","title":"Every time I run force the download_dataset rule all rules are executed"},{"location":"faq/#error-table-xxx-doesnt-exist-while-running-the-download_phone_data-or-download_fitbit_data-rule","text":"Problem Error in .local ( conn, statement, ... ) : could not run statement: Table 'db_name.table_name' doesn ' t exist Calls: colnames ... .local -> dbSendQuery -> dbSendQuery -> .local -> .Call Execution halted Solution Please make sure the sensors listed in [PHONE_VALID_SENSED_BINS][PHONE_SENSORS] and the [TABLE] of each sensor you activated in config.yaml match your database tables.","title":"Error Table XXX doesn't exist while running the download_phone_data or download_fitbit_data rule."},{"location":"faq/#how-do-i-install-rapids-on-ubuntu-1604","text":"Solution Install dependencies (Homebrew - if not installed): sudo apt-get install libmariadb-client-lgpl-dev libxml2-dev libssl-dev Install brew for linux and add the following line to ~/.bashrc : export PATH=$HOME/.linuxbrew/bin:$PATH source ~/.bashrc Install MySQL brew install mysql brew services start mysql Install R, pandoc and rmarkdown: brew install r brew install gcc@6 (needed due to this bug ) HOMEBREW_CC=gcc-6 brew install pandoc Install miniconda using these instructions Clone our repo: git clone https://github.com/carissalow/rapids Create a python virtual environment: cd rapids conda env create -f environment.yml -n MY_ENV_NAME conda activate MY_ENV_NAME Install R packages and virtual environment: snakemake renv_install snakemake renv_init snakemake renv_restore This step could take several minutes to complete. Please be patient and let it run until completion.","title":"How do I install RAPIDS on Ubuntu 16.04"},{"location":"faq/#mysqlh-cannot-be-found","text":"Problem -------------------------- [ ERROR MESSAGE ] ---------------------------- <stdin>:1:10: fatal error: mysql.h: No such file or directory compilation terminated. ----------------------------------------------------------------------- ERROR: configuration failed for package 'RMySQL' Solution sudo apt install libmariadbclient-dev","title":"mysql.h cannot be found"},{"location":"faq/#no-package-libcurl-found","text":"Problem libcurl cannot be found Solution Install libcurl sudo apt install libcurl4-openssl-dev","title":"No package libcurl found"},{"location":"faq/#configuration-failed-because-openssl-was-not-found","text":"Problem openssl cannot be found Solution Install openssl sudo apt install libssl-dev","title":"Configuration failed because openssl was not found."},{"location":"faq/#configuration-failed-because-libxml-20-was-not-found","text":"Problem libxml-2.0 cannot be found Solution Install libxml-2.0 sudo apt install libxml2-dev","title":"Configuration failed because libxml-2.0 was not found"},{"location":"faq/#ssl-connection-error-when-running-rapids","text":"Problem You are getting the following error message when running RAPIDS: Error: Failed to connect: SSL connection error: error:1425F102:SSL routines:ssl_choose_client_version:unsupported protocol. Solution This is a bug in Ubuntu 20.04 when trying to connect to an old MySQL server with MySQL client 8.0. You should get the same error message if you try to connect from the command line. There you can add the option --ssl-mode=DISABLED but we can't do this from the R connector. If you can't update your server, the quickest solution would be to import your database to another server or to a local environment. Alternatively, you could replace mysql-client and libmysqlclient-dev with mariadb-client and libmariadbclient-dev and reinstall renv. More info about this issue here","title":"SSL connection error when running RAPIDS"},{"location":"faq/#db_tables-key-not-found","text":"Problem If you get the following error KeyError in line 43 of preprocessing.smk: 'PHONE_SENSORS' , it means that the indentation of the key [PHONE_SENSORS] is not matching the other child elements of PHONE_VALID_SENSED_BINS Solution You need to add or remove any leading whitespaces as needed on that line. PHONE_VALID_SENSED_BINS : COMPUTE : False # This flag is automatically ignored (set to True) if you are extracting PHONE_VALID_SENSED_DAYS or screen or Barnett's location features BIN_SIZE : &bin_size 5 # (in minutes) PHONE_SENSORS : []","title":"DB_TABLES key not found"},{"location":"faq/#error-while-updating-your-conda-environment-in-ubuntu","text":"Problem You get the following error: CondaMultiError: CondaVerificationError: The package for tk located at /home/ubuntu/miniconda2/pkgs/tk-8.6.9-hed695b0_1003 appears to be corrupted. The path 'include/mysqlStubs.h' specified in the package manifest cannot be found. ClobberError: This transaction has incompatible packages due to a shared path. packages: conda-forge/linux-64::llvm-openmp-10.0.0-hc9558a2_0, anaconda/linux-64::intel-openmp-2019.4-243 path: 'lib/libiomp5.so' Solution Reinstall conda","title":"Error while updating your conda environment in Ubuntu"},{"location":"file-structure/","text":"File Structure \u00b6 Tip Read this page if you want to learn more about how RAPIDS is structured. If you want to start using it go to Installation , then to Configuration , and then to Execution All paths mentioned in this page are relative to RAPIDS\u2019 root folder. If you want to extract the behavioral features that RAPIDS offers, you will only have to create or modify the .env file , participants files , time segment files , and the config.yaml file as instructed in the Configuration page . The config.yaml file is the heart of RAPIDS and includes parameters to manage participants, data sources, sensor data, visualizations and more. All data is saved in data/ . The data/external/ folder stores any data imported or created by the user, data/raw/ stores sensor data as imported from your database, data/interim/ has intermediate files necessary to compute behavioral features from raw data, and data/processed/ has all the final files with the behavioral features in folders per participant and sensor. RAPIDS source code is saved in src/ . The src/data/ folder stores scripts to download, clean and pre-process sensor data, src/features has scripts to extract behavioral features organized in their respective sensor subfolders , src/models/ can host any script to create models or statistical analyses with the behavioral features you extract, and src/visualization/ has scripts to create plots of the raw and processed data. There are other files and folders but only relevant if you are interested in extending RAPIDS (e.g. virtual env files, docs, tests, Dockerfile, the Snakefile, etc.). In the figure below, we represent the interactions between users and files. After a user modifies the configuration files mentioned above, the Snakefile file will search for and execute the Snakemake rules that contain the Python or R scripts necessary to generate or update the required output files (behavioral features, plots, etc.). Interaction diagram between the user, and important files in RAPIDS","title":"File Structure"},{"location":"file-structure/#file-structure","text":"Tip Read this page if you want to learn more about how RAPIDS is structured. If you want to start using it go to Installation , then to Configuration , and then to Execution All paths mentioned in this page are relative to RAPIDS\u2019 root folder. If you want to extract the behavioral features that RAPIDS offers, you will only have to create or modify the .env file , participants files , time segment files , and the config.yaml file as instructed in the Configuration page . The config.yaml file is the heart of RAPIDS and includes parameters to manage participants, data sources, sensor data, visualizations and more. All data is saved in data/ . The data/external/ folder stores any data imported or created by the user, data/raw/ stores sensor data as imported from your database, data/interim/ has intermediate files necessary to compute behavioral features from raw data, and data/processed/ has all the final files with the behavioral features in folders per participant and sensor. RAPIDS source code is saved in src/ . The src/data/ folder stores scripts to download, clean and pre-process sensor data, src/features has scripts to extract behavioral features organized in their respective sensor subfolders , src/models/ can host any script to create models or statistical analyses with the behavioral features you extract, and src/visualization/ has scripts to create plots of the raw and processed data. There are other files and folders but only relevant if you are interested in extending RAPIDS (e.g. virtual env files, docs, tests, Dockerfile, the Snakefile, etc.). In the figure below, we represent the interactions between users and files. After a user modifies the configuration files mentioned above, the Snakefile file will search for and execute the Snakemake rules that contain the Python or R scripts necessary to generate or update the required output files (behavioral features, plots, etc.). Interaction diagram between the user, and important files in RAPIDS","title":"File Structure"},{"location":"team/","text":"RAPIDS Team \u00b6 If you are interested in contributing feel free to submit a pull request or contact us. Core Team \u00b6 Julio Vega (Designer and Lead Developer) \u00b6 About Julio Vega is a postdoctoral associate at the Mobile Sensing + Health Institute. He is interested in personalized methodologies to monitor chronic conditions that affect daily human behavior using mobile and wearable data. vegaju at upmc . edu Personal Website Meng Li \u00b6 About Meng Li received her Master of Science degree in Information Science from the University of Pittsburgh. She is interested in applying machine learning algorithms to the medical field. lim11 at upmc . edu Linkedin Profile Github Profile Abhineeth Reddy Kunta \u00b6 About Abhineeth Reddy Kunta is a Senior Software Engineer with the Mobile Sensing + Health Institute. He is experienced in software development and specializes in building solutions using machine learning. Abhineeth likes exploring ways to leverage technology in advancing medicine and education. Previously he worked as a Computer Programmer at Georgia Department of Public Health. He has a master\u2019s degree in Computer Science from George Mason University. Kwesi Aguillera \u00b6 About Kwesi Aguillera is currently in his first year at the University of Pittsburgh pursuing a Master of Sciences in Information Science specializing in Big Data Analytics. He received his Bachelor of Science degree in Computer Science and Management from the University of the West Indies. Kwesi considers himself a full stack developer and looks forward to applying this knowledge to big data analysis. Linkedin Profile Echhit Joshi \u00b6 About Echhit Joshi is a Masters student at the School of Computing and Information at University of Pittsburgh. His areas of interest are Machine/Deep Learning, Data Mining, and Analytics. Linkedin Profile Nicolas Leo \u00b6 About Nicolas is a rising senior studying computer science at the University of Pittsburgh. His academic interests include databases, machine learning, and application development. After completing his undergraduate degree, he plans to attend graduate school for a MS in Computer Science with a focus on Intelligent Systems. Nikunj Goel \u00b6 About Nik is a graduate student at the University of Pittsburgh pursuing Master of Science in Information Science. He earned his Bachelor of Technology degree in Information Technology from India. He is a Data Enthusiasts and passionate about finding the meaning out of raw data. In a long term, his goal is to create a breakthrough in Data Science and Deep Learning. Linkedin Profile Community Contributors \u00b6 Agam Kumar \u00b6 About Agam is a junior at Carnegie Mellon University studying Statistics and Machine Learning and pursuing an additional major in Computer Science. He is a member of the Data Science team in the Health and Human Performance Lab at CMU and has keen interests in software development and data science. His research interests include ML applications in medicine. Linkedin Profile Github Profile Yasaman S. Sefidgar \u00b6 About Linkedin Profile Advisors \u00b6 Afsaneh Doryab \u00b6 About Personal Website Carissa Low \u00b6 About Profile","title":"Team"},{"location":"team/#rapids-team","text":"If you are interested in contributing feel free to submit a pull request or contact us.","title":"RAPIDS Team"},{"location":"team/#core-team","text":"","title":"Core Team"},{"location":"team/#julio-vega-designer-and-lead-developer","text":"About Julio Vega is a postdoctoral associate at the Mobile Sensing + Health Institute. He is interested in personalized methodologies to monitor chronic conditions that affect daily human behavior using mobile and wearable data. vegaju at upmc . edu Personal Website","title":"Julio Vega (Designer and Lead Developer)"},{"location":"team/#meng-li","text":"About Meng Li received her Master of Science degree in Information Science from the University of Pittsburgh. She is interested in applying machine learning algorithms to the medical field. lim11 at upmc . edu Linkedin Profile Github Profile","title":"Meng Li"},{"location":"team/#abhineeth-reddy-kunta","text":"About Abhineeth Reddy Kunta is a Senior Software Engineer with the Mobile Sensing + Health Institute. He is experienced in software development and specializes in building solutions using machine learning. Abhineeth likes exploring ways to leverage technology in advancing medicine and education. Previously he worked as a Computer Programmer at Georgia Department of Public Health. He has a master\u2019s degree in Computer Science from George Mason University.","title":"Abhineeth Reddy Kunta"},{"location":"team/#kwesi-aguillera","text":"About Kwesi Aguillera is currently in his first year at the University of Pittsburgh pursuing a Master of Sciences in Information Science specializing in Big Data Analytics. He received his Bachelor of Science degree in Computer Science and Management from the University of the West Indies. Kwesi considers himself a full stack developer and looks forward to applying this knowledge to big data analysis. Linkedin Profile","title":"Kwesi Aguillera"},{"location":"team/#echhit-joshi","text":"About Echhit Joshi is a Masters student at the School of Computing and Information at University of Pittsburgh. His areas of interest are Machine/Deep Learning, Data Mining, and Analytics. Linkedin Profile","title":"Echhit Joshi"},{"location":"team/#nicolas-leo","text":"About Nicolas is a rising senior studying computer science at the University of Pittsburgh. His academic interests include databases, machine learning, and application development. After completing his undergraduate degree, he plans to attend graduate school for a MS in Computer Science with a focus on Intelligent Systems.","title":"Nicolas Leo"},{"location":"team/#nikunj-goel","text":"About Nik is a graduate student at the University of Pittsburgh pursuing Master of Science in Information Science. He earned his Bachelor of Technology degree in Information Technology from India. He is a Data Enthusiasts and passionate about finding the meaning out of raw data. In a long term, his goal is to create a breakthrough in Data Science and Deep Learning. Linkedin Profile","title":"Nikunj Goel"},{"location":"team/#community-contributors","text":"","title":"Community Contributors"},{"location":"team/#agam-kumar","text":"About Agam is a junior at Carnegie Mellon University studying Statistics and Machine Learning and pursuing an additional major in Computer Science. He is a member of the Data Science team in the Health and Human Performance Lab at CMU and has keen interests in software development and data science. His research interests include ML applications in medicine. Linkedin Profile Github Profile","title":"Agam Kumar"},{"location":"team/#yasaman-s-sefidgar","text":"About Linkedin Profile","title":"Yasaman S. Sefidgar"},{"location":"team/#advisors","text":"","title":"Advisors"},{"location":"team/#afsaneh-doryab","text":"About Personal Website","title":"Afsaneh Doryab"},{"location":"team/#carissa-low","text":"About Profile","title":"Carissa Low"},{"location":"developers/documentation/","text":"Documentation \u00b6 We use mkdocs with the material theme to write these docs. Whenever you make any changes, just push them back to the repo and the documentation will be deployed automatically. Set up development environment \u00b6 Make sure your conda environment is active pip install mkdocs pip install mkdocs-material Preview \u00b6 Run the following command in RAPIDS root folder and go to http://127.0.0.1:8000 : mkdocs serve File Structure \u00b6 The documentation config file is /mkdocs.yml , if you are adding new .md files to the docs modify the nav attribute at the bottom of that file. You can use the hierarchy there to find all the files that appear in the documentation. Reference \u00b6 Check this page to get familiar with the different visual elements we can use in the docs (admonitions, code blocks, tables, etc.) You can also refer to /docs/setup/installation.md and /docs/setup/configuration.md to see practical examples of these elements. Hint Any links to internal pages should be relative to the current page. For example, any link from this page (documentation) which is inside ./developers should begin with ../ to go one folder level up like: [ mylink ]( ../setup/installation.md ) Extras \u00b6 You can insert emojis using this syntax :[SOURCE]-[ICON_NAME] from the following sources: https://materialdesignicons.com/ https://fontawesome.com/icons/tasks?style=solid https://primer.style/octicons/ You can use this page to create markdown tables more easily","title":"Documentation"},{"location":"developers/documentation/#documentation","text":"We use mkdocs with the material theme to write these docs. Whenever you make any changes, just push them back to the repo and the documentation will be deployed automatically.","title":"Documentation"},{"location":"developers/documentation/#set-up-development-environment","text":"Make sure your conda environment is active pip install mkdocs pip install mkdocs-material","title":"Set up development environment"},{"location":"developers/documentation/#preview","text":"Run the following command in RAPIDS root folder and go to http://127.0.0.1:8000 : mkdocs serve","title":"Preview"},{"location":"developers/documentation/#file-structure","text":"The documentation config file is /mkdocs.yml , if you are adding new .md files to the docs modify the nav attribute at the bottom of that file. You can use the hierarchy there to find all the files that appear in the documentation.","title":"File Structure"},{"location":"developers/documentation/#reference","text":"Check this page to get familiar with the different visual elements we can use in the docs (admonitions, code blocks, tables, etc.) You can also refer to /docs/setup/installation.md and /docs/setup/configuration.md to see practical examples of these elements. Hint Any links to internal pages should be relative to the current page. For example, any link from this page (documentation) which is inside ./developers should begin with ../ to go one folder level up like: [ mylink ]( ../setup/installation.md )","title":"Reference"},{"location":"developers/documentation/#extras","text":"You can insert emojis using this syntax :[SOURCE]-[ICON_NAME] from the following sources: https://materialdesignicons.com/ https://fontawesome.com/icons/tasks?style=solid https://primer.style/octicons/ You can use this page to create markdown tables more easily","title":"Extras"},{"location":"developers/remote-support/","text":"Remote Support \u00b6 We use the Live Share extension of Visual Studio Code to debug bugs when sharing data or database credentials is not possible. Install Visual Studio Code Open you RAPIDS root folder in a new VSCode window Open a new Terminal Terminal > New terminal Install the Live Share extension pack Press Ctrl + P or Cmd + P and run this command: >live share: start collaboration session 6. Follow the instructions and share the session link you receive","title":"Remote Support"},{"location":"developers/remote-support/#remote-support","text":"We use the Live Share extension of Visual Studio Code to debug bugs when sharing data or database credentials is not possible. Install Visual Studio Code Open you RAPIDS root folder in a new VSCode window Open a new Terminal Terminal > New terminal Install the Live Share extension pack Press Ctrl + P or Cmd + P and run this command: >live share: start collaboration session 6. Follow the instructions and share the session link you receive","title":"Remote Support"},{"location":"developers/test-cases/","text":"Test Cases \u00b6 Along with the continued development and the addition of new sensors and features to the RAPIDS pipeline, tests for the currently available sensors and features are being implemented. Since this is a Work In Progress this page will be updated with the list of sensors and features for which testing is available. For each of the sensors listed a description of the data used for testing (test cases) are outline. Currently for all intent and testing purposes the tests/data/raw/test01/ contains all the test data files for testing android data formats and tests/data/raw/test02/ contains all the test data files for testing iOS data formats. It follows that the expected (verified output) are contained in the tests/data/processed/test01/ and tests/data/processed/test02/ for Android and iOS respectively. tests/data/raw/test03/ and tests/data/raw/test04/ contain data files for testing empty raw data files for android and iOS respectively. The following is a list of the sensors that testing is currently available. Messages (SMS) \u00b6 The raw message data file contains data for 2 separate days. The data for the first day contains records 5 records for every epoch . The second day's data contains 6 records for each of only 2 epoch (currently morning and evening ) The raw message data contains records for both message_types (i.e. recieved and sent ) in both days in all epochs. The number records with each message_types per epoch is randomly distributed There is at least one records with each message_types per epoch. There is one raw message data file each, as described above, for testing both iOS and Android data. There is also an additional empty data file for both android and iOS for testing empty data files Calls \u00b6 Due to the difference in the format of the raw call data for iOS and Android the following is the expected results the calls_with_datetime_unified.csv . This would give a better idea of the use cases being tested since the calls_with_datetime_unified.csv would make both the iOS and Android data comparable. The call data would contain data for 2 days. The data for the first day contains 6 records for every epoch . The second day's data contains 6 records for each of only 2 epoch (currently morning and evening ) The call data contains records for all call_types (i.e. incoming , outgoing and missed ) in both days in all epochs. The number records with each of the call_types per epoch is randomly distributed. There is at least one records with each call_types per epoch. There is one call data file each, as described above, for testing both iOS and Android data. There is also an additional empty data file for both android and iOS for testing empty data files Screen \u00b6 Due to the difference in the format of the raw screen data for iOS and Android the following is the expected results the screen_deltas.csv . This would give a better idea of the use cases being tested since the screen_eltas.csv would make both the iOS and Android data comparable These files are used to calculate the features for the screen sensor The screen delta data file contains data for 1 day. The screen delta data contains 1 record to represent an unlock episode that falls within an epoch for every epoch . The screen delta data contains 1 record to represent an unlock episode that falls across the boundary of 2 epochs. Namely the unlock episode starts in one epoch and ends in the next, thus there is a record for unlock episodes that fall across night to morning , morning to afternoon and finally afternoon to night The testing is done for unlock episode_type. There is one screen data file each for testing both iOS and Android data formats. There is also an additional empty data file for both android and iOS for testing empty data files Battery \u00b6 Due to the difference in the format of the raw battery data for iOS and Android as well as versions of iOS the following is the expected results the battery_deltas.csv . This would give a better idea of the use cases being tested since the battery_deltas.csv would make both the iOS and Android data comparable. These files are used to calculate the features for the battery sensor. The battery delta data file contains data for 1 day. The battery delta data contains 1 record each for a charging and discharging episode that falls within an epoch for every epoch . Thus, for the daily epoch there would be multiple charging and discharging episodes Since either a charging episode or a discharging episode and not both can occur across epochs, in order to test episodes that occur across epochs alternating episodes of charging and discharging episodes that fall across night to morning , morning to afternoon and finally afternoon to night are present in the battery delta data. This starts with a discharging episode that begins in night and end in morning . There is one battery data file each, for testing both iOS and Android data formats. There is also an additional empty data file for both android and iOS for testing empty data files Bluetooth \u00b6 The raw Bluetooth data file contains data for 1 day. The raw Bluetooth data contains at least 2 records for each epoch . Each epoch has a record with a timestamp for the beginning boundary for that epoch and a record with a timestamp for the ending boundary for that epoch . (e.g. For the morning epoch there is a record with a timestamp for 6:00AM and another record with a timestamp for 11:59:59AM . These are to test edge cases) An option of 5 Bluetooth devices are randomly distributed throughout the data records. There is one raw Bluetooth data file each, for testing both iOS and Android data formats. There is also an additional empty data file for both android and iOS for testing empty data files. WIFI \u00b6 There are 2 data files ( wifi_raw.csv and sensor_wifi_raw.csv ) for each fake participant for each phone platform. The raw WIFI data files contain data for 1 day. The sensor_wifi_raw.csv data contains at least 2 records for each epoch . Each epoch has a record with a timestamp for the beginning boundary for that epoch and a record with a timestamp for the ending boundary for that epoch . (e.g. For the morning epoch there is a record with a timestamp for 6:00AM and another record with a timestamp for 11:59:59AM . These are to test edge cases) The wifi_raw.csv data contains 3 records with random timestamps for each epoch to represent visible broadcasting WIFI network. This file is empty for the iOS phone testing data. An option of 10 access point devices is randomly distributed throughout the data records. 5 each for sensor_wifi_raw.csv and wifi_raw.csv . There data files for testing both iOS and Android data formats. There are also additional empty data files for both android and iOS for testing empty data files. Light \u00b6 The raw light data file contains data for 1 day. The raw light data contains 3 or 4 rows of data for each epoch except night . The single row of data for night is for testing features for single values inputs. (Example testing the standard deviation of one input value) Since light is only available for Android there is only one file that contains data for Android. All other files (i.e. for iPhone) are empty data files. Application Foreground \u00b6 The raw application foreground data file contains data for 1 day. The raw application foreground data contains 7 - 9 rows of data for each epoch . The records for each epoch contains apps that are randomly selected from a list of apps that are from the MULTIPLE_CATEGORIES and SINGLE_CATEGORIES (See testing_config.yaml ). There are also records in each epoch that have apps randomly selected from a list of apps that are from the EXCLUDED_CATEGORIES and EXCLUDED_APPS . This is to test that these apps are actually being excluded from the calculations of features. There are also records to test SINGLE_APPS calculations. Since application foreground is only available for Android there is only one file that contains data for Android. All other files (i.e. for iPhone) are empty data files. Activity Recognition \u00b6 The raw Activity Recognition data file contains data for 1 day. The raw Activity Recognition data each epoch period contains rows that records 2 - 5 different activity_types . The is such that durations of activities can be tested. Additionally, there are records that mimic the duration of an activity over the time boundary of neighboring epochs. (For example, there a set of records that mimic the participant in_vehicle from afternoon into evening ) There is one file each with raw Activity Recognition data for testing both iOS and Android data formats. (plugin_google_activity_recognition_raw.csv for android and plugin_ios_activity_recognition_raw.csv for iOS) There is also an additional empty data file for both android and iOS for testing empty data files. Conversation \u00b6 The raw conversation data file contains data for 2 day. The raw conversation data contains records with a sample of both datatypes (i.e. voice/noise = 0 , and conversation = 2 ) as well as rows with for samples of each of the inference values (i.e. silence = 0 , noise = 1 , voice = 2 , and unknown = 3 ) for each epoch . The different datatype and inference records are randomly distributed throughout the epoch . Additionally there are 2 - 5 records for conversations ( datatype = 2, and inference = -1) in each epoch and for each epoch except night, there is a conversation record that has a double_convo_start timestamp that is from the previous epoch . This is to test the calculations of features across epochs . There is a raw conversation data file for both android and iOS platforms ( plugin_studentlife_audio_android_raw.csv and plugin_studentlife_audio_raw.csv respectively). Finally, there are also additional empty data files for both android and iOS for testing empty data files","title":"Test cases"},{"location":"developers/test-cases/#test-cases","text":"Along with the continued development and the addition of new sensors and features to the RAPIDS pipeline, tests for the currently available sensors and features are being implemented. Since this is a Work In Progress this page will be updated with the list of sensors and features for which testing is available. For each of the sensors listed a description of the data used for testing (test cases) are outline. Currently for all intent and testing purposes the tests/data/raw/test01/ contains all the test data files for testing android data formats and tests/data/raw/test02/ contains all the test data files for testing iOS data formats. It follows that the expected (verified output) are contained in the tests/data/processed/test01/ and tests/data/processed/test02/ for Android and iOS respectively. tests/data/raw/test03/ and tests/data/raw/test04/ contain data files for testing empty raw data files for android and iOS respectively. The following is a list of the sensors that testing is currently available.","title":"Test Cases"},{"location":"developers/test-cases/#messages-sms","text":"The raw message data file contains data for 2 separate days. The data for the first day contains records 5 records for every epoch . The second day's data contains 6 records for each of only 2 epoch (currently morning and evening ) The raw message data contains records for both message_types (i.e. recieved and sent ) in both days in all epochs. The number records with each message_types per epoch is randomly distributed There is at least one records with each message_types per epoch. There is one raw message data file each, as described above, for testing both iOS and Android data. There is also an additional empty data file for both android and iOS for testing empty data files","title":"Messages (SMS)"},{"location":"developers/test-cases/#calls","text":"Due to the difference in the format of the raw call data for iOS and Android the following is the expected results the calls_with_datetime_unified.csv . This would give a better idea of the use cases being tested since the calls_with_datetime_unified.csv would make both the iOS and Android data comparable. The call data would contain data for 2 days. The data for the first day contains 6 records for every epoch . The second day's data contains 6 records for each of only 2 epoch (currently morning and evening ) The call data contains records for all call_types (i.e. incoming , outgoing and missed ) in both days in all epochs. The number records with each of the call_types per epoch is randomly distributed. There is at least one records with each call_types per epoch. There is one call data file each, as described above, for testing both iOS and Android data. There is also an additional empty data file for both android and iOS for testing empty data files","title":"Calls"},{"location":"developers/test-cases/#screen","text":"Due to the difference in the format of the raw screen data for iOS and Android the following is the expected results the screen_deltas.csv . This would give a better idea of the use cases being tested since the screen_eltas.csv would make both the iOS and Android data comparable These files are used to calculate the features for the screen sensor The screen delta data file contains data for 1 day. The screen delta data contains 1 record to represent an unlock episode that falls within an epoch for every epoch . The screen delta data contains 1 record to represent an unlock episode that falls across the boundary of 2 epochs. Namely the unlock episode starts in one epoch and ends in the next, thus there is a record for unlock episodes that fall across night to morning , morning to afternoon and finally afternoon to night The testing is done for unlock episode_type. There is one screen data file each for testing both iOS and Android data formats. There is also an additional empty data file for both android and iOS for testing empty data files","title":"Screen"},{"location":"developers/test-cases/#battery","text":"Due to the difference in the format of the raw battery data for iOS and Android as well as versions of iOS the following is the expected results the battery_deltas.csv . This would give a better idea of the use cases being tested since the battery_deltas.csv would make both the iOS and Android data comparable. These files are used to calculate the features for the battery sensor. The battery delta data file contains data for 1 day. The battery delta data contains 1 record each for a charging and discharging episode that falls within an epoch for every epoch . Thus, for the daily epoch there would be multiple charging and discharging episodes Since either a charging episode or a discharging episode and not both can occur across epochs, in order to test episodes that occur across epochs alternating episodes of charging and discharging episodes that fall across night to morning , morning to afternoon and finally afternoon to night are present in the battery delta data. This starts with a discharging episode that begins in night and end in morning . There is one battery data file each, for testing both iOS and Android data formats. There is also an additional empty data file for both android and iOS for testing empty data files","title":"Battery"},{"location":"developers/test-cases/#bluetooth","text":"The raw Bluetooth data file contains data for 1 day. The raw Bluetooth data contains at least 2 records for each epoch . Each epoch has a record with a timestamp for the beginning boundary for that epoch and a record with a timestamp for the ending boundary for that epoch . (e.g. For the morning epoch there is a record with a timestamp for 6:00AM and another record with a timestamp for 11:59:59AM . These are to test edge cases) An option of 5 Bluetooth devices are randomly distributed throughout the data records. There is one raw Bluetooth data file each, for testing both iOS and Android data formats. There is also an additional empty data file for both android and iOS for testing empty data files.","title":"Bluetooth"},{"location":"developers/test-cases/#wifi","text":"There are 2 data files ( wifi_raw.csv and sensor_wifi_raw.csv ) for each fake participant for each phone platform. The raw WIFI data files contain data for 1 day. The sensor_wifi_raw.csv data contains at least 2 records for each epoch . Each epoch has a record with a timestamp for the beginning boundary for that epoch and a record with a timestamp for the ending boundary for that epoch . (e.g. For the morning epoch there is a record with a timestamp for 6:00AM and another record with a timestamp for 11:59:59AM . These are to test edge cases) The wifi_raw.csv data contains 3 records with random timestamps for each epoch to represent visible broadcasting WIFI network. This file is empty for the iOS phone testing data. An option of 10 access point devices is randomly distributed throughout the data records. 5 each for sensor_wifi_raw.csv and wifi_raw.csv . There data files for testing both iOS and Android data formats. There are also additional empty data files for both android and iOS for testing empty data files.","title":"WIFI"},{"location":"developers/test-cases/#light","text":"The raw light data file contains data for 1 day. The raw light data contains 3 or 4 rows of data for each epoch except night . The single row of data for night is for testing features for single values inputs. (Example testing the standard deviation of one input value) Since light is only available for Android there is only one file that contains data for Android. All other files (i.e. for iPhone) are empty data files.","title":"Light"},{"location":"developers/test-cases/#application-foreground","text":"The raw application foreground data file contains data for 1 day. The raw application foreground data contains 7 - 9 rows of data for each epoch . The records for each epoch contains apps that are randomly selected from a list of apps that are from the MULTIPLE_CATEGORIES and SINGLE_CATEGORIES (See testing_config.yaml ). There are also records in each epoch that have apps randomly selected from a list of apps that are from the EXCLUDED_CATEGORIES and EXCLUDED_APPS . This is to test that these apps are actually being excluded from the calculations of features. There are also records to test SINGLE_APPS calculations. Since application foreground is only available for Android there is only one file that contains data for Android. All other files (i.e. for iPhone) are empty data files.","title":"Application Foreground"},{"location":"developers/test-cases/#activity-recognition","text":"The raw Activity Recognition data file contains data for 1 day. The raw Activity Recognition data each epoch period contains rows that records 2 - 5 different activity_types . The is such that durations of activities can be tested. Additionally, there are records that mimic the duration of an activity over the time boundary of neighboring epochs. (For example, there a set of records that mimic the participant in_vehicle from afternoon into evening ) There is one file each with raw Activity Recognition data for testing both iOS and Android data formats. (plugin_google_activity_recognition_raw.csv for android and plugin_ios_activity_recognition_raw.csv for iOS) There is also an additional empty data file for both android and iOS for testing empty data files.","title":"Activity Recognition"},{"location":"developers/test-cases/#conversation","text":"The raw conversation data file contains data for 2 day. The raw conversation data contains records with a sample of both datatypes (i.e. voice/noise = 0 , and conversation = 2 ) as well as rows with for samples of each of the inference values (i.e. silence = 0 , noise = 1 , voice = 2 , and unknown = 3 ) for each epoch . The different datatype and inference records are randomly distributed throughout the epoch . Additionally there are 2 - 5 records for conversations ( datatype = 2, and inference = -1) in each epoch and for each epoch except night, there is a conversation record that has a double_convo_start timestamp that is from the previous epoch . This is to test the calculations of features across epochs . There is a raw conversation data file for both android and iOS platforms ( plugin_studentlife_audio_android_raw.csv and plugin_studentlife_audio_raw.csv respectively). Finally, there are also additional empty data files for both android and iOS for testing empty data files","title":"Conversation"},{"location":"developers/testing/","text":"Testing \u00b6 The following is a simple guide to testing RAPIDS. All files necessary for testing are stored in the /tests directory Steps for Testing \u00b6 To begin testing RAPIDS place the fake raw input data csv files in tests/data/raw/ . The fake participant files should be placed in tests/data/external/ . The expected output files of RAPIDS after processing the input data should be placed in tests/data/processesd/ . The Snakemake rule(s) that are to be tested must be placed in the tests/Snakemake file. The current tests/Snakemake is a good example of how to define them. (At the time of writing this documentation the snakefile contains rules messages (SMS), calls and screen) Edit the tests/settings/config.yaml . Add and/or remove the rules to be run for testing from the forcerun list. Edit the tests/settings/testing_config.yaml with the necessary configuration settings for running the rules to be tested. Add any additional testscripts in tests/scripts . Uncomment or comment off lines in the testing shell script tests/scripts/run_tests.sh . Run the testing shell script. tests/scripts/run_tests.sh The following is a snippet of the output you should see after running your test. test_sensors_files_exist ( test_sensor_features.TestSensorFeatures ) ... ok test_sensors_features_calculations ( test_sensor_features.TestSensorFeatures ) ... FAIL ====================================================================== FAIL: test_sensors_features_calculations ( test_sensor_features.TestSensorFeatures ) ---------------------------------------------------------------------- The results above show that the first test test_sensors_files_exist passed while test_sensors_features_calculations failed. In addition you should get the traceback of the failure (not shown here). For more information on how to implement test scripts and use unittest please see Unittest Documentation Testing of the RAPIDS sensors and features is a work-in-progress. Please see test-cases for a list of sensors and features that have testing currently available. Currently the repository is set up to test a number of sensors out of the box by simply running the tests/scripts/run_tests.sh command once the RAPIDS python environment is active.","title":"Testing"},{"location":"developers/testing/#testing","text":"The following is a simple guide to testing RAPIDS. All files necessary for testing are stored in the /tests directory","title":"Testing"},{"location":"developers/testing/#steps-for-testing","text":"To begin testing RAPIDS place the fake raw input data csv files in tests/data/raw/ . The fake participant files should be placed in tests/data/external/ . The expected output files of RAPIDS after processing the input data should be placed in tests/data/processesd/ . The Snakemake rule(s) that are to be tested must be placed in the tests/Snakemake file. The current tests/Snakemake is a good example of how to define them. (At the time of writing this documentation the snakefile contains rules messages (SMS), calls and screen) Edit the tests/settings/config.yaml . Add and/or remove the rules to be run for testing from the forcerun list. Edit the tests/settings/testing_config.yaml with the necessary configuration settings for running the rules to be tested. Add any additional testscripts in tests/scripts . Uncomment or comment off lines in the testing shell script tests/scripts/run_tests.sh . Run the testing shell script. tests/scripts/run_tests.sh The following is a snippet of the output you should see after running your test. test_sensors_files_exist ( test_sensor_features.TestSensorFeatures ) ... ok test_sensors_features_calculations ( test_sensor_features.TestSensorFeatures ) ... FAIL ====================================================================== FAIL: test_sensors_features_calculations ( test_sensor_features.TestSensorFeatures ) ---------------------------------------------------------------------- The results above show that the first test test_sensors_files_exist passed while test_sensors_features_calculations failed. In addition you should get the traceback of the failure (not shown here). For more information on how to implement test scripts and use unittest please see Unittest Documentation Testing of the RAPIDS sensors and features is a work-in-progress. Please see test-cases for a list of sensors and features that have testing currently available. Currently the repository is set up to test a number of sensors out of the box by simply running the tests/scripts/run_tests.sh command once the RAPIDS python environment is active.","title":"Steps for Testing"},{"location":"developers/virtual-environments/","text":"Virtual Environments \u00b6 Add new packages \u00b6 Try to install any new package using conda . If a package is not available in one of conda \u2018s channels you can install it with pip but make sure your virtual environment is active. Update your conda environment.yaml \u00b6 After installing a new package you can use the following command in your terminal to update your environment.yaml before publishing your pipeline. Note that we ignore the package version for libfortran to keep compatibility with Linux: conda env export --no-builds | sed 's/^.*libgfortran.*$/ - libgfortran/' > environment.yml Update and prune your conda environment from a environment.yaml file \u00b6 Execute the following command in your terminal, see these docs for more information conda env update --prefix ./env --file environment.yml --prune","title":"Virtual Environments"},{"location":"developers/virtual-environments/#virtual-environments","text":"","title":"Virtual Environments"},{"location":"developers/virtual-environments/#add-new-packages","text":"Try to install any new package using conda . If a package is not available in one of conda \u2018s channels you can install it with pip but make sure your virtual environment is active.","title":"Add new packages"},{"location":"developers/virtual-environments/#update-your-conda-environmentyaml","text":"After installing a new package you can use the following command in your terminal to update your environment.yaml before publishing your pipeline. Note that we ignore the package version for libfortran to keep compatibility with Linux: conda env export --no-builds | sed 's/^.*libgfortran.*$/ - libgfortran/' > environment.yml","title":"Update your conda environment.yaml"},{"location":"developers/virtual-environments/#update-and-prune-your-conda-environment-from-a-environmentyaml-file","text":"Execute the following command in your terminal, see these docs for more information conda env update --prefix ./env --file environment.yml --prune","title":"Update and prune your conda environment from a environment.yaml file"},{"location":"features/add-new-features/","text":"Add New Features \u00b6 Hint We recommend reading the Behavioral Features Introduction before reading this page Hint You won\u2019t have to deal with time zones, dates, times, data cleaning or preprocessing. The data that RAPIDS pipes to your feature extraction code is ready to process. New Features for Existing Sensors \u00b6 You can add new features to any existing sensors (see list below) by adding a new provider in three steps: Modify the config.yaml file Create a provider folder, script and function Implement your features extraction code As a tutorial, we will add a new provider for PHONE_ACCELEROMETER called VEGA that extracts feature1 , feature2 , feature3 in Python and that it requires a parameter from the user called MY_PARAMETER . Existing Sensors An existing sensor is any of the phone or Fitbit sensors with a configuration entry in config.yaml : Phone Accelerometer Phone Activity Recognition Phone Applications Foreground Phone Battery Phone Bluetooth Phone Calls Phone Conversation Phone Data Yield Phone Light Phone Locations Phone Messages Phone Screen Phone WiFI Connected Phone WiFI Visible Fitbit Heart Rate Summary Fitbit Heart Rate Intraday Fitbit Sleep Summary Fitbit Steps Summary Fitbit Steps Intraday Modify the config.yaml file \u00b6 In this step you need to add your provider configuration section under the relevant sensor in config.yaml . See our example for our tutorial\u2019s VEGA provider for PHONE_ACCELEROMETER : Example configuration for a new accelerometer provider VEGA PHONE_ACCELEROMETER : TABLE : accelerometer PROVIDERS : RAPIDS : COMPUTE : False ... PANDA : COMPUTE : False ... VEGA : COMPUTE : False FEATURES : [ \"feature1\" , \"feature2\" , \"feature3\" ] MY_PARAMTER : a_string SRC_FOLDER : \"vega\" SRC_LANGUAGE : \"python\" Key Description [COMPUTE] Flag to activate/deactivate your provider [FEATURES] List of features your provider supports. Your provider code should only return the features on this list [MY_PARAMTER] An arbitrary parameter that our example provider VEGA needs. This can be a boolean, integer, float, string or an array of any of such types. [SRC_LANGUAGE] The programming language of your provider script, it can be python or r , in our example python [SRC_FOLDER] The name of your provider in lower case, in our example vega (this will be the name of your folder in the next step) Create a provider folder, script and function \u00b6 In this step you need to add a folder, script and function for your provider. Create your provider folder under src/feature/DEVICE_SENSOR/YOUR_PROVIDER , in our example src/feature/phone_accelerometer/vega (same as [SRC_FOLDER] in the step above). Create your provider script inside your provider folder, it can be a Python file called main.py or an R file called main.R . Add your provider function in your provider script. The name of such function should be [providername]_features , in our example vega_features Python function def [ providername ] _features ( sensor_data_files , time_segment , provider , filter_data_by_segment , * args , ** kwargs ): R function [ providername ] _ features <- function ( sensor_data , time_segment , provider ) Implement your feature extraction code \u00b6 The provider function that you created in the step above will receive the following parameters: Parameter Description sensor_data_files Path to the CSV file containing the data of a single participant. This data has been cleaned and preprocessed. Your function will be automatically called for each participant in your study (in the [PIDS] array in config.yaml ) time_segment The label of the time segment that should be processed. provider The parameters you configured for your provider in config.yaml will be available in this variable as a dictionary in Python or a list in R. In our example this dictionary contains {MY_PARAMETER:\"a_string\"} filter_data_by_segment Python only. A function that you will use to filter your data. In R this function is already available in the environment. *args Python only. Not used for now **kwargs Python only. Not used for now The code to extract your behavioral features should be implemented in your provider function and in general terms it will have three stages: 1. Read a participant\u2019s data by loading the CSV data stored in the file pointed by sensor_data_files acc_data = pd . read_csv ( sensor_data_files [ \"sensor_data\" ]) Note that phone\u2019s battery, screen, and activity recognition data is given as episodes instead of event rows (for example, start and end timestamps of the periods the phone screen was on) 2. Filter your data to process only those rows that belong to time_segment This step is only one line of code, but to undersand why we need it, keep reading. acc_data = filter_data_by_segment ( acc_data , time_segment ) You should use the filter_data_by_segment() function to process and group those rows that belong to each of the time segments RAPIDS could be configured with . Let\u2019s understand the filter_data_by_segment() function with an example. A RAPIDS user can extract features on any arbitrary time segment . A time segment is a period of time that has a label and one or more instances. For example, the user (or you) could have requested features on a daily, weekly, and week-end basis for p01 . The labels are arbritrary and the instances depend on the days a participant was monitored for: the daily segment could be named my_days and if p01 was monitored for 14 days, it would have 14 instances the weekly segment could be named my_weeks and if p01 was monitored for 14 days, it would have 2 instances. the weekend segment could be named my_weekends and if p01 was monitored for 14 days, it would have 2 instances. For this example, RAPIDS will call your provider function three times for p01 , once where time_segment is my_days , once where time_segment is my_weeks and once where time_segment is my_weekends . In this example not every row in p01 \u2018s data needs to take part in the feature computation for either segment and the rows need to be grouped differently. Thus filter_data_by_segment() comes in handy, it will return a data frame that contains the rows that were logged during a time segment plus an extra column called local_segment . This new column will have as many unique values as time segment instances exist (14, 2, and 2 for our p01 \u2018s my_days , my_weeks , and my_weekends examples). After filtering, you should group the data frame by this column and compute any desired features , for example: acc_features [ \"maxmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . max () The reason RAPIDS does not filter the participant\u2019s data set for you is because your code might need to compute something based on a participant\u2019s complete dataset before computing their features. For example, you might want to identify the number that called a participant the most throughout the study before computing a feature with the number of calls the participant received from this number. 3. Return a data frame with your features After filtering, grouping your data, and computing your features, your provider function should return a data frame that has: One row per time segment instance (e.g. 14 our p01 \u2018s my_days example) The local_segment column added by filter_data_by_segment() One column per feature. By convention the name of your features should only contain letters or numbers ( feature1 ). RAPIDS will automatically add the right sensor and provider prefix ( phone_accelerometr_vega_ ) PHONE_ACCELEROMETER Provider Example For your reference, this a short example of our own provider ( RAPIDS ) for PHONE_ACCELEROMETER that computes five acceleration features def rapids_features ( sensor_data_files , time_segment , provider , filter_data_by_segment , * args , ** kwargs ): acc_data = pd . read_csv ( sensor_data_files [ \"sensor_data\" ]) requested_features = provider [ \"FEATURES\" ] # name of the features this function can compute base_features_names = [ \"maxmagnitude\" , \"minmagnitude\" , \"avgmagnitude\" , \"medianmagnitude\" , \"stdmagnitude\" ] # the subset of requested features this function can compute features_to_compute = list ( set ( requested_features ) & set ( base_features_names )) acc_features = pd . DataFrame ( columns = [ \"local_segment\" ] + features_to_compute ) if not acc_data . empty : acc_data = filter_data_by_segment ( acc_data , time_segment ) if not acc_data . empty : acc_features = pd . DataFrame () # get magnitude related features: magnitude = sqrt(x^2+y^2+z^2) magnitude = acc_data . apply ( lambda row : np . sqrt ( row [ \"double_values_0\" ] ** 2 + row [ \"double_values_1\" ] ** 2 + row [ \"double_values_2\" ] ** 2 ), axis = 1 ) acc_data = acc_data . assign ( magnitude = magnitude . values ) if \"maxmagnitude\" in features_to_compute : acc_features [ \"maxmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . max () if \"minmagnitude\" in features_to_compute : acc_features [ \"minmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . min () if \"avgmagnitude\" in features_to_compute : acc_features [ \"avgmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . mean () if \"medianmagnitude\" in features_to_compute : acc_features [ \"medianmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . median () if \"stdmagnitude\" in features_to_compute : acc_features [ \"stdmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . std () acc_features = acc_features . reset_index () return acc_features New Features for Non-Existing Sensors \u00b6 If you want to add features for a device or a sensor that we do not support at the moment (those that do not appear in the \"Existing Sensors\" list above), contact us or request it on Slack and we can add the necessary code so you can follow the instructions above.","title":"Add New Features"},{"location":"features/add-new-features/#add-new-features","text":"Hint We recommend reading the Behavioral Features Introduction before reading this page Hint You won\u2019t have to deal with time zones, dates, times, data cleaning or preprocessing. The data that RAPIDS pipes to your feature extraction code is ready to process.","title":"Add New Features"},{"location":"features/add-new-features/#new-features-for-existing-sensors","text":"You can add new features to any existing sensors (see list below) by adding a new provider in three steps: Modify the config.yaml file Create a provider folder, script and function Implement your features extraction code As a tutorial, we will add a new provider for PHONE_ACCELEROMETER called VEGA that extracts feature1 , feature2 , feature3 in Python and that it requires a parameter from the user called MY_PARAMETER . Existing Sensors An existing sensor is any of the phone or Fitbit sensors with a configuration entry in config.yaml : Phone Accelerometer Phone Activity Recognition Phone Applications Foreground Phone Battery Phone Bluetooth Phone Calls Phone Conversation Phone Data Yield Phone Light Phone Locations Phone Messages Phone Screen Phone WiFI Connected Phone WiFI Visible Fitbit Heart Rate Summary Fitbit Heart Rate Intraday Fitbit Sleep Summary Fitbit Steps Summary Fitbit Steps Intraday","title":"New Features for Existing Sensors"},{"location":"features/add-new-features/#modify-the-configyaml-file","text":"In this step you need to add your provider configuration section under the relevant sensor in config.yaml . See our example for our tutorial\u2019s VEGA provider for PHONE_ACCELEROMETER : Example configuration for a new accelerometer provider VEGA PHONE_ACCELEROMETER : TABLE : accelerometer PROVIDERS : RAPIDS : COMPUTE : False ... PANDA : COMPUTE : False ... VEGA : COMPUTE : False FEATURES : [ \"feature1\" , \"feature2\" , \"feature3\" ] MY_PARAMTER : a_string SRC_FOLDER : \"vega\" SRC_LANGUAGE : \"python\" Key Description [COMPUTE] Flag to activate/deactivate your provider [FEATURES] List of features your provider supports. Your provider code should only return the features on this list [MY_PARAMTER] An arbitrary parameter that our example provider VEGA needs. This can be a boolean, integer, float, string or an array of any of such types. [SRC_LANGUAGE] The programming language of your provider script, it can be python or r , in our example python [SRC_FOLDER] The name of your provider in lower case, in our example vega (this will be the name of your folder in the next step)","title":"Modify the config.yaml file"},{"location":"features/add-new-features/#create-a-provider-folder-script-and-function","text":"In this step you need to add a folder, script and function for your provider. Create your provider folder under src/feature/DEVICE_SENSOR/YOUR_PROVIDER , in our example src/feature/phone_accelerometer/vega (same as [SRC_FOLDER] in the step above). Create your provider script inside your provider folder, it can be a Python file called main.py or an R file called main.R . Add your provider function in your provider script. The name of such function should be [providername]_features , in our example vega_features Python function def [ providername ] _features ( sensor_data_files , time_segment , provider , filter_data_by_segment , * args , ** kwargs ): R function [ providername ] _ features <- function ( sensor_data , time_segment , provider )","title":"Create a provider folder, script and function"},{"location":"features/add-new-features/#implement-your-feature-extraction-code","text":"The provider function that you created in the step above will receive the following parameters: Parameter Description sensor_data_files Path to the CSV file containing the data of a single participant. This data has been cleaned and preprocessed. Your function will be automatically called for each participant in your study (in the [PIDS] array in config.yaml ) time_segment The label of the time segment that should be processed. provider The parameters you configured for your provider in config.yaml will be available in this variable as a dictionary in Python or a list in R. In our example this dictionary contains {MY_PARAMETER:\"a_string\"} filter_data_by_segment Python only. A function that you will use to filter your data. In R this function is already available in the environment. *args Python only. Not used for now **kwargs Python only. Not used for now The code to extract your behavioral features should be implemented in your provider function and in general terms it will have three stages: 1. Read a participant\u2019s data by loading the CSV data stored in the file pointed by sensor_data_files acc_data = pd . read_csv ( sensor_data_files [ \"sensor_data\" ]) Note that phone\u2019s battery, screen, and activity recognition data is given as episodes instead of event rows (for example, start and end timestamps of the periods the phone screen was on) 2. Filter your data to process only those rows that belong to time_segment This step is only one line of code, but to undersand why we need it, keep reading. acc_data = filter_data_by_segment ( acc_data , time_segment ) You should use the filter_data_by_segment() function to process and group those rows that belong to each of the time segments RAPIDS could be configured with . Let\u2019s understand the filter_data_by_segment() function with an example. A RAPIDS user can extract features on any arbitrary time segment . A time segment is a period of time that has a label and one or more instances. For example, the user (or you) could have requested features on a daily, weekly, and week-end basis for p01 . The labels are arbritrary and the instances depend on the days a participant was monitored for: the daily segment could be named my_days and if p01 was monitored for 14 days, it would have 14 instances the weekly segment could be named my_weeks and if p01 was monitored for 14 days, it would have 2 instances. the weekend segment could be named my_weekends and if p01 was monitored for 14 days, it would have 2 instances. For this example, RAPIDS will call your provider function three times for p01 , once where time_segment is my_days , once where time_segment is my_weeks and once where time_segment is my_weekends . In this example not every row in p01 \u2018s data needs to take part in the feature computation for either segment and the rows need to be grouped differently. Thus filter_data_by_segment() comes in handy, it will return a data frame that contains the rows that were logged during a time segment plus an extra column called local_segment . This new column will have as many unique values as time segment instances exist (14, 2, and 2 for our p01 \u2018s my_days , my_weeks , and my_weekends examples). After filtering, you should group the data frame by this column and compute any desired features , for example: acc_features [ \"maxmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . max () The reason RAPIDS does not filter the participant\u2019s data set for you is because your code might need to compute something based on a participant\u2019s complete dataset before computing their features. For example, you might want to identify the number that called a participant the most throughout the study before computing a feature with the number of calls the participant received from this number. 3. Return a data frame with your features After filtering, grouping your data, and computing your features, your provider function should return a data frame that has: One row per time segment instance (e.g. 14 our p01 \u2018s my_days example) The local_segment column added by filter_data_by_segment() One column per feature. By convention the name of your features should only contain letters or numbers ( feature1 ). RAPIDS will automatically add the right sensor and provider prefix ( phone_accelerometr_vega_ ) PHONE_ACCELEROMETER Provider Example For your reference, this a short example of our own provider ( RAPIDS ) for PHONE_ACCELEROMETER that computes five acceleration features def rapids_features ( sensor_data_files , time_segment , provider , filter_data_by_segment , * args , ** kwargs ): acc_data = pd . read_csv ( sensor_data_files [ \"sensor_data\" ]) requested_features = provider [ \"FEATURES\" ] # name of the features this function can compute base_features_names = [ \"maxmagnitude\" , \"minmagnitude\" , \"avgmagnitude\" , \"medianmagnitude\" , \"stdmagnitude\" ] # the subset of requested features this function can compute features_to_compute = list ( set ( requested_features ) & set ( base_features_names )) acc_features = pd . DataFrame ( columns = [ \"local_segment\" ] + features_to_compute ) if not acc_data . empty : acc_data = filter_data_by_segment ( acc_data , time_segment ) if not acc_data . empty : acc_features = pd . DataFrame () # get magnitude related features: magnitude = sqrt(x^2+y^2+z^2) magnitude = acc_data . apply ( lambda row : np . sqrt ( row [ \"double_values_0\" ] ** 2 + row [ \"double_values_1\" ] ** 2 + row [ \"double_values_2\" ] ** 2 ), axis = 1 ) acc_data = acc_data . assign ( magnitude = magnitude . values ) if \"maxmagnitude\" in features_to_compute : acc_features [ \"maxmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . max () if \"minmagnitude\" in features_to_compute : acc_features [ \"minmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . min () if \"avgmagnitude\" in features_to_compute : acc_features [ \"avgmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . mean () if \"medianmagnitude\" in features_to_compute : acc_features [ \"medianmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . median () if \"stdmagnitude\" in features_to_compute : acc_features [ \"stdmagnitude\" ] = acc_data . groupby ([ \"local_segment\" ])[ \"magnitude\" ] . std () acc_features = acc_features . reset_index () return acc_features","title":"Implement your feature extraction code"},{"location":"features/add-new-features/#new-features-for-non-existing-sensors","text":"If you want to add features for a device or a sensor that we do not support at the moment (those that do not appear in the \"Existing Sensors\" list above), contact us or request it on Slack and we can add the necessary code so you can follow the instructions above.","title":"New Features for Non-Existing Sensors"},{"location":"features/feature-introduction/","text":"Behavioral Features Introduction \u00b6 Every phone or Fitbit sensor has a corresponding config section in config.yaml , these sections follow a similar structure and we\u2019ll use PHONE_ACCELEROMETER as an example to explain this structure. Hint We recommend reading this page if you are using RAPIDS for the first time All computed sensor features are stored under /data/processed/features on files per sensor, per participant and per study (all participants). Every time you change any sensor parameters, provider parameters or provider features, all the necessary files will be updated as soon as you execute RAPIDS. Config section example for PHONE_ACCELEROMETER # 1) Config section PHONE_ACCELEROMETER : # 2) Parameters for PHONE_ACCELEROMETER TABLE : accelerometer # 3) Providers for PHONE_ACCELEROMETER PROVIDERS : # 4) RAPIDS provider RAPIDS : # 4.1) Parameters of RAPIDS provider of PHONE_ACCELEROMETER COMPUTE : False # 4.2) Features of RAPIDS provider of PHONE_ACCELEROMETER FEATURES : [ \"maxmagnitude\" , \"minmagnitude\" , \"avgmagnitude\" , \"medianmagnitude\" , \"stdmagnitude\" ] SRC_FOLDER : \"rapids\" # inside src/features/phone_accelerometer SRC_LANGUAGE : \"python\" # 5) PANDA provider PANDA : # 5.1) Parameters of PANDA provider of PHONE_ACCELEROMETER COMPUTE : False VALID_SENSED_MINUTES : False # 5.2) Features of PANDA provider of PHONE_ACCELEROMETER FEATURES : exertional_activity_episode : [ \"sumduration\" , \"maxduration\" , \"minduration\" , \"avgduration\" , \"medianduration\" , \"stdduration\" ] nonexertional_activity_episode : [ \"sumduration\" , \"maxduration\" , \"minduration\" , \"avgduration\" , \"medianduration\" , \"stdduration\" ] SRC_FOLDER : \"panda\" # inside src/features/phone_accelerometer SRC_LANGUAGE : \"python\" Sensor Parameters \u00b6 Each sensor configuration section has a \u201cparameters\u201d subsection (see #2 in the example). These are parameters that affect different aspects of how the raw data is downloaded, and processed. The TABLE parameter exists for every sensor, but some sensors will have extra parameters like [PHONE_LOCATIONS] . We explain these parameters in a table at the top of each sensor documentation page. Sensor Providers \u00b6 Each sensor configuration section can have zero, one or more behavioral feature providers (see #3 in the example). A provider is a script created by the core RAPIDS team or other researchers that extracts behavioral features for that sensor. In this example, accelerometer has two providers: RAPIDS (see #4 ) and PANDA (see #5 ). Provider Parameters \u00b6 Each provider has parameters that affect the computation of the behavioral features it offers (see #4.1 or #5.1 in the example). These parameters will include at least a [COMPUTE] flag that you switch to True to extract a provider\u2019s behavioral features. We explain every provider\u2019s parameter in a table under the Parameters description heading on each provider documentation page. Provider Features \u00b6 Each provider offers a set of behavioral features (see #4.2 or #5.2 in the example). For some providers these features are grouped in an array (like those for RAPIDS provider in #4.2 ) but for others they are grouped in a collection of arrays depending on the meaning and purpose of those features (like those for PANDAS provider in #5.2 ). In either case, you can delete the features you are not interested in and they will not be included in the sensor\u2019s output feature file. We explain each behavioral feature in a table under the Features description heading on each provider documentation page.","title":"Introduction"},{"location":"features/feature-introduction/#behavioral-features-introduction","text":"Every phone or Fitbit sensor has a corresponding config section in config.yaml , these sections follow a similar structure and we\u2019ll use PHONE_ACCELEROMETER as an example to explain this structure. Hint We recommend reading this page if you are using RAPIDS for the first time All computed sensor features are stored under /data/processed/features on files per sensor, per participant and per study (all participants). Every time you change any sensor parameters, provider parameters or provider features, all the necessary files will be updated as soon as you execute RAPIDS. Config section example for PHONE_ACCELEROMETER # 1) Config section PHONE_ACCELEROMETER : # 2) Parameters for PHONE_ACCELEROMETER TABLE : accelerometer # 3) Providers for PHONE_ACCELEROMETER PROVIDERS : # 4) RAPIDS provider RAPIDS : # 4.1) Parameters of RAPIDS provider of PHONE_ACCELEROMETER COMPUTE : False # 4.2) Features of RAPIDS provider of PHONE_ACCELEROMETER FEATURES : [ \"maxmagnitude\" , \"minmagnitude\" , \"avgmagnitude\" , \"medianmagnitude\" , \"stdmagnitude\" ] SRC_FOLDER : \"rapids\" # inside src/features/phone_accelerometer SRC_LANGUAGE : \"python\" # 5) PANDA provider PANDA : # 5.1) Parameters of PANDA provider of PHONE_ACCELEROMETER COMPUTE : False VALID_SENSED_MINUTES : False # 5.2) Features of PANDA provider of PHONE_ACCELEROMETER FEATURES : exertional_activity_episode : [ \"sumduration\" , \"maxduration\" , \"minduration\" , \"avgduration\" , \"medianduration\" , \"stdduration\" ] nonexertional_activity_episode : [ \"sumduration\" , \"maxduration\" , \"minduration\" , \"avgduration\" , \"medianduration\" , \"stdduration\" ] SRC_FOLDER : \"panda\" # inside src/features/phone_accelerometer SRC_LANGUAGE : \"python\"","title":"Behavioral Features Introduction"},{"location":"features/feature-introduction/#sensor-parameters","text":"Each sensor configuration section has a \u201cparameters\u201d subsection (see #2 in the example). These are parameters that affect different aspects of how the raw data is downloaded, and processed. The TABLE parameter exists for every sensor, but some sensors will have extra parameters like [PHONE_LOCATIONS] . We explain these parameters in a table at the top of each sensor documentation page.","title":"Sensor Parameters"},{"location":"features/feature-introduction/#sensor-providers","text":"Each sensor configuration section can have zero, one or more behavioral feature providers (see #3 in the example). A provider is a script created by the core RAPIDS team or other researchers that extracts behavioral features for that sensor. In this example, accelerometer has two providers: RAPIDS (see #4 ) and PANDA (see #5 ).","title":"Sensor Providers"},{"location":"features/feature-introduction/#provider-parameters","text":"Each provider has parameters that affect the computation of the behavioral features it offers (see #4.1 or #5.1 in the example). These parameters will include at least a [COMPUTE] flag that you switch to True to extract a provider\u2019s behavioral features. We explain every provider\u2019s parameter in a table under the Parameters description heading on each provider documentation page.","title":"Provider Parameters"},{"location":"features/feature-introduction/#provider-features","text":"Each provider offers a set of behavioral features (see #4.2 or #5.2 in the example). For some providers these features are grouped in an array (like those for RAPIDS provider in #4.2 ) but for others they are grouped in a collection of arrays depending on the meaning and purpose of those features (like those for PANDAS provider in #5.2 ). In either case, you can delete the features you are not interested in and they will not be included in the sensor\u2019s output feature file. We explain each behavioral feature in a table under the Features description heading on each provider documentation page.","title":"Provider Features"},{"location":"features/fitbit-heartrate-intraday/","text":"Fitbit Heart Rate Intraday \u00b6 Sensor parameters description for [FITBIT_HEARTRATE_INTRADAY] : Key Description [TABLE] Database table name or file path where the heart rate intraday data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-07\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:1200.6102,\u201dmax\u201d:88,\u201dmin\u201d:31,\u201dminutes\u201d:1058,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:760.3020,\u201dmax\u201d:120,\u201dmin\u201d:86,\u201dminutes\u201d:366,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:15.2048,\u201dmax\u201d:146,\u201dmin\u201d:120,\u201dminutes\u201d:2,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:221,\u201dmin\u201d:148,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:72}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:68},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:67},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:67},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-08\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:1100.1120,\u201dmax\u201d:89,\u201dmin\u201d:30,\u201dminutes\u201d:921,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:660.0012,\u201dmax\u201d:118,\u201dmin\u201d:82,\u201dminutes\u201d:361,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:23.7088,\u201dmax\u201d:142,\u201dmin\u201d:108,\u201dminutes\u201d:3,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:221,\u201dmin\u201d:148,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:70}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:77},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:75},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:73},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-09\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:750.3615,\u201dmax\u201d:77,\u201dmin\u201d:30,\u201dminutes\u201d:851,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:734.1516,\u201dmax\u201d:107,\u201dmin\u201d:77,\u201dminutes\u201d:550,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:131.8579,\u201dmax\u201d:130,\u201dmin\u201d:107,\u201dminutes\u201d:29,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:220,\u201dmin\u201d:130,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:69}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:90},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:89},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:88},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} PLAIN_TEXT device_id local_date_time heartrate heartrate_zone a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:00:00 68 outofrange a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:01:00 67 outofrange a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:02:00 67 outofrange RAPIDS provider \u00b6 Available time segments Available for all time segments File Sequence - data/raw/ { pid } /fitbit_heartrate_intraday_raw.csv - data/raw/ { pid } /fitbit_heartrate_intraday_parsed.csv - data/raw/ { pid } /fitbit_heartrate_intraday_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_heartrate_intraday_features/fitbit_heartrate_intraday_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_heartrate_intraday.csv Parameters description for [FITBIT_HEARTRATE_INTRADAY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_HEARTRATE_INTRADAY features from the RAPIDS provider [FEATURES] Features to be computed from heart rate intraday data, see table below Features description for [FITBIT_HEARTRATE_INTRADAY][PROVIDERS][RAPIDS] : Feature Units Description maxhr beats/mins The maximum heart rate during a time segment. minhr beats/mins The minimum heart rate during a time segment. avghr beats/mins The average heart rate during a time segment. medianhr beats/mins The median of heart rate during a time segment. modehr beats/mins The mode of heart rate during a time segment. stdhr beats/mins The standard deviation of heart rate during a time segment. diffmaxmodehr beats/mins The difference between the maximum and mode heart rate during a time segment. diffminmodehr beats/mins The difference between the mode and minimum heart rate during a time segment. entropyhr nats Shannon\u2019s entropy measurement based on heart rate during a time segment. minutesonZONE minutes Number of minutes the user\u2019s heart rate fell within each heartrate_zone during a time segment. Assumptions/Observations There are four heart rate zones (ZONE): outofrange , fatburn , cardio , and peak . Please refer to Fitbit documentation for more information about the way they are computed.","title":"Fitbit Heart Rate Intraday"},{"location":"features/fitbit-heartrate-intraday/#fitbit-heart-rate-intraday","text":"Sensor parameters description for [FITBIT_HEARTRATE_INTRADAY] : Key Description [TABLE] Database table name or file path where the heart rate intraday data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-07\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:1200.6102,\u201dmax\u201d:88,\u201dmin\u201d:31,\u201dminutes\u201d:1058,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:760.3020,\u201dmax\u201d:120,\u201dmin\u201d:86,\u201dminutes\u201d:366,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:15.2048,\u201dmax\u201d:146,\u201dmin\u201d:120,\u201dminutes\u201d:2,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:221,\u201dmin\u201d:148,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:72}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:68},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:67},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:67},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-08\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:1100.1120,\u201dmax\u201d:89,\u201dmin\u201d:30,\u201dminutes\u201d:921,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:660.0012,\u201dmax\u201d:118,\u201dmin\u201d:82,\u201dminutes\u201d:361,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:23.7088,\u201dmax\u201d:142,\u201dmin\u201d:108,\u201dminutes\u201d:3,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:221,\u201dmin\u201d:148,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:70}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:77},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:75},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:73},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-09\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:750.3615,\u201dmax\u201d:77,\u201dmin\u201d:30,\u201dminutes\u201d:851,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:734.1516,\u201dmax\u201d:107,\u201dmin\u201d:77,\u201dminutes\u201d:550,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:131.8579,\u201dmax\u201d:130,\u201dmin\u201d:107,\u201dminutes\u201d:29,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:220,\u201dmin\u201d:130,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:69}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:90},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:89},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:88},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} PLAIN_TEXT device_id local_date_time heartrate heartrate_zone a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:00:00 68 outofrange a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:01:00 67 outofrange a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:02:00 67 outofrange","title":"Fitbit Heart Rate Intraday"},{"location":"features/fitbit-heartrate-intraday/#rapids-provider","text":"Available time segments Available for all time segments File Sequence - data/raw/ { pid } /fitbit_heartrate_intraday_raw.csv - data/raw/ { pid } /fitbit_heartrate_intraday_parsed.csv - data/raw/ { pid } /fitbit_heartrate_intraday_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_heartrate_intraday_features/fitbit_heartrate_intraday_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_heartrate_intraday.csv Parameters description for [FITBIT_HEARTRATE_INTRADAY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_HEARTRATE_INTRADAY features from the RAPIDS provider [FEATURES] Features to be computed from heart rate intraday data, see table below Features description for [FITBIT_HEARTRATE_INTRADAY][PROVIDERS][RAPIDS] : Feature Units Description maxhr beats/mins The maximum heart rate during a time segment. minhr beats/mins The minimum heart rate during a time segment. avghr beats/mins The average heart rate during a time segment. medianhr beats/mins The median of heart rate during a time segment. modehr beats/mins The mode of heart rate during a time segment. stdhr beats/mins The standard deviation of heart rate during a time segment. diffmaxmodehr beats/mins The difference between the maximum and mode heart rate during a time segment. diffminmodehr beats/mins The difference between the mode and minimum heart rate during a time segment. entropyhr nats Shannon\u2019s entropy measurement based on heart rate during a time segment. minutesonZONE minutes Number of minutes the user\u2019s heart rate fell within each heartrate_zone during a time segment. Assumptions/Observations There are four heart rate zones (ZONE): outofrange , fatburn , cardio , and peak . Please refer to Fitbit documentation for more information about the way they are computed.","title":"RAPIDS provider"},{"location":"features/fitbit-heartrate-summary/","text":"Fitbit Heart Rate Summary \u00b6 Sensor parameters description for [FITBIT_HEARTRATE_SUMMARY] : Key Description [TABLE] Database table name or file path where the heart rate summary data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-07\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:1200.6102,\u201dmax\u201d:88,\u201dmin\u201d:31,\u201dminutes\u201d:1058,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:760.3020,\u201dmax\u201d:120,\u201dmin\u201d:86,\u201dminutes\u201d:366,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:15.2048,\u201dmax\u201d:146,\u201dmin\u201d:120,\u201dminutes\u201d:2,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:221,\u201dmin\u201d:148,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:72}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:68},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:67},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:67},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-08\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:1100.1120,\u201dmax\u201d:89,\u201dmin\u201d:30,\u201dminutes\u201d:921,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:660.0012,\u201dmax\u201d:118,\u201dmin\u201d:82,\u201dminutes\u201d:361,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:23.7088,\u201dmax\u201d:142,\u201dmin\u201d:108,\u201dminutes\u201d:3,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:221,\u201dmin\u201d:148,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:70}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:77},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:75},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:73},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-09\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:750.3615,\u201dmax\u201d:77,\u201dmin\u201d:30,\u201dminutes\u201d:851,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:734.1516,\u201dmax\u201d:107,\u201dmin\u201d:77,\u201dminutes\u201d:550,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:131.8579,\u201dmax\u201d:130,\u201dmin\u201d:107,\u201dminutes\u201d:29,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:220,\u201dmin\u201d:130,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:69}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:90},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:89},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:88},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} PLAIN_TEXT device_id local_date_time heartrate_daily_restinghr heartrate_daily_caloriesoutofrange heartrate_daily_caloriesfatburn heartrate_daily_caloriescardio heartrate_daily_caloriespeak a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 72 1200.6102 760.3020 15.2048 0 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-08 70 1100.1120 660.0012 23.7088 0 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-09 69 750.3615 734.1516 131.8579 0 RAPIDS provider \u00b6 Available time segments Only available for segments that span 1 or more complete days (e.g. Jan 1 st 00:00 to Jan 3 rd 23:59) File Sequence - data/raw/ { pid } /fitbit_heartrate_summary_raw.csv - data/raw/ { pid } /fitbit_heartrate_summary_parsed.csv - data/raw/ { pid } /fitbit_heartrate_summary_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_heartrate_summary_features/fitbit_heartrate_summary_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_heartrate_summary.csv Parameters description for [FITBIT_HEARTRATE_SUMMARY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_HEARTRATE_SUMMARY features from the RAPIDS provider [FEATURES] Features to be computed from heart rate summary data, see table below Features description for [FITBIT_HEARTRATE_SUMMARY][PROVIDERS][RAPIDS] : Feature Units Description maxrestinghr beats/mins The maximum daily resting heart rate during a time segment. minrestinghr beats/mins The minimum daily resting heart rate during a time segment. avgrestinghr beats/mins The average daily resting heart rate during a time segment. medianrestinghr beats/mins The median of daily resting heart rate during a time segment. moderestinghr beats/mins The mode of daily resting heart rate during a time segment. stdrestinghr beats/mins The standard deviation of daily resting heart rate during a time segment. diffmaxmoderestinghr beats/mins The difference between the maximum and mode daily resting heart rate during a time segment. diffminmoderestinghr beats/mins The difference between the mode and minimum daily resting heart rate during a time segment. entropyrestinghr nats Shannon\u2019s entropy measurement based on daily resting heart rate during a time segment. sumcaloriesZONE cals The total daily calories burned within heartrate_zone during a time segment. maxcaloriesZONE cals The maximum daily calories burned within heartrate_zone during a time segment. mincaloriesZONE cals The minimum daily calories burned within heartrate_zone during a time segment. avgcaloriesZONE cals The average daily calories burned within heartrate_zone during a time segment. mediancaloriesZONE cals The median of daily calories burned within heartrate_zone during a time segment. stdcaloriesZONE cals The standard deviation of daily calories burned within heartrate_zone during a time segment. entropycaloriesZONE nats Shannon\u2019s entropy measurement based on daily calories burned within heartrate_zone during a time segment. Assumptions/Observations There are four heart rate zones (ZONE): outofrange , fatburn , cardio , and peak . Please refer to Fitbit documentation for more information about the way they are computed. Calories\u2019 accuracy depends on the users\u2019 Fitbit profile (weight, height, etc.).","title":"Fitbit Heart Rate Summary"},{"location":"features/fitbit-heartrate-summary/#fitbit-heart-rate-summary","text":"Sensor parameters description for [FITBIT_HEARTRATE_SUMMARY] : Key Description [TABLE] Database table name or file path where the heart rate summary data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-07\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:1200.6102,\u201dmax\u201d:88,\u201dmin\u201d:31,\u201dminutes\u201d:1058,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:760.3020,\u201dmax\u201d:120,\u201dmin\u201d:86,\u201dminutes\u201d:366,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:15.2048,\u201dmax\u201d:146,\u201dmin\u201d:120,\u201dminutes\u201d:2,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:221,\u201dmin\u201d:148,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:72}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:68},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:67},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:67},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-08\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:1100.1120,\u201dmax\u201d:89,\u201dmin\u201d:30,\u201dminutes\u201d:921,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:660.0012,\u201dmax\u201d:118,\u201dmin\u201d:82,\u201dminutes\u201d:361,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:23.7088,\u201dmax\u201d:142,\u201dmin\u201d:108,\u201dminutes\u201d:3,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:221,\u201dmin\u201d:148,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:70}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:77},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:75},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:73},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201cactivities-heart\u201d:[{\u201cdateTime\u201d:\u201d2020-10-09\u201d,\u201dvalue\u201d:{\u201ccustomHeartRateZones\u201d:[],\u201dheartRateZones\u201d:[{\u201ccaloriesOut\u201d:750.3615,\u201dmax\u201d:77,\u201dmin\u201d:30,\u201dminutes\u201d:851,\u201dname\u201d:\u201dOut of Range\u201d},{\u201ccaloriesOut\u201d:734.1516,\u201dmax\u201d:107,\u201dmin\u201d:77,\u201dminutes\u201d:550,\u201dname\u201d:\u201dFat Burn\u201d},{\u201ccaloriesOut\u201d:131.8579,\u201dmax\u201d:130,\u201dmin\u201d:107,\u201dminutes\u201d:29,\u201dname\u201d:\u201dCardio\u201d},{\u201ccaloriesOut\u201d:0,\u201dmax\u201d:220,\u201dmin\u201d:130,\u201dminutes\u201d:0,\u201dname\u201d:\u201dPeak\u201d}],\u201drestingHeartRate\u201d:69}}],\u201dactivities-heart-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:90},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:89},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:88},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} PLAIN_TEXT device_id local_date_time heartrate_daily_restinghr heartrate_daily_caloriesoutofrange heartrate_daily_caloriesfatburn heartrate_daily_caloriescardio heartrate_daily_caloriespeak a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 72 1200.6102 760.3020 15.2048 0 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-08 70 1100.1120 660.0012 23.7088 0 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-09 69 750.3615 734.1516 131.8579 0","title":"Fitbit Heart Rate Summary"},{"location":"features/fitbit-heartrate-summary/#rapids-provider","text":"Available time segments Only available for segments that span 1 or more complete days (e.g. Jan 1 st 00:00 to Jan 3 rd 23:59) File Sequence - data/raw/ { pid } /fitbit_heartrate_summary_raw.csv - data/raw/ { pid } /fitbit_heartrate_summary_parsed.csv - data/raw/ { pid } /fitbit_heartrate_summary_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_heartrate_summary_features/fitbit_heartrate_summary_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_heartrate_summary.csv Parameters description for [FITBIT_HEARTRATE_SUMMARY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_HEARTRATE_SUMMARY features from the RAPIDS provider [FEATURES] Features to be computed from heart rate summary data, see table below Features description for [FITBIT_HEARTRATE_SUMMARY][PROVIDERS][RAPIDS] : Feature Units Description maxrestinghr beats/mins The maximum daily resting heart rate during a time segment. minrestinghr beats/mins The minimum daily resting heart rate during a time segment. avgrestinghr beats/mins The average daily resting heart rate during a time segment. medianrestinghr beats/mins The median of daily resting heart rate during a time segment. moderestinghr beats/mins The mode of daily resting heart rate during a time segment. stdrestinghr beats/mins The standard deviation of daily resting heart rate during a time segment. diffmaxmoderestinghr beats/mins The difference between the maximum and mode daily resting heart rate during a time segment. diffminmoderestinghr beats/mins The difference between the mode and minimum daily resting heart rate during a time segment. entropyrestinghr nats Shannon\u2019s entropy measurement based on daily resting heart rate during a time segment. sumcaloriesZONE cals The total daily calories burned within heartrate_zone during a time segment. maxcaloriesZONE cals The maximum daily calories burned within heartrate_zone during a time segment. mincaloriesZONE cals The minimum daily calories burned within heartrate_zone during a time segment. avgcaloriesZONE cals The average daily calories burned within heartrate_zone during a time segment. mediancaloriesZONE cals The median of daily calories burned within heartrate_zone during a time segment. stdcaloriesZONE cals The standard deviation of daily calories burned within heartrate_zone during a time segment. entropycaloriesZONE nats Shannon\u2019s entropy measurement based on daily calories burned within heartrate_zone during a time segment. Assumptions/Observations There are four heart rate zones (ZONE): outofrange , fatburn , cardio , and peak . Please refer to Fitbit documentation for more information about the way they are computed. Calories\u2019 accuracy depends on the users\u2019 Fitbit profile (weight, height, etc.).","title":"RAPIDS provider"},{"location":"features/fitbit-sleep-summary/","text":"Fitbit Sleep Summary \u00b6 Sensor parameters description for [FITBIT_SLEEP_SUMMARY] : Key Description [TABLE] Database table name or file path where the sleep summary data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data with Fitbit\u2019s sleep API Version 1 JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d: [{\u201cawakeCount\u201d: 2, \u201cawakeDuration\u201d: 3, \u201cawakeningsCount\u201d: 10, \u201cdateOfSleep\u201d: \u201c2020-10-07\u201d, \u201cduration\u201d: 8100000, \u201cefficiency\u201d: 91, \u201cendTime\u201d: \u201c2020-10-07T18:10:00.000\u201d, \u201cisMainSleep\u201d: true, \u201clogId\u201d: 14147921940, \u201cminuteData\u201d: [{\u201cdateTime\u201d: \u201c15:55:00\u201d, \u201cvalue\u201d: \u201c3\u201d}, {\u201cdateTime\u201d: \u201c15:56:00\u201d, \u201cvalue\u201d: \u201c3\u201d}, {\u201cdateTime\u201d: \u201c15:57:00\u201d, \u201cvalue\u201d: \u201c2\u201d},\u2026], \u201cminutesAfterWakeup\u201d: 0, \u201cminutesAsleep\u201d: 123, \u201cminutesAwake\u201d: 12, \u201cminutesToFallAsleep\u201d: 0, \u201crestlessCount\u201d: 8, \u201crestlessDuration\u201d: 9, \u201cstartTime\u201d: \u201c2020-10-07T15:55:00.000\u201d, \u201ctimeInBed\u201d: 135}, {\u201cawakeCount\u201d: 0, \u201cawakeDuration\u201d: 0, \u201cawakeningsCount\u201d: 1, \u201cdateOfSleep\u201d: \u201c2020-10-07\u201d, \u201cduration\u201d: 3780000, \u201cefficiency\u201d: 100, \u201cendTime\u201d: \u201c2020-10-07T10:52:30.000\u201d, \u201cisMainSleep\u201d: false, \u201clogId\u201d: 14144903977, \u201cminuteData\u201d: [{\u201cdateTime\u201d: \u201c09:49:00\u201d, \u201cvalue\u201d: \u201c1\u201d}, {\u201cdateTime\u201d: \u201c09:50:00\u201d, \u201cvalue\u201d: \u201c1\u201d}, {\u201cdateTime\u201d: \u201c09:51:00\u201d, \u201cvalue\u201d: \u201c1\u201d},\u2026], \u201cminutesAfterWakeup\u201d: 1, \u201cminutesAsleep\u201d: 62, \u201cminutesAwake\u201d: 0, \u201cminutesToFallAsleep\u201d: 0, \u201crestlessCount\u201d: 1, \u201crestlessDuration\u201d: 1, \u201cstartTime\u201d: \u201c2020-10-07T09:49:00.000\u201d, \u201ctimeInBed\u201d: 63}], \u201csummary\u201d: {\u201ctotalMinutesAsleep\u201d: 185, \u201ctotalSleepRecords\u201d: 2, \u201ctotalTimeInBed\u201d: 198}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d: [{\u201cawakeCount\u201d: 3, \u201cawakeDuration\u201d: 21, \u201cawakeningsCount\u201d: 16, \u201cdateOfSleep\u201d: \u201c2020-10-08\u201d, \u201cduration\u201d: 19260000, \u201cefficiency\u201d: 89, \u201cendTime\u201d: \u201c2020-10-08T06:01:30.000\u201d, \u201cisMainSleep\u201d: true, \u201clogId\u201d: 14150613895, \u201cminuteData\u201d: [{\u201cdateTime\u201d: \u201c00:40:00\u201d, \u201cvalue\u201d: \u201c3\u201d}, {\u201cdateTime\u201d: \u201c00:41:00\u201d, \u201cvalue\u201d: \u201c3\u201d}, {\u201cdateTime\u201d: \u201c00:42:00\u201d, \u201cvalue\u201d: \u201c3\u201d},\u2026], \u201cminutesAfterWakeup\u201d: 0, \u201cminutesAsleep\u201d: 275, \u201cminutesAwake\u201d: 33, \u201cminutesToFallAsleep\u201d: 0, \u201crestlessCount\u201d: 13, \u201crestlessDuration\u201d: 25, \u201cstartTime\u201d: \u201c2020-10-08T00:40:00.000\u201d, \u201ctimeInBed\u201d: 321}], \u201csummary\u201d: {\u201ctotalMinutesAsleep\u201d: 275, \u201ctotalSleepRecords\u201d: 1, \u201ctotalTimeInBed\u201d: 321}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d: [{\u201cawakeCount\u201d: 1, \u201cawakeDuration\u201d: 3, \u201cawakeningsCount\u201d: 8, \u201cdateOfSleep\u201d: \u201c2020-10-09\u201d, \u201cduration\u201d: 19320000, \u201cefficiency\u201d: 96, \u201cendTime\u201d: \u201c2020-10-09T05:57:30.000\u201d, \u201cisMainSleep\u201d: true, \u201clogId\u201d: 14161136803, \u201cminuteData\u201d: [{\u201cdateTime\u201d: \u201c00:35:30\u201d, \u201cvalue\u201d: \u201c2\u201d}, {\u201cdateTime\u201d: \u201c00:36:30\u201d, \u201cvalue\u201d: \u201c1\u201d}, {\u201cdateTime\u201d: \u201c00:37:30\u201d, \u201cvalue\u201d: \u201c1\u201d},\u2026], \u201cminutesAfterWakeup\u201d: 0, \u201cminutesAsleep\u201d: 309, \u201cminutesAwake\u201d: 13, \u201cminutesToFallAsleep\u201d: 0, \u201crestlessCount\u201d: 7, \u201crestlessDuration\u201d: 10, \u201cstartTime\u201d: \u201c2020-10-09T00:35:30.000\u201d, \u201ctimeInBed\u201d: 322}], \u201csummary\u201d: {\u201ctotalMinutesAsleep\u201d: 309, \u201ctotalSleepRecords\u201d: 1, \u201ctotalTimeInBed\u201d: 322}} PLAIN_TEXT device_id local_start_date_time local_end_date_time efficiency minutes_after_wakeup minutes_asleep minutes_awake minutes_to_fall_asleep minutes_in_bed is_main_sleep type count_awake duration_awake count_awakenings count_restless duration_restless a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 15:55:00 2020-10-07 18:10:00 91 0 123 12 0 135 1 classic 2 3 10 8 9 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 09:49:00 2020-10-07 10:52:30 100 1 62 0 0 63 0 classic 0 0 1 1 1 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-08 00:40:00 2020-10-08 06:01:30 89 0 275 33 0 321 1 classic 3 21 16 13 25 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-09 00:35:30 2020-10-09 05:57:30 96 0 309 13 0 322 1 classic 1 3 8 7 10 Example of the structure of source data with Fitbit\u2019s sleep API Version 1.2 JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d:[{\u201cdateOfSleep\u201d:\u201d2020-10-10\u201d,\u201dduration\u201d:3600000,\u201defficiency\u201d:92,\u201dendTime\u201d:\u201d2020-10-10T16:37:00.000\u201d,\u201dinfoCode\u201d:2,\u201disMainSleep\u201d:false,\u201dlevels\u201d:{\u201cdata\u201d:[{\u201cdateTime\u201d:\u201d2020-10-10T15:36:30.000\u201d,\u201dlevel\u201d:\u201drestless\u201d,\u201dseconds\u201d:60},{\u201cdateTime\u201d:\u201d2020-10-10T15:37:30.000\u201d,\u201dlevel\u201d:\u201dasleep\u201d,\u201dseconds\u201d:660},{\u201cdateTime\u201d:\u201d2020-10-10T15:48:30.000\u201d,\u201dlevel\u201d:\u201drestless\u201d,\u201dseconds\u201d:60},\u2026], \u201csummary\u201d:{\u201casleep\u201d:{\u201ccount\u201d:0,\u201dminutes\u201d:56},\u201dawake\u201d:{\u201ccount\u201d:0,\u201dminutes\u201d:0},\u201drestless\u201d:{\u201ccount\u201d:3,\u201dminutes\u201d:4}}},\u201dlogId\u201d:26315914306,\u201dminutesAfterWakeup\u201d:0,\u201dminutesAsleep\u201d:55,\u201dminutesAwake\u201d:5,\u201dminutesToFallAsleep\u201d:0,\u201dstartTime\u201d:\u201d2020-10-10T15:36:30.000\u201d,\u201dtimeInBed\u201d:60,\u201dtype\u201d:\u201dclassic\u201d},{\u201cdateOfSleep\u201d:\u201d2020-10-10\u201d,\u201dduration\u201d:22980000,\u201defficiency\u201d:88,\u201dendTime\u201d:\u201d2020-10-10T08:10:00.000\u201d,\u201dinfoCode\u201d:0,\u201disMainSleep\u201d:true,\u201dlevels\u201d:{\u201cdata\u201d:[{\u201cdateTime\u201d:\u201d2020-10-10T01:46:30.000\u201d,\u201dlevel\u201d:\u201dlight\u201d,\u201dseconds\u201d:420},{\u201cdateTime\u201d:\u201d2020-10-10T01:53:30.000\u201d,\u201dlevel\u201d:\u201ddeep\u201d,\u201dseconds\u201d:1230},{\u201cdateTime\u201d:\u201d2020-10-10T02:14:00.000\u201d,\u201dlevel\u201d:\u201dlight\u201d,\u201dseconds\u201d:360},\u2026], \u201csummary\u201d:{\u201cdeep\u201d:{\u201ccount\u201d:3,\u201dminutes\u201d:92,\u201dthirtyDayAvgMinutes\u201d:0},\u201dlight\u201d:{\u201ccount\u201d:29,\u201dminutes\u201d:193,\u201dthirtyDayAvgMinutes\u201d:0},\u201drem\u201d:{\u201ccount\u201d:4,\u201dminutes\u201d:33,\u201dthirtyDayAvgMinutes\u201d:0},\u201dwake\u201d:{\u201ccount\u201d:28,\u201dminutes\u201d:65,\u201dthirtyDayAvgMinutes\u201d:0}}},\u201dlogId\u201d:26311786557,\u201dminutesAfterWakeup\u201d:0,\u201dminutesAsleep\u201d:318,\u201dminutesAwake\u201d:65,\u201dminutesToFallAsleep\u201d:0,\u201dstartTime\u201d:\u201d2020-10-10T01:46:30.000\u201d,\u201dtimeInBed\u201d:383,\u201dtype\u201d:\u201dstages\u201d}],\u201dsummary\u201d:{\u201cstages\u201d:{\u201cdeep\u201d:92,\u201dlight\u201d:193,\u201drem\u201d:33,\u201dwake\u201d:65},\u201dtotalMinutesAsleep\u201d:373,\u201dtotalSleepRecords\u201d:2,\u201dtotalTimeInBed\u201d:443}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d:[{\u201cdateOfSleep\u201d:\u201d2020-10-11\u201d,\u201dduration\u201d:41640000,\u201defficiency\u201d:89,\u201dendTime\u201d:\u201d2020-10-11T11:47:00.000\u201d,\u201dinfoCode\u201d:0,\u201disMainSleep\u201d:true,\u201dlevels\u201d:{\u201cdata\u201d:[{\u201cdateTime\u201d:\u201d2020-10-11T00:12:30.000\u201d,\u201dlevel\u201d:\u201dwake\u201d,\u201dseconds\u201d:450},{\u201cdateTime\u201d:\u201d2020-10-11T00:20:00.000\u201d,\u201dlevel\u201d:\u201dlight\u201d,\u201dseconds\u201d:870},{\u201cdateTime\u201d:\u201d2020-10-11T00:34:30.000\u201d,\u201dlevel\u201d:\u201dwake\u201d,\u201dseconds\u201d:780},\u2026], \u201csummary\u201d:{\u201cdeep\u201d:{\u201ccount\u201d:4,\u201dminutes\u201d:52,\u201dthirtyDayAvgMinutes\u201d:62},\u201dlight\u201d:{\u201ccount\u201d:32,\u201dminutes\u201d:442,\u201dthirtyDayAvgMinutes\u201d:364},\u201drem\u201d:{\u201ccount\u201d:6,\u201dminutes\u201d:68,\u201dthirtyDayAvgMinutes\u201d:58},\u201dwake\u201d:{\u201ccount\u201d:29,\u201dminutes\u201d:132,\u201dthirtyDayAvgMinutes\u201d:94}}},\u201dlogId\u201d:26589710670,\u201dminutesAfterWakeup\u201d:1,\u201dminutesAsleep\u201d:562,\u201dminutesAwake\u201d:132,\u201dminutesToFallAsleep\u201d:0,\u201dstartTime\u201d:\u201d2020-10-11T00:12:30.000\u201d,\u201dtimeInBed\u201d:694,\u201dtype\u201d:\u201dstages\u201d}],\u201dsummary\u201d:{\u201cstages\u201d:{\u201cdeep\u201d:52,\u201dlight\u201d:442,\u201drem\u201d:68,\u201dwake\u201d:132},\u201dtotalMinutesAsleep\u201d:562,\u201dtotalSleepRecords\u201d:1,\u201dtotalTimeInBed\u201d:694}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d:[{\u201cdateOfSleep\u201d:\u201d2020-10-12\u201d,\u201dduration\u201d:28980000,\u201defficiency\u201d:93,\u201dendTime\u201d:\u201d2020-10-12T09:34:30.000\u201d,\u201dinfoCode\u201d:0,\u201disMainSleep\u201d:true,\u201dlevels\u201d:{\u201cdata\u201d:[{\u201cdateTime\u201d:\u201d2020-10-12T01:31:00.000\u201d,\u201dlevel\u201d:\u201dwake\u201d,\u201dseconds\u201d:600},{\u201cdateTime\u201d:\u201d2020-10-12T01:41:00.000\u201d,\u201dlevel\u201d:\u201dlight\u201d,\u201dseconds\u201d:60},{\u201cdateTime\u201d:\u201d2020-10-12T01:42:00.000\u201d,\u201dlevel\u201d:\u201ddeep\u201d,\u201dseconds\u201d:2340},\u2026], \u201csummary\u201d:{\u201cdeep\u201d:{\u201ccount\u201d:4,\u201dminutes\u201d:63,\u201dthirtyDayAvgMinutes\u201d:59},\u201dlight\u201d:{\u201ccount\u201d:27,\u201dminutes\u201d:257,\u201dthirtyDayAvgMinutes\u201d:364},\u201drem\u201d:{\u201ccount\u201d:5,\u201dminutes\u201d:94,\u201dthirtyDayAvgMinutes\u201d:58},\u201dwake\u201d:{\u201ccount\u201d:24,\u201dminutes\u201d:69,\u201dthirtyDayAvgMinutes\u201d:95}}},\u201dlogId\u201d:26589710673,\u201dminutesAfterWakeup\u201d:0,\u201dminutesAsleep\u201d:415,\u201dminutesAwake\u201d:68,\u201dminutesToFallAsleep\u201d:0,\u201dstartTime\u201d:\u201d2020-10-12T01:31:00.000\u201d,\u201dtimeInBed\u201d:483,\u201dtype\u201d:\u201dstages\u201d}],\u201dsummary\u201d:{\u201cstages\u201d:{\u201cdeep\u201d:63,\u201dlight\u201d:257,\u201drem\u201d:94,\u201dwake\u201d:69},\u201dtotalMinutesAsleep\u201d:415,\u201dtotalSleepRecords\u201d:1,\u201dtotalTimeInBed\u201d:483}} PLAIN_TEXT device_id local_start_date_time local_end_date_time efficiency minutes_after_wakeup minutes_asleep minutes_awake minutes_to_fall_asleep minutes_in_bed is_main_sleep type a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-10 15:36:30 2020-10-10 16:37:00 92 0 55 5 0 60 0 classic a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-10 01:46:30 2020-10-10 08:10:00 88 0 318 65 0 383 1 stages a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-11 00:12:30 2020-10-11 11:47:00 89 1 562 132 0 694 1 stages a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-12 01:31:00 2020-10-12 09:34:30 93 0 415 68 0 483 1 stages RAPIDS provider \u00b6 Available time segments Only available for segments that span 1 or more complete days (e.g. Jan 1 st 00:00 to Jan 3 rd 23:59) File Sequence - data/raw/ { pid } /fitbit_sleep_summary_raw.csv - data/raw/ { pid } /fitbit_sleep_summary_parsed.csv - data/raw/ { pid } /fitbit_sleep_summary_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_sleep_summary_features/fitbit_sleep_summary_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_sleep_summary.csv Parameters description for [FITBIT_SLEEP_SUMMARY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_SLEEP_SUMMARY features from the RAPIDS provider [SLEEP_TYPES] Types of sleep to be included in the feature extraction computation. Fitbit provides 3 types of sleep: main , nap , all . [FEATURES] Features to be computed from sleep summary data, see table below Features description for [FITBIT_SLEEP_SUMMARY][PROVIDERS][RAPIDS] : Feature Units Description countepisodeTYPE episodes Number of sleep episodes for a certain sleep type during a time segment. avgefficiencyTYPE scores Average sleep efficiency for a certain sleep type during a time segment. sumdurationafterwakeupTYPE minutes Total duration the user stayed in bed after waking up for a certain sleep type during a time segment. sumdurationasleepTYPE minutes Total sleep duration for a certain sleep type during a time segment. sumdurationawakeTYPE minutes Total duration the user stayed awake but still in bed for a certain sleep type during a time segment. sumdurationtofallasleepTYPE minutes Total duration the user spent to fall asleep for a certain sleep type during a time segment. sumdurationinbedTYPE minutes Total duration the user stayed in bed (sumdurationtofallasleep + sumdurationawake + sumdurationasleep + sumdurationafterwakeup) for a certain sleep type during a time segment. avgdurationafterwakeupTYPE minutes Average duration the user stayed in bed after waking up for a certain sleep type during a time segment. avgdurationasleepTYPE minutes Average sleep duration for a certain sleep type during a time segment. avgdurationawakeTYPE minutes Average duration the user stayed awake but still in bed for a certain sleep type during a time segment. avgdurationtofallasleepTYPE minutes Average duration the user spent to fall asleep for a certain sleep type during a time segment. avgdurationinbedTYPE minutes Average duration the user stayed in bed (sumdurationtofallasleep + sumdurationawake + sumdurationasleep + sumdurationafterwakeup) for a certain sleep type during a time segment. Assumptions/Observations There are three sleep types (TYPE): main , nap , all . The all type contains both main sleep and naps. There are two versions of Fitbit\u2019s sleep API ( version 1 and version 1.2 ), and each provides raw sleep data in a different format: Count & duration summaries . v1 contains count_awake , duration_awake , count_awakenings , count_restless , and duration_restless fields for every sleep record but v1.2 does not. API columns . Features are computed based on the values provided by Fitbit\u2019s API: efficiency , minutes_after_wakeup , minutes_asleep , minutes_awake , minutes_to_fall_asleep , minutes_in_bed , is_main_sleep and type .","title":"Fitbit Sleep Summary"},{"location":"features/fitbit-sleep-summary/#fitbit-sleep-summary","text":"Sensor parameters description for [FITBIT_SLEEP_SUMMARY] : Key Description [TABLE] Database table name or file path where the sleep summary data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data with Fitbit\u2019s sleep API Version 1 JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d: [{\u201cawakeCount\u201d: 2, \u201cawakeDuration\u201d: 3, \u201cawakeningsCount\u201d: 10, \u201cdateOfSleep\u201d: \u201c2020-10-07\u201d, \u201cduration\u201d: 8100000, \u201cefficiency\u201d: 91, \u201cendTime\u201d: \u201c2020-10-07T18:10:00.000\u201d, \u201cisMainSleep\u201d: true, \u201clogId\u201d: 14147921940, \u201cminuteData\u201d: [{\u201cdateTime\u201d: \u201c15:55:00\u201d, \u201cvalue\u201d: \u201c3\u201d}, {\u201cdateTime\u201d: \u201c15:56:00\u201d, \u201cvalue\u201d: \u201c3\u201d}, {\u201cdateTime\u201d: \u201c15:57:00\u201d, \u201cvalue\u201d: \u201c2\u201d},\u2026], \u201cminutesAfterWakeup\u201d: 0, \u201cminutesAsleep\u201d: 123, \u201cminutesAwake\u201d: 12, \u201cminutesToFallAsleep\u201d: 0, \u201crestlessCount\u201d: 8, \u201crestlessDuration\u201d: 9, \u201cstartTime\u201d: \u201c2020-10-07T15:55:00.000\u201d, \u201ctimeInBed\u201d: 135}, {\u201cawakeCount\u201d: 0, \u201cawakeDuration\u201d: 0, \u201cawakeningsCount\u201d: 1, \u201cdateOfSleep\u201d: \u201c2020-10-07\u201d, \u201cduration\u201d: 3780000, \u201cefficiency\u201d: 100, \u201cendTime\u201d: \u201c2020-10-07T10:52:30.000\u201d, \u201cisMainSleep\u201d: false, \u201clogId\u201d: 14144903977, \u201cminuteData\u201d: [{\u201cdateTime\u201d: \u201c09:49:00\u201d, \u201cvalue\u201d: \u201c1\u201d}, {\u201cdateTime\u201d: \u201c09:50:00\u201d, \u201cvalue\u201d: \u201c1\u201d}, {\u201cdateTime\u201d: \u201c09:51:00\u201d, \u201cvalue\u201d: \u201c1\u201d},\u2026], \u201cminutesAfterWakeup\u201d: 1, \u201cminutesAsleep\u201d: 62, \u201cminutesAwake\u201d: 0, \u201cminutesToFallAsleep\u201d: 0, \u201crestlessCount\u201d: 1, \u201crestlessDuration\u201d: 1, \u201cstartTime\u201d: \u201c2020-10-07T09:49:00.000\u201d, \u201ctimeInBed\u201d: 63}], \u201csummary\u201d: {\u201ctotalMinutesAsleep\u201d: 185, \u201ctotalSleepRecords\u201d: 2, \u201ctotalTimeInBed\u201d: 198}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d: [{\u201cawakeCount\u201d: 3, \u201cawakeDuration\u201d: 21, \u201cawakeningsCount\u201d: 16, \u201cdateOfSleep\u201d: \u201c2020-10-08\u201d, \u201cduration\u201d: 19260000, \u201cefficiency\u201d: 89, \u201cendTime\u201d: \u201c2020-10-08T06:01:30.000\u201d, \u201cisMainSleep\u201d: true, \u201clogId\u201d: 14150613895, \u201cminuteData\u201d: [{\u201cdateTime\u201d: \u201c00:40:00\u201d, \u201cvalue\u201d: \u201c3\u201d}, {\u201cdateTime\u201d: \u201c00:41:00\u201d, \u201cvalue\u201d: \u201c3\u201d}, {\u201cdateTime\u201d: \u201c00:42:00\u201d, \u201cvalue\u201d: \u201c3\u201d},\u2026], \u201cminutesAfterWakeup\u201d: 0, \u201cminutesAsleep\u201d: 275, \u201cminutesAwake\u201d: 33, \u201cminutesToFallAsleep\u201d: 0, \u201crestlessCount\u201d: 13, \u201crestlessDuration\u201d: 25, \u201cstartTime\u201d: \u201c2020-10-08T00:40:00.000\u201d, \u201ctimeInBed\u201d: 321}], \u201csummary\u201d: {\u201ctotalMinutesAsleep\u201d: 275, \u201ctotalSleepRecords\u201d: 1, \u201ctotalTimeInBed\u201d: 321}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d: [{\u201cawakeCount\u201d: 1, \u201cawakeDuration\u201d: 3, \u201cawakeningsCount\u201d: 8, \u201cdateOfSleep\u201d: \u201c2020-10-09\u201d, \u201cduration\u201d: 19320000, \u201cefficiency\u201d: 96, \u201cendTime\u201d: \u201c2020-10-09T05:57:30.000\u201d, \u201cisMainSleep\u201d: true, \u201clogId\u201d: 14161136803, \u201cminuteData\u201d: [{\u201cdateTime\u201d: \u201c00:35:30\u201d, \u201cvalue\u201d: \u201c2\u201d}, {\u201cdateTime\u201d: \u201c00:36:30\u201d, \u201cvalue\u201d: \u201c1\u201d}, {\u201cdateTime\u201d: \u201c00:37:30\u201d, \u201cvalue\u201d: \u201c1\u201d},\u2026], \u201cminutesAfterWakeup\u201d: 0, \u201cminutesAsleep\u201d: 309, \u201cminutesAwake\u201d: 13, \u201cminutesToFallAsleep\u201d: 0, \u201crestlessCount\u201d: 7, \u201crestlessDuration\u201d: 10, \u201cstartTime\u201d: \u201c2020-10-09T00:35:30.000\u201d, \u201ctimeInBed\u201d: 322}], \u201csummary\u201d: {\u201ctotalMinutesAsleep\u201d: 309, \u201ctotalSleepRecords\u201d: 1, \u201ctotalTimeInBed\u201d: 322}} PLAIN_TEXT device_id local_start_date_time local_end_date_time efficiency minutes_after_wakeup minutes_asleep minutes_awake minutes_to_fall_asleep minutes_in_bed is_main_sleep type count_awake duration_awake count_awakenings count_restless duration_restless a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 15:55:00 2020-10-07 18:10:00 91 0 123 12 0 135 1 classic 2 3 10 8 9 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 09:49:00 2020-10-07 10:52:30 100 1 62 0 0 63 0 classic 0 0 1 1 1 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-08 00:40:00 2020-10-08 06:01:30 89 0 275 33 0 321 1 classic 3 21 16 13 25 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-09 00:35:30 2020-10-09 05:57:30 96 0 309 13 0 322 1 classic 1 3 8 7 10 Example of the structure of source data with Fitbit\u2019s sleep API Version 1.2 JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d:[{\u201cdateOfSleep\u201d:\u201d2020-10-10\u201d,\u201dduration\u201d:3600000,\u201defficiency\u201d:92,\u201dendTime\u201d:\u201d2020-10-10T16:37:00.000\u201d,\u201dinfoCode\u201d:2,\u201disMainSleep\u201d:false,\u201dlevels\u201d:{\u201cdata\u201d:[{\u201cdateTime\u201d:\u201d2020-10-10T15:36:30.000\u201d,\u201dlevel\u201d:\u201drestless\u201d,\u201dseconds\u201d:60},{\u201cdateTime\u201d:\u201d2020-10-10T15:37:30.000\u201d,\u201dlevel\u201d:\u201dasleep\u201d,\u201dseconds\u201d:660},{\u201cdateTime\u201d:\u201d2020-10-10T15:48:30.000\u201d,\u201dlevel\u201d:\u201drestless\u201d,\u201dseconds\u201d:60},\u2026], \u201csummary\u201d:{\u201casleep\u201d:{\u201ccount\u201d:0,\u201dminutes\u201d:56},\u201dawake\u201d:{\u201ccount\u201d:0,\u201dminutes\u201d:0},\u201drestless\u201d:{\u201ccount\u201d:3,\u201dminutes\u201d:4}}},\u201dlogId\u201d:26315914306,\u201dminutesAfterWakeup\u201d:0,\u201dminutesAsleep\u201d:55,\u201dminutesAwake\u201d:5,\u201dminutesToFallAsleep\u201d:0,\u201dstartTime\u201d:\u201d2020-10-10T15:36:30.000\u201d,\u201dtimeInBed\u201d:60,\u201dtype\u201d:\u201dclassic\u201d},{\u201cdateOfSleep\u201d:\u201d2020-10-10\u201d,\u201dduration\u201d:22980000,\u201defficiency\u201d:88,\u201dendTime\u201d:\u201d2020-10-10T08:10:00.000\u201d,\u201dinfoCode\u201d:0,\u201disMainSleep\u201d:true,\u201dlevels\u201d:{\u201cdata\u201d:[{\u201cdateTime\u201d:\u201d2020-10-10T01:46:30.000\u201d,\u201dlevel\u201d:\u201dlight\u201d,\u201dseconds\u201d:420},{\u201cdateTime\u201d:\u201d2020-10-10T01:53:30.000\u201d,\u201dlevel\u201d:\u201ddeep\u201d,\u201dseconds\u201d:1230},{\u201cdateTime\u201d:\u201d2020-10-10T02:14:00.000\u201d,\u201dlevel\u201d:\u201dlight\u201d,\u201dseconds\u201d:360},\u2026], \u201csummary\u201d:{\u201cdeep\u201d:{\u201ccount\u201d:3,\u201dminutes\u201d:92,\u201dthirtyDayAvgMinutes\u201d:0},\u201dlight\u201d:{\u201ccount\u201d:29,\u201dminutes\u201d:193,\u201dthirtyDayAvgMinutes\u201d:0},\u201drem\u201d:{\u201ccount\u201d:4,\u201dminutes\u201d:33,\u201dthirtyDayAvgMinutes\u201d:0},\u201dwake\u201d:{\u201ccount\u201d:28,\u201dminutes\u201d:65,\u201dthirtyDayAvgMinutes\u201d:0}}},\u201dlogId\u201d:26311786557,\u201dminutesAfterWakeup\u201d:0,\u201dminutesAsleep\u201d:318,\u201dminutesAwake\u201d:65,\u201dminutesToFallAsleep\u201d:0,\u201dstartTime\u201d:\u201d2020-10-10T01:46:30.000\u201d,\u201dtimeInBed\u201d:383,\u201dtype\u201d:\u201dstages\u201d}],\u201dsummary\u201d:{\u201cstages\u201d:{\u201cdeep\u201d:92,\u201dlight\u201d:193,\u201drem\u201d:33,\u201dwake\u201d:65},\u201dtotalMinutesAsleep\u201d:373,\u201dtotalSleepRecords\u201d:2,\u201dtotalTimeInBed\u201d:443}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d:[{\u201cdateOfSleep\u201d:\u201d2020-10-11\u201d,\u201dduration\u201d:41640000,\u201defficiency\u201d:89,\u201dendTime\u201d:\u201d2020-10-11T11:47:00.000\u201d,\u201dinfoCode\u201d:0,\u201disMainSleep\u201d:true,\u201dlevels\u201d:{\u201cdata\u201d:[{\u201cdateTime\u201d:\u201d2020-10-11T00:12:30.000\u201d,\u201dlevel\u201d:\u201dwake\u201d,\u201dseconds\u201d:450},{\u201cdateTime\u201d:\u201d2020-10-11T00:20:00.000\u201d,\u201dlevel\u201d:\u201dlight\u201d,\u201dseconds\u201d:870},{\u201cdateTime\u201d:\u201d2020-10-11T00:34:30.000\u201d,\u201dlevel\u201d:\u201dwake\u201d,\u201dseconds\u201d:780},\u2026], \u201csummary\u201d:{\u201cdeep\u201d:{\u201ccount\u201d:4,\u201dminutes\u201d:52,\u201dthirtyDayAvgMinutes\u201d:62},\u201dlight\u201d:{\u201ccount\u201d:32,\u201dminutes\u201d:442,\u201dthirtyDayAvgMinutes\u201d:364},\u201drem\u201d:{\u201ccount\u201d:6,\u201dminutes\u201d:68,\u201dthirtyDayAvgMinutes\u201d:58},\u201dwake\u201d:{\u201ccount\u201d:29,\u201dminutes\u201d:132,\u201dthirtyDayAvgMinutes\u201d:94}}},\u201dlogId\u201d:26589710670,\u201dminutesAfterWakeup\u201d:1,\u201dminutesAsleep\u201d:562,\u201dminutesAwake\u201d:132,\u201dminutesToFallAsleep\u201d:0,\u201dstartTime\u201d:\u201d2020-10-11T00:12:30.000\u201d,\u201dtimeInBed\u201d:694,\u201dtype\u201d:\u201dstages\u201d}],\u201dsummary\u201d:{\u201cstages\u201d:{\u201cdeep\u201d:52,\u201dlight\u201d:442,\u201drem\u201d:68,\u201dwake\u201d:132},\u201dtotalMinutesAsleep\u201d:562,\u201dtotalSleepRecords\u201d:1,\u201dtotalTimeInBed\u201d:694}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 {\u201csleep\u201d:[{\u201cdateOfSleep\u201d:\u201d2020-10-12\u201d,\u201dduration\u201d:28980000,\u201defficiency\u201d:93,\u201dendTime\u201d:\u201d2020-10-12T09:34:30.000\u201d,\u201dinfoCode\u201d:0,\u201disMainSleep\u201d:true,\u201dlevels\u201d:{\u201cdata\u201d:[{\u201cdateTime\u201d:\u201d2020-10-12T01:31:00.000\u201d,\u201dlevel\u201d:\u201dwake\u201d,\u201dseconds\u201d:600},{\u201cdateTime\u201d:\u201d2020-10-12T01:41:00.000\u201d,\u201dlevel\u201d:\u201dlight\u201d,\u201dseconds\u201d:60},{\u201cdateTime\u201d:\u201d2020-10-12T01:42:00.000\u201d,\u201dlevel\u201d:\u201ddeep\u201d,\u201dseconds\u201d:2340},\u2026], \u201csummary\u201d:{\u201cdeep\u201d:{\u201ccount\u201d:4,\u201dminutes\u201d:63,\u201dthirtyDayAvgMinutes\u201d:59},\u201dlight\u201d:{\u201ccount\u201d:27,\u201dminutes\u201d:257,\u201dthirtyDayAvgMinutes\u201d:364},\u201drem\u201d:{\u201ccount\u201d:5,\u201dminutes\u201d:94,\u201dthirtyDayAvgMinutes\u201d:58},\u201dwake\u201d:{\u201ccount\u201d:24,\u201dminutes\u201d:69,\u201dthirtyDayAvgMinutes\u201d:95}}},\u201dlogId\u201d:26589710673,\u201dminutesAfterWakeup\u201d:0,\u201dminutesAsleep\u201d:415,\u201dminutesAwake\u201d:68,\u201dminutesToFallAsleep\u201d:0,\u201dstartTime\u201d:\u201d2020-10-12T01:31:00.000\u201d,\u201dtimeInBed\u201d:483,\u201dtype\u201d:\u201dstages\u201d}],\u201dsummary\u201d:{\u201cstages\u201d:{\u201cdeep\u201d:63,\u201dlight\u201d:257,\u201drem\u201d:94,\u201dwake\u201d:69},\u201dtotalMinutesAsleep\u201d:415,\u201dtotalSleepRecords\u201d:1,\u201dtotalTimeInBed\u201d:483}} PLAIN_TEXT device_id local_start_date_time local_end_date_time efficiency minutes_after_wakeup minutes_asleep minutes_awake minutes_to_fall_asleep minutes_in_bed is_main_sleep type a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-10 15:36:30 2020-10-10 16:37:00 92 0 55 5 0 60 0 classic a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-10 01:46:30 2020-10-10 08:10:00 88 0 318 65 0 383 1 stages a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-11 00:12:30 2020-10-11 11:47:00 89 1 562 132 0 694 1 stages a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-12 01:31:00 2020-10-12 09:34:30 93 0 415 68 0 483 1 stages","title":"Fitbit Sleep Summary"},{"location":"features/fitbit-sleep-summary/#rapids-provider","text":"Available time segments Only available for segments that span 1 or more complete days (e.g. Jan 1 st 00:00 to Jan 3 rd 23:59) File Sequence - data/raw/ { pid } /fitbit_sleep_summary_raw.csv - data/raw/ { pid } /fitbit_sleep_summary_parsed.csv - data/raw/ { pid } /fitbit_sleep_summary_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_sleep_summary_features/fitbit_sleep_summary_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_sleep_summary.csv Parameters description for [FITBIT_SLEEP_SUMMARY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_SLEEP_SUMMARY features from the RAPIDS provider [SLEEP_TYPES] Types of sleep to be included in the feature extraction computation. Fitbit provides 3 types of sleep: main , nap , all . [FEATURES] Features to be computed from sleep summary data, see table below Features description for [FITBIT_SLEEP_SUMMARY][PROVIDERS][RAPIDS] : Feature Units Description countepisodeTYPE episodes Number of sleep episodes for a certain sleep type during a time segment. avgefficiencyTYPE scores Average sleep efficiency for a certain sleep type during a time segment. sumdurationafterwakeupTYPE minutes Total duration the user stayed in bed after waking up for a certain sleep type during a time segment. sumdurationasleepTYPE minutes Total sleep duration for a certain sleep type during a time segment. sumdurationawakeTYPE minutes Total duration the user stayed awake but still in bed for a certain sleep type during a time segment. sumdurationtofallasleepTYPE minutes Total duration the user spent to fall asleep for a certain sleep type during a time segment. sumdurationinbedTYPE minutes Total duration the user stayed in bed (sumdurationtofallasleep + sumdurationawake + sumdurationasleep + sumdurationafterwakeup) for a certain sleep type during a time segment. avgdurationafterwakeupTYPE minutes Average duration the user stayed in bed after waking up for a certain sleep type during a time segment. avgdurationasleepTYPE minutes Average sleep duration for a certain sleep type during a time segment. avgdurationawakeTYPE minutes Average duration the user stayed awake but still in bed for a certain sleep type during a time segment. avgdurationtofallasleepTYPE minutes Average duration the user spent to fall asleep for a certain sleep type during a time segment. avgdurationinbedTYPE minutes Average duration the user stayed in bed (sumdurationtofallasleep + sumdurationawake + sumdurationasleep + sumdurationafterwakeup) for a certain sleep type during a time segment. Assumptions/Observations There are three sleep types (TYPE): main , nap , all . The all type contains both main sleep and naps. There are two versions of Fitbit\u2019s sleep API ( version 1 and version 1.2 ), and each provides raw sleep data in a different format: Count & duration summaries . v1 contains count_awake , duration_awake , count_awakenings , count_restless , and duration_restless fields for every sleep record but v1.2 does not. API columns . Features are computed based on the values provided by Fitbit\u2019s API: efficiency , minutes_after_wakeup , minutes_asleep , minutes_awake , minutes_to_fall_asleep , minutes_in_bed , is_main_sleep and type .","title":"RAPIDS provider"},{"location":"features/fitbit-steps-intraday/","text":"Fitbit Steps Intraday \u00b6 Sensor parameters description for [FITBIT_STEPS_INTRADAY] : Key Description [TABLE] Database table name or file path where the steps intraday data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-07\u201d,\u201dvalue\u201d:\u201d1775\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:5},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:3},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:0},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-08\u201d,\u201dvalue\u201d:\u201d3201\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:14},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:11},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:10},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-09\u201d,\u201dvalue\u201d:\u201d998\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:0},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:0},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:0},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} PLAIN_TEXT device_id local_date_time steps a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:00:00 5 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:01:00 3 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:02:00 0 RAPIDS provider \u00b6 Available time segments Available for all time segments File Sequence - data/raw/ { pid } /fitbit_steps_intraday_raw.csv - data/raw/ { pid } /fitbit_steps_intraday_parsed.csv - data/raw/ { pid } /fitbit_steps_intraday_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_steps_intraday_features/fitbit_steps_intraday_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_steps_intraday.csv Parameters description for [FITBIT_STEPS_INTRADAY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_STEPS_INTRADAY features from the RAPIDS provider [FEATURES] Features to be computed from steps intraday data, see table below [THRESHOLD_ACTIVE_BOUT] Every minute with Fitbit steps data wil be labelled as sedentary if its step count is below this threshold, otherwise, active . [INCLUDE_ZERO_STEP_ROWS] Whether or not to include time segments with a 0 step count during the whole day. Features description for [FITBIT_STEPS_INTRADAY][PROVIDERS][RAPIDS] : Feature Units Description sumsteps steps The total step count during a time segment. maxsteps steps The maximum step count during a time segment. minsteps steps The minimum step count during a time segment. avgsteps steps The average step count during a time segment. stdsteps steps The standard deviation of step count during a time segment. countepisodesedentarybout bouts Number of sedentary bouts during a time segment. sumdurationsedentarybout minutes Total duration of all sedentary bouts during a time segment. maxdurationsedentarybout minutes The maximum duration of any sedentary bout during a time segment. mindurationsedentarybout minutes The minimum duration of any sedentary bout during a time segment. avgdurationsedentarybout minutes The average duration of sedentary bouts during a time segment. stddurationsedentarybout minutes The standard deviation of the duration of sedentary bouts during a time segment. countepisodeactivebout bouts Number of active bouts during a time segment. sumdurationactivebout minutes Total duration of all active bouts during a time segment. maxdurationactivebout minutes The maximum duration of any active bout during a time segment. mindurationactivebout minutes The minimum duration of any active bout during a time segment. avgdurationactivebout minutes The average duration of active bouts during a time segment. stddurationactivebout minutes The standard deviation of the duration of active bouts during a time segment. Assumptions/Observations Active and sedentary bouts . If the step count per minute is smaller than THRESHOLD_ACTIVE_BOUT (default value is 10), that minute is labelled as sedentary, otherwise, is labelled as active. Active and sedentary bouts are periods of consecutive minutes labelled as active or sedentary .","title":"Fitbit Steps Intraday"},{"location":"features/fitbit-steps-intraday/#fitbit-steps-intraday","text":"Sensor parameters description for [FITBIT_STEPS_INTRADAY] : Key Description [TABLE] Database table name or file path where the steps intraday data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-07\u201d,\u201dvalue\u201d:\u201d1775\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:5},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:3},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:0},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-08\u201d,\u201dvalue\u201d:\u201d3201\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:14},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:11},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:10},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-09\u201d,\u201dvalue\u201d:\u201d998\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:0},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:0},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:0},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} PLAIN_TEXT device_id local_date_time steps a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:00:00 5 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:01:00 3 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 00:02:00 0","title":"Fitbit Steps Intraday"},{"location":"features/fitbit-steps-intraday/#rapids-provider","text":"Available time segments Available for all time segments File Sequence - data/raw/ { pid } /fitbit_steps_intraday_raw.csv - data/raw/ { pid } /fitbit_steps_intraday_parsed.csv - data/raw/ { pid } /fitbit_steps_intraday_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_steps_intraday_features/fitbit_steps_intraday_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_steps_intraday.csv Parameters description for [FITBIT_STEPS_INTRADAY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_STEPS_INTRADAY features from the RAPIDS provider [FEATURES] Features to be computed from steps intraday data, see table below [THRESHOLD_ACTIVE_BOUT] Every minute with Fitbit steps data wil be labelled as sedentary if its step count is below this threshold, otherwise, active . [INCLUDE_ZERO_STEP_ROWS] Whether or not to include time segments with a 0 step count during the whole day. Features description for [FITBIT_STEPS_INTRADAY][PROVIDERS][RAPIDS] : Feature Units Description sumsteps steps The total step count during a time segment. maxsteps steps The maximum step count during a time segment. minsteps steps The minimum step count during a time segment. avgsteps steps The average step count during a time segment. stdsteps steps The standard deviation of step count during a time segment. countepisodesedentarybout bouts Number of sedentary bouts during a time segment. sumdurationsedentarybout minutes Total duration of all sedentary bouts during a time segment. maxdurationsedentarybout minutes The maximum duration of any sedentary bout during a time segment. mindurationsedentarybout minutes The minimum duration of any sedentary bout during a time segment. avgdurationsedentarybout minutes The average duration of sedentary bouts during a time segment. stddurationsedentarybout minutes The standard deviation of the duration of sedentary bouts during a time segment. countepisodeactivebout bouts Number of active bouts during a time segment. sumdurationactivebout minutes Total duration of all active bouts during a time segment. maxdurationactivebout minutes The maximum duration of any active bout during a time segment. mindurationactivebout minutes The minimum duration of any active bout during a time segment. avgdurationactivebout minutes The average duration of active bouts during a time segment. stddurationactivebout minutes The standard deviation of the duration of active bouts during a time segment. Assumptions/Observations Active and sedentary bouts . If the step count per minute is smaller than THRESHOLD_ACTIVE_BOUT (default value is 10), that minute is labelled as sedentary, otherwise, is labelled as active. Active and sedentary bouts are periods of consecutive minutes labelled as active or sedentary .","title":"RAPIDS provider"},{"location":"features/fitbit-steps-summary/","text":"Fitbit Steps Summary \u00b6 Sensor parameters description for [FITBIT_STEPS_SUMMARY] : Key Description [TABLE] Database table name or file path where the steps summary data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-07\u201d,\u201dvalue\u201d:\u201d1775\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:5},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:3},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:0},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-08\u201d,\u201dvalue\u201d:\u201d3201\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:14},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:11},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:10},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-09\u201d,\u201dvalue\u201d:\u201d998\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:0},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:0},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:0},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} PLAIN_TEXT device_id local_date_time steps a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 1775 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-08 3201 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-09 998 RAPIDS provider \u00b6 Available time segments Only available for segments that span 1 or more complete days (e.g. Jan 1 st 00:00 to Jan 3 rd 23:59) File Sequence - data/raw/ { pid } /fitbit_steps_summary_raw.csv - data/raw/ { pid } /fitbit_steps_summary_parsed.csv - data/raw/ { pid } /fitbit_steps_summary_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_steps_summary_features/fitbit_steps_summary_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_steps_summary.csv Parameters description for [FITBIT_STEPS_SUMMARY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_STEPS_SUMMARY features from the RAPIDS provider [FEATURES] Features to be computed from steps summary data, see table below Features description for [FITBIT_STEPS_SUMMARY][PROVIDERS][RAPIDS] : Feature Units Description maxsumsteps steps The maximum daily step count during a time segment. minsumsteps steps The minimum daily step count during a time segment. avgsumsteps steps The average daily step count during a time segment. mediansumsteps steps The median of daily step count during a time segment. stdsumsteps steps The standard deviation of daily step count during a time segment. Assumptions/Observations NA","title":"Fitbit Steps Summary"},{"location":"features/fitbit-steps-summary/#fitbit-steps-summary","text":"Sensor parameters description for [FITBIT_STEPS_SUMMARY] : Key Description [TABLE] Database table name or file path where the steps summary data is stored. The configuration keys in Device Data Source Configuration control whether this parameter is interpreted as table or file. The format of the column(s) containing the Fitbit sensor data can be JSON or PLAIN_TEXT . The data in JSON format is obtained directly from the Fitbit API. We support PLAIN_TEXT in case you already parsed your data and don\u2019t have access to your participants\u2019 Fitbit accounts anymore. If your data is in JSON format then summary and intraday data come packed together. We provide examples of the input format that RAPIDS expects, note that both examples for JSON and PLAIN_TEXT are tabular and the actual format difference comes in the fitbit_data column (we truncate the JSON example for brevity). Example of the structure of source data JSON device_id fitbit_data a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-07\u201d,\u201dvalue\u201d:\u201d1775\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:5},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:3},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:0},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-08\u201d,\u201dvalue\u201d:\u201d3201\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:14},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:11},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:10},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} a748ee1a-1d0b-4ae9-9074-279a2b6ba524 \u201cactivities-steps\u201d:[{\u201cdateTime\u201d:\u201d2020-10-09\u201d,\u201dvalue\u201d:\u201d998\u201d}],\u201dactivities-steps-intraday\u201d:{\u201cdataset\u201d:[{\u201ctime\u201d:\u201d00:00:00\u201d,\u201dvalue\u201d:0},{\u201ctime\u201d:\u201d00:01:00\u201d,\u201dvalue\u201d:0},{\u201ctime\u201d:\u201d00:02:00\u201d,\u201dvalue\u201d:0},\u2026],\u201ddatasetInterval\u201d:1,\u201ddatasetType\u201d:\u201dminute\u201d}} PLAIN_TEXT device_id local_date_time steps a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-07 1775 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-08 3201 a748ee1a-1d0b-4ae9-9074-279a2b6ba524 2020-10-09 998","title":"Fitbit Steps Summary"},{"location":"features/fitbit-steps-summary/#rapids-provider","text":"Available time segments Only available for segments that span 1 or more complete days (e.g. Jan 1 st 00:00 to Jan 3 rd 23:59) File Sequence - data/raw/ { pid } /fitbit_steps_summary_raw.csv - data/raw/ { pid } /fitbit_steps_summary_parsed.csv - data/raw/ { pid } /fitbit_steps_summary_parsed_with_datetime.csv - data/interim/ { pid } /fitbit_steps_summary_features/fitbit_steps_summary_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /fitbit_steps_summary.csv Parameters description for [FITBIT_STEPS_SUMMARY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract FITBIT_STEPS_SUMMARY features from the RAPIDS provider [FEATURES] Features to be computed from steps summary data, see table below Features description for [FITBIT_STEPS_SUMMARY][PROVIDERS][RAPIDS] : Feature Units Description maxsumsteps steps The maximum daily step count during a time segment. minsumsteps steps The minimum daily step count during a time segment. avgsumsteps steps The average daily step count during a time segment. mediansumsteps steps The median of daily step count during a time segment. stdsumsteps steps The standard deviation of daily step count during a time segment. Assumptions/Observations NA","title":"RAPIDS provider"},{"location":"features/phone-accelerometer/","text":"Phone Accelerometer \u00b6 Sensor parameters description for [PHONE_ACCELEROMETER] : Key Description [TABLE] Database table where the accelerometer data is stored RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_accelerometer_raw.csv - data/raw/ { pid } /phone_accelerometer_with_datetime.csv - data/interim/ { pid } /phone_accelerometer_features/phone_accelerometer_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_accelerometer.csv Parameters description for [PHONE_ACCELEROMETER][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_ACCELEROMETER features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_ACCELEROMETER][PROVIDERS][RAPIDS] : Feature Units Description maxmagnitude m/s 2 The maximum magnitude of acceleration ( \\(\\|acceleration\\| = \\sqrt{x^2 + y^2 + z^2}\\) ). minmagnitude m/s 2 The minimum magnitude of acceleration. avgmagnitude m/s 2 The average magnitude of acceleration. medianmagnitude m/s 2 The median magnitude of acceleration. stdmagnitude m/s 2 The standard deviation of acceleration. Assumptions/Observations Analyzing accelerometer data is a memory intensive task. If RAPIDS crashes is likely because the accelerometer dataset for a participant is to big to fit in memory. We are considering different alternatives to overcome this problem. PANDA provider \u00b6 These features are based on the work by Panda et al . Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_accelerometer_raw.csv - data/raw/ { pid } /phone_accelerometer_with_datetime.csv - data/interim/ { pid } /phone_accelerometer_features/phone_accelerometer_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_accelerometer.csv Parameters description for [PHONE_ACCELEROMETER][PROVIDERS][PANDA] : Key Description [COMPUTE] Set to True to extract PHONE_ACCELEROMETER features from the PANDA provider [FEATURES] Features to be computed for exertional and non-exertional activity episodes, see table below Features description for [PHONE_ACCELEROMETER][PROVIDERS][PANDA] : Feature Units Description sumduration minutes Total duration of all exertional or non-exertional activity episodes. maxduration minutes Longest duration of any exertional or non-exertional activity episode. minduration minutes Shortest duration of any exertional or non-exertional activity episode. avgduration minutes Average duration of any exertional or non-exertional activity episode. medianduration minutes Median duration of any exertional or non-exertional activity episode. stdduration minutes Standard deviation of the duration of all exertional or non-exertional activity episodes. Assumptions/Observations Analyzing accelerometer data is a memory intensive task. If RAPIDS crashes is likely because the accelerometer dataset for a participant is to big to fit in memory. We are considering different alternatives to overcome this problem. See Panda et al for a definition of exertional and non-exertional activity episodes","title":"Phone Accelerometer"},{"location":"features/phone-accelerometer/#phone-accelerometer","text":"Sensor parameters description for [PHONE_ACCELEROMETER] : Key Description [TABLE] Database table where the accelerometer data is stored","title":"Phone Accelerometer"},{"location":"features/phone-accelerometer/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_accelerometer_raw.csv - data/raw/ { pid } /phone_accelerometer_with_datetime.csv - data/interim/ { pid } /phone_accelerometer_features/phone_accelerometer_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_accelerometer.csv Parameters description for [PHONE_ACCELEROMETER][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_ACCELEROMETER features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_ACCELEROMETER][PROVIDERS][RAPIDS] : Feature Units Description maxmagnitude m/s 2 The maximum magnitude of acceleration ( \\(\\|acceleration\\| = \\sqrt{x^2 + y^2 + z^2}\\) ). minmagnitude m/s 2 The minimum magnitude of acceleration. avgmagnitude m/s 2 The average magnitude of acceleration. medianmagnitude m/s 2 The median magnitude of acceleration. stdmagnitude m/s 2 The standard deviation of acceleration. Assumptions/Observations Analyzing accelerometer data is a memory intensive task. If RAPIDS crashes is likely because the accelerometer dataset for a participant is to big to fit in memory. We are considering different alternatives to overcome this problem.","title":"RAPIDS provider"},{"location":"features/phone-accelerometer/#panda-provider","text":"These features are based on the work by Panda et al . Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_accelerometer_raw.csv - data/raw/ { pid } /phone_accelerometer_with_datetime.csv - data/interim/ { pid } /phone_accelerometer_features/phone_accelerometer_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_accelerometer.csv Parameters description for [PHONE_ACCELEROMETER][PROVIDERS][PANDA] : Key Description [COMPUTE] Set to True to extract PHONE_ACCELEROMETER features from the PANDA provider [FEATURES] Features to be computed for exertional and non-exertional activity episodes, see table below Features description for [PHONE_ACCELEROMETER][PROVIDERS][PANDA] : Feature Units Description sumduration minutes Total duration of all exertional or non-exertional activity episodes. maxduration minutes Longest duration of any exertional or non-exertional activity episode. minduration minutes Shortest duration of any exertional or non-exertional activity episode. avgduration minutes Average duration of any exertional or non-exertional activity episode. medianduration minutes Median duration of any exertional or non-exertional activity episode. stdduration minutes Standard deviation of the duration of all exertional or non-exertional activity episodes. Assumptions/Observations Analyzing accelerometer data is a memory intensive task. If RAPIDS crashes is likely because the accelerometer dataset for a participant is to big to fit in memory. We are considering different alternatives to overcome this problem. See Panda et al for a definition of exertional and non-exertional activity episodes","title":"PANDA provider"},{"location":"features/phone-activity-recognition/","text":"Phone Activity Recognition \u00b6 Sensor parameters description for [PHONE_ACTIVITY_RECOGNITION] : Key Description [TABLE][ANDROID] Database table where the activity data from Android devices is stored (the AWARE client saves this data on different tables for Android and iOS) [TABLE][IOS] Database table where the activity data from iOS devices is stored (the AWARE client saves this data on different tables for Android and iOS) [EPISODE_THRESHOLD_BETWEEN_ROWS] Difference in minutes between any two rows for them to be considered part of the same activity episode RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_activity_recognition_raw.csv - data/raw/ { pid } /phone_activity_recognition_with_datetime.csv - data/raw/ { pid } /phone_activity_recognition_with_datetime_unified.csv - data/interim/ { pid } /phone_activity_recognition_episodes.csv - data/interim/ { pid } /phone_activity_recognition_episodes_resampled.csv - data/interim/ { pid } /phone_activity_recognition_episodes_resampled_with_datetime.csv - data/interim/ { pid } /phone_activity_recognition_features/phone_activity_recognition_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_activity_recognition.csv Parameters description for [PHONE_ACTIVITY_RECOGNITION][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_ACTIVITY_RECOGNITION features from the RAPIDS provider [FEATURES] Features to be computed, see table below [ACTIVITY_CLASSES][STATIONARY] An array of the activity labels to be considered in the STATIONARY category choose any of still , tilting [ACTIVITY_CLASSES][MOBILE] An array of the activity labels to be considered in the MOBILE category choose any of on_foot , walking , running , on_bicycle [ACTIVITY_CLASSES][VEHICLE] An array of the activity labels to be considered in the VEHICLE category choose any of in_vehicule Features description for [PHONE_ACTIVITY_RECOGNITION][PROVIDERS][RAPIDS] : Feature Units Description count rows Number of episodes. mostcommonactivity activity type The most common activity type (e.g. still , on_foot , etc.). If there is a tie, the first one is chosen. countuniqueactivities activity type Number of unique activities. durationstationary minutes The total duration of [ACTIVITY_CLASSES][STATIONARY] episodes durationmobile minutes The total duration of [ACTIVITY_CLASSES][MOBILE] episodes of on foot, running, and on bicycle activities durationvehicle minutes The total duration of [ACTIVITY_CLASSES][VEHICLE] episodes of on vehicle activity Assumptions/Observations iOS Activity Recognition names and types are unified with Android labels: iOS Activity Name Android Activity Name Android Activity Type walking walking 7 running running 8 cycling on_bicycle 1 automotive in_vehicle 0 stationary still 3 unknown unknown 4 In AWARE, Activity Recognition data for Android and iOS are stored in two different database tables, RAPIDS automatically infers what platform each participant belongs to based on their participant file .","title":"Phone Activity Recognition"},{"location":"features/phone-activity-recognition/#phone-activity-recognition","text":"Sensor parameters description for [PHONE_ACTIVITY_RECOGNITION] : Key Description [TABLE][ANDROID] Database table where the activity data from Android devices is stored (the AWARE client saves this data on different tables for Android and iOS) [TABLE][IOS] Database table where the activity data from iOS devices is stored (the AWARE client saves this data on different tables for Android and iOS) [EPISODE_THRESHOLD_BETWEEN_ROWS] Difference in minutes between any two rows for them to be considered part of the same activity episode","title":"Phone Activity Recognition"},{"location":"features/phone-activity-recognition/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_activity_recognition_raw.csv - data/raw/ { pid } /phone_activity_recognition_with_datetime.csv - data/raw/ { pid } /phone_activity_recognition_with_datetime_unified.csv - data/interim/ { pid } /phone_activity_recognition_episodes.csv - data/interim/ { pid } /phone_activity_recognition_episodes_resampled.csv - data/interim/ { pid } /phone_activity_recognition_episodes_resampled_with_datetime.csv - data/interim/ { pid } /phone_activity_recognition_features/phone_activity_recognition_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_activity_recognition.csv Parameters description for [PHONE_ACTIVITY_RECOGNITION][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_ACTIVITY_RECOGNITION features from the RAPIDS provider [FEATURES] Features to be computed, see table below [ACTIVITY_CLASSES][STATIONARY] An array of the activity labels to be considered in the STATIONARY category choose any of still , tilting [ACTIVITY_CLASSES][MOBILE] An array of the activity labels to be considered in the MOBILE category choose any of on_foot , walking , running , on_bicycle [ACTIVITY_CLASSES][VEHICLE] An array of the activity labels to be considered in the VEHICLE category choose any of in_vehicule Features description for [PHONE_ACTIVITY_RECOGNITION][PROVIDERS][RAPIDS] : Feature Units Description count rows Number of episodes. mostcommonactivity activity type The most common activity type (e.g. still , on_foot , etc.). If there is a tie, the first one is chosen. countuniqueactivities activity type Number of unique activities. durationstationary minutes The total duration of [ACTIVITY_CLASSES][STATIONARY] episodes durationmobile minutes The total duration of [ACTIVITY_CLASSES][MOBILE] episodes of on foot, running, and on bicycle activities durationvehicle minutes The total duration of [ACTIVITY_CLASSES][VEHICLE] episodes of on vehicle activity Assumptions/Observations iOS Activity Recognition names and types are unified with Android labels: iOS Activity Name Android Activity Name Android Activity Type walking walking 7 running running 8 cycling on_bicycle 1 automotive in_vehicle 0 stationary still 3 unknown unknown 4 In AWARE, Activity Recognition data for Android and iOS are stored in two different database tables, RAPIDS automatically infers what platform each participant belongs to based on their participant file .","title":"RAPIDS provider"},{"location":"features/phone-applications-foreground/","text":"Phone Applications Foreground \u00b6 Sensor parameters description for [PHONE_APPLICATIONS_FOREGROUND] (these parameters are used by the only provider available at the moment, RAPIDS): Key Description [TABLE] Database table where the applications foreground data is stored [APPLICATION_CATEGORIES][CATALOGUE_SOURCE] FILE or GOOGLE . If FILE , app categories (genres) are read from [CATALOGUE_FILE] . If [GOOGLE] , app categories (genres) are scrapped from the Play Store [APPLICATION_CATEGORIES][CATALOGUE_FILE] CSV file with a package_name and genre column. By default we provide the catalogue created by Stachl et al in data/external/stachl_application_genre_catalogue.csv [APPLICATION_CATEGORIES][UPDATE_CATALOGUE_FILE] if [CATALOGUE_SOURCE] is equal to FILE , this flag signals whether or not to update [CATALOGUE_FILE] , if [CATALOGUE_SOURCE] is equal to GOOGLE all scraped genres will be saved to [CATALOGUE_FILE] [APPLICATION_CATEGORIES][SCRAPE_MISSING_CATEGORIES] This flag signals whether or not to scrape categories (genres) missing from the [CATALOGUE_FILE] . If [CATALOGUE_SOURCE] is equal to GOOGLE , all genres are scraped anyway (this flag is ignored) RAPIDS provider \u00b6 The app category (genre) catalogue used in these features was originally created by Stachl et al . Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_applications_foreground_raw.csv - data/raw/ { pid } /phone_applications_foreground_with_datetime.csv - data/raw/ { pid } /phone_applications_foreground_with_datetime_with_categories.csv - data/interim/ { pid } /phone_applications_foreground_features/phone_applications_foreground_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_applications_foreground.csv Parameters description for [PHONE_APPLICATIONS_FOREGROUND][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_APPLICATIONS_FOREGROUND features from the RAPIDS provider [FEATURES] Features to be computed, see table below [SINGLE_CATEGORIES] An array of app categories to be included in the feature extraction computation. The special keyword all represents a category with all the apps from each participant. By default we use the category catalogue pointed by [APPLICATION_CATEGORIES][CATALOGUE_FILE] (see the Sensor parameters description table above) [MULTIPLE_CATEGORIES] An array of collections representing meta-categories (a group of categories). They key of each element is the name of the meta-category and the value is an array of member app categories. By default we use the category catalogue pointed by [APPLICATION_CATEGORIES][CATALOGUE_FILE] (see the Sensor parameters description table above) [SINGLE_APPS] An array of apps to be included in the feature extraction computation. Use their package name (e.g. com.google.android.youtube ) or the reserved keyword top1global (the most used app by a participant over the whole monitoring study) [EXCLUDED_CATEGORIES] An array of app categories to be excluded from the feature extraction computation. By default we use the category catalogue pointed by [APPLICATION_CATEGORIES][CATALOGUE_FILE] (see the Sensor parameters description table above) [EXCLUDED_APPS] An array of apps to be excluded from the feature extraction computation. Use their package name, for example: com.google.android.youtube Features description for [PHONE_APPLICATIONS_FOREGROUND][PROVIDERS][RAPIDS] : Feature Units Description count apps Number of times a single app or apps within a category were used (i.e. they were brought to the foreground either by tapping their icon or switching to it from another app) timeoffirstuse minutes The time in minutes between 12:00am (midnight) and the first use of a single app or apps within a category during a time_segment timeoflastuse minutes The time in minutes between 12:00am (midnight) and the last use of a single app or apps within a category during a time_segment frequencyentropy nats The entropy of the used apps within a category during a time_segment (each app is seen as a unique event, the more apps were used, the higher the entropy). This is especially relevant when computed over all apps. Entropy cannot be obtained for a single app Assumptions/Observations Features can be computed by app, by apps grouped under a single category (genre) and by multiple categories grouped together (meta-categories). For example, we can get features for Facebook (single app), for Social Network apps (a category including Facebook and other social media apps) or for Social (a meta-category formed by Social Network and Social Media Tools categories). Apps installed by default like YouTube are considered systems apps on some phones. We do an exact match to exclude apps where \u201cgenre\u201d == EXCLUDED_CATEGORIES or \u201cpackage_name\u201d == EXCLUDED_APPS . We provide three ways of classifying and app within a category (genre): a) by automatically scraping its official category from the Google Play Store, b) by using the catalogue created by Stachl et al. which we provide in RAPIDS ( data/external/stachl_application_genre_catalogue.csv ), or c) by manually creating a personalized catalogue. You can choose a, b or c by modifying [APPLICATION_GENRES] keys and values (see the Sensor parameters description table above).","title":"Phone Applications Foreground"},{"location":"features/phone-applications-foreground/#phone-applications-foreground","text":"Sensor parameters description for [PHONE_APPLICATIONS_FOREGROUND] (these parameters are used by the only provider available at the moment, RAPIDS): Key Description [TABLE] Database table where the applications foreground data is stored [APPLICATION_CATEGORIES][CATALOGUE_SOURCE] FILE or GOOGLE . If FILE , app categories (genres) are read from [CATALOGUE_FILE] . If [GOOGLE] , app categories (genres) are scrapped from the Play Store [APPLICATION_CATEGORIES][CATALOGUE_FILE] CSV file with a package_name and genre column. By default we provide the catalogue created by Stachl et al in data/external/stachl_application_genre_catalogue.csv [APPLICATION_CATEGORIES][UPDATE_CATALOGUE_FILE] if [CATALOGUE_SOURCE] is equal to FILE , this flag signals whether or not to update [CATALOGUE_FILE] , if [CATALOGUE_SOURCE] is equal to GOOGLE all scraped genres will be saved to [CATALOGUE_FILE] [APPLICATION_CATEGORIES][SCRAPE_MISSING_CATEGORIES] This flag signals whether or not to scrape categories (genres) missing from the [CATALOGUE_FILE] . If [CATALOGUE_SOURCE] is equal to GOOGLE , all genres are scraped anyway (this flag is ignored)","title":"Phone Applications Foreground"},{"location":"features/phone-applications-foreground/#rapids-provider","text":"The app category (genre) catalogue used in these features was originally created by Stachl et al . Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_applications_foreground_raw.csv - data/raw/ { pid } /phone_applications_foreground_with_datetime.csv - data/raw/ { pid } /phone_applications_foreground_with_datetime_with_categories.csv - data/interim/ { pid } /phone_applications_foreground_features/phone_applications_foreground_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_applications_foreground.csv Parameters description for [PHONE_APPLICATIONS_FOREGROUND][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_APPLICATIONS_FOREGROUND features from the RAPIDS provider [FEATURES] Features to be computed, see table below [SINGLE_CATEGORIES] An array of app categories to be included in the feature extraction computation. The special keyword all represents a category with all the apps from each participant. By default we use the category catalogue pointed by [APPLICATION_CATEGORIES][CATALOGUE_FILE] (see the Sensor parameters description table above) [MULTIPLE_CATEGORIES] An array of collections representing meta-categories (a group of categories). They key of each element is the name of the meta-category and the value is an array of member app categories. By default we use the category catalogue pointed by [APPLICATION_CATEGORIES][CATALOGUE_FILE] (see the Sensor parameters description table above) [SINGLE_APPS] An array of apps to be included in the feature extraction computation. Use their package name (e.g. com.google.android.youtube ) or the reserved keyword top1global (the most used app by a participant over the whole monitoring study) [EXCLUDED_CATEGORIES] An array of app categories to be excluded from the feature extraction computation. By default we use the category catalogue pointed by [APPLICATION_CATEGORIES][CATALOGUE_FILE] (see the Sensor parameters description table above) [EXCLUDED_APPS] An array of apps to be excluded from the feature extraction computation. Use their package name, for example: com.google.android.youtube Features description for [PHONE_APPLICATIONS_FOREGROUND][PROVIDERS][RAPIDS] : Feature Units Description count apps Number of times a single app or apps within a category were used (i.e. they were brought to the foreground either by tapping their icon or switching to it from another app) timeoffirstuse minutes The time in minutes between 12:00am (midnight) and the first use of a single app or apps within a category during a time_segment timeoflastuse minutes The time in minutes between 12:00am (midnight) and the last use of a single app or apps within a category during a time_segment frequencyentropy nats The entropy of the used apps within a category during a time_segment (each app is seen as a unique event, the more apps were used, the higher the entropy). This is especially relevant when computed over all apps. Entropy cannot be obtained for a single app Assumptions/Observations Features can be computed by app, by apps grouped under a single category (genre) and by multiple categories grouped together (meta-categories). For example, we can get features for Facebook (single app), for Social Network apps (a category including Facebook and other social media apps) or for Social (a meta-category formed by Social Network and Social Media Tools categories). Apps installed by default like YouTube are considered systems apps on some phones. We do an exact match to exclude apps where \u201cgenre\u201d == EXCLUDED_CATEGORIES or \u201cpackage_name\u201d == EXCLUDED_APPS . We provide three ways of classifying and app within a category (genre): a) by automatically scraping its official category from the Google Play Store, b) by using the catalogue created by Stachl et al. which we provide in RAPIDS ( data/external/stachl_application_genre_catalogue.csv ), or c) by manually creating a personalized catalogue. You can choose a, b or c by modifying [APPLICATION_GENRES] keys and values (see the Sensor parameters description table above).","title":"RAPIDS provider"},{"location":"features/phone-battery/","text":"Phone Battery \u00b6 Sensor parameters description for [PHONE_BATTERY] : Key Description [TABLE] Database table where the battery data is stored [EPISODE_THRESHOLD_BETWEEN_ROWS] Difference in minutes between any two rows for them to be considered part of the same battery charge or discharge episode RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_battery_raw.csv - data/interim/ { pid } /phone_battery_episodes.csv - data/interim/ { pid } /phone_battery_episodes_resampled.csv - data/interim/ { pid } /phone_battery_episodes_resampled_with_datetime.csv - data/interim/ { pid } /phone_battery_features/phone_battery_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_battery.csv Parameters description for [PHONE_BATTERY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_BATTERY features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_BATTERY][PROVIDERS][RAPIDS] : Feature Units Description countdischarge episodes Number of discharging episodes. sumdurationdischarge minutes The total duration of all discharging episodes. countcharge episodes Number of battery charging episodes. sumdurationcharge minutes The total duration of all charging episodes. avgconsumptionrate episodes/minutes The average of all episodes\u2019 consumption rates. An episode\u2019s consumption rate is defined as the ratio between its battery delta and duration maxconsumptionrate episodes/minutes The highest of all episodes\u2019 consumption rates. An episode\u2019s consumption rate is defined as the ratio between its battery delta and duration Assumptions/Observations We convert battery data collected with iOS client v1 (autodetected because battery status 4 do not exist) to match Android battery format: we swap status 3 for 5 and 1 for 3 We group battery data into discharge or charge episodes considering any contiguous rows with consecutive reductions or increases of the battery level if they are logged within [EPISODE_THRESHOLD_BETWEEN_ROWS] minutes from each other.","title":"Phone Battery"},{"location":"features/phone-battery/#phone-battery","text":"Sensor parameters description for [PHONE_BATTERY] : Key Description [TABLE] Database table where the battery data is stored [EPISODE_THRESHOLD_BETWEEN_ROWS] Difference in minutes between any two rows for them to be considered part of the same battery charge or discharge episode","title":"Phone Battery"},{"location":"features/phone-battery/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_battery_raw.csv - data/interim/ { pid } /phone_battery_episodes.csv - data/interim/ { pid } /phone_battery_episodes_resampled.csv - data/interim/ { pid } /phone_battery_episodes_resampled_with_datetime.csv - data/interim/ { pid } /phone_battery_features/phone_battery_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_battery.csv Parameters description for [PHONE_BATTERY][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_BATTERY features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_BATTERY][PROVIDERS][RAPIDS] : Feature Units Description countdischarge episodes Number of discharging episodes. sumdurationdischarge minutes The total duration of all discharging episodes. countcharge episodes Number of battery charging episodes. sumdurationcharge minutes The total duration of all charging episodes. avgconsumptionrate episodes/minutes The average of all episodes\u2019 consumption rates. An episode\u2019s consumption rate is defined as the ratio between its battery delta and duration maxconsumptionrate episodes/minutes The highest of all episodes\u2019 consumption rates. An episode\u2019s consumption rate is defined as the ratio between its battery delta and duration Assumptions/Observations We convert battery data collected with iOS client v1 (autodetected because battery status 4 do not exist) to match Android battery format: we swap status 3 for 5 and 1 for 3 We group battery data into discharge or charge episodes considering any contiguous rows with consecutive reductions or increases of the battery level if they are logged within [EPISODE_THRESHOLD_BETWEEN_ROWS] minutes from each other.","title":"RAPIDS provider"},{"location":"features/phone-bluetooth/","text":"Phone Bluetooth \u00b6 Sensor parameters description for [PHONE_BLUETOOTH] : Key Description [TABLE] Database table where the bluetooth data is stored RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_bluetooth_raw.csv - data/raw/ { pid } /phone_bluetooth_with_datetime.csv - data/interim/ { pid } /phone_bluetooth_features/phone_bluetooth_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_bluetooth.csv \" Parameters description for [PHONE_BLUETOOTH][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_BLUETOOTH features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_BLUETOOTH][PROVIDERS][RAPIDS] : Feature Units Description countscans devices Number of scanned devices during a time_segment , a device can be detected multiple times over time and these appearances are counted separately uniquedevices devices Number of unique devices during a time_segment as identified by their hardware ( bt_address ) address countscansmostuniquedevice scans Number of scans of the most scanned device during a time_segment across the whole monitoring period Assumptions/Observations NA","title":"Phone Bluetooth"},{"location":"features/phone-bluetooth/#phone-bluetooth","text":"Sensor parameters description for [PHONE_BLUETOOTH] : Key Description [TABLE] Database table where the bluetooth data is stored","title":"Phone Bluetooth"},{"location":"features/phone-bluetooth/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_bluetooth_raw.csv - data/raw/ { pid } /phone_bluetooth_with_datetime.csv - data/interim/ { pid } /phone_bluetooth_features/phone_bluetooth_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_bluetooth.csv \" Parameters description for [PHONE_BLUETOOTH][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_BLUETOOTH features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_BLUETOOTH][PROVIDERS][RAPIDS] : Feature Units Description countscans devices Number of scanned devices during a time_segment , a device can be detected multiple times over time and these appearances are counted separately uniquedevices devices Number of unique devices during a time_segment as identified by their hardware ( bt_address ) address countscansmostuniquedevice scans Number of scans of the most scanned device during a time_segment across the whole monitoring period Assumptions/Observations NA","title":"RAPIDS provider"},{"location":"features/phone-calls/","text":"Phone Calls \u00b6 Sensor parameters description for [PHONE_CALLS] : Key Description [TABLE] Database table where the calls data is stored RAPIDS Provider \u00b6 Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_calls_raw.csv - data/raw/ { pid } /phone_calls_with_datetime.csv - data/raw/ { pid } /phone_calls_with_datetime_unified.csv - data/interim/ { pid } /phone_calls_features/phone_calls_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_calls.csv Parameters description for [PHONE_CALLS][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_CALLS features from the RAPIDS provider [CALL_TYPES] The particular call_type that will be analyzed. The options for this parameter are incoming, outgoing or missed. [FEATURES] Features to be computed for outgoing , incoming , and missed calls. Note that the same features are available for both incoming and outgoing calls, while missed calls has its own set of features. See the tables below. Features description for [PHONE_CALLS][PROVIDERS][RAPIDS] incoming and outgoing calls: Feature Units Description count calls Number of calls of a particular call_type occurred during a particular time_segment . distinctcontacts contacts Number of distinct contacts that are associated with a particular call_type for a particular time_segment meanduration seconds The mean duration of all calls of a particular call_type during a particular time_segment . sumduration seconds The sum of the duration of all calls of a particular call_type during a particular time_segment . minduration seconds The duration of the shortest call of a particular call_type during a particular time_segment . maxduration seconds The duration of the longest call of a particular call_type during a particular time_segment . stdduration seconds The standard deviation of the duration of all the calls of a particular call_type during a particular time_segment . modeduration seconds The mode of the duration of all the calls of a particular call_type during a particular time_segment . entropyduration nats The estimate of the Shannon entropy for the the duration of all the calls of a particular call_type during a particular time_segment . timefirstcall minutes The time in minutes between 12:00am (midnight) and the first call of call_type . timelastcall minutes The time in minutes between 12:00am (midnight) and the last call of call_type . countmostfrequentcontact calls The number of calls of a particular call_type during a particular time_segment of the most frequent contact throughout the monitored period. Features description for [PHONE_CALLS][PROVIDERS][RAPIDS] missed calls: Feature Units Description count calls Number of missed calls that occurred during a particular time_segment . distinctcontacts contacts Number of distinct contacts that are associated with missed calls for a particular time_segment timefirstcall minutes The time in hours from 12:00am (Midnight) that the first missed call occurred. timelastcall minutes The time in hours from 12:00am (Midnight) that the last missed call occurred. countmostfrequentcontact calls The number of missed calls during a particular time_segment of the most frequent contact throughout the monitored period. Assumptions/Observations Traces for iOS calls are unique even for the same contact calling a participant more than once which renders countmostfrequentcontact meaningless and distinctcontacts equal to the total number of traces. [CALL_TYPES] and [FEATURES] keys in config.yaml need to match. For example, [CALL_TYPES] outgoing matches the [FEATURES] key outgoing iOS calls data is transformed to match Android calls data format. See our algorithm","title":"Phone Calls"},{"location":"features/phone-calls/#phone-calls","text":"Sensor parameters description for [PHONE_CALLS] : Key Description [TABLE] Database table where the calls data is stored","title":"Phone Calls"},{"location":"features/phone-calls/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_calls_raw.csv - data/raw/ { pid } /phone_calls_with_datetime.csv - data/raw/ { pid } /phone_calls_with_datetime_unified.csv - data/interim/ { pid } /phone_calls_features/phone_calls_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_calls.csv Parameters description for [PHONE_CALLS][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_CALLS features from the RAPIDS provider [CALL_TYPES] The particular call_type that will be analyzed. The options for this parameter are incoming, outgoing or missed. [FEATURES] Features to be computed for outgoing , incoming , and missed calls. Note that the same features are available for both incoming and outgoing calls, while missed calls has its own set of features. See the tables below. Features description for [PHONE_CALLS][PROVIDERS][RAPIDS] incoming and outgoing calls: Feature Units Description count calls Number of calls of a particular call_type occurred during a particular time_segment . distinctcontacts contacts Number of distinct contacts that are associated with a particular call_type for a particular time_segment meanduration seconds The mean duration of all calls of a particular call_type during a particular time_segment . sumduration seconds The sum of the duration of all calls of a particular call_type during a particular time_segment . minduration seconds The duration of the shortest call of a particular call_type during a particular time_segment . maxduration seconds The duration of the longest call of a particular call_type during a particular time_segment . stdduration seconds The standard deviation of the duration of all the calls of a particular call_type during a particular time_segment . modeduration seconds The mode of the duration of all the calls of a particular call_type during a particular time_segment . entropyduration nats The estimate of the Shannon entropy for the the duration of all the calls of a particular call_type during a particular time_segment . timefirstcall minutes The time in minutes between 12:00am (midnight) and the first call of call_type . timelastcall minutes The time in minutes between 12:00am (midnight) and the last call of call_type . countmostfrequentcontact calls The number of calls of a particular call_type during a particular time_segment of the most frequent contact throughout the monitored period. Features description for [PHONE_CALLS][PROVIDERS][RAPIDS] missed calls: Feature Units Description count calls Number of missed calls that occurred during a particular time_segment . distinctcontacts contacts Number of distinct contacts that are associated with missed calls for a particular time_segment timefirstcall minutes The time in hours from 12:00am (Midnight) that the first missed call occurred. timelastcall minutes The time in hours from 12:00am (Midnight) that the last missed call occurred. countmostfrequentcontact calls The number of missed calls during a particular time_segment of the most frequent contact throughout the monitored period. Assumptions/Observations Traces for iOS calls are unique even for the same contact calling a participant more than once which renders countmostfrequentcontact meaningless and distinctcontacts equal to the total number of traces. [CALL_TYPES] and [FEATURES] keys in config.yaml need to match. For example, [CALL_TYPES] outgoing matches the [FEATURES] key outgoing iOS calls data is transformed to match Android calls data format. See our algorithm","title":"RAPIDS Provider"},{"location":"features/phone-conversation/","text":"Phone Conversation \u00b6 Sensor parameters description for [PHONE_CONVERSATION] : Key Description [TABLE][ANDROID] Database table where the conversation data from Android devices is stored (the AWARE client saves this data on different tables for Android and iOS) [TABLE][IOS] Database table where the conversation data from iOS devices is stored (the AWARE client saves this data on different tables for Android and iOS) RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_conversation_raw.csv - data/raw/ { pid } /phone_conversation_with_datetime.csv - data/raw/ { pid } /phone_conversation_with_datetime_unified.csv - data/interim/ { pid } /phone_conversation_features/phone_conversation_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_conversation.csv Parameters description for [PHONE_CONVERSATION][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_CONVERSATION features from the RAPIDS provider [FEATURES] Features to be computed, see table below [RECORDING_MINUTES] Minutes the plugin was recording audio (default 1 min) [PAUSED_MINUTES] Minutes the plugin was NOT recording audio (default 3 min) Features description for [PHONE_CONVERSATION][PROVIDERS][RAPIDS] : Feature Units Description minutessilence minutes Minutes labeled as silence minutesnoise minutes Minutes labeled as noise minutesvoice minutes Minutes labeled as voice minutesunknown minutes Minutes labeled as unknown sumconversationduration minutes Total duration of all conversations maxconversationduration minutes Longest duration of all conversations minconversationduration minutes Shortest duration of all conversations avgconversationduration minutes Average duration of all conversations sdconversationduration minutes Standard Deviation of the duration of all conversations timefirstconversation minutes Minutes since midnight when the first conversation for a time segment was detected timelastconversation minutes Minutes since midnight when the last conversation for a time segment was detected noisesumenergy L2-norm Sum of all energy values when inference is noise noiseavgenergy L2-norm Average of all energy values when inference is noise noisesdenergy L2-norm Standard Deviation of all energy values when inference is noise noiseminenergy L2-norm Minimum of all energy values when inference is noise noisemaxenergy L2-norm Maximum of all energy values when inference is noise voicesumenergy L2-norm Sum of all energy values when inference is voice voiceavgenergy L2-norm Average of all energy values when inference is voice voicesdenergy L2-norm Standard Deviation of all energy values when inference is voice voiceminenergy L2-norm Minimum of all energy values when inference is voice voicemaxenergy L2-norm Maximum of all energy values when inference is voice silencesensedfraction - Ratio between minutessilence and the sum of (minutessilence, minutesnoise, minutesvoice, minutesunknown) noisesensedfraction - Ratio between minutesnoise and the sum of (minutessilence, minutesnoise, minutesvoice, minutesunknown) voicesensedfraction - Ratio between minutesvoice and the sum of (minutessilence, minutesnoise, minutesvoice, minutesunknown) unknownsensedfraction - Ratio between minutesunknown and the sum of (minutessilence, minutesnoise, minutesvoice, minutesunknown) silenceexpectedfraction - Ration between minutessilence and the number of minutes that in theory should have been sensed based on the record and pause cycle of the plugin (1440 / recordingMinutes+pausedMinutes) noiseexpectedfraction - Ration between minutesnoise and the number of minutes that in theory should have been sensed based on the record and pause cycle of the plugin (1440 / recordingMinutes+pausedMinutes) voiceexpectedfraction - Ration between minutesvoice and the number of minutes that in theory should have been sensed based on the record and pause cycle of the plugin (1440 / recordingMinutes+pausedMinutes) unknownexpectedfraction - Ration between minutesunknown and the number of minutes that in theory should have been sensed based on the record and pause cycle of the plugin (1440 / recordingMinutes+pausedMinutes) Assumptions/Observations The timestamp of conversation rows in iOS is in seconds so we convert it to milliseconds to match Android\u2019s format","title":"Phone Conversation"},{"location":"features/phone-conversation/#phone-conversation","text":"Sensor parameters description for [PHONE_CONVERSATION] : Key Description [TABLE][ANDROID] Database table where the conversation data from Android devices is stored (the AWARE client saves this data on different tables for Android and iOS) [TABLE][IOS] Database table where the conversation data from iOS devices is stored (the AWARE client saves this data on different tables for Android and iOS)","title":"Phone Conversation"},{"location":"features/phone-conversation/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_conversation_raw.csv - data/raw/ { pid } /phone_conversation_with_datetime.csv - data/raw/ { pid } /phone_conversation_with_datetime_unified.csv - data/interim/ { pid } /phone_conversation_features/phone_conversation_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_conversation.csv Parameters description for [PHONE_CONVERSATION][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_CONVERSATION features from the RAPIDS provider [FEATURES] Features to be computed, see table below [RECORDING_MINUTES] Minutes the plugin was recording audio (default 1 min) [PAUSED_MINUTES] Minutes the plugin was NOT recording audio (default 3 min) Features description for [PHONE_CONVERSATION][PROVIDERS][RAPIDS] : Feature Units Description minutessilence minutes Minutes labeled as silence minutesnoise minutes Minutes labeled as noise minutesvoice minutes Minutes labeled as voice minutesunknown minutes Minutes labeled as unknown sumconversationduration minutes Total duration of all conversations maxconversationduration minutes Longest duration of all conversations minconversationduration minutes Shortest duration of all conversations avgconversationduration minutes Average duration of all conversations sdconversationduration minutes Standard Deviation of the duration of all conversations timefirstconversation minutes Minutes since midnight when the first conversation for a time segment was detected timelastconversation minutes Minutes since midnight when the last conversation for a time segment was detected noisesumenergy L2-norm Sum of all energy values when inference is noise noiseavgenergy L2-norm Average of all energy values when inference is noise noisesdenergy L2-norm Standard Deviation of all energy values when inference is noise noiseminenergy L2-norm Minimum of all energy values when inference is noise noisemaxenergy L2-norm Maximum of all energy values when inference is noise voicesumenergy L2-norm Sum of all energy values when inference is voice voiceavgenergy L2-norm Average of all energy values when inference is voice voicesdenergy L2-norm Standard Deviation of all energy values when inference is voice voiceminenergy L2-norm Minimum of all energy values when inference is voice voicemaxenergy L2-norm Maximum of all energy values when inference is voice silencesensedfraction - Ratio between minutessilence and the sum of (minutessilence, minutesnoise, minutesvoice, minutesunknown) noisesensedfraction - Ratio between minutesnoise and the sum of (minutessilence, minutesnoise, minutesvoice, minutesunknown) voicesensedfraction - Ratio between minutesvoice and the sum of (minutessilence, minutesnoise, minutesvoice, minutesunknown) unknownsensedfraction - Ratio between minutesunknown and the sum of (minutessilence, minutesnoise, minutesvoice, minutesunknown) silenceexpectedfraction - Ration between minutessilence and the number of minutes that in theory should have been sensed based on the record and pause cycle of the plugin (1440 / recordingMinutes+pausedMinutes) noiseexpectedfraction - Ration between minutesnoise and the number of minutes that in theory should have been sensed based on the record and pause cycle of the plugin (1440 / recordingMinutes+pausedMinutes) voiceexpectedfraction - Ration between minutesvoice and the number of minutes that in theory should have been sensed based on the record and pause cycle of the plugin (1440 / recordingMinutes+pausedMinutes) unknownexpectedfraction - Ration between minutesunknown and the number of minutes that in theory should have been sensed based on the record and pause cycle of the plugin (1440 / recordingMinutes+pausedMinutes) Assumptions/Observations The timestamp of conversation rows in iOS is in seconds so we convert it to milliseconds to match Android\u2019s format","title":"RAPIDS provider"},{"location":"features/phone-data-yield/","text":"Phone Data Yield \u00b6 This is a combinatorial sensor which means that we use the data from multiple sensors to extract data yield features. Data yield features can be used to remove rows ( time segments ) that do not contain enough data. You should decide what is your \u201cenough\u201d threshold depending on the type of sensors you collected (frequency vs event based, e.g. acceleroemter vs calls), the length of your study, and the rates of missing data that your analysis could handle. Why is data yield important? Imagine that you want to extract PHONE_CALL features on daily segments ( 00:00 to 23:59 ). Let\u2019s say that on day 1 the phone logged 10 calls and 23 hours of data from other sensors and on day 2 the phone logged 10 calls and only 2 hours of data from other sensors. It\u2019s more likely that other calls were placed on the 22 hours of data that you didn\u2019t log on day 2 than on the 1 hour of data you didn\u2019t log on day 1, and so including day 2 in your analysis could bias your results. Sensor parameters description for [PHONE_DATA_YIELD] : Key Description [SENSORS] One or more phone sensor config keys (e.g. PHONE_MESSAGE ). The more keys you include the more accurately RAPIDS can approximate the time an smartphone was sensing data. The supported phone sensors you can include in this list are outlined below ( do NOT include Fitbit sensors ). Supported phone sensors for [PHONE_DATA_YIELD][SENSORS] PHONE_ACCELEROMETER PHONE_ACTIVITY_RECOGNITION PHONE_APPLICATIONS_FOREGROUND PHONE_BATTERY PHONE_BLUETOOTH PHONE_CALLS PHONE_CONVERSATION PHONE_MESSAGES PHONE_LIGHT PHONE_LOCATIONS PHONE_SCREEN PHONE_WIFI_VISIBLE PHONE_WIFI_CONNECTED RAPIDS provider \u00b6 Before explaining the data yield features, let\u2019s define the following relevant concepts: A valid minute is any 60 second window when any phone sensor logged at least 1 row of data A valid hour is any 60 minute window with at least X valid minutes. The X or threshold is given by [MINUTE_RATIO_THRESHOLD_FOR_VALID_YIELDED_HOURS] The timestamps of all sensors are concatenated and then grouped per time segment. Minute and hour windows are created from the beginning of each time segment instance and these windows are marked as valid based on the definitions above. The duration of each time segment is taken into account to compute the features described below. Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } / { sensor } _raw.csv # one for every [PHONE_DATA_YIELD][SENSORS] - data/interim/ { pid } /phone_yielded_timestamps.csv - data/interim/ { pid } /phone_yielded_timestamps_with_datetime.csv - data/interim/ { pid } /phone_data_yield_features/phone_data_yield_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_data_yield.csv Parameters description for [PHONE_DATA_YIELD][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_DATA_YIELD features from the RAPIDS provider [FEATURES] Features to be computed, see table below [MINUTE_RATIO_THRESHOLD_FOR_VALID_YIELDED_HOURS] The proportion [0.0 ,1.0] of valid minutes in a 60-minute window necessary to flag that window as valid. Features description for [PHONE_DATA_YIELD][PROVIDERS][RAPIDS] : Feature Units Description ratiovalidyieldedminutes rows The ratio between the number of valid minutes and the duration in minutes of a time segment. ratiovalidyieldedhours lux The ratio between the number of valid hours and the duration in hours of a time segment. If the time segment is shorter than 1 hour this feature will always be 1. Assumptions/Observations We recommend using ratiovalidyieldedminutes on time segments that are shorter than two or three hours and ratiovalidyieldedhours for longer segments. This is because relying on yielded minutes only can be misleading when a big chunk of those missing minutes are clustered together. For example, let\u2019s assume we are working with a 24-hour time segment that is missing 12 hours of data. Two extreme cases can occur: the 12 missing hours are from the beginning of the segment or 30 minutes could be missing from every hour (24 * 30 minutes = 12 hours). ratiovalidyieldedminutes would be 0.5 for both a and b (hinting the missing circumstances are similar). However, ratiovalidyieldedhours would be 0.5 for a and 1.0 for b if [MINUTE_RATIO_THRESHOLD_FOR_VALID_YIELDED_HOURS] is between [0.0 and 0.49] (hinting that the missing circumstances might be more favorable for b . In other words, sensed data for b is more evenly spread compared to a .","title":"Phone Data Yield"},{"location":"features/phone-data-yield/#phone-data-yield","text":"This is a combinatorial sensor which means that we use the data from multiple sensors to extract data yield features. Data yield features can be used to remove rows ( time segments ) that do not contain enough data. You should decide what is your \u201cenough\u201d threshold depending on the type of sensors you collected (frequency vs event based, e.g. acceleroemter vs calls), the length of your study, and the rates of missing data that your analysis could handle. Why is data yield important? Imagine that you want to extract PHONE_CALL features on daily segments ( 00:00 to 23:59 ). Let\u2019s say that on day 1 the phone logged 10 calls and 23 hours of data from other sensors and on day 2 the phone logged 10 calls and only 2 hours of data from other sensors. It\u2019s more likely that other calls were placed on the 22 hours of data that you didn\u2019t log on day 2 than on the 1 hour of data you didn\u2019t log on day 1, and so including day 2 in your analysis could bias your results. Sensor parameters description for [PHONE_DATA_YIELD] : Key Description [SENSORS] One or more phone sensor config keys (e.g. PHONE_MESSAGE ). The more keys you include the more accurately RAPIDS can approximate the time an smartphone was sensing data. The supported phone sensors you can include in this list are outlined below ( do NOT include Fitbit sensors ). Supported phone sensors for [PHONE_DATA_YIELD][SENSORS] PHONE_ACCELEROMETER PHONE_ACTIVITY_RECOGNITION PHONE_APPLICATIONS_FOREGROUND PHONE_BATTERY PHONE_BLUETOOTH PHONE_CALLS PHONE_CONVERSATION PHONE_MESSAGES PHONE_LIGHT PHONE_LOCATIONS PHONE_SCREEN PHONE_WIFI_VISIBLE PHONE_WIFI_CONNECTED","title":"Phone Data Yield"},{"location":"features/phone-data-yield/#rapids-provider","text":"Before explaining the data yield features, let\u2019s define the following relevant concepts: A valid minute is any 60 second window when any phone sensor logged at least 1 row of data A valid hour is any 60 minute window with at least X valid minutes. The X or threshold is given by [MINUTE_RATIO_THRESHOLD_FOR_VALID_YIELDED_HOURS] The timestamps of all sensors are concatenated and then grouped per time segment. Minute and hour windows are created from the beginning of each time segment instance and these windows are marked as valid based on the definitions above. The duration of each time segment is taken into account to compute the features described below. Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } / { sensor } _raw.csv # one for every [PHONE_DATA_YIELD][SENSORS] - data/interim/ { pid } /phone_yielded_timestamps.csv - data/interim/ { pid } /phone_yielded_timestamps_with_datetime.csv - data/interim/ { pid } /phone_data_yield_features/phone_data_yield_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_data_yield.csv Parameters description for [PHONE_DATA_YIELD][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_DATA_YIELD features from the RAPIDS provider [FEATURES] Features to be computed, see table below [MINUTE_RATIO_THRESHOLD_FOR_VALID_YIELDED_HOURS] The proportion [0.0 ,1.0] of valid minutes in a 60-minute window necessary to flag that window as valid. Features description for [PHONE_DATA_YIELD][PROVIDERS][RAPIDS] : Feature Units Description ratiovalidyieldedminutes rows The ratio between the number of valid minutes and the duration in minutes of a time segment. ratiovalidyieldedhours lux The ratio between the number of valid hours and the duration in hours of a time segment. If the time segment is shorter than 1 hour this feature will always be 1. Assumptions/Observations We recommend using ratiovalidyieldedminutes on time segments that are shorter than two or three hours and ratiovalidyieldedhours for longer segments. This is because relying on yielded minutes only can be misleading when a big chunk of those missing minutes are clustered together. For example, let\u2019s assume we are working with a 24-hour time segment that is missing 12 hours of data. Two extreme cases can occur: the 12 missing hours are from the beginning of the segment or 30 minutes could be missing from every hour (24 * 30 minutes = 12 hours). ratiovalidyieldedminutes would be 0.5 for both a and b (hinting the missing circumstances are similar). However, ratiovalidyieldedhours would be 0.5 for a and 1.0 for b if [MINUTE_RATIO_THRESHOLD_FOR_VALID_YIELDED_HOURS] is between [0.0 and 0.49] (hinting that the missing circumstances might be more favorable for b . In other words, sensed data for b is more evenly spread compared to a .","title":"RAPIDS provider"},{"location":"features/phone-light/","text":"Phone Light \u00b6 Sensor parameters description for [PHONE_LIGHT] : Key Description [TABLE] Database table where the light data is stored RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_light_raw.csv - data/raw/ { pid } /phone_light_with_datetime.csv - data/interim/ { pid } /phone_light_features/phone_light_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_light.csv Parameters description for [PHONE_LIGHT][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_LIGHT features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_LIGHT][PROVIDERS][RAPIDS] : Feature Units Description count rows Number light sensor rows recorded. maxlux lux The maximum ambient luminance. minlux lux The minimum ambient luminance. avglux lux The average ambient luminance. medianlux lux The median ambient luminance. stdlux lux The standard deviation of ambient luminance. Assumptions/Observations NA","title":"Phone Light"},{"location":"features/phone-light/#phone-light","text":"Sensor parameters description for [PHONE_LIGHT] : Key Description [TABLE] Database table where the light data is stored","title":"Phone Light"},{"location":"features/phone-light/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_light_raw.csv - data/raw/ { pid } /phone_light_with_datetime.csv - data/interim/ { pid } /phone_light_features/phone_light_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_light.csv Parameters description for [PHONE_LIGHT][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_LIGHT features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_LIGHT][PROVIDERS][RAPIDS] : Feature Units Description count rows Number light sensor rows recorded. maxlux lux The maximum ambient luminance. minlux lux The minimum ambient luminance. avglux lux The average ambient luminance. medianlux lux The median ambient luminance. stdlux lux The standard deviation of ambient luminance. Assumptions/Observations NA","title":"RAPIDS provider"},{"location":"features/phone-locations/","text":"Phone Locations \u00b6 Sensor parameters description for [PHONE_LOCATIONS] : Key Description [TABLE] Database table where the location data is stored [LOCATIONS_TO_USE] Type of location data to use, one of ALL , GPS or FUSED_RESAMPLED . This filter is based on the provider column of the AWARE locations table, ALL includes every row, GPS only includes rows where provider is gps, and FUSED_RESAMPLED only includes rows where provider is fused after being resampled. [FUSED_RESAMPLED_CONSECUTIVE_THRESHOLD] if FUSED_RESAMPLED is used, the original fused data has to be resampled, a location row will be resampled to the next valid timestamp (see the Assumptions/Observations below) only if the time difference between them is less or equal than this threshold (in minutes). [FUSED_RESAMPLED_TIME_SINCE_VALID_LOCATION] if FUSED_RESAMPLED is used, the original fused data has to be resampled, a location row will be resampled at most for this long (in minutes) Assumptions/Observations Types of location data to use AWARE Android and iOS clients can collect location coordinates through the phone's GPS, the network cellular towers around the phone or Google's fused location API. If you want to use only the GPS provider set [LOCATIONS_TO_USE] to GPS , if you want to use all providers (not recommended due to the difference in accuracy) set [LOCATIONS_TO_USE] to ALL , if your AWARE client was configured to use fused location only or want to focus only on this provider, set [LOCATIONS_TO_USE] to RESAMPLE_FUSED . RESAMPLE_FUSED takes the original fused location coordinates and replicates each pair forward in time as long as the phone was sensing data as indicated by the joined timestamps of [PHONE_DATA_YIELD][SENSORS] , this is done because Google's API only logs a new location coordinate pair when it is sufficiently different in time or space from the previous one. There are two parameters associated with resampling fused location. FUSED_RESAMPLED_CONSECUTIVE_THRESHOLD (in minutes, default 30) controls the maximum gap between any two coordinate pairs to replicate the last known pair (for example, participant A's phone did not collect data between 10.30am and 10:50am and between 11:05am and 11:40am, the last known coordinate pair will be replicated during the first period but not the second, in other words, we assume that we cannot longer guarantee the participant stayed at the last known location if the phone did not sense data for more than 30 minutes). FUSED_RESAMPLED_TIME_SINCE_VALID_LOCATION (in minutes, default 720 or 12 hours) stops the last known fused location from being replicated longer that this threshold even if the phone was sensing data continuously (for example, participant A went home at 9pm and their phone was sensing data without gaps until 11am the next morning, the last known location will only be replicated until 9am). If you have suggestions to modify or improve this resampling, let us know. BARNETT provider \u00b6 These features are based on the original open-source implementation by Barnett et al and some features created by Canzian et al . Available time segments and platforms Available only for segments that start at 00:00:00 and end at 23:59:59 of the same day (daily segments) Available for Android and iOS File Sequence - data/raw/ { pid } /phone_locations_raw.csv - data/interim/ { pid } /phone_locations_processed.csv - data/interim/ { pid } /phone_locations_processed_with_datetime.csv - data/interim/ { pid } /phone_locations_features/phone_locations_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_locations.csv Parameters description for [PHONE_LOCATIONS][PROVIDERS][BARNETT] : Key Description [COMPUTE] Set to True to extract PHONE_LOCATIONS features from the BARNETT provider [FEATURES] Features to be computed, see table below [ACCURACY_LIMIT] An integer in meters, any location rows with an accuracy higher than this will be dropped. This number means there\u2019s a 68% probability the true location is within this radius [TIMEZONE] Timezone where the location data was collected. By default points to the one defined in the Configuration [MINUTES_DATA_USED] Set to True to include an extra column in the final location feature file containing the number of minutes used to compute the features on each time segment. Use this for quality control purposes, the more data minutes exist for a period, the more reliable its features should be. For fused location, a single minute can contain more than one coordinate pair if the participant is moving fast enough. Features description for [PHONE_LOCATIONS][PROVIDERS][BARNETT] adapted from Beiwe Summary Statistics : Feature Units Description hometime minutes Time at home. Time spent at home in minutes. Home is the most visited significant location between 8 pm and 8 am including any pauses within a 200-meter radius. disttravelled meters Total distance travelled over a day (flights). rog meters The Radius of Gyration (rog) is a measure in meters of the area covered by a person over a day. A centroid is calculated for all the places (pauses) visited during a day and a weighted distance between all the places and that centroid is computed. The weights are proportional to the time spent in each place. maxdiam meters The maximum diameter is the largest distance between any two pauses. maxhomedist meters The maximum distance from home in meters. siglocsvisited locations The number of significant locations visited during the day. Significant locations are computed using k-means clustering over pauses found in the whole monitoring period. The number of clusters is found iterating k from 1 to 200 stopping until the centroids of two significant locations are within 400 meters of one another. avgflightlen meters Mean length of all flights. stdflightlen meters Standard deviation of the length of all flights. avgflightdur seconds Mean duration of all flights. stdflightdur seconds The standard deviation of the duration of all flights. probpause - The fraction of a day spent in a pause (as opposed to a flight) siglocentropy nats Shannon\u2019s entropy measurement based on the proportion of time spent at each significant location visited during a day. circdnrtn - A continuous metric quantifying a person\u2019s circadian routine that can take any value between 0 and 1, where 0 represents a daily routine completely different from any other sensed days and 1 a routine the same as every other sensed day. wkenddayrtn - Same as circdnrtn but computed separately for weekends and weekdays. Assumptions/Observations Barnett's et al features These features are based on a Pause-Flight model. A pause is defined as a mobiity trace (location pings) within a certain duration and distance (by default 300 seconds and 60 meters). A flight is any mobility trace between two pauses. Data is resampled and imputed before the features are computed. See Barnett et al for more information. In RAPIDS we only expose two parameters for these features (timezone and accuracy limit). You can change other parameters in src/features/phone_locations/barnett/library/MobilityFeatures.R . Significant Locations Significant locations are determined using K-means clustering on pauses longer than 10 minutes. The number of clusters (K) is increased until no two clusters are within 400 meters from each other. After this, pauses within a certain range of a cluster (200 meters by default) will count as a visit to that significant location. This description was adapted from the Supplementary Materials of Barnett et al . The Circadian Calculation For a detailed description of how this is calculated, see Canzian et al . DORYAB provider \u00b6 These features are based on the original implementation by Doryab et al. . Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_locations_raw.csv - data/interim/ { pid } /phone_locations_processed.csv - data/interim/ { pid } /phone_locations_processed_with_datetime.csv - data/interim/ { pid } /phone_locations_features/phone_locations_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_locations.csv Parameters description for [PHONE_LOCATIONS][PROVIDERS][BARNETT] : Key Description [COMPUTE] Set to True to extract PHONE_LOCATIONS features from the BARNETT provider [FEATURES] Features to be computed, see table below [DBSCAN_EPS] The maximum distance in meters between two samples for one to be considered as in the neighborhood of the other. This is not a maximum bound on the distances of points within a cluster. This is the most important DBSCAN parameter to choose appropriately for your data set and distance function. [DBSCAN_MINSAMPLES] The number of samples (or total weight) in a neighborhood for a point to be considered as a core point of a cluster. This includes the point itself. [THRESHOLD_STATIC] It is the threshold value in km/hr which labels a row as Static or Moving. [MAXIMUM_GAP_ALLOWED] The maximum gap (in seconds) allowed between any two consecutive rows for them to be considered part of the same displacement. If this threshold is too high, it can throw speed and distance calculations off for periods when the the phone was not sensing. [MINUTES_DATA_USED] Set to True to include an extra column in the final location feature file containing the number of minutes used to compute the features on each time segment. Use this for quality control purposes, the more data minutes exist for a period, the more reliable its features should be. For fused location, a single minute can contain more than one coordinate pair if the participant is moving fast enough. [SAMPLING_FREQUENCY] Expected time difference between any two location rows in minutes. If set to 0 , the sampling frequency will be inferred automatically as the median of all the differences between any two consecutive row timestamps (recommended if you are using FUSED_RESAMPLED data). This parameter impacts all the time calculations. Features description for [PHONE_LOCATIONS][PROVIDERS][BARNETT] : Feature Units Description locationvariance \\(meters^2\\) The sum of the variances of the latitude and longitude columns. loglocationvariance - Log of the sum of the variances of the latitude and longitude columns. totaldistance meters Total distance travelled in a time segment using the haversine formula. averagespeed km/hr Average speed in a time segment considering only the instances labeled as Moving. varspeed km/hr Speed variance in a time segment considering only the instances labeled as Moving. circadianmovement - \"It encodes the extent to which a person\u2019s location patterns follow a 24-hour circadian cycle.\" Doryab et al. . numberofsignificantplaces places Number of significant locations visited. It is calculated using the DBSCAN clustering algorithm which takes in EPS and MIN_SAMPLES as parameters to identify clusters. Each cluster is a significant place. numberlocationtransitions transitions Number of movements between any two clusters in a time segment. radiusgyration meters Quantifies the area covered by a participant timeattop1location minutes Time spent at the most significant location. timeattop2location minutes Time spent at the 2 nd most significant location. timeattop3location minutes Time spent at the 3 rd most significant location. movingtostaticratio - Ratio between the number of rows labeled Moving versus Static outlierstimepercent - Ratio between the number of rows that belong to non-significant clusters divided by the total number of rows in a time segment. maxlengthstayatclusters minutes Maximum time spent in a cluster (significant location). minlengthstayatclusters minutes Minimum time spent in a cluster (significant location). meanlengthstayatclusters minutes Average time spent in a cluster (significant location). stdlengthstayatclusters minutes Standard deviation of time spent in a cluster (significant location). locationentropy nats Shannon Entropy computed over the row count of each cluster (significant location), it will be higher the more rows belong to a cluster (i.e. the more time a participant spent at a significant location). normalizedlocationentropy nats Shannon Entropy computed over the row count of each cluster (significant location) divided by the number of clusters, it will be higher the more rows belong to a cluster (i.e. the more time a participant spent at a significant location). Assumptions/Observations Significant Locations Identified Significant locations are determined using DBSCAN clustering on locations that a patient visit over the course of the period of data collection. The Circadian Calculation For a detailed description of how this is calculated, see Canzian et al .","title":"Phone Locations"},{"location":"features/phone-locations/#phone-locations","text":"Sensor parameters description for [PHONE_LOCATIONS] : Key Description [TABLE] Database table where the location data is stored [LOCATIONS_TO_USE] Type of location data to use, one of ALL , GPS or FUSED_RESAMPLED . This filter is based on the provider column of the AWARE locations table, ALL includes every row, GPS only includes rows where provider is gps, and FUSED_RESAMPLED only includes rows where provider is fused after being resampled. [FUSED_RESAMPLED_CONSECUTIVE_THRESHOLD] if FUSED_RESAMPLED is used, the original fused data has to be resampled, a location row will be resampled to the next valid timestamp (see the Assumptions/Observations below) only if the time difference between them is less or equal than this threshold (in minutes). [FUSED_RESAMPLED_TIME_SINCE_VALID_LOCATION] if FUSED_RESAMPLED is used, the original fused data has to be resampled, a location row will be resampled at most for this long (in minutes) Assumptions/Observations Types of location data to use AWARE Android and iOS clients can collect location coordinates through the phone's GPS, the network cellular towers around the phone or Google's fused location API. If you want to use only the GPS provider set [LOCATIONS_TO_USE] to GPS , if you want to use all providers (not recommended due to the difference in accuracy) set [LOCATIONS_TO_USE] to ALL , if your AWARE client was configured to use fused location only or want to focus only on this provider, set [LOCATIONS_TO_USE] to RESAMPLE_FUSED . RESAMPLE_FUSED takes the original fused location coordinates and replicates each pair forward in time as long as the phone was sensing data as indicated by the joined timestamps of [PHONE_DATA_YIELD][SENSORS] , this is done because Google's API only logs a new location coordinate pair when it is sufficiently different in time or space from the previous one. There are two parameters associated with resampling fused location. FUSED_RESAMPLED_CONSECUTIVE_THRESHOLD (in minutes, default 30) controls the maximum gap between any two coordinate pairs to replicate the last known pair (for example, participant A's phone did not collect data between 10.30am and 10:50am and between 11:05am and 11:40am, the last known coordinate pair will be replicated during the first period but not the second, in other words, we assume that we cannot longer guarantee the participant stayed at the last known location if the phone did not sense data for more than 30 minutes). FUSED_RESAMPLED_TIME_SINCE_VALID_LOCATION (in minutes, default 720 or 12 hours) stops the last known fused location from being replicated longer that this threshold even if the phone was sensing data continuously (for example, participant A went home at 9pm and their phone was sensing data without gaps until 11am the next morning, the last known location will only be replicated until 9am). If you have suggestions to modify or improve this resampling, let us know.","title":"Phone Locations"},{"location":"features/phone-locations/#barnett-provider","text":"These features are based on the original open-source implementation by Barnett et al and some features created by Canzian et al . Available time segments and platforms Available only for segments that start at 00:00:00 and end at 23:59:59 of the same day (daily segments) Available for Android and iOS File Sequence - data/raw/ { pid } /phone_locations_raw.csv - data/interim/ { pid } /phone_locations_processed.csv - data/interim/ { pid } /phone_locations_processed_with_datetime.csv - data/interim/ { pid } /phone_locations_features/phone_locations_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_locations.csv Parameters description for [PHONE_LOCATIONS][PROVIDERS][BARNETT] : Key Description [COMPUTE] Set to True to extract PHONE_LOCATIONS features from the BARNETT provider [FEATURES] Features to be computed, see table below [ACCURACY_LIMIT] An integer in meters, any location rows with an accuracy higher than this will be dropped. This number means there\u2019s a 68% probability the true location is within this radius [TIMEZONE] Timezone where the location data was collected. By default points to the one defined in the Configuration [MINUTES_DATA_USED] Set to True to include an extra column in the final location feature file containing the number of minutes used to compute the features on each time segment. Use this for quality control purposes, the more data minutes exist for a period, the more reliable its features should be. For fused location, a single minute can contain more than one coordinate pair if the participant is moving fast enough. Features description for [PHONE_LOCATIONS][PROVIDERS][BARNETT] adapted from Beiwe Summary Statistics : Feature Units Description hometime minutes Time at home. Time spent at home in minutes. Home is the most visited significant location between 8 pm and 8 am including any pauses within a 200-meter radius. disttravelled meters Total distance travelled over a day (flights). rog meters The Radius of Gyration (rog) is a measure in meters of the area covered by a person over a day. A centroid is calculated for all the places (pauses) visited during a day and a weighted distance between all the places and that centroid is computed. The weights are proportional to the time spent in each place. maxdiam meters The maximum diameter is the largest distance between any two pauses. maxhomedist meters The maximum distance from home in meters. siglocsvisited locations The number of significant locations visited during the day. Significant locations are computed using k-means clustering over pauses found in the whole monitoring period. The number of clusters is found iterating k from 1 to 200 stopping until the centroids of two significant locations are within 400 meters of one another. avgflightlen meters Mean length of all flights. stdflightlen meters Standard deviation of the length of all flights. avgflightdur seconds Mean duration of all flights. stdflightdur seconds The standard deviation of the duration of all flights. probpause - The fraction of a day spent in a pause (as opposed to a flight) siglocentropy nats Shannon\u2019s entropy measurement based on the proportion of time spent at each significant location visited during a day. circdnrtn - A continuous metric quantifying a person\u2019s circadian routine that can take any value between 0 and 1, where 0 represents a daily routine completely different from any other sensed days and 1 a routine the same as every other sensed day. wkenddayrtn - Same as circdnrtn but computed separately for weekends and weekdays. Assumptions/Observations Barnett's et al features These features are based on a Pause-Flight model. A pause is defined as a mobiity trace (location pings) within a certain duration and distance (by default 300 seconds and 60 meters). A flight is any mobility trace between two pauses. Data is resampled and imputed before the features are computed. See Barnett et al for more information. In RAPIDS we only expose two parameters for these features (timezone and accuracy limit). You can change other parameters in src/features/phone_locations/barnett/library/MobilityFeatures.R . Significant Locations Significant locations are determined using K-means clustering on pauses longer than 10 minutes. The number of clusters (K) is increased until no two clusters are within 400 meters from each other. After this, pauses within a certain range of a cluster (200 meters by default) will count as a visit to that significant location. This description was adapted from the Supplementary Materials of Barnett et al . The Circadian Calculation For a detailed description of how this is calculated, see Canzian et al .","title":"BARNETT provider"},{"location":"features/phone-locations/#doryab-provider","text":"These features are based on the original implementation by Doryab et al. . Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_locations_raw.csv - data/interim/ { pid } /phone_locations_processed.csv - data/interim/ { pid } /phone_locations_processed_with_datetime.csv - data/interim/ { pid } /phone_locations_features/phone_locations_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_locations.csv Parameters description for [PHONE_LOCATIONS][PROVIDERS][BARNETT] : Key Description [COMPUTE] Set to True to extract PHONE_LOCATIONS features from the BARNETT provider [FEATURES] Features to be computed, see table below [DBSCAN_EPS] The maximum distance in meters between two samples for one to be considered as in the neighborhood of the other. This is not a maximum bound on the distances of points within a cluster. This is the most important DBSCAN parameter to choose appropriately for your data set and distance function. [DBSCAN_MINSAMPLES] The number of samples (or total weight) in a neighborhood for a point to be considered as a core point of a cluster. This includes the point itself. [THRESHOLD_STATIC] It is the threshold value in km/hr which labels a row as Static or Moving. [MAXIMUM_GAP_ALLOWED] The maximum gap (in seconds) allowed between any two consecutive rows for them to be considered part of the same displacement. If this threshold is too high, it can throw speed and distance calculations off for periods when the the phone was not sensing. [MINUTES_DATA_USED] Set to True to include an extra column in the final location feature file containing the number of minutes used to compute the features on each time segment. Use this for quality control purposes, the more data minutes exist for a period, the more reliable its features should be. For fused location, a single minute can contain more than one coordinate pair if the participant is moving fast enough. [SAMPLING_FREQUENCY] Expected time difference between any two location rows in minutes. If set to 0 , the sampling frequency will be inferred automatically as the median of all the differences between any two consecutive row timestamps (recommended if you are using FUSED_RESAMPLED data). This parameter impacts all the time calculations. Features description for [PHONE_LOCATIONS][PROVIDERS][BARNETT] : Feature Units Description locationvariance \\(meters^2\\) The sum of the variances of the latitude and longitude columns. loglocationvariance - Log of the sum of the variances of the latitude and longitude columns. totaldistance meters Total distance travelled in a time segment using the haversine formula. averagespeed km/hr Average speed in a time segment considering only the instances labeled as Moving. varspeed km/hr Speed variance in a time segment considering only the instances labeled as Moving. circadianmovement - \"It encodes the extent to which a person\u2019s location patterns follow a 24-hour circadian cycle.\" Doryab et al. . numberofsignificantplaces places Number of significant locations visited. It is calculated using the DBSCAN clustering algorithm which takes in EPS and MIN_SAMPLES as parameters to identify clusters. Each cluster is a significant place. numberlocationtransitions transitions Number of movements between any two clusters in a time segment. radiusgyration meters Quantifies the area covered by a participant timeattop1location minutes Time spent at the most significant location. timeattop2location minutes Time spent at the 2 nd most significant location. timeattop3location minutes Time spent at the 3 rd most significant location. movingtostaticratio - Ratio between the number of rows labeled Moving versus Static outlierstimepercent - Ratio between the number of rows that belong to non-significant clusters divided by the total number of rows in a time segment. maxlengthstayatclusters minutes Maximum time spent in a cluster (significant location). minlengthstayatclusters minutes Minimum time spent in a cluster (significant location). meanlengthstayatclusters minutes Average time spent in a cluster (significant location). stdlengthstayatclusters minutes Standard deviation of time spent in a cluster (significant location). locationentropy nats Shannon Entropy computed over the row count of each cluster (significant location), it will be higher the more rows belong to a cluster (i.e. the more time a participant spent at a significant location). normalizedlocationentropy nats Shannon Entropy computed over the row count of each cluster (significant location) divided by the number of clusters, it will be higher the more rows belong to a cluster (i.e. the more time a participant spent at a significant location). Assumptions/Observations Significant Locations Identified Significant locations are determined using DBSCAN clustering on locations that a patient visit over the course of the period of data collection. The Circadian Calculation For a detailed description of how this is calculated, see Canzian et al .","title":"DORYAB provider"},{"location":"features/phone-messages/","text":"Phone Messages \u00b6 Sensor parameters description for [PHONE_MESSAGES] : Key Description [TABLE] Database table where the messages data is stored RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_messages_raw.csv - data/raw/ { pid } /phone_messages_with_datetime.csv - data/interim/ { pid } /phone_messages_features/phone_messages_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_messages.csv Parameters description for [PHONE_MESSAGES][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_MESSAGES features from the RAPIDS provider [MESSAGES_TYPES] The messages_type that will be analyzed. The options for this parameter are received or sent . [FEATURES] Features to be computed, see table below for [MESSAGES_TYPES] received and sent Features description for [PHONE_MESSAGES][PROVIDERS][RAPIDS] : Feature Units Description count messages Number of messages of type messages_type that occurred during a particular time_segment . distinctcontacts contacts Number of distinct contacts that are associated with a particular messages_type during a particular time_segment . timefirstmessages minutes Number of minutes between 12:00am (midnight) and the first message of a particular messages_type during a particular time_segment . timelastmessages minutes Number of minutes between 12:00am (midnight) and the last message of a particular messages_type during a particular time_segment . countmostfrequentcontact messages Number of messages from the contact with the most messages of messages_type during a time_segment throughout the whole dataset of each participant. Assumptions/Observations [MESSAGES_TYPES] and [FEATURES] keys in config.yaml need to match. For example, [MESSAGES_TYPES] sent matches the [FEATURES] key sent","title":"Phone Messages"},{"location":"features/phone-messages/#phone-messages","text":"Sensor parameters description for [PHONE_MESSAGES] : Key Description [TABLE] Database table where the messages data is stored","title":"Phone Messages"},{"location":"features/phone-messages/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_messages_raw.csv - data/raw/ { pid } /phone_messages_with_datetime.csv - data/interim/ { pid } /phone_messages_features/phone_messages_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_messages.csv Parameters description for [PHONE_MESSAGES][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_MESSAGES features from the RAPIDS provider [MESSAGES_TYPES] The messages_type that will be analyzed. The options for this parameter are received or sent . [FEATURES] Features to be computed, see table below for [MESSAGES_TYPES] received and sent Features description for [PHONE_MESSAGES][PROVIDERS][RAPIDS] : Feature Units Description count messages Number of messages of type messages_type that occurred during a particular time_segment . distinctcontacts contacts Number of distinct contacts that are associated with a particular messages_type during a particular time_segment . timefirstmessages minutes Number of minutes between 12:00am (midnight) and the first message of a particular messages_type during a particular time_segment . timelastmessages minutes Number of minutes between 12:00am (midnight) and the last message of a particular messages_type during a particular time_segment . countmostfrequentcontact messages Number of messages from the contact with the most messages of messages_type during a time_segment throughout the whole dataset of each participant. Assumptions/Observations [MESSAGES_TYPES] and [FEATURES] keys in config.yaml need to match. For example, [MESSAGES_TYPES] sent matches the [FEATURES] key sent","title":"RAPIDS provider"},{"location":"features/phone-screen/","text":"Phone Screen \u00b6 Sensor parameters description for [PHONE_SCREEN] : Key Description [TABLE] Database table where the screen data is stored RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_screen_raw.csv - data/raw/ { pid } /phone_screen_with_datetime.csv - data/raw/ { pid } /phone_screen_with_datetime_unified.csv - data/interim/ { pid } /phone_screen_episodes.csv - data/interim/ { pid } /phone_screen_episodes_resampled.csv - data/interim/ { pid } /phone_screen_episodes_resampled_with_datetime.csv - data/interim/ { pid } /phone_screen_features/phone_screen_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_screen.csv Parameters description for [PHONE_SCREEN][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_SCREEN features from the RAPIDS provider [FEATURES] Features to be computed, see table below [REFERENCE_HOUR_FIRST_USE] The reference point from which firstuseafter is to be computed, default is midnight [IGNORE_EPISODES_SHORTER_THAN] Ignore episodes that are shorter than this threshold (minutes). Set to 0 to disable this filter. [IGNORE_EPISODES_LONGER_THAN] Ignore episodes that are longer than this threshold (minutes). Set to 0 to disable this filter. [EPISODE_TYPES] Currently we only support unlock episodes (from when the phone is unlocked until the screen is off) Features description for [PHONE_SCREEN][PROVIDERS][RAPIDS] : Feature Units Description sumduration minutes Total duration of all unlock episodes. maxduration minutes Longest duration of any unlock episode. minduration minutes Shortest duration of any unlock episode. avgduration minutes Average duration of all unlock episodes. stdduration minutes Standard deviation duration of all unlock episodes. countepisode episodes Number of all unlock episodes |firstuseafter |minutes |Minutes until the first unlock episode. Assumptions/Observations In Android, lock events can happen right after an off event, after a few seconds of an off event, or never happen depending on the phone's settings, therefore, an unlock episode is defined as the time between an unlock and a off event. In iOS, on and off events do not exist, so an unlock episode is defined as the time between an unlock and a lock event. Events in iOS are recorded reliably albeit some duplicated lock events within milliseconds from each other, so we only keep consecutive unlock/lock pairs. In Android you cand find multiple consecutive unlock or lock events, so we only keep consecutive unlock/off pairs. In our experiments these cases are less than 10% of the screen events collected and this happens because ACTION_SCREEN_OFF and ACTION_SCREEN_ON are sent when the device becomes non-interactive which may have nothing to do with the screen turning off . In addition to unlock/off episodes, in Android it is possible to measure the time spent on the lock screen before an unlock event as well as the total screen time (i.e. ON to OFF ) but these are not implemented at the moment. We transform iOS screen events to match Android\u2019s format, we replace lock episodes with off episodes (2 with 0) in iOS. However, as mentioned above this is still computing unlock to lock episodes.","title":"Phone Screen"},{"location":"features/phone-screen/#phone-screen","text":"Sensor parameters description for [PHONE_SCREEN] : Key Description [TABLE] Database table where the screen data is stored","title":"Phone Screen"},{"location":"features/phone-screen/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_screen_raw.csv - data/raw/ { pid } /phone_screen_with_datetime.csv - data/raw/ { pid } /phone_screen_with_datetime_unified.csv - data/interim/ { pid } /phone_screen_episodes.csv - data/interim/ { pid } /phone_screen_episodes_resampled.csv - data/interim/ { pid } /phone_screen_episodes_resampled_with_datetime.csv - data/interim/ { pid } /phone_screen_features/phone_screen_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_screen.csv Parameters description for [PHONE_SCREEN][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_SCREEN features from the RAPIDS provider [FEATURES] Features to be computed, see table below [REFERENCE_HOUR_FIRST_USE] The reference point from which firstuseafter is to be computed, default is midnight [IGNORE_EPISODES_SHORTER_THAN] Ignore episodes that are shorter than this threshold (minutes). Set to 0 to disable this filter. [IGNORE_EPISODES_LONGER_THAN] Ignore episodes that are longer than this threshold (minutes). Set to 0 to disable this filter. [EPISODE_TYPES] Currently we only support unlock episodes (from when the phone is unlocked until the screen is off) Features description for [PHONE_SCREEN][PROVIDERS][RAPIDS] : Feature Units Description sumduration minutes Total duration of all unlock episodes. maxduration minutes Longest duration of any unlock episode. minduration minutes Shortest duration of any unlock episode. avgduration minutes Average duration of all unlock episodes. stdduration minutes Standard deviation duration of all unlock episodes. countepisode episodes Number of all unlock episodes |firstuseafter |minutes |Minutes until the first unlock episode. Assumptions/Observations In Android, lock events can happen right after an off event, after a few seconds of an off event, or never happen depending on the phone's settings, therefore, an unlock episode is defined as the time between an unlock and a off event. In iOS, on and off events do not exist, so an unlock episode is defined as the time between an unlock and a lock event. Events in iOS are recorded reliably albeit some duplicated lock events within milliseconds from each other, so we only keep consecutive unlock/lock pairs. In Android you cand find multiple consecutive unlock or lock events, so we only keep consecutive unlock/off pairs. In our experiments these cases are less than 10% of the screen events collected and this happens because ACTION_SCREEN_OFF and ACTION_SCREEN_ON are sent when the device becomes non-interactive which may have nothing to do with the screen turning off . In addition to unlock/off episodes, in Android it is possible to measure the time spent on the lock screen before an unlock event as well as the total screen time (i.e. ON to OFF ) but these are not implemented at the moment. We transform iOS screen events to match Android\u2019s format, we replace lock episodes with off episodes (2 with 0) in iOS. However, as mentioned above this is still computing unlock to lock episodes.","title":"RAPIDS provider"},{"location":"features/phone-wifi-connected/","text":"Phone WiFi Connected \u00b6 Sensor parameters description for [PHONE_WIFI_CONNECTED] : Key Description [TABLE] Database table where the wifi (connected) data is stored RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_wifi_connected_raw.csv - data/raw/ { pid } /phone_wifi_connected_with_datetime.csv - data/interim/ { pid } /phone_wifi_connected_features/phone_wifi_connected_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_wifi_connected.csv Parameters description for [PHONE_WIFI_CONNECTED][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_WIFI_CONNECTED features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_WIFI_CONNECTED][PROVIDERS][RAPIDS] : Feature Units Description countscans devices Number of scanned WiFi access points connected during a time_segment, an access point can be detected multiple times over time and these appearances are counted separately uniquedevices devices Number of unique access point during a time_segment as identified by their hardware address countscansmostuniquedevice scans Number of scans of the most scanned access point during a time_segment across the whole monitoring period Assumptions/Observations A connected WiFI access point is one that a phone was connected to. By default AWARE stores this data in the sensor_wifi table.","title":"Phone WiFI Connected"},{"location":"features/phone-wifi-connected/#phone-wifi-connected","text":"Sensor parameters description for [PHONE_WIFI_CONNECTED] : Key Description [TABLE] Database table where the wifi (connected) data is stored","title":"Phone WiFi Connected"},{"location":"features/phone-wifi-connected/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android and iOS File Sequence - data/raw/ { pid } /phone_wifi_connected_raw.csv - data/raw/ { pid } /phone_wifi_connected_with_datetime.csv - data/interim/ { pid } /phone_wifi_connected_features/phone_wifi_connected_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_wifi_connected.csv Parameters description for [PHONE_WIFI_CONNECTED][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_WIFI_CONNECTED features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_WIFI_CONNECTED][PROVIDERS][RAPIDS] : Feature Units Description countscans devices Number of scanned WiFi access points connected during a time_segment, an access point can be detected multiple times over time and these appearances are counted separately uniquedevices devices Number of unique access point during a time_segment as identified by their hardware address countscansmostuniquedevice scans Number of scans of the most scanned access point during a time_segment across the whole monitoring period Assumptions/Observations A connected WiFI access point is one that a phone was connected to. By default AWARE stores this data in the sensor_wifi table.","title":"RAPIDS provider"},{"location":"features/phone-wifi-visible/","text":"Phone WiFi Visible \u00b6 Sensor parameters description for [PHONE_WIFI_VISIBLE] : Key Description [TABLE] Database table where the wifi (visible) data is stored RAPIDS provider \u00b6 Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_wifi_visible_raw.csv - data/raw/ { pid } /phone_wifi_visible_with_datetime.csv - data/interim/ { pid } /phone_wifi_visible_features/phone_wifi_visible_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_wifi_visible.csv Parameters description for [PHONE_WIFI_VISIBLE][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_WIFI_VISIBLE features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_WIFI_VISIBLE][PROVIDERS][RAPIDS] : Feature Units Description countscans devices Number of scanned WiFi access points visible during a time_segment, an access point can be detected multiple times over time and these appearances are counted separately uniquedevices devices Number of unique access point during a time_segment as identified by their hardware address countscansmostuniquedevice scans Number of scans of the most scanned access point during a time_segment across the whole monitoring period Assumptions/Observations A visible WiFI access point is one that a phone sensed around itself but that it was not connected to. Due to API restrictions, this sensor is not available on iOS. By default AWARE stores this data in the wifi table.","title":"Phone WiFI Visible"},{"location":"features/phone-wifi-visible/#phone-wifi-visible","text":"Sensor parameters description for [PHONE_WIFI_VISIBLE] : Key Description [TABLE] Database table where the wifi (visible) data is stored","title":"Phone WiFi Visible"},{"location":"features/phone-wifi-visible/#rapids-provider","text":"Available time segments and platforms Available for all time segments Available for Android only File Sequence - data/raw/ { pid } /phone_wifi_visible_raw.csv - data/raw/ { pid } /phone_wifi_visible_with_datetime.csv - data/interim/ { pid } /phone_wifi_visible_features/phone_wifi_visible_ { language } _ { provider_key } .csv - data/processed/features/ { pid } /phone_wifi_visible.csv Parameters description for [PHONE_WIFI_VISIBLE][PROVIDERS][RAPIDS] : Key Description [COMPUTE] Set to True to extract PHONE_WIFI_VISIBLE features from the RAPIDS provider [FEATURES] Features to be computed, see table below Features description for [PHONE_WIFI_VISIBLE][PROVIDERS][RAPIDS] : Feature Units Description countscans devices Number of scanned WiFi access points visible during a time_segment, an access point can be detected multiple times over time and these appearances are counted separately uniquedevices devices Number of unique access point during a time_segment as identified by their hardware address countscansmostuniquedevice scans Number of scans of the most scanned access point during a time_segment across the whole monitoring period Assumptions/Observations A visible WiFI access point is one that a phone sensed around itself but that it was not connected to. Due to API restrictions, this sensor is not available on iOS. By default AWARE stores this data in the wifi table.","title":"RAPIDS provider"},{"location":"setup/configuration/","text":"Configuration \u00b6 You need to follow these steps to configure your RAPIDS deployment before you can extract behavioral features Add your database credentials Choose the timezone of your study Create your participants files Select what time segments you want to extract features on Modify your device data source configuration Select what sensors and features you want to process When you are done with this configuration, go to executing RAPIDS . Hint Every time you see config[\"KEY\"] or [KEY] in these docs we are referring to the corresponding key in the config.yaml file. Database credentials \u00b6 Create an empty file called .env in your RAPIDS root directory Add the following lines and replace your database-specific credentials (user, password, host, and database): [ MY_GROUP ] user=MY_USER password=MY_PASSWORD host=MY_HOST port=3306 database=MY_DATABASE Warning The label MY_GROUP is arbitrary but it has to match the following config.yaml key: DATABASE_GROUP : &database_group MY_GROUP Note You can ignore this step if you are only processing Fitbit data in CSV files. Timezone of your study \u00b6 Single timezone \u00b6 If your study only happened in a single time zone, select the appropriate code form this list and change the following config key. Double check your timezone code pick, for example US Eastern Time is America/New_York not EST TIMEZONE : &timezone America/New_York Multiple timezones \u00b6 Support coming soon. Participant files \u00b6 Participant files link together multiple devices (smartphones and wearables) to specific participants and identify them throughout RAPIDS. You can create these files manually or automatically . Participant files are stored in data/external/participant_files/pxx.yaml and follow a unified structure . Note The list PIDS in config.yaml needs to have the participant file names of the people you want to process. For example, if you created p01.yaml , p02.yaml and p03.yaml files in /data/external/participant_files/ , then PIDS should be: PIDS : [ p01 , p02 , p03 ] Tip Attribute values of the [PHONE] and [FITBIT] sections in every participant file are optional which allows you to analyze data from participants that only carried smartphones, only Fitbit devices, or both. Optional: Migrating participants files with the old format If you were using the pre-release version of RAPIDS with participant files in plain text (as opposed to yaml), you can run the following command and your old files will be converted into yaml files stored in data/external/participant_files/ python tools/update_format_participant_files.py Structure of participants files \u00b6 Example of the structure of a participant file In this example, the participant used an android phone, an ios phone, and a fitbit device throughout the study between Apr 23 rd 2020 and Oct 28 th 2020 PHONE : DEVICE_IDS : [ a748ee1a-1d0b-4ae9-9074-279a2b6ba524 , dsadas-2324-fgsf-sdwr-gdfgs4rfsdf43 ] PLATFORMS : [ android , ios ] LABEL : test01 START_DATE : 2020-04-23 END_DATE : 2020-10-28 FITBIT : DEVICE_IDS : [ fitbit1 ] LABEL : test01 START_DATE : 2020-04-23 END_DATE : 2020-10-28 For [PHONE] Key Description [DEVICE_IDS] An array of the strings that uniquely identify each smartphone, you can have more than one for when participants changed phones in the middle of the study, in this case, data from all their devices will be joined and relabeled with the last 1 on this list. [PLATFORMS] An array that specifies the OS of each smartphone in [DEVICE_IDS] , use a combination of android or ios (we support participants that changed platforms in the middle of your study!). If you have an aware_device table in your database you can set [PLATFORMS]: [multiple] and RAPIDS will infer them automatically. [LABEL] A string that is used in reports and visualizations. [START_DATE] A string with format YYY-MM-DD . Only data collected after this date will be included in the analysis [END_DATE] A string with format YYY-MM-DD . Only data collected before this date will be included in the analysis For [FITBIT] Key Description [DEVICE_IDS] An array of the strings that uniquely identify each Fitbit, you can have more than one in case the participant changed devices in the middle of the study, in this case, data from all devices will be joined and relabeled with the last device_id on this list. [LABEL] A string that is used in reports and visualizations. [START_DATE] A string with format YYY-MM-DD . Only data collected after this date will be included in the analysis [END_DATE] A string with format YYY-MM-DD . Only data collected before this date will be included in the analysis Automatic creation of participant files \u00b6 You have two options a) use the aware_device table in your database or b) use a CSV file. In either case, in your config.yaml , set [PHONE_SECTION][ADD] or [FITBIT_SECTION][ADD] to TRUE depending on what devices you used in your study. Set [DEVICE_ID_COLUMN] to the name of the column that uniquely identifies each device and include any device ids you want to ignore in [IGNORED_DEVICE_IDS] . aware_device table Set the following keys in your config.yaml CREATE_PARTICIPANT_FILES : SOURCE : TYPE : AWARE_DEVICE_TABLE DATABASE_GROUP : *database_group CSV_FILE_PATH : \"\" TIMEZONE : *timezone PHONE_SECTION : ADD : TRUE # or FALSE DEVICE_ID_COLUMN : device_id # column name IGNORED_DEVICE_IDS : [] FITBIT_SECTION : ADD : TRUE # or FALSE DEVICE_ID_COLUMN : fitbit_id # column name IGNORED_DEVICE_IDS : [] Then run snakemake -j1 create_participants_files CSV file Set the following keys in your config.yaml . CREATE_PARTICIPANT_FILES : SOURCE : TYPE : CSV_FILE DATABASE_GROUP : \"\" CSV_FILE_PATH : \"your_path/to_your.csv\" TIMEZONE : *timezone PHONE_SECTION : ADD : TRUE # or FALSE DEVICE_ID_COLUMN : device_id # column name IGNORED_DEVICE_IDS : [] FITBIT_SECTION : ADD : TRUE # or FALSE DEVICE_ID_COLUMN : fitbit_id # column name IGNORED_DEVICE_IDS : [] Your CSV file ( [SOURCE][CSV_FILE_PATH] ) should have the following columns but you can omit any values you don\u2019t have on each column: Column Description phone device id The name of this column has to match [PHONE_SECTION][DEVICE_ID_COLUMN] . Separate multiple ids with ; fitbit device id The name of this column has to match [FITBIT_SECTION][DEVICE_ID_COLUMN] . Separate multiple ids with ; pid Unique identifiers with the format pXXX (your participant files will be named with this string platform Use android , ios or multiple as explained above, separate values with ; label A human readable string that is used in reports and visualizations. start_date A string with format YYY-MM-DD . end_date A string with format YYY-MM-DD . Example device_id,pid,label,platform,start_date,end_date,fitbit_id a748ee1a-1d0b-4ae9-9074-279a2b6ba524;dsadas-2324-fgsf-sdwr-gdfgs4rfsdf43,p01,julio,android;ios,2020-01-01,2021-01-01,fitbit1 4c4cf7a1-0340-44bc-be0f-d5053bf7390c,p02,meng,ios,2021-01-01,2022-01-01,fitbit2 Then run snakemake -j1 create_participants_files Time Segments \u00b6 Time segments (or epochs) are the time windows on which you want to extract behavioral features. For example, you might want to process data on every day, every morning, or only during weekends. RAPIDS offers three categories of time segments that are flexible enough to cover most use cases: frequency (short time windows every day), periodic (arbitrary time windows on any day), and event (arbitrary time windows around events of interest). See also our examples . Frequency Segments These segments are computed on every day and all have the same duration (for example 30 minutes). Set the following keys in your config.yaml TIME_SEGMENTS : &time_segments TYPE : FREQUENCY FILE : \"data/external/your_frequency_segments.csv\" INCLUDE_PAST_PERIODIC_SEGMENTS : FALSE The file pointed by [TIME_SEGMENTS][FILE] should have the following format and can only have 1 row. Column Description label A string that is used as a prefix in the name of your time segments length An integer representing the duration of your time segments in minutes Example label,length thirtyminutes,30 This configuration will compute 48 time segments for every day when any data from any participant was sensed. For example: start_time,length,label 00:00,30,thirtyminutes0000 00:30,30,thirtyminutes0001 01:00,30,thirtyminutes0002 01:30,30,thirtyminutes0003 ... Periodic Segments These segments can be computed every day, or on specific days of the week, month, quarter, and year. Their minimum duration is 1 minute but they can be as long as you want. Set the following keys in your config.yaml . TIME_SEGMENTS : &time_segments TYPE : PERIODIC FILE : \"data/external/your_periodic_segments.csv\" INCLUDE_PAST_PERIODIC_SEGMENTS : FALSE # or TRUE If [INCLUDE_PAST_PERIODIC_SEGMENTS] is set to TRUE , RAPIDS will consider instances of your segments back enough in the past as to include the first row of data of each participant. For example, if the first row of data from a participant happened on Saturday March 7 th 2020 and the requested segment duration is 7 days starting on every Sunday, the first segment to be considered would start on Sunday March 1 st if [INCLUDE_PAST_PERIODIC_SEGMENTS] is TRUE or on Sunday March 8 th if FALSE . The file pointed by [TIME_SEGMENTS][FILE] should have the following format and can have multiple rows. Column Description label A string that is used as a prefix in the name of your time segments. It has to be unique between rows start_time A string with format HH:MM:SS representing the starting time of this segment on any day length A string representing the length of this segment.It can have one or more of the following strings XXD XXH XXM XXS to represent days, hours, minutes and seconds. For example 7D 23H 59M 59S repeats_on One of the follow options every_day , wday , qday , mday , and yday . The last four represent a week, quarter, month and year day repeats_value An integer complementing repeats_on . If you set repeats_on to every_day set this to 0 , otherwise 1-7 represent a wday starting from Mondays, 1-31 represent a mday , 1-91 represent a qday , and 1-366 represent a yday Example label,start_time,length,repeats_on,repeats_value daily,00:00:00,23H 59M 59S,every_day,0 morning,06:00:00,5H 59M 59S,every_day,0 afternoon,12:00:00,5H 59M 59S,every_day,0 evening,18:00:00,5H 59M 59S,every_day,0 night,00:00:00,5H 59M 59S,every_day,0 This configuration will create five segments instances ( daily , morning , afternoon , evening , night ) on any given day ( every_day set to 0). The daily segment will start at midnight and will last 23:59:59 , the other four segments will start at 6am, 12pm, 6pm, and 12am respectively and last for 05:59:59 . Event segments These segments can be computed before or after an event of interest (defined as any UNIX timestamp). Their minimum duration is 1 minute but they can be as long as you want. The start of each segment can be shifted backwards or forwards from the specified timestamp. Set the following keys in your config.yaml . TIME_SEGMENTS : &time_segments TYPE : EVENT FILE : \"data/external/your_event_segments.csv\" INCLUDE_PAST_PERIODIC_SEGMENTS : FALSE # or TRUE The file pointed by [TIME_SEGMENTS][FILE] should have the following format and can have multiple rows. Column Description label A string that is used as a prefix in the name of your time segments. If labels are unique, every segment is independent; if two or more segments have the same label, their data will be grouped when computing auxiliary data for features like the most frequent contact for calls (the most frequent contact will be computed across all these segments). There cannot be two overlaping event segments with the same label (RAPIDS will throw an error) event_timestamp A UNIX timestamp that represents the moment an event of interest happened (clinical relapse, survey, readmission, etc.). The corresponding time segment will be computed around this moment using length , shift , and shift_direction length A string representing the length of this segment. It can have one or more of the following keys XXD XXH XXM XXS to represent a number of days, hours, minutes, and seconds. For example 7D 23H 59M 59S shift A string representing the time shift from event_timestamp . It can have one or more of the following keys XXD XXH XXM XXS to represent a number of days, hours, minutes and seconds. For example 7D 23H 59M 59S . Use this value to change the start of a segment with respect to its event_timestamp . For example, set this variable to 1H to create a segment that starts 1 hour from an event of interest ( shift_direction determines if it\u2019s before or after). shift_direction An integer representing whether the shift is before ( -1 ) or after ( 1 ) an event_timestamp device_id The device id (smartphone or fitbit) to whom this segment belongs to. You have to create a line in this event segment file for each event of a participant that you want to analyse. If you have participants with multiple device ids you can choose any of them Example label,event_timestamp,length,shift,shift_direction,device_id stress1,1587661220000,1H,5M,1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 stress2,1587747620000,4H,4H,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 stress3,1587906020000,3H,5M,1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 stress4,1584291600000,7H,4H,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 stress5,1588172420000,9H,5M,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 mood,1587661220000,1H,0,0,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 mood,1587747620000,1D,0,0,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 mood,1587906020000,7D,0,0,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 This example will create eight segments for a single participant ( a748ee1a... ), five independent stressX segments with various lengths (1,4,3,7, and 9 hours). Segments stress1 , stress3 , and stress5 are shifted forwards by 5 minutes and stress2 and stress4 are shifted backwards by 4 hours (that is, if the stress4 event happened on March 15 th at 1pm EST ( 1584291600000 ), the time segment will start on that day at 9am and end at 4pm). The three mood segments are 1 hour, 1 day and 7 days long and have no shift. In addition, these mood segments are grouped together, meaning that although RAPIDS will compute features on each one of them, some necessary information to compute a few of such features will be extracted from all three segments, for example the phone contact that called a participant the most or the location clusters visited by a participant. Segment Examples \u00b6 5-minutes Use the following Frequency segment file to create 288 (12 * 60 * 24) 5-minute segments starting from midnight of every day in your study label,length fiveminutes,5 Daily Use the following Periodic segment file to create daily segments starting from midnight of every day in your study label,start_time,length,repeats_on,repeats_value daily,00:00:00,23H 59M 59S,every_day,0 Morning Use the following Periodic segment file to create morning segments starting at 06:00:00 and ending at 11:59:59 of every day in your study label,start_time,length,repeats_on,repeats_value morning,06:00:00,5H 59M 59S,every_day,0 Overnight Use the following Periodic segment file to create overnight segments starting at 20:00:00 and ending at 07:59:59 (next day) of every day in your study label,start_time,length,repeats_on,repeats_value morning,20:00:00,11H 59M 59S,every_day,0 Weekly Use the following Periodic segment file to create non-overlapping weekly segments starting at midnight of every Monday in your study label,start_time,length,repeats_on,repeats_value weekly,00:00:00,6D 23H 59M 59S,wday,1 Use the following Periodic segment file to create overlapping weekly segments starting at midnight of every day in your study label,start_time,length,repeats_on,repeats_value weekly,00:00:00,6D 23H 59M 59S,every_day,0 Week-ends Use the following Periodic segment file to create week-end segments starting at midnight of every Saturday in your study label,start_time,length,repeats_on,repeats_value weekend,00:00:00,1D 23H 59M 59S,wday,6 Around surveys Use the following Event segment file to create two 2-hour segments that start 1 hour before surveys answered by 3 participants label,event_timestamp,length,shift,shift_direction,device_id survey1,1587661220000,2H,1H,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 survey2,1587747620000,2H,1H,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 survey1,1587906020000,2H,1H,-1,rqtertsd-43ff-34fr-3eeg-efe4fergregr survey2,1584291600000,2H,1H,-1,rqtertsd-43ff-34fr-3eeg-efe4fergregr survey1,1588172420000,2H,1H,-1,klj34oi2-8frk-2343-21kk-324ljklewlr3 survey2,1584291600000,2H,1H,-1,klj34oi2-8frk-2343-21kk-324ljklewlr3 Device Data Source Configuration \u00b6 You might need to modify the following config keys in your config.yaml depending on what devices your participants used and where you are storing your data. You can ignore [PHONE_DATA_CONFIGURATION] or [FITBIT_DATA_CONFIGURATION] if you are not working with either devices. Phone The relevant config.yaml section looks like this by default: PHONE_DATA_CONFIGURATION : SOURCE : TYPE : DATABASE DATABASE_GROUP : *database_group DEVICE_ID_COLUMN : device_id # column name TIMEZONE : TYPE : SINGLE # SINGLE (MULTIPLE support coming soon) VALUE : *timezone Parameters for [PHONE_DATA_CONFIGURATION] Key Description [SOURCE] [TYPE] Only DATABASE is supported (phone data will be pulled from a database) [SOURCE] [DATABASE_GROUP] *database_group points to the value defined before in Database credentials [SOURCE] [DEVICE_ID_COLUMN] A column that contains strings that uniquely identify smartphones. For data collected with AWARE this is usually device_id [TIMEZONE] [TYPE] Only SINGLE is supported for now [TIMEZONE] [VALUE] *timezone points to the value defined before in Timezone of your study Fitbit The relevant config.yaml section looks like this by default: FITBIT_DATA_CONFIGURATION : SOURCE : TYPE : DATABASE # DATABASE or FILES (set each [FITBIT_SENSOR][TABLE] attribute with a table name or a file path accordingly) COLUMN_FORMAT : JSON # JSON or PLAIN_TEXT DATABASE_GROUP : *database_group DEVICE_ID_COLUMN : device_id # column name TIMEZONE : TYPE : SINGLE # Fitbit devices don't support time zones so we read this data in the timezone indicated by VALUE VALUE : *timezone Parameters for For [FITBIT_DATA_CONFIGURATION] Key Description [SOURCE] [TYPE] DATABASE or FILES (set each [FITBIT_SENSOR] [TABLE] attribute accordingly with a table name or a file path) [SOURCE] [COLUMN_FORMAT] JSON or PLAIN_TEXT . Column format of the source data. If you pulled your data directly from the Fitbit API the column containing the sensor data will be in JSON format [SOURCE] [DATABASE_GROUP] *database_group points to the value defined before in Database credentials . Only used if [TYPE] is DATABASE . [SOURCE] [DEVICE_ID_COLUMN] A column that contains strings that uniquely identify Fitbit devices. [TIMEZONE] [TYPE] Only SINGLE is supported (Fitbit devices always store data in local time). [TIMEZONE] [VALUE] *timezone points to the value defined before in Timezone of your study Sensor and Features to Process \u00b6 Finally, you need to modify the config.yaml section of the sensors you want to extract behavioral features from. All sensors follow the same naming nomenclature ( DEVICE_SENSOR ) and parameter structure which we explain in the Behavioral Features Introduction . Done Head over to Execution to learn how to execute RAPIDS.","title":"Configuration"},{"location":"setup/configuration/#configuration","text":"You need to follow these steps to configure your RAPIDS deployment before you can extract behavioral features Add your database credentials Choose the timezone of your study Create your participants files Select what time segments you want to extract features on Modify your device data source configuration Select what sensors and features you want to process When you are done with this configuration, go to executing RAPIDS . Hint Every time you see config[\"KEY\"] or [KEY] in these docs we are referring to the corresponding key in the config.yaml file.","title":"Configuration"},{"location":"setup/configuration/#database-credentials","text":"Create an empty file called .env in your RAPIDS root directory Add the following lines and replace your database-specific credentials (user, password, host, and database): [ MY_GROUP ] user=MY_USER password=MY_PASSWORD host=MY_HOST port=3306 database=MY_DATABASE Warning The label MY_GROUP is arbitrary but it has to match the following config.yaml key: DATABASE_GROUP : &database_group MY_GROUP Note You can ignore this step if you are only processing Fitbit data in CSV files.","title":"Database credentials"},{"location":"setup/configuration/#timezone-of-your-study","text":"","title":"Timezone of your study"},{"location":"setup/configuration/#single-timezone","text":"If your study only happened in a single time zone, select the appropriate code form this list and change the following config key. Double check your timezone code pick, for example US Eastern Time is America/New_York not EST TIMEZONE : &timezone America/New_York","title":"Single timezone"},{"location":"setup/configuration/#multiple-timezones","text":"Support coming soon.","title":"Multiple timezones"},{"location":"setup/configuration/#participant-files","text":"Participant files link together multiple devices (smartphones and wearables) to specific participants and identify them throughout RAPIDS. You can create these files manually or automatically . Participant files are stored in data/external/participant_files/pxx.yaml and follow a unified structure . Note The list PIDS in config.yaml needs to have the participant file names of the people you want to process. For example, if you created p01.yaml , p02.yaml and p03.yaml files in /data/external/participant_files/ , then PIDS should be: PIDS : [ p01 , p02 , p03 ] Tip Attribute values of the [PHONE] and [FITBIT] sections in every participant file are optional which allows you to analyze data from participants that only carried smartphones, only Fitbit devices, or both. Optional: Migrating participants files with the old format If you were using the pre-release version of RAPIDS with participant files in plain text (as opposed to yaml), you can run the following command and your old files will be converted into yaml files stored in data/external/participant_files/ python tools/update_format_participant_files.py","title":"Participant files"},{"location":"setup/configuration/#structure-of-participants-files","text":"Example of the structure of a participant file In this example, the participant used an android phone, an ios phone, and a fitbit device throughout the study between Apr 23 rd 2020 and Oct 28 th 2020 PHONE : DEVICE_IDS : [ a748ee1a-1d0b-4ae9-9074-279a2b6ba524 , dsadas-2324-fgsf-sdwr-gdfgs4rfsdf43 ] PLATFORMS : [ android , ios ] LABEL : test01 START_DATE : 2020-04-23 END_DATE : 2020-10-28 FITBIT : DEVICE_IDS : [ fitbit1 ] LABEL : test01 START_DATE : 2020-04-23 END_DATE : 2020-10-28 For [PHONE] Key Description [DEVICE_IDS] An array of the strings that uniquely identify each smartphone, you can have more than one for when participants changed phones in the middle of the study, in this case, data from all their devices will be joined and relabeled with the last 1 on this list. [PLATFORMS] An array that specifies the OS of each smartphone in [DEVICE_IDS] , use a combination of android or ios (we support participants that changed platforms in the middle of your study!). If you have an aware_device table in your database you can set [PLATFORMS]: [multiple] and RAPIDS will infer them automatically. [LABEL] A string that is used in reports and visualizations. [START_DATE] A string with format YYY-MM-DD . Only data collected after this date will be included in the analysis [END_DATE] A string with format YYY-MM-DD . Only data collected before this date will be included in the analysis For [FITBIT] Key Description [DEVICE_IDS] An array of the strings that uniquely identify each Fitbit, you can have more than one in case the participant changed devices in the middle of the study, in this case, data from all devices will be joined and relabeled with the last device_id on this list. [LABEL] A string that is used in reports and visualizations. [START_DATE] A string with format YYY-MM-DD . Only data collected after this date will be included in the analysis [END_DATE] A string with format YYY-MM-DD . Only data collected before this date will be included in the analysis","title":"Structure of participants files"},{"location":"setup/configuration/#automatic-creation-of-participant-files","text":"You have two options a) use the aware_device table in your database or b) use a CSV file. In either case, in your config.yaml , set [PHONE_SECTION][ADD] or [FITBIT_SECTION][ADD] to TRUE depending on what devices you used in your study. Set [DEVICE_ID_COLUMN] to the name of the column that uniquely identifies each device and include any device ids you want to ignore in [IGNORED_DEVICE_IDS] . aware_device table Set the following keys in your config.yaml CREATE_PARTICIPANT_FILES : SOURCE : TYPE : AWARE_DEVICE_TABLE DATABASE_GROUP : *database_group CSV_FILE_PATH : \"\" TIMEZONE : *timezone PHONE_SECTION : ADD : TRUE # or FALSE DEVICE_ID_COLUMN : device_id # column name IGNORED_DEVICE_IDS : [] FITBIT_SECTION : ADD : TRUE # or FALSE DEVICE_ID_COLUMN : fitbit_id # column name IGNORED_DEVICE_IDS : [] Then run snakemake -j1 create_participants_files CSV file Set the following keys in your config.yaml . CREATE_PARTICIPANT_FILES : SOURCE : TYPE : CSV_FILE DATABASE_GROUP : \"\" CSV_FILE_PATH : \"your_path/to_your.csv\" TIMEZONE : *timezone PHONE_SECTION : ADD : TRUE # or FALSE DEVICE_ID_COLUMN : device_id # column name IGNORED_DEVICE_IDS : [] FITBIT_SECTION : ADD : TRUE # or FALSE DEVICE_ID_COLUMN : fitbit_id # column name IGNORED_DEVICE_IDS : [] Your CSV file ( [SOURCE][CSV_FILE_PATH] ) should have the following columns but you can omit any values you don\u2019t have on each column: Column Description phone device id The name of this column has to match [PHONE_SECTION][DEVICE_ID_COLUMN] . Separate multiple ids with ; fitbit device id The name of this column has to match [FITBIT_SECTION][DEVICE_ID_COLUMN] . Separate multiple ids with ; pid Unique identifiers with the format pXXX (your participant files will be named with this string platform Use android , ios or multiple as explained above, separate values with ; label A human readable string that is used in reports and visualizations. start_date A string with format YYY-MM-DD . end_date A string with format YYY-MM-DD . Example device_id,pid,label,platform,start_date,end_date,fitbit_id a748ee1a-1d0b-4ae9-9074-279a2b6ba524;dsadas-2324-fgsf-sdwr-gdfgs4rfsdf43,p01,julio,android;ios,2020-01-01,2021-01-01,fitbit1 4c4cf7a1-0340-44bc-be0f-d5053bf7390c,p02,meng,ios,2021-01-01,2022-01-01,fitbit2 Then run snakemake -j1 create_participants_files","title":"Automatic creation of participant files"},{"location":"setup/configuration/#time-segments","text":"Time segments (or epochs) are the time windows on which you want to extract behavioral features. For example, you might want to process data on every day, every morning, or only during weekends. RAPIDS offers three categories of time segments that are flexible enough to cover most use cases: frequency (short time windows every day), periodic (arbitrary time windows on any day), and event (arbitrary time windows around events of interest). See also our examples . Frequency Segments These segments are computed on every day and all have the same duration (for example 30 minutes). Set the following keys in your config.yaml TIME_SEGMENTS : &time_segments TYPE : FREQUENCY FILE : \"data/external/your_frequency_segments.csv\" INCLUDE_PAST_PERIODIC_SEGMENTS : FALSE The file pointed by [TIME_SEGMENTS][FILE] should have the following format and can only have 1 row. Column Description label A string that is used as a prefix in the name of your time segments length An integer representing the duration of your time segments in minutes Example label,length thirtyminutes,30 This configuration will compute 48 time segments for every day when any data from any participant was sensed. For example: start_time,length,label 00:00,30,thirtyminutes0000 00:30,30,thirtyminutes0001 01:00,30,thirtyminutes0002 01:30,30,thirtyminutes0003 ... Periodic Segments These segments can be computed every day, or on specific days of the week, month, quarter, and year. Their minimum duration is 1 minute but they can be as long as you want. Set the following keys in your config.yaml . TIME_SEGMENTS : &time_segments TYPE : PERIODIC FILE : \"data/external/your_periodic_segments.csv\" INCLUDE_PAST_PERIODIC_SEGMENTS : FALSE # or TRUE If [INCLUDE_PAST_PERIODIC_SEGMENTS] is set to TRUE , RAPIDS will consider instances of your segments back enough in the past as to include the first row of data of each participant. For example, if the first row of data from a participant happened on Saturday March 7 th 2020 and the requested segment duration is 7 days starting on every Sunday, the first segment to be considered would start on Sunday March 1 st if [INCLUDE_PAST_PERIODIC_SEGMENTS] is TRUE or on Sunday March 8 th if FALSE . The file pointed by [TIME_SEGMENTS][FILE] should have the following format and can have multiple rows. Column Description label A string that is used as a prefix in the name of your time segments. It has to be unique between rows start_time A string with format HH:MM:SS representing the starting time of this segment on any day length A string representing the length of this segment.It can have one or more of the following strings XXD XXH XXM XXS to represent days, hours, minutes and seconds. For example 7D 23H 59M 59S repeats_on One of the follow options every_day , wday , qday , mday , and yday . The last four represent a week, quarter, month and year day repeats_value An integer complementing repeats_on . If you set repeats_on to every_day set this to 0 , otherwise 1-7 represent a wday starting from Mondays, 1-31 represent a mday , 1-91 represent a qday , and 1-366 represent a yday Example label,start_time,length,repeats_on,repeats_value daily,00:00:00,23H 59M 59S,every_day,0 morning,06:00:00,5H 59M 59S,every_day,0 afternoon,12:00:00,5H 59M 59S,every_day,0 evening,18:00:00,5H 59M 59S,every_day,0 night,00:00:00,5H 59M 59S,every_day,0 This configuration will create five segments instances ( daily , morning , afternoon , evening , night ) on any given day ( every_day set to 0). The daily segment will start at midnight and will last 23:59:59 , the other four segments will start at 6am, 12pm, 6pm, and 12am respectively and last for 05:59:59 . Event segments These segments can be computed before or after an event of interest (defined as any UNIX timestamp). Their minimum duration is 1 minute but they can be as long as you want. The start of each segment can be shifted backwards or forwards from the specified timestamp. Set the following keys in your config.yaml . TIME_SEGMENTS : &time_segments TYPE : EVENT FILE : \"data/external/your_event_segments.csv\" INCLUDE_PAST_PERIODIC_SEGMENTS : FALSE # or TRUE The file pointed by [TIME_SEGMENTS][FILE] should have the following format and can have multiple rows. Column Description label A string that is used as a prefix in the name of your time segments. If labels are unique, every segment is independent; if two or more segments have the same label, their data will be grouped when computing auxiliary data for features like the most frequent contact for calls (the most frequent contact will be computed across all these segments). There cannot be two overlaping event segments with the same label (RAPIDS will throw an error) event_timestamp A UNIX timestamp that represents the moment an event of interest happened (clinical relapse, survey, readmission, etc.). The corresponding time segment will be computed around this moment using length , shift , and shift_direction length A string representing the length of this segment. It can have one or more of the following keys XXD XXH XXM XXS to represent a number of days, hours, minutes, and seconds. For example 7D 23H 59M 59S shift A string representing the time shift from event_timestamp . It can have one or more of the following keys XXD XXH XXM XXS to represent a number of days, hours, minutes and seconds. For example 7D 23H 59M 59S . Use this value to change the start of a segment with respect to its event_timestamp . For example, set this variable to 1H to create a segment that starts 1 hour from an event of interest ( shift_direction determines if it\u2019s before or after). shift_direction An integer representing whether the shift is before ( -1 ) or after ( 1 ) an event_timestamp device_id The device id (smartphone or fitbit) to whom this segment belongs to. You have to create a line in this event segment file for each event of a participant that you want to analyse. If you have participants with multiple device ids you can choose any of them Example label,event_timestamp,length,shift,shift_direction,device_id stress1,1587661220000,1H,5M,1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 stress2,1587747620000,4H,4H,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 stress3,1587906020000,3H,5M,1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 stress4,1584291600000,7H,4H,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 stress5,1588172420000,9H,5M,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 mood,1587661220000,1H,0,0,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 mood,1587747620000,1D,0,0,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 mood,1587906020000,7D,0,0,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 This example will create eight segments for a single participant ( a748ee1a... ), five independent stressX segments with various lengths (1,4,3,7, and 9 hours). Segments stress1 , stress3 , and stress5 are shifted forwards by 5 minutes and stress2 and stress4 are shifted backwards by 4 hours (that is, if the stress4 event happened on March 15 th at 1pm EST ( 1584291600000 ), the time segment will start on that day at 9am and end at 4pm). The three mood segments are 1 hour, 1 day and 7 days long and have no shift. In addition, these mood segments are grouped together, meaning that although RAPIDS will compute features on each one of them, some necessary information to compute a few of such features will be extracted from all three segments, for example the phone contact that called a participant the most or the location clusters visited by a participant.","title":"Time Segments"},{"location":"setup/configuration/#segment-examples","text":"5-minutes Use the following Frequency segment file to create 288 (12 * 60 * 24) 5-minute segments starting from midnight of every day in your study label,length fiveminutes,5 Daily Use the following Periodic segment file to create daily segments starting from midnight of every day in your study label,start_time,length,repeats_on,repeats_value daily,00:00:00,23H 59M 59S,every_day,0 Morning Use the following Periodic segment file to create morning segments starting at 06:00:00 and ending at 11:59:59 of every day in your study label,start_time,length,repeats_on,repeats_value morning,06:00:00,5H 59M 59S,every_day,0 Overnight Use the following Periodic segment file to create overnight segments starting at 20:00:00 and ending at 07:59:59 (next day) of every day in your study label,start_time,length,repeats_on,repeats_value morning,20:00:00,11H 59M 59S,every_day,0 Weekly Use the following Periodic segment file to create non-overlapping weekly segments starting at midnight of every Monday in your study label,start_time,length,repeats_on,repeats_value weekly,00:00:00,6D 23H 59M 59S,wday,1 Use the following Periodic segment file to create overlapping weekly segments starting at midnight of every day in your study label,start_time,length,repeats_on,repeats_value weekly,00:00:00,6D 23H 59M 59S,every_day,0 Week-ends Use the following Periodic segment file to create week-end segments starting at midnight of every Saturday in your study label,start_time,length,repeats_on,repeats_value weekend,00:00:00,1D 23H 59M 59S,wday,6 Around surveys Use the following Event segment file to create two 2-hour segments that start 1 hour before surveys answered by 3 participants label,event_timestamp,length,shift,shift_direction,device_id survey1,1587661220000,2H,1H,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 survey2,1587747620000,2H,1H,-1,a748ee1a-1d0b-4ae9-9074-279a2b6ba524 survey1,1587906020000,2H,1H,-1,rqtertsd-43ff-34fr-3eeg-efe4fergregr survey2,1584291600000,2H,1H,-1,rqtertsd-43ff-34fr-3eeg-efe4fergregr survey1,1588172420000,2H,1H,-1,klj34oi2-8frk-2343-21kk-324ljklewlr3 survey2,1584291600000,2H,1H,-1,klj34oi2-8frk-2343-21kk-324ljklewlr3","title":"Segment Examples"},{"location":"setup/configuration/#device-data-source-configuration","text":"You might need to modify the following config keys in your config.yaml depending on what devices your participants used and where you are storing your data. You can ignore [PHONE_DATA_CONFIGURATION] or [FITBIT_DATA_CONFIGURATION] if you are not working with either devices. Phone The relevant config.yaml section looks like this by default: PHONE_DATA_CONFIGURATION : SOURCE : TYPE : DATABASE DATABASE_GROUP : *database_group DEVICE_ID_COLUMN : device_id # column name TIMEZONE : TYPE : SINGLE # SINGLE (MULTIPLE support coming soon) VALUE : *timezone Parameters for [PHONE_DATA_CONFIGURATION] Key Description [SOURCE] [TYPE] Only DATABASE is supported (phone data will be pulled from a database) [SOURCE] [DATABASE_GROUP] *database_group points to the value defined before in Database credentials [SOURCE] [DEVICE_ID_COLUMN] A column that contains strings that uniquely identify smartphones. For data collected with AWARE this is usually device_id [TIMEZONE] [TYPE] Only SINGLE is supported for now [TIMEZONE] [VALUE] *timezone points to the value defined before in Timezone of your study Fitbit The relevant config.yaml section looks like this by default: FITBIT_DATA_CONFIGURATION : SOURCE : TYPE : DATABASE # DATABASE or FILES (set each [FITBIT_SENSOR][TABLE] attribute with a table name or a file path accordingly) COLUMN_FORMAT : JSON # JSON or PLAIN_TEXT DATABASE_GROUP : *database_group DEVICE_ID_COLUMN : device_id # column name TIMEZONE : TYPE : SINGLE # Fitbit devices don't support time zones so we read this data in the timezone indicated by VALUE VALUE : *timezone Parameters for For [FITBIT_DATA_CONFIGURATION] Key Description [SOURCE] [TYPE] DATABASE or FILES (set each [FITBIT_SENSOR] [TABLE] attribute accordingly with a table name or a file path) [SOURCE] [COLUMN_FORMAT] JSON or PLAIN_TEXT . Column format of the source data. If you pulled your data directly from the Fitbit API the column containing the sensor data will be in JSON format [SOURCE] [DATABASE_GROUP] *database_group points to the value defined before in Database credentials . Only used if [TYPE] is DATABASE . [SOURCE] [DEVICE_ID_COLUMN] A column that contains strings that uniquely identify Fitbit devices. [TIMEZONE] [TYPE] Only SINGLE is supported (Fitbit devices always store data in local time). [TIMEZONE] [VALUE] *timezone points to the value defined before in Timezone of your study","title":"Device Data Source Configuration"},{"location":"setup/configuration/#sensor-and-features-to-process","text":"Finally, you need to modify the config.yaml section of the sensors you want to extract behavioral features from. All sensors follow the same naming nomenclature ( DEVICE_SENSOR ) and parameter structure which we explain in the Behavioral Features Introduction . Done Head over to Execution to learn how to execute RAPIDS.","title":"Sensor and Features to Process"},{"location":"setup/execution/","text":"Execution \u00b6 After you have installed and configured RAPIDS, use the following command to execute it. ./rapids -j1 Ready to extract behavioral features If you are ready to extract features head over to the Behavioral Features Introduction Info The script ./rapids is a wrapper around Snakemake so you can pass any parameters that Snakemake accepts (e.g. -j1 ). Updating RAPIDS output after modifying config.yaml Any changes to the config.yaml file will be applied automatically and only the relevant files will be updated. This means that after modifying the features list for PHONE_MESSAGE for example, RAPIDS will update the output file with the correct features. Multi-core You can run RAPIDS over multiple cores by modifying the -j argument (e.g. use -j8 to use 8 cores). However , take into account that this means multiple sensor datasets for different participants will be load in memory at the same time. If RAPIDS crashes because it ran out of memory reduce the number of cores and try again. As reference, we have run RAPIDS over 12 cores and 32 Gb of RAM without problems for a study with 200 participants with 14 days of low-frequency smartphone data (no accelerometer, gyroscope, or magnetometer). Forcing a complete rerun If you want to update your data from your database or rerun the whole pipeline from scratch run one or both of the following commands depending on the devices you are using: ./rapids -j1 -R download_phone_data ./rapids -j1 -R download_fitbit_data Deleting RAPIDS output If you want to delete all the output files RAPIDS produces you can execute the following command: ./rapids -j1 --delete-all-output","title":"Execution"},{"location":"setup/execution/#execution","text":"After you have installed and configured RAPIDS, use the following command to execute it. ./rapids -j1 Ready to extract behavioral features If you are ready to extract features head over to the Behavioral Features Introduction Info The script ./rapids is a wrapper around Snakemake so you can pass any parameters that Snakemake accepts (e.g. -j1 ). Updating RAPIDS output after modifying config.yaml Any changes to the config.yaml file will be applied automatically and only the relevant files will be updated. This means that after modifying the features list for PHONE_MESSAGE for example, RAPIDS will update the output file with the correct features. Multi-core You can run RAPIDS over multiple cores by modifying the -j argument (e.g. use -j8 to use 8 cores). However , take into account that this means multiple sensor datasets for different participants will be load in memory at the same time. If RAPIDS crashes because it ran out of memory reduce the number of cores and try again. As reference, we have run RAPIDS over 12 cores and 32 Gb of RAM without problems for a study with 200 participants with 14 days of low-frequency smartphone data (no accelerometer, gyroscope, or magnetometer). Forcing a complete rerun If you want to update your data from your database or rerun the whole pipeline from scratch run one or both of the following commands depending on the devices you are using: ./rapids -j1 -R download_phone_data ./rapids -j1 -R download_fitbit_data Deleting RAPIDS output If you want to delete all the output files RAPIDS produces you can execute the following command: ./rapids -j1 --delete-all-output","title":"Execution"},{"location":"setup/installation/","text":"Installation \u00b6 You can install RAPIDS using Docker (the fastest), or native instructions for MacOS and Ubuntu Docker Install Docker Pull our RAPIDS container docker pull agamk/rapids:latest ` Run RAPIDS' container (after this step is done you should see a prompt in the main RAPIDS folder with its python environment active) docker run -it agamk/rapids:latest Pull the latest version of RAPIDS git pull Make RAPIDS script executable chmod +x rapids Check that RAPIDS is working ./rapids -j1 Optional . You can edit RAPIDS files with vim but we recommend using Visual Studio Code and its Remote Containers extension How to configure Remote Containers extension Make sure RAPIDS container is running Install the Remote - Containers extension Go to the Remote Explorer panel on the left hand sidebar On the top right dropdown menu choose Containers Double click on the agamk/rapids container in the CONTAINERS tree A new VS Code session should open on RAPIDS main folder insidethe container. MacOS We tested these instructions in Catalina Install brew Install MySQL brew install mysql brew services start mysql Install R 4.0, pandoc and rmarkdown. If you have other instances of R, we recommend uninstalling them brew install r brew install pandoc Rscript --vanilla -e 'install.packages(\"rmarkdown\", repos=\"http://cran.us.r-project.org\")' Install miniconda (restart your terminal afterwards) brew cask install miniconda conda init zsh # (or conda init bash) Clone our repo git clone https://github.com/carissalow/rapids Create a python virtual environment cd rapids conda env create -f environment.yml -n rapids conda activate rapids Install R packages and virtual environment: snakemake -j1 renv_install snakemake -j1 renv_restore Note This step could take several minutes to complete, especially if you have less than 3Gb of RAM or packages need to be compiled from source. Please be patient and let it run until completion. Make RAPIDS script executable chmod +x rapids Check that RAPIDS is working ./rapids -j1 Ubuntu We tested on Ubuntu 18.04 & 20.04 Install dependencies sudo apt install libcurl4-openssl-dev sudo apt install libssl-dev sudo apt install libxml2-dev Install MySQL sudo apt install libmysqlclient-dev sudo apt install mysql-server Add key for R\u2019s repository. sudo apt-key adv --keyserver keyserver.ubuntu.com --recv-keys E298A3A825C0D65DFD57CBB651716619E084DAB9 Add R\u2019s repository For 18.04 sudo add-apt-repository 'deb https://cloud.r-project.org/bin/linux/ubuntu bionic-cran40/' For 20.04 sudo add-apt-repository 'deb https://cloud.r-project.org/bin/linux/ubuntu focal-cran40/' Install R 4.0. If you have other instances of R, we recommend uninstalling them sudo apt update sudo apt install r-base Install Pandoc and rmarkdown sudo apt install pandoc Rscript --vanilla -e 'install.packages(\"rmarkdown\", repos=\"http://cran.us.r-project.org\")' Install git sudo apt install git Install miniconda Restart your current shell Clone our repo: git clone https://github.com/carissalow/rapids Create a python virtual environment: cd rapids conda env create -f environment.yml -n MY_ENV_NAME conda activate MY_ENV_NAME Install R packages and virtual environment: snakemake -j1 renv_install snakemake -j1 renv_restore Note This step could take several minutes to complete, especially if you have less than 3Gb of RAM or packages need to be compiled from source. Please be patient and let it run until completion. Make RAPIDS script executable chmod +x rapids Check that RAPIDS is working ./rapids -j1","title":"Installation"},{"location":"setup/installation/#installation","text":"You can install RAPIDS using Docker (the fastest), or native instructions for MacOS and Ubuntu Docker Install Docker Pull our RAPIDS container docker pull agamk/rapids:latest ` Run RAPIDS' container (after this step is done you should see a prompt in the main RAPIDS folder with its python environment active) docker run -it agamk/rapids:latest Pull the latest version of RAPIDS git pull Make RAPIDS script executable chmod +x rapids Check that RAPIDS is working ./rapids -j1 Optional . You can edit RAPIDS files with vim but we recommend using Visual Studio Code and its Remote Containers extension How to configure Remote Containers extension Make sure RAPIDS container is running Install the Remote - Containers extension Go to the Remote Explorer panel on the left hand sidebar On the top right dropdown menu choose Containers Double click on the agamk/rapids container in the CONTAINERS tree A new VS Code session should open on RAPIDS main folder insidethe container. MacOS We tested these instructions in Catalina Install brew Install MySQL brew install mysql brew services start mysql Install R 4.0, pandoc and rmarkdown. If you have other instances of R, we recommend uninstalling them brew install r brew install pandoc Rscript --vanilla -e 'install.packages(\"rmarkdown\", repos=\"http://cran.us.r-project.org\")' Install miniconda (restart your terminal afterwards) brew cask install miniconda conda init zsh # (or conda init bash) Clone our repo git clone https://github.com/carissalow/rapids Create a python virtual environment cd rapids conda env create -f environment.yml -n rapids conda activate rapids Install R packages and virtual environment: snakemake -j1 renv_install snakemake -j1 renv_restore Note This step could take several minutes to complete, especially if you have less than 3Gb of RAM or packages need to be compiled from source. Please be patient and let it run until completion. Make RAPIDS script executable chmod +x rapids Check that RAPIDS is working ./rapids -j1 Ubuntu We tested on Ubuntu 18.04 & 20.04 Install dependencies sudo apt install libcurl4-openssl-dev sudo apt install libssl-dev sudo apt install libxml2-dev Install MySQL sudo apt install libmysqlclient-dev sudo apt install mysql-server Add key for R\u2019s repository. sudo apt-key adv --keyserver keyserver.ubuntu.com --recv-keys E298A3A825C0D65DFD57CBB651716619E084DAB9 Add R\u2019s repository For 18.04 sudo add-apt-repository 'deb https://cloud.r-project.org/bin/linux/ubuntu bionic-cran40/' For 20.04 sudo add-apt-repository 'deb https://cloud.r-project.org/bin/linux/ubuntu focal-cran40/' Install R 4.0. If you have other instances of R, we recommend uninstalling them sudo apt update sudo apt install r-base Install Pandoc and rmarkdown sudo apt install pandoc Rscript --vanilla -e 'install.packages(\"rmarkdown\", repos=\"http://cran.us.r-project.org\")' Install git sudo apt install git Install miniconda Restart your current shell Clone our repo: git clone https://github.com/carissalow/rapids Create a python virtual environment: cd rapids conda env create -f environment.yml -n MY_ENV_NAME conda activate MY_ENV_NAME Install R packages and virtual environment: snakemake -j1 renv_install snakemake -j1 renv_restore Note This step could take several minutes to complete, especially if you have less than 3Gb of RAM or packages need to be compiled from source. Please be patient and let it run until completion. Make RAPIDS script executable chmod +x rapids Check that RAPIDS is working ./rapids -j1","title":"Installation"},{"location":"workflow-examples/minimal/","text":"Minimal Working Example \u00b6 This is a quick guide for creating and running a simple pipeline to extract missing, outgoing, and incoming call features for daily and night epochs of one participant monitored on the US East coast. Install RAPIDS and make sure your conda environment is active (see Installation ) Make the changes listed below for the corresponding Configuration step (we provide an example of what the relevant sections in your config.yml will look like after you are done) Things to change on each configuration step 1. Setup your database connection credentials in .env . We assume your credentials group is called MY_GROUP . 2. America/New_York should be the default timezone 3. Create a participant file p01.yaml based on one of your participants and add p01 to [PIDS] in config.yaml . The following would be the content of your p01.yaml participant file: PHONE : DEVICE_IDS : [ aaaaaaaa-1111-bbbb-2222-cccccccccccc ] # your participant's AWARE device id PLATFORMS : [ android ] # or ios LABEL : MyTestP01 # any string START_DATE : 2020-01-01 # this can also be empty END_DATE : 2021-01-01 # this can also be empty 4. [TIME_SEGMENTS][TYPE] should be the default PERIODIC . Change [TIME_SEGMENTS][FILE] with the path of a file containing the following lines: label,start_time,length,repeats_on,repeats_value daily,00:00:00,23H 59M 59S,every_day,0 night,00:00:00,5H 59M 59S,every_day,0 5. If you collected data with AWARE you won\u2019t need to modify the attributes of [DEVICE_DATA][PHONE] 6. Set [PHONE_CALLS][PROVIDERS][RAPIDS][COMPUTE] to True Example of the config.yaml sections after the changes outlined above PIDS: [p01] TIMEZONE: &timezone America/New_York DATABASE_GROUP: &database_group MY_GROUP # ... other irrelevant sections TIME_SEGMENTS: &time_segments TYPE: PERIODIC FILE: \"data/external/timesegments_periodic.csv\" # make sure the three lines specified above are in the file INCLUDE_PAST_PERIODIC_SEGMENTS: FALSE # No need to change this if you collected AWARE data on a database and your credentials are grouped under `MY_GROUP` in `.env` DEVICE_DATA: PHONE: SOURCE: TYPE: DATABASE DATABASE_GROUP: *database_group DEVICE_ID_COLUMN: device_id # column name TIMEZONE: TYPE: SINGLE # SINGLE or MULTIPLE VALUE: *timezone ############## PHONE ########################################################### ################################################################################ # ... other irrelevant sections # Communication call features config, TYPES and FEATURES keys need to match PHONE_CALLS: TABLE: calls # change if your calls table has a different name PROVIDERS: RAPIDS: COMPUTE: True # set this to True! CALL_TYPES: ... Run RAPIDS ./rapids -j1 The call features for daily and morning time segments will be in /data/processed/features/p01/phone_calls.csv","title":"Minimal"},{"location":"workflow-examples/minimal/#minimal-working-example","text":"This is a quick guide for creating and running a simple pipeline to extract missing, outgoing, and incoming call features for daily and night epochs of one participant monitored on the US East coast. Install RAPIDS and make sure your conda environment is active (see Installation ) Make the changes listed below for the corresponding Configuration step (we provide an example of what the relevant sections in your config.yml will look like after you are done) Things to change on each configuration step 1. Setup your database connection credentials in .env . We assume your credentials group is called MY_GROUP . 2. America/New_York should be the default timezone 3. Create a participant file p01.yaml based on one of your participants and add p01 to [PIDS] in config.yaml . The following would be the content of your p01.yaml participant file: PHONE : DEVICE_IDS : [ aaaaaaaa-1111-bbbb-2222-cccccccccccc ] # your participant's AWARE device id PLATFORMS : [ android ] # or ios LABEL : MyTestP01 # any string START_DATE : 2020-01-01 # this can also be empty END_DATE : 2021-01-01 # this can also be empty 4. [TIME_SEGMENTS][TYPE] should be the default PERIODIC . Change [TIME_SEGMENTS][FILE] with the path of a file containing the following lines: label,start_time,length,repeats_on,repeats_value daily,00:00:00,23H 59M 59S,every_day,0 night,00:00:00,5H 59M 59S,every_day,0 5. If you collected data with AWARE you won\u2019t need to modify the attributes of [DEVICE_DATA][PHONE] 6. Set [PHONE_CALLS][PROVIDERS][RAPIDS][COMPUTE] to True Example of the config.yaml sections after the changes outlined above PIDS: [p01] TIMEZONE: &timezone America/New_York DATABASE_GROUP: &database_group MY_GROUP # ... other irrelevant sections TIME_SEGMENTS: &time_segments TYPE: PERIODIC FILE: \"data/external/timesegments_periodic.csv\" # make sure the three lines specified above are in the file INCLUDE_PAST_PERIODIC_SEGMENTS: FALSE # No need to change this if you collected AWARE data on a database and your credentials are grouped under `MY_GROUP` in `.env` DEVICE_DATA: PHONE: SOURCE: TYPE: DATABASE DATABASE_GROUP: *database_group DEVICE_ID_COLUMN: device_id # column name TIMEZONE: TYPE: SINGLE # SINGLE or MULTIPLE VALUE: *timezone ############## PHONE ########################################################### ################################################################################ # ... other irrelevant sections # Communication call features config, TYPES and FEATURES keys need to match PHONE_CALLS: TABLE: calls # change if your calls table has a different name PROVIDERS: RAPIDS: COMPUTE: True # set this to True! CALL_TYPES: ... Run RAPIDS ./rapids -j1 The call features for daily and morning time segments will be in /data/processed/features/p01/phone_calls.csv","title":"Minimal Working Example"}]}
\ No newline at end of file
diff --git a/latest/sitemap.xml b/latest/sitemap.xml
index 7e02bf4e..d018a09e 100644
--- a/latest/sitemap.xml
+++ b/latest/sitemap.xml
@@ -1,147 +1,147 @@
 <?xml version="1.0" encoding="UTF-8"?>
 <urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9"><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url><url>
      <loc>None</loc>
-     <lastmod>2020-12-02</lastmod>
+     <lastmod>2020-12-03</lastmod>
      <changefreq>daily</changefreq>
     </url>
 </urlset>
\ No newline at end of file
diff --git a/latest/sitemap.xml.gz b/latest/sitemap.xml.gz
index 40efebc501f25b75b836f0ba54feec55cb08c729..4c7772715a3212064978fd3da90d2c06afb06091 100644
GIT binary patch
delta 195
zcmcc0c#V-=zMF$%srbo>?6UPy;-*vb`uwUGoHnX14B(F1yZ6GCBib`;#2!4a|Cv_t
zTS~X;m8HacsTjewiDjqCkA3Cc7w7j>Zu_DK{WsZU1GRQvc%XRfrN_%nIw~_(xLdM5
zdOnZMbcy$ZXb(xtgo){{jh`M}(=PnxS3JxAen6YWFTX2m(o#dG=S-isJ!ka_N8?>_
t_tR>{=bC&w`_g|?nZ;S2h=fK~F=XOzcj%hvU%B1<8@V%%&*f)e008$VSkV9g

delta 194
zcmcb{c$JY|zMF$XT<gR{cG>z<OWH1-z39pBps`IwH^gD}c5xryJxwQ*8SDN{&vz?1
zU+Kgzyy)w?qr8a{lFMuJ?_PLx`e(%Z0(I>|{$LZ%r<=NU<$|O5rmTDIwa7Ou%v;5d
z$K;h^ywY8#Qzzt38)<8YOq;Q1n)>^m1y8^5hWc!NubMvRD(`y9^XH?tg??Fa$0%Xl
s!8<AI-+U4JqW-glMVBWcp^;S#nfTisTD$Gf?P>Yv^TR6U@-r|10C(nA`2YX_